Compare commits

..

106 Commits

Author SHA1 Message Date
David Calavera
28220acecc Bump to version 1.8.2.
Signed-off-by: David Calavera <david.calavera@gmail.com>
2015-09-03 11:26:10 -07:00
Tianon Gravi
10f250c72b Swap "ubuntu-debootstrap" for just "ubuntu"
See https://github.com/docker-library/official-images/pull/982#issuecomment-133207587.

Signed-off-by: Andrew "Tianon" Page <admwiggin@gmail.com>
(cherry picked from commit a5eec723f8)
2015-09-03 11:26:07 -07:00
Jessica Frazelle
3e8da36017 use apt-ftparchive and reprepro to enable apt-pinning;
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit 12a71c8954)
2015-09-03 05:20:58 -04:00
David Calavera
1cce9a26a3 Bump libnetwork version to bc565c2d295067c1a43674a23a473ec6336d7fd4
Signed-off-by: David Calavera <david.calavera@gmail.com>
2015-09-02 17:24:37 -04:00
Tonis Tiigi
d7f8b4d43e Fix goroutine leak on pull
Close the pipeWriter even if there was no error.

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
2015-09-01 09:31:43 -04:00
Derek McGowan
6f7bbc3171 Fix sanitize URL bug on layer upload
Update the distribution version to include sanitize URL fix

Fixes #15875

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
2015-08-31 03:37:50 -04:00
Harald Albers
947087fb24 Update bash completion for docker run
Also fixed sort order of options using `sort -d`

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit 644c158837)
2015-08-28 05:03:01 -04:00
Harald Albers
ffe7e48ed6 Add options for the json-file logging driver to bash completion
Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit 5c1ad6f90c)
2015-08-28 05:02:55 -04:00
Harald Albers
eeecd1cf59 Add storage options to bash completion
Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit e4d8a8e1ca)

Conflicts:
	contrib/completion/bash/docker
2015-08-28 05:02:45 -04:00
Harald Albers
3f411db15b Add missing storage drivers to bash completion
Signed-off-by: Harald Albers <github@albersweb.de>
2015-08-28 04:59:30 -04:00
Harald Albers
789197f33d Remove -h flag from completion and daemon reference
All docker subcommands support `-h` as an alias for `--help`
unless they have `-h` aliased to something else like `docker run`,
which uses `-h` for `--hostname`.

`-h` is not included in the help messages of the commands, though.

It ist visible in
* reference: only in `docker daemon` reference,
  see output of `grep -Rse --help=false docs`
* man pages: only in `docker` man page
  see output of `grep -RF '**-h**' man`

For consistency reasons, this commit removes `-h` as an alias for
`--help` from the reference page, man page and the bash completion.

Signed-off-by: Harald Albers <github@albersweb.de>
2015-08-28 04:59:09 -04:00
Harald Albers
0c71d09921 Add docker ps --format to bash completion
Signed-off-by: Harald Albers <github@albersweb.de>
2015-08-28 04:57:54 -04:00
Tonis Tiigi
cc8320cb58 Fix pull on client disconnect
Fixes #15589

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
2015-08-28 04:26:22 -04:00
Arnaud Porterie
c22b292719 Update vendoring
Signed-off-by: Arnaud Porterie <arnaud.porterie@docker.com>
2015-08-26 04:45:56 -04:00
Tonis Tiigi
14d2083f14 Remove nil contexts
Causes daemon panic because loggers can’t be found.

Fixes #15724

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
(cherry picked from commit 0c08913d52)
2015-08-25 12:25:38 -04:00
Tonis Tiigi
ea56c5e1ce Update volumes userguide
Fixes #15644

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
2015-08-25 12:25:37 -04:00
Harald Albers
341ff018a2 Fix bash completion for log driver options
This option was incorrectly ported to the new `daemon` subcommand
structure.

Beside the obvious effect that completion of `docker daemon --log-opt`
did not work, this also caused completion of `docker` and `docker xxx`
to fail on macs with

> bash: words: bad array subscript

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit 18381faee6)

Conflicts:
	contrib/completion/bash/docker
2015-08-25 12:25:37 -04:00
Shijiang Wei
e07819293a a quick fix to #15626
Signed-off-by: Shijiang Wei <mountkin@gmail.com>
2015-08-25 12:25:37 -04:00
Marius Sturm
6ec8d40ae7 Initialize LogConfig in daemon mode
Signed-off-by: Marius Sturm <marius@graylog.com>
(cherry picked from commit e904cbec03)

Conflicts:
	docker/daemon.go
2015-08-25 12:25:37 -04:00
Jana Radhakrishnan
16d64608f3 Vendoring in vishvananda/netlink
Updating netlink package to 4b5dce31de6d42af5bb9811c6d265472199e0fec
to fix certain wierd netlink issues seen.

Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
(cherry picked from commit 7948b755c7)

Conflicts:
	hack/vendor.sh
2015-08-25 12:25:37 -04:00
David Calavera
00a27b6872 Fix ignore -q flag in docker ps when there is a default format.
Docker ps default format should not take precedence over cli flags.
This happens effectively for other flags except `-q`.
We need to let the cli to set the format as table to print the
expected output with `-q`.

Signed-off-by: David Calavera <david.calavera@gmail.com>
2015-08-25 04:32:20 -04:00
Vincent Batts
fc12b9ddce vendor: update tar-split to v0.9.6
Fixes rare edge case of handling GNU LongLink and LongName entries.
Perf improvements. /dev/null writes were taking CPU time during docker
push. Thanks @LK4D4
Various cleanup too.

Signed-off-by: Vincent Batts <vbatts@redhat.com>
2015-08-25 04:32:07 -04:00
Jessie Frazelle
b66e5ef208 Merge pull request #15544 from icecrime/bump_1.8.1
Bump 1.8.1
2015-08-12 20:19:06 -07:00
Arnaud Porterie
d12ea79c9d Update VERSION and CHANGELOG.md
Signed-off-by: Arnaud Porterie <arnaud.porterie@docker.com>
2015-08-12 18:54:41 -07:00
Arnaud Porterie
a9aaa66780 Update regression test
Signed-off-by: Arnaud Porterie <arnaud.porterie@docker.com>
2015-08-12 18:52:01 -07:00
Arnaud Porterie
e19060dcea Revert #14884
This reverts commit 810d3b2642.

Signed-off-by: Arnaud Porterie <arnaud.porterie@docker.com>
2015-08-12 18:51:57 -07:00
David Calavera
b0e0dbb33b Merge pull request #15091 from calavera/bump_v1.8.0
Bump v1.8.0
2015-08-11 13:16:04 -05:00
David Calavera
0d03096b65 Bump version to 1.8.0
Signed-off-by: David Calavera <david.calavera@gmail.com>
2015-08-11 11:35:27 -05:00
Jessica Frazelle
55e9551aaa minor cosmetic change to client output on pull
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit 8bd6322760)
2015-08-11 11:35:27 -05:00
Mary Anthony
c65afe6ba8 Updating new vm and with migrate
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit e5c160e1d6)
2015-08-11 11:35:27 -05:00
Mary Anthony
5745aaed22 Initial work
Removing references to regsitry 1.0, pointing to distribution
Updating links and title
Adding in comments

Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit d550d729e1)
2015-08-11 11:35:27 -05:00
Diogo Monica
b6f0f93c94 Docker content trust documentation
- started from Diogo's work
- updated after discussions with team
- Updating with new key names
- fixing weight
- adding in sandbox
- adding in gliffy for images
- backing out to old names for now
- Copy edit pass
- Entering comments from the content trust team
- Update name of branch and image name
- Removing the last diogo reference
- Updating with Seb's comments

Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 753bf40f15)
2015-08-11 11:35:26 -05:00
Youcef YEKHLEF
33b16fef43 Add fedora 22 install docs warning about systemd-network
With systemd in version 219 IP forwarding should be forced in the interface configuration file.

Signed-off-by: yyekhlef <yyekhlef@gmail.com>
(cherry picked from commit 35aebdc803)
2015-08-10 12:11:42 -05:00
Youcef YEKHLEF
9705c349c5 Add fedora 22 install docs warning about systemd-network
Following @cles commit #6bd15a68c for Arch

Signed-off-by: yyekhlef <yyekhlef@gmail.com>
(cherry picked from commit f224a9f3d6)
2015-08-10 12:11:42 -05:00
Ankush Agarwal
3de34af5d1 Remove note from docker version 1.3 on Env variables
Fixes #14734

Signed-off-by: Ankush Agarwal <ankushagarwal11@gmail.com>
(cherry picked from commit 89ec5399af)
2015-08-10 12:11:42 -05:00
Ankush Agarwal
b3f3abfc94 Make official install instructions use curl instead of wget
YOU GO CURL

Fixes #14650

Signed-off-by: Ankush Agarwal <ankushagarwal11@gmail.com>
(cherry picked from commit c76fef1791)
2015-08-10 12:10:42 -05:00
Mary Anthony
783baec49c Fixes #15349 too many apis
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 1d5f479e01)
2015-08-10 12:10:42 -05:00
Vincent Bernat
c1d9e7c6fb doc: add a notice about memory accounting overhead
The documentation for Debian and Ubuntu explains how to enable memory
and swap accounting but doesn't explain why it is disabled in the first
place. The problem with those subsystems is that they incurs a
performance hit even when not used at all. Add this explanation.

The provided figure are quite vague. The memory overhead is easily
verifiable. It is for example cited in [RedHat documentation][]. For the
performance hit, maybe the performance is better now, but a few years
ago, it was said to be [between 6 and 15%][].

The goal is that people don't just enable memory accounting if they
don't have a use for it.

[RedHat documentation]: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Resource_Management_Guide/sec-memory.html
[between 6 and 15%]: https://lwn.net/Articles/517562/

Signed-off-by: Vincent Bernat <vincent@bernat.im>
(cherry picked from commit 56b33e9f27)
2015-08-10 12:10:42 -05:00
Derek McGowan
83f6dbe30a Skip notary tests which update system clock
Currently some notary tests change the system clock to check for expiration.
Skip these tests until the code can be refactored to not rely on updating the system clock.

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
(cherry picked from commit bf3c1e6a3a)
2015-08-07 10:08:37 -07:00
Jessica Frazelle
a97b89b585 remove docker-unconfined profile we were not using it and it breaks apparmor on wheezy
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit e542238f2a)
2015-08-06 20:08:25 -07:00
Tibor Vass
29ea36a880 registry: Do not push to mirrors
This patch splits LookupEndpoints into LookupPullEndpoints and
LookupPushEndpoints so that mirrors added with --registry-mirror are
skipped in the list returned by LookupPushEndpoints.

Fixes https://github.com/docker/distribution/issues/823

Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit b899977ee2)
2015-08-06 19:55:06 -07:00
Alessandro Boch
ed672d1609 Vendoring libnetwork bd3eecc96f3c05a4acef1bedcf74397bc6850d22
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit e35a5ae463)
2015-08-06 19:55:05 -07:00
Jessica Frazelle
5916664220 revert apparmor changes back to how it was in 1.7.1, but keep tests
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit ed248207d7)
2015-08-06 19:55:05 -07:00
Lei
da4b336233 Remove redundant ip_forward check
Signed-off-by: Lei Jitang <leijitang@huawei.com>
(cherry picked from commit 6a0050d0f0)
2015-08-06 09:55:02 -07:00
Josh Hawn
74df05ccaa [graph] Use a pipe for downloads to write progress
The process of pulling an image spawns a new goroutine for each layer in the
image manifest. If any of these downloads fail we would stop everything and
return the error, even though other goroutines would still be running and
writing output through a progress reader which is attached to an http response
writer. Since the request handler had already returned from the first error,
the http server panics when one of these download goroutines makes a write to
the response writer buffer.

This patch prevents this crash in the daemon http server by waiting for all of
the download goroutines to complete, even if one of them fails. Only then does
it return, terminating the request handler.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)

(cherry picked from commit d80c4244d3)
2015-08-06 09:14:16 -07:00
Dimitri John Ledkov
2c875215b1 systemd: set service type to notify.
Currently the service type is 'simple', the default, meaning that
docker.service is considered to be started straight after
spawning. This is incorrect as there is significant amount of time
between spawning and docker ready to accept connections on the passed
sockets. Docker does implement systemd socket activate and
notification protocol, and send the ready signal to systemd, once it
is ready. However for systemd to take those notifications into
account, the service file type should be set to notify.

Signed-off-by: Dimitri John Ledkov <dimitri.j.ledkov@intel.com>
(cherry picked from commit d3e5179c29)
2015-08-06 08:39:11 -07:00
Mary Anthony
be40a48c12 - Remove references to sudo in basics.md; see sudo instructions top of file
- Removing references to Boot2Docker replacing with Docker Machine
- Removing sudo warnings in instances where appropriate (no sudo in file)
- Updating with comments

Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit cc375a1e48)
2015-08-06 08:39:10 -07:00
Filipe Oliveira
85f7f7cfc7 Adding support to forked distributions in installer script.
Signed-off-by: Filipe Oliveira <contato@fmoliveira.com.br>
(cherry picked from commit f618de1543)
2015-08-06 08:35:41 -07:00
David Calavera
e15f6fca3f Fail fail when the ps format template is invalid.
Fixes error continuing execution when the parsing fails.

Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 3d3db0d4af)
2015-08-05 16:21:30 -07:00
Vincent Demeester
d3bbaa70cd Update some contributions documentations
- Add a golint entry to coding-style.md

Signed-off-by: Vincent Demeester <vincent@sbr.pm>
(cherry picked from commit b304920021)
2015-08-05 10:54:39 -07:00
Harald Albers
cc6f6cb2e2 Add --config to bash completion
The custom configuration will also be used in docker invocations made
by the completion script itself, just like `-H`.

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit b898111d3a)
2015-08-05 09:52:08 -07:00
Jessica Frazelle
c967dd289f update systemd article to reference dropin file
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit 35e7a7c3e2)
2015-08-04 14:11:36 -07:00
Sally O'Malley
7895ec25ea make man/docker.1.md consistent with docker --help
"Options:" listed when you run "docker --help" and "docker daemon
--help" do not match the options listed in "man/docker.1.md".  This PR
makes 'docker --help', 'docker daemon --help' and 'man docker' consistent.
Also 2 typo fixes.

Signed-off-by: Sally O'Malley <somalley@redhat.com>
(cherry picked from commit f3bea61c80)
2015-08-04 09:35:25 -07:00
Charles Chan
5b06c94701 Fix #15212: Add "Labels" key to output of /containers/json
Applied retroactively from API v1.18 - v1.21.

Signed-off-by: Charles Chan <charleswhchan@users.noreply.github.com>
(cherry picked from commit b245bcd458)
2015-08-04 09:35:25 -07:00
Eric Windisch
5851e2da60 Remove container AA profile from packaging
Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit 0f4e5f7149)
2015-08-03 17:49:58 -07:00
Eric Windisch
9eff33735a Fix the proc integration test & include missing AA profile
Integration tests were failing due to proc filter behavior
changes with new apparmor policies.

Also include the missing docker-unconfined policy resolving
potential startup errors. This policy is complain-only so
it should behave identically to the standard unconfined policy,
but will not apply system path-based policies within containers.

Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit 5832715052)
2015-08-03 17:49:58 -07:00
Alessandro Boch
fc7697b050 Fix preallocated bridge networks
- Because of a bug, all the statically preallocated
  bridge networks have /24 as network mask.

Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit dab0447ae0)
2015-08-03 17:21:31 -07:00
David Calavera
1bf8954d0d Remove key file when migration fails.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 07c45e499d)
2015-08-03 16:22:34 -07:00
Michael Crosby
dfd9f5989a Add LXC built in support deprecation notice
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
(cherry picked from commit 06f6c0c7e5)
2015-08-03 16:22:34 -07:00
Josh Hawn
d9581e861d [graph] Enforce manifest/layer digest verification
We noticed a regression since the 1.7.1 patch after some refactoring. This
patch corrects the behavior and adds integration tests for modified manifest
and rootfs layer blobs.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)

(cherry picked from commit de52a3bcaa)
2015-08-03 16:22:33 -07:00
Stephen Rust
c383ceaf37 Docs: Add Blockbridge volume plugin
Signed-off-by: Stephen Rust <srust@blockbridge.com>
(cherry picked from commit 8b15b7958a)
2015-08-03 11:29:25 -07:00
Harald Albers
948912f692 Filter completions of docker inspect by --type
Completion now filters the images and containers by given
`--type`.

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit 69cde5a302)
2015-08-03 11:29:25 -07:00
jrabbit
d19b1b927b Actually link to the information
Signed-off-by: jrabbit <jackjrabbit@gmail.com>
(cherry picked from commit d2cd142ce1)
2015-08-03 09:44:11 -07:00
Tibor Vass
53f5905379 Add missing LICENSE files for docker/notary
Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit 084af30f39)
2015-08-03 09:43:19 -07:00
Tibor Vass
60cbf4da6c Vendor docker/distribution to 7dc8d4a26b689bd4892f2f2322dbce0b7119d686
Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit 0cce1fb37f)
2015-08-03 09:43:19 -07:00
evalle
183628388c Fix wrong path to YaST's 'Routing' menu for openSUSE Tumbleweed
Signed-off-by: evalle <shmarnev@gmail.com>
(cherry picked from commit 5f1593c7b8)
2015-08-03 09:37:35 -07:00
Charles Chan
fbd2267e7d Update docker_remote_api.md
Minor fixes:
* v1.19: GET /containers/(id)/logs - add missing '/'
* v1.18: Break up POST /containers/create and POST /containers/(id)/start into separate lines.

Signed-off-by: Charles Chan <charleswhchan@users.noreply.github.com>
(cherry picked from commit 084d464081)
2015-08-03 09:37:35 -07:00
Derek McGowan
a16ab243e5 Updated to use latest version of notary
Update UX to use aliases for root, snapshot, and target key

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
(cherry picked from commit 6ce76cd9ed)
2015-08-03 09:37:35 -07:00
Derek McGowan
b3c3c4cddc Vendor latest notary
Use updated notary to pick up updates from security review

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
(cherry picked from commit d594c6fcd8)
2015-08-03 09:37:34 -07:00
Qiang Huang
0fe5aad984 Remove unused variable
Introduced by #15209 unintentionally.

Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
(cherry picked from commit f5557f4f43)
2015-08-03 09:37:34 -07:00
Brian Goff
0f5e2fd479 Ensure reader position is at the end after tailing
After tailing a file, if the number of lines requested is > the number
of lines in the file, this would cause a json unmarshalling error to
occur when we later try to go follow the file.
So brute force set it to the end if any tailing occurred.

There is potential that there could be some missing log messages if logs
are being written very quickly, however I was not able to make this
happen even with `while true; do echo hello; done`, so this is probably
acceptable.

While testing this I also found a panic in LogWatcher.Close can be
called twice due to a race. Fix channel close to only close when there
has been no signal to the channel.

Signed-off-by: Brian Goff <cpuguy83@gmail.com>
(cherry picked from commit c57faa91e2)
2015-08-03 09:37:34 -07:00
David Calavera
2f7145b1c5 Remove read index that causes dead lock.
Let the iterator to lock the index when it needs it.

Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 5ed84009b3)
2015-07-31 14:56:17 -07:00
Harald Albers
81efe1f32e Fix completion of commands after a global option with arg
Without this fix, `docker -l info ` would not complete the commands.

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit aab82c5c22)
2015-07-31 12:20:42 -07:00
Harald Albers
5ba75ac343 Add completion of global options to docker daemon
It's a bit confusing: the "global options" are valid as "global options"
for all client commands (i.e. all but daemon).
Example: `docker --log-level info run`

For `docker daemon`, these "global options" are only valid as "command
options".
Example: `docker daemon --log-level info`

As command completion cannot tell which command the user is going to
type next, completion for the daemon command has to allow illegal
syntaxes like
`docker --log-level info daemon --log-level info`

Signed-off-by: Harald Albers <github@albersweb.de>
(cherry picked from commit e0dad9a153)
2015-07-31 12:20:42 -07:00
Alessandro Boch
290987fcb4 Add test code to cover issue #14859
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit 4964ab0821)
2015-07-31 08:59:42 -07:00
Alessandro Boch
98855c863d Vendoring libnetwork 31139cdb513aea5ad1ed08b60d4350a68b4c96db
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit 91274625ba)
2015-07-31 08:59:42 -07:00
Jessica Frazelle
b1f394a247 fix regression
Signed-off-by: Jessica Frazelle <acidburn@docker.com>
(cherry picked from commit b0af811272)
2015-07-30 17:21:22 -07:00
Avi Miller
a819a60a94 Fix for #14924. Seperates lsb_dist detection from dist_version detection
so that the latter can be distro specific.

Signed-off-by: Avi Miller <avi.miller@oracle.com>
(cherry picked from commit 5c6446f335)
2015-07-30 17:21:21 -07:00
Tibor Vass
33cdc7f2c4 registry: allow fallback on unknown errors
This patch fixes a bug where a user specifies a v1 mirror for
--registry-mirror and pull an image from the Hub.

It used to not fallback because of an unexpected error returned when
trying to JSON marshal nginx output.

We now ensure that any unexpected error falls back to the next endpoint
in the list.

Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit a21ba12f4e)
2015-07-30 17:21:21 -07:00
Jay Kamat
117860577c Fix a couple broken links
Signed-off-by: Jay Kamat <github@jgkamat.33mail.com>
(cherry picked from commit 4ca0aad855)
2015-07-30 17:21:21 -07:00
Dan Walsh
b0ac5df367 Labels on network content need to be shared if shared network namespace
If I run two containers with the same network they share the same /etc/resolv.conf.
The current code changes the labels of the /etc/resolv.conf currently to the
private label which causes it to be unusable in the first container.

This patch changes the labels to a shared label if more then one container
will use the content.

Docker-DCO-1.1-Signed-off-by: Dan Walsh dwalsh@redhat.com (github: rhatdan)

Docker-DCO-1.1-Signed-off-by: Dan Walsh <dwalsh@redhat.com> (github: rhatdan)

(cherry picked from commit 90b8cebda6)
2015-07-30 17:21:21 -07:00
Josh Hawn
c109095a58 Fix docker cp Behavior With Symlinks
[pkg/archive] Update archive/copy path handling

  - Remove unused TarOptions.Name field.
  - Add new TarOptions.RebaseNames field.
  - Update some of the logic around path dir/base splitting.
  - Update some of the logic behind archive entry name rebasing.

[api/types] Add LinkTarget field to PathStat

[daemon] Fix stat, archive, extract of symlinks

  These operations *should* resolve symlinks that are in the path but if the
  resource itself is a symlink then it *should not* be resolved. This patch
  puts this logic into a common function `resolvePath` which resolves symlinks
  of the path's dir in scope of the container rootfs but does not resolve the
  final element of the path. Now archive, extract, and stat operations will
  return symlinks if the path is indeed a symlink.

[api/client] Update cp path hanling

[docs/reference/api] Update description of stat

  Add the linkTarget field to the header of the archive endpoint.
  Remove path field.

[integration-cli] Fix/Add cp symlink test cases

  Copying a symlink should do just that: copy the symlink NOT
  copy the target of the symlink. Also, the resulting file from
  the copy should have the name of the symlink NOT the name of
  the target file.

  Copying to a symlink should copy to the symlink target and not
  modify the symlink itself.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)

(cherry picked from commit 75f6929b44)
2015-07-30 17:21:20 -07:00
Stephen Rust
d394113dfe Check for nil before using HostConfig to adjustCpuShares
Fix #14915. Add unit test for #14915.
Thanks @runcom for the test case: when the client calls 1.18 api
version w/o hostconfig it results in a nil pointer dereference.

Signed-off-by: Stephen Rust <srust@blockbridge.com>
(cherry picked from commit c358a4cd35)
2015-07-30 17:21:20 -07:00
Michael Crosby
2af7f63173 Don't mount /proc as ro
This caused a regression with LSM labeling.

Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
(cherry picked from commit bfc51cf660)
2015-07-30 17:21:20 -07:00
Eric Windisch
f156fb7be5 Only explicitly deny ptrace for container-originated procs
The 'deny ptrace' statement was supposed to only ignore
ptrace failures in the AUDIT log. However, ptrace was implicitly
allowed from unconfined processes (such as the docker daemon and
its integration tests) due to the abstractions/base include.

This rule narrows the definition such that it will only ignore
the failures originating inside of the container and will not
cause denials when the daemon or its tests ptrace inside processes.

Introduces positive and negative tests for ptrace /w apparmor.

Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit f5c388b35a)
2015-07-30 12:54:14 -07:00
Tibor Vass
559043b953 tlsconfig: better format for error message in tlsconfig
Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit c7a04fda2a)
2015-07-30 12:54:14 -07:00
Harald Albers
ba8abcb3dd Add docker daemon to bash completion
Signed-off-by: Harald Albers <github@albersweb.de>
Signed-off-by: Tibor Vass <tibor@docker.com>
(cherry picked from commit 8cc8ee9254)
2015-07-30 12:54:14 -07:00
David Calavera
ebf396c6e8 Log each release step.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 776600fabb)
2015-07-30 09:07:15 -07:00
Josh Hawn
47d52fb872 [api/client] Tag resolved digest from Dockerfile
Builds where the base images have been resolved to trusted digest
references will now be tagged with the original tag reference from
the Dockerfile on a successful build.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)

(cherry picked from commit bb2e6c72d2)
2015-07-29 16:53:10 -07:00
Vincent Batts
d167338876 graphdriver/*: expect uncompressed tar for ApplyDiff
The `ApplyDiff` function takes a tar archive stream that is
automagically decompressed later. This was causing a double
decompression, and when the layer was empty, that causes an early EOF.

Signed-off-by: Vincent Batts <vbatts@redhat.com>
(cherry picked from commit 273f50c741)
2015-07-29 16:48:42 -07:00
Vincent Batts
e6844381f0 archive, chrootarchive: split out decompression
In `ApplyLayer` and `Untar`, the stream is magically decompressed. Since
this is not able to be toggled, rather than break this ./pkg/ API, add
an `ApplyUncompressedLayer` and `UntarUncompressed` that does not
magically decompress the layer stream.

Signed-off-by: Vincent Batts <vbatts@redhat.com>
(cherry picked from commit 56bf275e32)
2015-07-29 16:48:41 -07:00
Derek McGowan
589922adf0 Fix login and search TLS configuration
Currently login and search do not load per registry certificates.
This is a regression caused by the last refactor since this was recently fixed.

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
(cherry picked from commit e863a07b89)
2015-07-29 16:45:16 -07:00
David Calavera
689c4e6075 Keep backwards compatibility in kill api.
Return an error when the container is stopped only in api versions
equal or greater than 1.20 (docker 1.8).

Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 621e3d8587)
2015-07-29 16:44:21 -07:00
David Calavera
43da1adedb Use *int64 for MemorySwappiness.
So we marshal/unmarshal its value properly when it's empty.

Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 4e25d2982b)
2015-07-29 16:44:21 -07:00
Antonio Murdaca
686fe02020 Fix install script to handle debian 8.1 apt repo string
Fix #14911

Signed-off-by: Antonio Murdaca <runcom@linux.com>
(cherry picked from commit 98f15cae89)
2015-07-29 10:46:26 -07:00
Eric Windisch
1d02be1c7a Mark engine AA policy as complain-only
The engine policy will now only complain
as a temporary measure to ensure we do not
cause breakages while users exercise this
policy.

This is NOT the policy for containers, but
for the newly-introduced policy for the
daemon itself.

Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit 6c887be769)
2015-07-29 09:54:16 -07:00
Eric Windisch
edb60b950a AA: Eliminate 'file' permission
Implements the policies for the remaining binaries
called by the Docker engine and eliminates the
giant whitelisted 'all files' permission in favor
of granular whitelisting and child-specific policies.

It should be possible now to remove the 'file' permission,
but for the sake of keeping Docker unbroken, we'll try
to gradually tighten the policy.

Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit 8b2fcddcd2)
2015-07-29 09:54:16 -07:00
Eric Windisch
e0e852ee6f Restore AppArmor profile generation
Will attempt to load profiles automatically. If loading fails
but the profiles are already loaded, execution will continue.

A hard failure will only occur if Docker cannot load
the profiles *and* they have not already been loaded via
some other means.

Also introduces documentation for AppArmor.

Signed-off-by: Eric Windisch <eric@windisch.us>
(cherry picked from commit 3edc88f76d)
2015-07-29 09:54:16 -07:00
Jessica Frazelle
b537508f8c only write distributions file if dne
Signed-off-by: Jessica Frazelle <princess@docker.com>
(cherry picked from commit 2eee192366)
2015-07-28 17:48:56 -07:00
Avi Miller
37e886eb7b Fix typo.
Signed-off-by: Avi Miller <avi.miller@oracle.com>
(cherry picked from commit 59e48b3468)
2015-07-28 09:39:55 -07:00
Avi Miller
50f65742ef Switch to using only the RPM command to determine the distro version.
Signed-off-by: Avi Miller <avi.miller@oracle.com>
(cherry picked from commit a43199f143)
2015-07-28 09:39:50 -07:00
Alexander Morozov
56d859d052 Use math.MaxInt32 instead of math.MaxUint32
I think it was original intention, because even half of a comment was about
MaxInt32.

Fix #15038

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
(cherry picked from commit eb45602d2f)
2015-07-28 09:35:56 -07:00
Josh Hawn
546a704c63 [api/client] Fix build when context dir is symlink
Symbolic links in the context directory path are now evaluated.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)

(cherry picked from commit 01d570ad30)
2015-07-28 09:35:38 -07:00
Aaron Lehmann
fa85dc0030 Update vendored distribution repo to new version
This version includes a fix that avoids checking against specific HTTP
status codes. The previous behavior violated the registry API spec.

Fixes #14975

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
(cherry picked from commit 091dbc1034)
2015-07-28 09:35:26 -07:00
Antonio Murdaca
36b6e5884d Format times in inspect command with a template as RFC3339Nano
In 1.6.2 we were decoding inspect API response into interface{}.
time.Time fields were JSON encoded as RFC3339Nano in the response
and when decoded into interface{} they were just strings so the inspect
template treated them as just strings.
From 1.7 we are decoding into types.ContainerJSON and when the template
gets executed it now gets a time.Time and it's formatted as
2015-07-22 05:02:38.091530369 +0000 UTC.
This patch brings back the old behavior by typing time.Time fields
as string so they gets formatted as they were encoded in JSON -- RCF3339Nano

Signed-off-by: Antonio Murdaca <runcom@linux.com>
(cherry picked from commit c9207bc0aa)
2015-07-28 09:35:14 -07:00
Jessica Frazelle
90991ddb9b fix deb packaging systemd files
Signed-off-by: Jessica Frazelle <princess@docker.com>
(cherry picked from commit a2ea8f2ad8)
2015-07-28 09:34:48 -07:00
2904 changed files with 117694 additions and 306266 deletions

View File

@@ -1,3 +1,2 @@
bundles
.gopath
vendor/pkg

View File

@@ -1,51 +0,0 @@
<!--
If you are reporting a new issue, make sure that we do not have any duplicates
already open. You can ensure this by searching the issue list for this
repository. If there is a duplicate, please close your issue and add a comment
to the existing issue instead.
If you suspect your issue is a bug, please edit your issue description to
include the BUG REPORT INFORMATION shown below. If you fail to provide this
information within 7 days, we cannot debug your issue and will close it. We
will, however, reopen it if you later provide the information.
For more information about reporting issues, see
https://github.com/docker/docker/blob/master/CONTRIBUTING.md#reporting-other-issues
---------------------------------------------------
BUG REPORT INFORMATION
---------------------------------------------------
Use the commands below to provide key information from your environment:
You do NOT have to include this information if this is a FEATURE REQUEST
-->
**Output of `docker version`:**
```
(paste your output here)
```
**Output of `docker info`:**
```
(paste your output here)
```
**Additional environment details (AWS, VirtualBox, physical, etc.):**
**Steps to reproduce the issue:**
1.
2.
3.
**Describe the results you received:**
**Describe the results you expected:**
**Additional information you deem important (e.g. issue happens only occasionally):**

View File

@@ -1,23 +0,0 @@
<!--
Please make sure you've read and understood our contributing guidelines;
https://github.com/docker/docker/blob/master/CONTRIBUTING.md
** Make sure all your commits include a signature generated with `git commit -s` **
For additional information on our contributing process, read our contributing
guide https://docs.docker.com/opensource/code/
If this is a bug fix, make sure your description includes "fixes #xxxx", or
"closes #xxxx"
Please provide the following information:
-->
**- What I did**
**- How I did it**
**- How to verify it**
**- A picture of a cute animal (not mandatory but encouraged)**

16
.gitignore vendored
View File

@@ -4,14 +4,24 @@
*.exe
*.exe~
*.orig
*.rej
*.test
.*.swp
.DS_Store
.bashrc
.dotcloud
.flymake*
.git/
.gopath/
.hg/
.vagrant*
Vagrantfile
a.out
autogen/
bin
build_src
bundles/
docker/docker
dockerversion/version_autogen.go
docs/AWS_S3_BUCKET
docs/GITCOMMIT
docs/GIT_BRANCH
@@ -20,8 +30,8 @@ docs/_build
docs/_static
docs/_templates
docs/changed-files
# generated by man/md2man-all.sh
# generated by man/man/md2man-all.sh
man/man1
man/man5
man/man8
pyenv
vendor/pkg/

View File

@@ -93,8 +93,7 @@ Sven Dowideit <SvenDowideit@home.org.au> <¨SvenDowideit@home.org.au¨>
Sven Dowideit <SvenDowideit@home.org.au> <SvenDowideit@users.noreply.github.com>
Sven Dowideit <SvenDowideit@home.org.au> <sven@t440s.home.gateway>
<alexl@redhat.com> <alexander.larsson@gmail.com>
Alexander Morozov <lk4d4@docker.com> <lk4d4math@gmail.com>
Alexander Morozov <lk4d4@docker.com>
Alexandr Morozov <lk4d4math@gmail.com>
<git.nivoc@neverbox.com> <kuehnle@online.de>
O.S. Tezer <ostezer@gmail.com>
<ostezer@gmail.com> <ostezer@users.noreply.github.com>
@@ -107,9 +106,7 @@ Roberto G. Hashioka <roberto.hashioka@docker.com> <roberto_hashioka@hotmail.com>
Sridhar Ratnakumar <sridharr@activestate.com>
Sridhar Ratnakumar <sridharr@activestate.com> <github@srid.name>
Liang-Chi Hsieh <viirya@gmail.com>
Aleksa Sarai <asarai@suse.de>
Aleksa Sarai <asarai@suse.de> <asarai@suse.com>
Aleksa Sarai <asarai@suse.de> <cyphar@cyphar.com>
Aleksa Sarai <cyphar@cyphar.com>
Will Weaver <monkey@buildingbananas.com>
Timothy Hobbs <timothyhobbs@seznam.cz>
Nathan LeClaire <nathan.leclaire@docker.com> <nathan.leclaire@gmail.com>
@@ -120,27 +117,24 @@ Nathan LeClaire <nathan.leclaire@docker.com> <nathanleclaire@gmail.com>
<marc@marc-abramowitz.com> <msabramo@gmail.com>
Matthew Heon <mheon@redhat.com> <mheon@mheonlaptop.redhat.com>
<bernat@luffy.cx> <vincent@bernat.im>
<bernat@luffy.cx> <Vincent.Bernat@exoscale.ch>
<p@pwaller.net> <peter@scraperwiki.com>
<andrew.weiss@outlook.com> <andrew.weiss@microsoft.com>
Francisco Carriedo <fcarriedo@gmail.com>
<julienbordellier@gmail.com> <git@julienbordellier.com>
<ahmetb@microsoft.com> <ahmetalpbalkan@gmail.com>
<lk4d4@docker.com> <lk4d4math@gmail.com>
<arnaud.porterie@docker.com> <icecrime@gmail.com>
<baloo@gandi.net> <superbaloo+registrations.github@superbaloo.net>
Brian Goff <cpuguy83@gmail.com>
<cpuguy83@gmail.com> <bgoff@cpuguy83-mbp.home>
<eric@windisch.us> <ewindisch@docker.com>
<ewindisch@docker.com> <eric@windisch.us>
<frank.rosquin+github@gmail.com> <frank.rosquin@gmail.com>
Hollie Teal <hollie@docker.com>
<hollie@docker.com> <hollie.teal@docker.com>
<hollie@docker.com> <hollietealok@users.noreply.github.com>
<huu@prismskylabs.com> <whoshuu@gmail.com>
Jessica Frazelle <jess@mesosphere.com>
Jessica Frazelle <jess@mesosphere.com> <jfrazelle@users.noreply.github.com>
Jessica Frazelle <jess@mesosphere.com> <acidburn@docker.com>
Jessica Frazelle <jess@mesosphere.com> <jess@docker.com>
Jessica Frazelle <jess@mesosphere.com> <princess@docker.com>
Jessica Frazelle <jess@docker.com> Jessie Frazelle <jfrazelle@users.noreply.github.com>
<jess@docker.com> <jfrazelle@users.noreply.github.com>
<konrad.wilhelm.kleine@gmail.com> <kwk@users.noreply.github.com>
<tintypemolly@gmail.com> <tintypemolly@Ohui-MacBook-Pro.local>
<estesp@linux.vnet.ibm.com> <estesp@gmail.com>
@@ -148,8 +142,6 @@ Jessica Frazelle <jess@mesosphere.com> <princess@docker.com>
Thomas LEVEIL <thomasleveil@gmail.com> Thomas LÉVEIL <thomasleveil@users.noreply.github.com>
<oi@truffles.me.uk> <timruffles@googlemail.com>
<Vincent.Bernat@exoscale.ch> <bernat@luffy.cx>
Antonio Murdaca <antonio.murdaca@gmail.com> <amurdaca@redhat.com>
Antonio Murdaca <antonio.murdaca@gmail.com> <runcom@redhat.com>
Antonio Murdaca <antonio.murdaca@gmail.com> <me@runcom.ninja>
Antonio Murdaca <antonio.murdaca@gmail.com> <runcom@linux.com>
Antonio Murdaca <antonio.murdaca@gmail.com> <runcom@users.noreply.github.com>
@@ -159,9 +151,8 @@ Deshi Xiao <dxiao@redhat.com> <xiaods@gmail.com>
Doug Davis <dug@us.ibm.com> <duglin@users.noreply.github.com>
Jacob Atzen <jacob@jacobatzen.dk> <jatzen@gmail.com>
Jeff Nickoloff <jeff.nickoloff@gmail.com> <jeff@allingeek.com>
John Howard (VM) <John.Howard@microsoft.com>
John Howard (VM) <John.Howard@microsoft.com> <john.howard@microsoft.com>
John Howard (VM) <John.Howard@microsoft.com> <jhoward@microsoft.com>
<jess@docker.com> <princess@docker.com>
John Howard (VM) <John.Howard@microsoft.com> John Howard <jhoward@microsoft.com>
Madhu Venugopal <madhu@socketplane.io> <madhu@docker.com>
Mary Anthony <mary.anthony@docker.com> <mary@docker.com>
Mary Anthony <mary.anthony@docker.com> moxiegirl <mary@docker.com>
@@ -178,60 +169,3 @@ bin liu <liubin0329@users.noreply.github.com> <liubin0329@gmail.com>
John Howard (VM) <John.Howard@microsoft.com> jhowardmsft <jhoward@microsoft.com>
Ankush Agarwal <ankushagarwal11@gmail.com> <ankushagarwal@users.noreply.github.com>
Tangi COLIN <tangicolin@gmail.com> tangicolin <tangicolin@gmail.com>
Allen Sun <allen.sun@daocloud.io>
Adrien Gallouët <adrien@gallouet.fr> <angt@users.noreply.github.com>
<aanm90@gmail.com> <martins@noironetworks.com>
Anuj Bahuguna <anujbahuguna.dev@gmail.com>
Anusha Ragunathan <anusha.ragunathan@docker.com> <anusha@docker.com>
Avi Miller <avi.miller@oracle.com> <avi.miller@gmail.com>
Brent Salisbury <brent.salisbury@docker.com> <brent@docker.com>
Chander G <chandergovind@gmail.com>
Chun Chen <ramichen@tencent.com> <chenchun.feed@gmail.com>
Ying Li <cyli@twistedmatrix.com>
Daehyeok Mun <daehyeok@gmail.com> <daehyeok@daehyeok-ui-MacBook-Air.local>
<dqminh@cloudflare.com> <dqminh89@gmail.com>
Daniel, Dao Quang Minh <dqminh@cloudflare.com>
Daniel Nephin <dnephin@docker.com> <dnephin@gmail.com>
Dave Tucker <dt@docker.com> <dave@dtucker.co.uk>
Doug Tangren <d.tangren@gmail.com>
Frederick F. Kautz IV <fkautz@redhat.com> <fkautz@alumni.cmu.edu>
Ben Golub <ben.golub@dotcloud.com>
Harold Cooper <hrldcpr@gmail.com>
hsinko <21551195@zju.edu.cn> <hsinko@users.noreply.github.com>
Josh Hawn <josh.hawn@docker.com> <jlhawn@berkeley.edu>
Justin Cormack <justin.cormack@docker.com>
<justin.cormack@docker.com> <justin.cormack@unikernel.com>
<justin.cormack@docker.com> <justin@specialbusservice.com>
Kamil Domański <kamil@domanski.co>
Lei Jitang <leijitang@huawei.com>
<leijitang@huawei.com> <leijitang@gmail.com>
Linus Heckemann <lheckemann@twig-world.com>
<lheckemann@twig-world.com> <anonymouse2048@gmail.com>
Lynda O'Leary <lyndaoleary29@gmail.com>
<lyndaoleary29@gmail.com> <lyndaoleary@hotmail.com>
Marianna Tessel <mtesselh@gmail.com>
Michael Huettermann <michael@huettermann.net>
Moysés Borges <moysesb@gmail.com>
<moysesb@gmail.com> <moyses.furtado@wplex.com.br>
Nigel Poulton <nigelpoulton@hotmail.com>
Qiang Huang <h.huangqiang@huawei.com>
<h.huangqiang@huawei.com> <qhuang@10.0.2.15>
Boaz Shuster <ripcurld.github@gmail.com>
Shuwei Hao <haosw@cn.ibm.com>
<haosw@cn.ibm.com> <haoshuwei24@gmail.com>
Soshi Katsuta <soshi.katsuta@gmail.com>
<soshi.katsuta@gmail.com> <katsuta_soshi@cyberagent.co.jp>
Stefan Berger <stefanb@linux.vnet.ibm.com>
<stefanb@linux.vnet.ibm.com> <stefanb@us.ibm.com>
Stephen Day <stephen.day@docker.com>
<stephen.day@docker.com> <stevvooe@users.noreply.github.com>
Toli Kuznets <toli@docker.com>
Tristan Carel <tristan@cogniteev.com>
<tristan@cogniteev.com> <tristan.carel@gmail.com>
Vincent Demeester <vincent@sbr.pm>
<vincent@sbr.pm> <vincent+github@demeester.fr>
Vishnu Kannan <vishnuk@google.com>
xlgao-zju <xlgao@zju.edu.cn> xlgao <xlgao@zju.edu.cn>
yuchangchun <yuchangchun1@huawei.com> y00277921 <yuchangchun1@huawei.com>
<zij@case.edu> <zjaffee@us.ibm.com>

461
AUTHORS

File diff suppressed because it is too large Load Diff

View File

@@ -1,556 +1,16 @@
# Changelog
Items starting with `DEPRECATE` are important deprecation notices. For more
information on the list of deprecated flags and APIs please have a look at
https://docs.docker.com/engine/deprecated/ where target removal dates can also
be found.
## 1.8.2 (2015-09-03)
## 1.11.0 (2016-04-13)
**IMPORTANT**: With Docker 1.11, a Linux docker installation is now made of 4 binaries (`docker`, [`docker-containerd`](https://github.com/docker/containerd), [`docker-containerd-shim`](https://github.com/docker/containerd) and [`docker-runc`](https://github.com/opencontainers/runc)). If you have scripts relying on docker being a single static binaries, please make sure to update them. Interaction with the daemon stay the same otherwise, the usage of the other binaries should be transparent. A Windows docker installation remains a single binary, `docker.exe`.
### Builder
- Fix a bug where Docker would not used the correct uid/gid when processing the `WORKDIR` command ([#21033](https://github.com/docker/docker/pull/21033))
- Fix a bug where copy operations with userns would not use the proper uid/gid ([#20782](https://github.com/docker/docker/pull/20782), [#21162](https://github.com/docker/docker/pull/21162))
### Client
* Usage of the `:` separator for security option has been deprecated. `=` should be used instead ([#21232](https://github.com/docker/docker/pull/21232))
+ The client user agent is now passed to the registry on `pull`, `build`, `push`, `login` and `search` operations ([#21306](https://github.com/docker/docker/pull/21306), [#21373](https://github.com/docker/docker/pull/21373))
* Allow setting the Domainname and Hostname separately through the API ([#20200](https://github.com/docker/docker/pull/20200))
* Docker info will now warn users if it can not detect the kernel version or the operating system ([#21128](https://github.com/docker/docker/pull/21128))
- Fix an issue where `docker stats --no-stream` output could be all 0s ([#20803](https://github.com/docker/docker/pull/20803))
- Fix a bug where some newly started container would not appear in a running `docker stats` command ([#20792](https://github.com/docker/docker/pull/20792))
* Post processing is no longer enabled for linux-cgo terminals ([#20587](https://github.com/docker/docker/pull/20587))
- Values to `--hostname` are now refused if they do not comply with [RFC1123](https://tools.ietf.org/html/rfc1123) ([#20566](https://github.com/docker/docker/pull/20566))
+ Docker learned how to use a SOCKS proxy ([#20366](https://github.com/docker/docker/pull/20366), [#18373](https://github.com/docker/docker/pull/18373))
+ Docker now supports external credential stores ([#20107](https://github.com/docker/docker/pull/20107))
* `docker ps` now supports displaying the list of volumes mounted inside a container ([#20017](https://github.com/docker/docker/pull/20017))
* `docker info` now also report Docker's root directory location ([#19986](https://github.com/docker/docker/pull/19986))
- Docker now prohibits login in with an empty username (spaces are trimmed) ([#19806](https://github.com/docker/docker/pull/19806))
* Docker events attributes are now sorted by key ([#19761](https://github.com/docker/docker/pull/19761))
* `docker ps` no longer show exported port for stopped containers ([#19483](https://github.com/docker/docker/pull/19483))
- Docker now cleans after itself if a save/export command fails ([#17849](https://github.com/docker/docker/pull/17849))
* Docker load learned how to display a progress bar ([#17329](https://github.com/docker/docker/pull/17329), [#120078](https://github.com/docker/docker/pull/20078))
### Distribution
- Fix a panic that occurred when pulling an images with 0 layers ([#21222](https://github.com/docker/docker/pull/21222))
- Fix a panic that could occur on error while pushing to a registry with a misconfigured token service ([#21212](https://github.com/docker/docker/pull/21212))
+ All first-level delegation roles are now signed when doing a trusted push ([#21046](https://github.com/docker/docker/pull/21046))
+ OAuth support for registries was added ([#20970](https://github.com/docker/docker/pull/20970))
* `docker login` now handles token using the implementation found in [docker/distribution](https://github.com/docker/distribution) ([#20832](https://github.com/docker/docker/pull/20832))
* `docker login` will no longer prompt for an email ([#20565](https://github.com/docker/docker/pull/20565))
* Docker will now fallback to registry V1 if no basic auth credentials are available ([#20241](https://github.com/docker/docker/pull/20241))
* Docker will now try to resume layer download where it left off after a network error/timeout ([#19840](https://github.com/docker/docker/pull/19840))
- Fix generated manifest mediaType when pushing cross-repository ([#19509](https://github.com/docker/docker/pull/19509))
- Fix docker requesting additional push credentials when pulling an image if Content Trust is enabled ([#20382](https://github.com/docker/docker/pull/20382))
### Logging
- Fix a race in the journald log driver ([#21311](https://github.com/docker/docker/pull/21311))
* Docker syslog driver now uses the RFC-5424 format when emitting logs ([#20121](https://github.com/docker/docker/pull/20121))
* Docker GELF log driver now allows to specify the compression algorithm and level via the `gelf-compression-type` and `gelf-compression-level` options ([#19831](https://github.com/docker/docker/pull/19831))
* Docker daemon learned to output uncolorized logs via the `--raw-logs` options ([#19794](https://github.com/docker/docker/pull/19794))
+ Docker, on Windows platform, now includes an ETW (Event Tracing in Windows) logging driver named `etwlogs` ([#19689](https://github.com/docker/docker/pull/19689))
* Journald log driver learned how to handle tags ([#19564](https://github.com/docker/docker/pull/19564))
+ The fluentd log driver learned the following options: `fluentd-address`, `fluentd-buffer-limit`, `fluentd-retry-wait`, `fluentd-max-retries` and `fluentd-async-connect` ([#19439](https://github.com/docker/docker/pull/19439))
+ Docker learned to send log to Google Cloud via the new `gcplogs` logging driver. ([#18766](https://github.com/docker/docker/pull/18766))
### Misc
+ When saving linked images together with `docker save` a subsequent `docker load` will correctly restore their parent/child relationship ([#21385](https://github.com/docker/docker/pull/c))
+ Support for building the Docker cli for OpenBSD was added ([#21325](https://github.com/docker/docker/pull/21325))
+ Labels can now be applied at network, volume and image creation ([#21270](https://github.com/docker/docker/pull/21270))
* The `dockremap` is now created as a system user ([#21266](https://github.com/docker/docker/pull/21266))
- Fix a few response body leaks ([#21258](https://github.com/docker/docker/pull/21258))
- Docker, when run as a service with systemd, will now properly manage its processes cgroups ([#20633](https://github.com/docker/docker/pull/20633))
* Docker info now reports the value of cgroup KernelMemory or emits a warning if it is not supported ([#20863](https://github.com/docker/docker/pull/20863))
* Docker info now also reports the cgroup driver in use ([#20388](https://github.com/docker/docker/pull/20388))
* Docker completion is now available on PowerShell ([#19894](https://github.com/docker/docker/pull/19894))
* `dockerinit` is no more ([#19490](https://github.com/docker/docker/pull/19490),[#19851](https://github.com/docker/docker/pull/19851))
+ Support for building Docker on arm64 was added ([#19013](https://github.com/docker/docker/pull/19013))
+ Experimental support for building docker.exe in a native Windows Docker installation ([#18348](https://github.com/docker/docker/pull/18348))
### Networking
- Fix panic if a node is forcibly removed from the cluster ([#21671](https://github.com/docker/docker/pull/21671))
- Fix "error creating vxlan interface" when starting a container in a Swarm cluster ([#21671](https://github.com/docker/docker/pull/21671))
* `docker network inspect` will now report all endpoints whether they have an active container or not ([#21160](https://github.com/docker/docker/pull/21160))
+ Experimental support for the MacVlan and IPVlan network drivers have been added ([#21122](https://github.com/docker/docker/pull/21122))
* Output of `docker network ls` is now sorted by network name ([#20383](https://github.com/docker/docker/pull/20383))
- Fix a bug where Docker would allow a network to be created with the reserved `default` name ([#19431](https://github.com/docker/docker/pull/19431))
* `docker network inspect` returns whether a network is internal or not ([#19357](https://github.com/docker/docker/pull/19357))
+ Control IPv6 via explicit option when creating a network (`docker network create --ipv6`). This shows up as a new `EnableIPv6` field in `docker network inspect` ([#17513](https://github.com/docker/docker/pull/17513))
* Support for AAAA Records (aka IPv6 Service Discovery) in embedded DNS Server ([#21396](https://github.com/docker/docker/pull/21396))
- Fix to not forward docker domain IPv6 queries to external servers ([#21396](https://github.com/docker/docker/pull/21396))
* Multiple A/AAAA records from embedded DNS Server for DNS Round robin ([#21019](https://github.com/docker/docker/pull/21019))
- Fix endpoint count inconsistency after an ungraceful dameon restart ([#21261](https://github.com/docker/docker/pull/21261))
- Move the ownership of exposed ports and port-mapping options from Endpoint to Sandbox ([#21019](https://github.com/docker/docker/pull/21019))
- Fixed a bug which prevents docker reload when host is configured with ipv6.disable=1 ([#21019](https://github.com/docker/docker/pull/21019))
- Added inbuilt nil IPAM driver ([#21019](https://github.com/docker/docker/pull/21019))
- Fixed bug in iptables.Exists() logic [#21019](https://github.com/docker/docker/pull/21019)
- Fixed a Veth interface leak when using overlay network ([#21019](https://github.com/docker/docker/pull/21019))
- Fixed a bug which prevents docker reload after a network delete during shutdown ([#20214](https://github.com/docker/docker/pull/20214))
- Make sure iptables chains are recreated on firewalld reload ([#20419](https://github.com/docker/docker/pull/20419))
- Allow to pass global datastore during config reload ([#20419](https://github.com/docker/docker/pull/20419))
- For anonymous containers use the alias name for IP to name mapping, ie:DNS PTR record ([#21019](https://github.com/docker/docker/pull/21019))
- Fix a panic when deleting an entry from /etc/hosts file ([#21019](https://github.com/docker/docker/pull/21019))
- Source the forwarded DNS queries from the container net namespace ([#21019](https://github.com/docker/docker/pull/21019))
- Fix to retain the network internal mode config for bridge networks on daemon reload ([#21780] (https://github.com/docker/docker/pull/21780))
- Fix to retain IPAM driver option configs on daemon reload ([#21914] (https://github.com/docker/docker/pull/21914))
### Plugins
- Fix a file descriptor leak that would occur every time plugins were enumerated ([#20686](https://github.com/docker/docker/pull/20686))
- Fix an issue where Authz plugin would corrupt the payload body when faced with a large amount of data ([#20602](https://github.com/docker/docker/pull/20602))
### Runtime
- Fix a panic that could occur when cleanup after a container started with invalid parameters ([#21716](https://github.com/docker/docker/pull/21716))
- Fix a race with event timers stopping early ([#21692](https://github.com/docker/docker/pull/21692))
- Fix race conditions in the layer store, potentially corrupting the map and crashing the process ([#21677](https://github.com/docker/docker/pull/21677))
- Un-deprecate auto-creation of host directories for mounts. This feature was marked deprecated in ([#21666](https://github.com/docker/docker/pull/21666))
Docker 1.9, but was decided to be too much of an backward-incompatible change, so it was decided to keep the feature.
+ It is now possible for containers to share the NET and IPC namespaces when `userns` is enabled ([#21383](https://github.com/docker/docker/pull/21383))
+ `docker inspect <image-id>` will now expose the rootfs layers ([#21370](https://github.com/docker/docker/pull/21370))
+ Docker Windows gained a minimal `top` implementation ([#21354](https://github.com/docker/docker/pull/21354))
* Docker learned to report the faulty exe when a container cannot be started due to its condition ([#21345](https://github.com/docker/docker/pull/21345))
* Docker with device mapper will now refuse to run if `udev sync` is not available ([#21097](https://github.com/docker/docker/pull/21097))
- Fix a bug where Docker would not validate the config file upon configuration reload ([#21089](https://github.com/docker/docker/pull/21089))
- Fix a hang that would happen on attach if initial start was to fail ([#21048](https://github.com/docker/docker/pull/21048))
- Fix an issue where registry service options in the daemon configuration file were not properly taken into account ([#21045](https://github.com/docker/docker/pull/21045))
- Fix a race between the exec and resize operations ([#21022](https://github.com/docker/docker/pull/21022))
- Fix an issue where nanoseconds were not correctly taken in account when filtering Docker events ([#21013](https://github.com/docker/docker/pull/21013))
- Fix the handling of Docker command when passed a 64 bytes id ([#21002](https://github.com/docker/docker/pull/21002))
* Docker will now return a `204` (i.e http.StatusNoContent) code when it successfully deleted a network ([#20977](https://github.com/docker/docker/pull/20977))
- Fix a bug where the daemon would wait indefinitely in case the process it was about to killed had already exited on its own ([#20967](https://github.com/docker/docker/pull/20967)
* The devmapper driver learned the `dm.min_free_space` option. If the mapped device free space reaches the passed value, new device creation will be prohibited. ([#20786](https://github.com/docker/docker/pull/20786))
+ Docker can now prevent processes in container to gain new privileges via the `--security-opt=no-new-privileges` flag ([#20727](https://github.com/docker/docker/pull/20727))
- Starting a container with the `--device` option will now correctly resolves symlinks ([#20684](https://github.com/docker/docker/pull/20684))
+ Docker now relies on [`containerd`](https://github.com/docker/containerd) and [`runc`](https://github.com/opencontainers/runc) to spawn containers. ([#20662](https://github.com/docker/docker/pull/20662))
- Fix docker configuration reloading to only alter value present in the given config file ([#20604](https://github.com/docker/docker/pull/20604))
+ Docker now allows setting a container hostname via the `--hostname` flag when `--net=host` ([#20177](https://github.com/docker/docker/pull/20177))
+ Docker now allows executing privileged container while running with `--userns-remap` if both `--privileged` and the new `--userns=host` flag are specified ([#20111](https://github.com/docker/docker/pull/20111))
- Fix Docker not cleaning up correctly old containers upon restarting after a crash ([#19679](https://github.com/docker/docker/pull/19679))
* Docker will now error out if it doesn't recognize a configuration key within the config file ([#19517](https://github.com/docker/docker/pull/19517))
- Fix container loading, on daemon startup, when they depends on a plugin running within a container ([#19500](https://github.com/docker/docker/pull/19500))
* `docker update` learned how to change a container restart policy ([#19116](https://github.com/docker/docker/pull/19116))
* `docker inspect` now also returns a new `State` field containing the container state in a human readable way (i.e. one of `created`, `restarting`, `running`, `paused`, `exited` or `dead`)([#18966](https://github.com/docker/docker/pull/18966))
+ Docker learned to limit the number of active pids (i.e. processes) within the container via the `pids-limit` flags. NOTE: This requires `CGROUP_PIDS=y` to be in the kernel configuration. ([#18697](https://github.com/docker/docker/pull/18697))
- `docker load` now has a `--quiet` option to suppress the load output ([#20078](https://github.com/docker/docker/pull/20078))
- Fix a bug in neighbor discovery for IPv6 peers ([#20842](https://github.com/docker/docker/pull/20842))
- Fix a panic during cleanup if a container was started with invalid options ([#21802](https://github.com/docker/docker/pull/21802))
- Fix a situation where a container cannot be stopped if the terminal is closed ([#21840](https://github.com/docker/docker/pull/21840))
### Security
* Object with the `pcp_pmcd_t` selinux type were given management access to `/var/lib/docker(/.*)?` ([#21370](https://github.com/docker/docker/pull/21370))
* `restart_syscall`, `copy_file_range`, `mlock2` joined the list of allowed calls in the default seccomp profile ([#21117](https://github.com/docker/docker/pull/21117), [#21262](https://github.com/docker/docker/pull/21262))
* `send`, `recv` and `x32` were added to the list of allowed syscalls and arch in the default seccomp profile ([#19432](https://github.com/docker/docker/pull/19432))
* Docker Content Trust now requests the server to perform snapshot signing ([#21046](https://github.com/docker/docker/pull/21046))
* Support for using YubiKeys for Content Trust signing has been moved out of experimental ([#21591](https://github.com/docker/docker/pull/21591))
### Volumes
* Output of `docker volume ls` is now sorted by volume name ([#20389](https://github.com/docker/docker/pull/20389))
* Local volumes can now accepts options similar to the unix `mount` tool ([#20262](https://github.com/docker/docker/pull/20262))
- Fix an issue where one letter directory name could not be used as source for volumes ([#21106](https://github.com/docker/docker/pull/21106))
+ `docker run -v` now accepts a new flag `nocopy`. This tell the runtime not to copy the container path content into the volume (which is the default behavior) ([#21223](https://github.com/docker/docker/pull/21223))
## 1.10.3 (2016-03-10)
### Runtime
- Fix Docker client exiting with an "Unrecognized input header" error [#20706](https://github.com/docker/docker/pull/20706)
- Fix Docker exiting if Exec is started with both `AttachStdin` and `Detach` [#20647](https://github.com/docker/docker/pull/20647)
### Distribution
- Fix a crash when pushing multiple images sharing the same layers to the same repository in parallel [#20831](https://github.com/docker/docker/pull/20831)
- Fix a panic when pushing images to a registry which uses a misconfigured token service [#21030](https://github.com/docker/docker/pull/21030)
### Plugin system
- Fix issue preventing volume plugins to start when SELinux is enabled [#20834](https://github.com/docker/docker/pull/20834)
- Prevent Docker from exiting if a volume plugin returns a null response for Get requests [#20682](https://github.com/docker/docker/pull/20682)
- Fix plugin system leaking file descriptors if a plugin has an error [#20680](https://github.com/docker/docker/pull/20680)
### Security
- Fix linux32 emulation to fail during docker build [#20672](https://github.com/docker/docker/pull/20672)
It was due to the `personality` syscall being blocked by the default seccomp profile.
- Fix Oracle XE 10g failing to start in a container [#20981](https://github.com/docker/docker/pull/20981)
It was due to the `ipc` syscall being blocked by the default seccomp profile.
- Fix user namespaces not working on Linux From Scratch [#20685](https://github.com/docker/docker/pull/20685)
- Fix issue preventing daemon to start if userns is enabled and the `subuid` or `subgid` files contain comments [#20725](https://github.com/docker/docker/pull/20725)
## 1.10.2 (2016-02-22)
### Runtime
- Prevent systemd from deleting containers' cgroups when its configuration is reloaded [#20518](https://github.com/docker/docker/pull/20518)
- Fix SELinux issues by disregarding `--read-only` when mounting `/dev/mqueue` [#20333](https://github.com/docker/docker/pull/20333)
- Fix chown permissions used during `docker cp` when userns is used [#20446](https://github.com/docker/docker/pull/20446)
- Fix configuration loading issue with all booleans defaulting to `true` [#20471](https://github.com/docker/docker/pull/20471)
- Fix occasional panic with `docker logs -f` [#20522](https://github.com/docker/docker/pull/20522)
### Distribution
- Keep layer reference if deletion failed to avoid a badly inconsistent state [#20513](https://github.com/docker/docker/pull/20513)
- Handle gracefully a corner case when canceling migration [#20372](https://github.com/docker/docker/pull/20372)
- Fix docker import on compressed data [#20367](https://github.com/docker/docker/pull/20367)
- Fix tar-split files corruption during migration that later cause docker push and docker save to fail [#20458](https://github.com/docker/docker/pull/20458)
### Networking
- Fix daemon crash if embedded DNS is sent garbage [#20510](https://github.com/docker/docker/pull/20510)
### Volumes
- Fix issue with multiple volume references with same name [#20381](https://github.com/docker/docker/pull/20381)
### Security
- Fix potential cache corruption and delegation conflict issues [#20523](https://github.com/docker/docker/pull/20523)
## 1.10.1 (2016-02-11)
### Runtime
* Do not stop daemon on migration hard failure [#20156](https://github.com/docker/docker/pull/20156)
- Fix various issues with migration to content-addressable images [#20058](https://github.com/docker/docker/pull/20058)
- Fix ZFS permission bug with user namespaces [#20045](https://github.com/docker/docker/pull/20045)
- Do not leak /dev/mqueue from the host to all containers, keep it container-specific [#19876](https://github.com/docker/docker/pull/19876) [#20133](https://github.com/docker/docker/pull/20133)
- Fix `docker ps --filter before=...` to not show stopped containers without providing `-a` flag [#20135](https://github.com/docker/docker/pull/20135)
### Security
- Fix issue preventing docker events to work properly with authorization plugin [#20002](https://github.com/docker/docker/pull/20002)
### Distribution
* Add additional verifications and prevent from uploading invalid data to registries [#20164](https://github.com/docker/docker/pull/20164)
- Fix regression preventing uppercase characters in image reference hostname [#20175](https://github.com/docker/docker/pull/20175)
### Networking
- Fix embedded DNS for user-defined networks in the presence of firewalld [#20060](https://github.com/docker/docker/pull/20060)
- Fix issue where removing a network during shutdown left Docker inoperable [#20181](https://github.com/docker/docker/issues/20181) [#20235](https://github.com/docker/docker/issues/20235)
- Embedded DNS is now able to return compressed results [#20181](https://github.com/docker/docker/issues/20181)
- Fix port-mapping issue with `userland-proxy=false` [#20181](https://github.com/docker/docker/issues/20181)
### Logging
- Fix bug where tcp+tls protocol would be rejected [#20109](https://github.com/docker/docker/pull/20109)
### Volumes
- Fix issue whereby older volume drivers would not receive volume options [#19983](https://github.com/docker/docker/pull/19983)
### Misc
- Remove TasksMax from Docker systemd service [#20167](https://github.com/docker/docker/pull/20167)
## 1.10.0 (2016-02-04)
**IMPORTANT**: Docker 1.10 uses a new content-addressable storage for images and layers.
A migration is performed the first time docker is run, and can take a significant amount of time depending on the number of images present.
Refer to this page on the wiki for more information: https://github.com/docker/docker/wiki/Engine-v1.10.0-content-addressability-migration
We also released a cool migration utility that enables you to perform the migration before updating to reduce downtime.
Engine 1.10 migrator can be found on Docker Hub: https://hub.docker.com/r/docker/v1.10-migrator/
### Runtime
+ New `docker update` command that allows updating resource constraints on running containers [#15078](https://github.com/docker/docker/pull/15078)
+ Add `--tmpfs` flag to `docker run` to create a tmpfs mount in a container [#13587](https://github.com/docker/docker/pull/13587)
+ Add `--format` flag to `docker images` command [#17692](https://github.com/docker/docker/pull/17692)
+ Allow to set daemon configuration in a file and hot-reload it with the `SIGHUP` signal [#18587](https://github.com/docker/docker/pull/18587)
+ Updated docker events to include more meta-data and event types [#18888](https://github.com/docker/docker/pull/18888)
This change is backward compatible in the API, but not on the CLI.
+ Add `--blkio-weight-device` flag to `docker run` [#13959](https://github.com/docker/docker/pull/13959)
+ Add `--device-read-bps` and `--device-write-bps` flags to `docker run` [#14466](https://github.com/docker/docker/pull/14466)
+ Add `--device-read-iops` and `--device-write-iops` flags to `docker run` [#15879](https://github.com/docker/docker/pull/15879)
+ Add `--oom-score-adj` flag to `docker run` [#16277](https://github.com/docker/docker/pull/16277)
+ Add `--detach-keys` flag to `attach`, `run`, `start` and `exec` commands to override the default key sequence that detaches from a container [#15666](https://github.com/docker/docker/pull/15666)
+ Add `--shm-size` flag to `run`, `create` and `build` to set the size of `/dev/shm` [#16168](https://github.com/docker/docker/pull/16168)
+ Show the number of running, stopped, and paused containers in `docker info` [#19249](https://github.com/docker/docker/pull/19249)
+ Show the `OSType` and `Architecture` in `docker info` [#17478](https://github.com/docker/docker/pull/17478)
+ Add `--cgroup-parent` flag on `daemon` to set cgroup parent for all containers [#19062](https://github.com/docker/docker/pull/19062)
+ Add `-L` flag to docker cp to follow symlinks [#16613](https://github.com/docker/docker/pull/16613)
+ New `status=dead` filter for `docker ps` [#17908](https://github.com/docker/docker/pull/17908)
* Change `docker run` exit codes to distinguish between runtime and application errors [#14012](https://github.com/docker/docker/pull/14012)
* Enhance `docker events --since` and `--until` to support nanoseconds and timezones [#17495](https://github.com/docker/docker/pull/17495)
* Add `--all`/`-a` flag to `stats` to include both running and stopped containers [#16742](https://github.com/docker/docker/pull/16742)
* Change the default cgroup-driver to `cgroupfs` [#17704](https://github.com/docker/docker/pull/17704)
* Emit a "tag" event when tagging an image with `build -t` [#17115](https://github.com/docker/docker/pull/17115)
* Best effort for linked containers' start order when starting the daemon [#18208](https://github.com/docker/docker/pull/18208)
* Add ability to add multiple tags on `build` [#15780](https://github.com/docker/docker/pull/15780)
* Permit `OPTIONS` request against any url, thus fixing issue with CORS [#19569](https://github.com/docker/docker/pull/19569)
- Fix the `--quiet` flag on `docker build` to actually be quiet [#17428](https://github.com/docker/docker/pull/17428)
- Fix `docker images --filter dangling=false` to now show all non-dangling images [#19326](https://github.com/docker/docker/pull/19326)
- Fix race condition causing autorestart turning off on restart [#17629](https://github.com/docker/docker/pull/17629)
- Recognize GPFS filesystems [#19216](https://github.com/docker/docker/pull/19216)
- Fix obscure bug preventing to start containers [#19751](https://github.com/docker/docker/pull/19751)
- Forbid `exec` during container restart [#19722](https://github.com/docker/docker/pull/19722)
- devicemapper: Increasing `--storage-opt dm.basesize` will now increase the base device size on daemon restart [#19123](https://github.com/docker/docker/pull/19123)
### Security
+ Add `--userns-remap` flag to `daemon` to support user namespaces (previously in experimental) [#19187](https://github.com/docker/docker/pull/19187)
+ Add support for custom seccomp profiles in `--security-opt` [#17989](https://github.com/docker/docker/pull/17989)
+ Add default seccomp profile [#18780](https://github.com/docker/docker/pull/18780)
+ Add `--authorization-plugin` flag to `daemon` to customize ACLs [#15365](https://github.com/docker/docker/pull/15365)
+ Docker Content Trust now supports the ability to read and write user delegations [#18887](https://github.com/docker/docker/pull/18887)
This is an optional, opt-in feature that requires the explicit use of the Notary command-line utility in order to be enabled.
Enabling delegation support in a specific repository will break the ability of Docker 1.9 and 1.8 to pull from that repository, if content trust is enabled.
* Allow SELinux to run in a container when using the BTRFS storage driver [#16452](https://github.com/docker/docker/pull/16452)
### Distribution
* Use content-addressable storage for images and layers [#17924](https://github.com/docker/docker/pull/17924)
Note that a migration is performed the first time docker is run; it can take a significant amount of time depending on the number of images and containers present.
Images no longer depend on the parent chain but contain a list of layer references.
`docker load`/`docker save` tarballs now also contain content-addressable image configurations.
For more information: https://github.com/docker/docker/wiki/Engine-v1.10.0-content-addressability-migration
* Add support for the new [manifest format ("schema2")](https://github.com/docker/distribution/blob/master/docs/spec/manifest-v2-2.md) [#18785](https://github.com/docker/docker/pull/18785)
* Lots of improvements for push and pull: performance++, retries on failed downloads, cancelling on client disconnect [#18353](https://github.com/docker/docker/pull/18353), [#18418](https://github.com/docker/docker/pull/18418), [#19109](https://github.com/docker/docker/pull/19109), [#18353](https://github.com/docker/docker/pull/18353)
* Limit v1 protocol fallbacks [#18590](https://github.com/docker/docker/pull/18590)
- Fix issue where docker could hang indefinitely waiting for a nonexistent process to pull an image [#19743](https://github.com/docker/docker/pull/19743)
### Networking
+ Use DNS-based discovery instead of `/etc/hosts` [#19198](https://github.com/docker/docker/pull/19198)
+ Support for network-scoped alias using `--net-alias` on `run` and `--alias` on `network connect` [#19242](https://github.com/docker/docker/pull/19242)
+ Add `--ip` and `--ip6` on `run` and `network connect` to support custom IP addresses for a container in a network [#19001](https://github.com/docker/docker/pull/19001)
+ Add `--ipam-opt` to `network create` for passing custom IPAM options [#17316](https://github.com/docker/docker/pull/17316)
+ Add `--internal` flag to `network create` to restrict external access to and from the network [#19276](https://github.com/docker/docker/pull/19276)
+ Add `kv.path` option to `--cluster-store-opt` [#19167](https://github.com/docker/docker/pull/19167)
+ Add `discovery.heartbeat` and `discovery.ttl` options to `--cluster-store-opt` to configure discovery TTL and heartbeat timer [#18204](https://github.com/docker/docker/pull/18204)
+ Add `--format` flag to `network inspect` [#17481](https://github.com/docker/docker/pull/17481)
+ Add `--link` to `network connect` to provide a container-local alias [#19229](https://github.com/docker/docker/pull/19229)
+ Support for Capability exchange with remote IPAM plugins [#18775](https://github.com/docker/docker/pull/18775)
+ Add `--force` to `network disconnect` to force container to be disconnected from network [#19317](https://github.com/docker/docker/pull/19317)
* Support for multi-host networking using built-in overlay driver for all engine supported kernels: 3.10+ [#18775](https://github.com/docker/docker/pull/18775)
* `--link` is now supported on `docker run` for containers in user-defined network [#19229](https://github.com/docker/docker/pull/19229)
* Enhance `docker network rm` to allow removing multiple networks [#17489](https://github.com/docker/docker/pull/17489)
* Include container names in `network inspect` [#17615](https://github.com/docker/docker/pull/17615)
* Include auto-generated subnets for user-defined networks in `network inspect` [#17316](https://github.com/docker/docker/pull/17316)
* Add `--filter` flag to `network ls` to hide predefined networks [#17782](https://github.com/docker/docker/pull/17782)
* Add support for network connect/disconnect to stopped containers [#18906](https://github.com/docker/docker/pull/18906)
* Add network ID to container inspect [#19323](https://github.com/docker/docker/pull/19323)
- Fix MTU issue where Docker would not start with two or more default routes [#18108](https://github.com/docker/docker/pull/18108)
- Fix duplicate IP address for containers [#18106](https://github.com/docker/docker/pull/18106)
- Fix issue preventing sometimes docker from creating the bridge network [#19338](https://github.com/docker/docker/pull/19338)
- Do not substitute 127.0.0.1 name server when using `--net=host` [#19573](https://github.com/docker/docker/pull/19573)
### Logging
+ New logging driver for Splunk [#16488](https://github.com/docker/docker/pull/16488)
+ Add support for syslog over TCP+TLS [#18998](https://github.com/docker/docker/pull/18998)
* Enhance `docker logs --since` and `--until` to support nanoseconds and time [#17495](https://github.com/docker/docker/pull/17495)
* Enhance AWS logs to auto-detect region [#16640](https://github.com/docker/docker/pull/16640)
### Volumes
+ Add support to set the mount propagation mode for a volume [#17034](https://github.com/docker/docker/pull/17034)
* Add `ls` and `inspect` endpoints to volume plugin API [#16534](https://github.com/docker/docker/pull/16534)
Existing plugins need to make use of these new APIs to satisfy users' expectation
For that, please use the new MIME type `application/vnd.docker.plugins.v1.2+json` [#19549](https://github.com/docker/docker/pull/19549)
- Fix data not being copied to named volumes [#19175](https://github.com/docker/docker/pull/19175)
- Fix issues preventing volume drivers from being containerized [#19500](https://github.com/docker/docker/pull/19500)
- Fix `docker volumes ls --dangling=false` to now show all non-dangling volumes [#19671](https://github.com/docker/docker/pull/19671)
- Do not remove named volumes on container removal [#19568](https://github.com/docker/docker/pull/19568)
- Allow external volume drivers to host anonymous volumes [#19190](https://github.com/docker/docker/pull/19190)
### Builder
+ Add support for `**` in `.dockerignore` to wildcard multiple levels of directories [#17090](https://github.com/docker/docker/pull/17090)
- Fix handling of UTF-8 characters in Dockerfiles [#17055](https://github.com/docker/docker/pull/17055)
- Fix permissions problem when reading from STDIN [#19283](https://github.com/docker/docker/pull/19283)
### Client
+ Add support for overriding the API version to use via an `DOCKER_API_VERSION` environment-variable [#15964](https://github.com/docker/docker/pull/15964)
- Fix a bug preventing Windows clients to log in to Docker Hub [#19891](https://github.com/docker/docker/pull/19891)
### Misc
* systemd: Set TasksMax in addition to LimitNPROC in systemd service file [#19391](https://github.com/docker/docker/pull/19391)
### Deprecations
* Remove LXC support. The LXC driver was deprecated in Docker 1.8, and has now been removed [#17700](https://github.com/docker/docker/pull/17700)
* Remove `--exec-driver` daemon flag, because it is no longer in use [#17700](https://github.com/docker/docker/pull/17700)
* Remove old deprecated single-dashed long CLI flags (such as `-rm`; use `--rm` instead) [#17724](https://github.com/docker/docker/pull/17724)
* Deprecate HostConfig at API container start [#17799](https://github.com/docker/docker/pull/17799)
* Deprecate docker packages for newly EOL'd Linux distributions: Fedora 21 and Ubuntu 15.04 (Vivid) [#18794](https://github.com/docker/docker/pull/18794), [#18809](https://github.com/docker/docker/pull/18809)
* Deprecate `-f` flag for docker tag [#18350](https://github.com/docker/docker/pull/18350)
## 1.9.1 (2015-11-21)
### Runtime
- Do not prevent daemon from booting if images could not be restored (#17695)
- Force IPC mount to unmount on daemon shutdown/init (#17539)
- Turn IPC unmount errors into warnings (#17554)
- Fix `docker stats` performance regression (#17638)
- Clarify cryptic error message upon `docker logs` if `--log-driver=none` (#17767)
- Fix seldom panics (#17639, #17634, #17703)
- Fix opq whiteouts problems for files with dot prefix (#17819)
- devicemapper: try defaulting to xfs instead of ext4 for performance reasons (#17903, #17918)
- devicemapper: fix displayed fs in docker info (#17974)
- selinux: only relabel if user requested so with the `z` option (#17450, #17834)
- Do not make network calls when normalizing names (#18014)
### Client
- Fix `docker login` on windows (#17738)
- Fix bug with `docker inspect` output when not connected to daemon (#17715)
- Fix `docker inspect -f {{.HostConfig.Dns}} somecontainer` (#17680)
### Builder
- Fix regression with symlink behavior in ADD/COPY (#17710)
### Networking
- Allow passing a network ID as an argument for `--net` (#17558)
- Fix connect to host and prevent disconnect from host for `host` network (#17476)
- Fix `--fixed-cidr` issue when gateway ip falls in ip-range and ip-range is
not the first block in the network (#17853)
- Restore deterministic `IPv6` generation from `MAC` address on default `bridge` network (#17890)
- Allow port-mapping only for endpoints created on docker run (#17858)
- Fixed an endpoint delete issue with a possible stale sbox (#18102)
### Distribution
- Correct parent chain in v2 push when v1Compatibility files on the disk are inconsistent (#18047)
## 1.9.0 (2015-11-03)
### Runtime
+ `docker stats` now returns block IO metrics (#15005)
+ `docker stats` now details network stats per interface (#15786)
+ Add `ancestor=<image>` filter to `docker ps --filter` flag to filter
containers based on their ancestor images (#14570)
+ Add `label=<somelabel>` filter to `docker ps --filter` to filter containers
based on label (#16530)
+ Add `--kernel-memory` flag to `docker run` (#14006)
+ Add `--message` flag to `docker import` allowing to specify an optional
message (#15711)
+ Add `--privileged` flag to `docker exec` (#14113)
+ Add `--stop-signal` flag to `docker run` allowing to replace the container
process stopping signal (#15307)
+ Add a new `unless-stopped` restart policy (#15348)
+ Inspecting an image now returns tags (#13185)
+ Add container size information to `docker inspect` (#15796)
+ Add `RepoTags` and `RepoDigests` field to `/images/{name:.*}/json` (#17275)
- Remove the deprecated `/container/ps` endpoint from the API (#15972)
- Send and document correct HTTP codes for `/exec/<name>/start` (#16250)
- Share shm and mqueue between containers sharing IPC namespace (#15862)
- Event stream now shows OOM status when `--oom-kill-disable` is set (#16235)
- Ensure special network files (/etc/hosts etc.) are read-only if bind-mounted
with `ro` option (#14965)
- Improve `rmi` performance (#16890)
- Do not update /etc/hosts for the default bridge network, except for links (#17325)
- Fix conflict with duplicate container names (#17389)
- Fix an issue with incorrect template execution in `docker inspect` (#17284)
- DEPRECATE `-c` short flag variant for `--cpu-shares` in docker run (#16271)
### Client
+ Allow `docker import` to import from local files (#11907)
### Builder
+ Add a `STOPSIGNAL` Dockerfile instruction allowing to set a different
stop-signal for the container process (#15307)
+ Add an `ARG` Dockerfile instruction and a `--build-arg` flag to `docker build`
that allows to add build-time environment variables (#15182)
- Improve cache miss performance (#16890)
### Storage
- devicemapper: Implement deferred deletion capability (#16381)
## Networking
+ `docker network` exits experimental and is part of standard release (#16645)
+ New network top-level concept, with associated subcommands and API (#16645)
WARNING: the API is different from the experimental API
+ Support for multiple isolated/micro-segmented networks (#16645)
+ Built-in multihost networking using VXLAN based overlay driver (#14071)
+ Support for third-party network plugins (#13424)
+ Ability to dynamically connect containers to multiple networks (#16645)
+ Support for user-defined IP address management via pluggable IPAM drivers (#16910)
+ Add daemon flags `--cluster-store` and `--cluster-advertise` for built-in nodes discovery (#16229)
+ Add `--cluster-store-opt` for setting up TLS settings (#16644)
+ Add `--dns-opt` to the daemon (#16031)
- DEPRECATE following container `NetworkSettings` fields in API v1.21: `EndpointID`, `Gateway`,
`GlobalIPv6Address`, `GlobalIPv6PrefixLen`, `IPAddress`, `IPPrefixLen`, `IPv6Gateway` and `MacAddress`.
Those are now specific to the `bridge` network. Use `NetworkSettings.Networks` to inspect
the networking settings of a container per network.
### Volumes
+ New top-level `volume` subcommand and API (#14242)
- Move API volume driver settings to host-specific config (#15798)
- Print an error message if volume name is not unique (#16009)
- Ensure volumes created from Dockerfiles always use the local volume driver
(#15507)
- DEPRECATE auto-creating missing host paths for bind mounts (#16349)
### Logging
+ Add `awslogs` logging driver for Amazon CloudWatch (#15495)
+ Add generic `tag` log option to allow customizing container/image
information passed to driver (e.g. show container names) (#15384)
- Implement the `docker logs` endpoint for the journald driver (#13707)
- DEPRECATE driver-specific log tags (e.g. `syslog-tag`, etc.) (#15384)
### Distribution
+ `docker search` now works with partial names (#16509)
- Push optimization: avoid buffering to file (#15493)
- The daemon will display progress for images that were already being pulled
by another client (#15489)
- Only permissions required for the current action being performed are requested (#)
+ Renaming trust keys (and respective environment variables) from `offline` to
`root` and `tagging` to `repository` (#16894)
- DEPRECATE trust key environment variables
`DOCKER_CONTENT_TRUST_OFFLINE_PASSPHRASE` and
`DOCKER_CONTENT_TRUST_TAGGING_PASSPHRASE` (#16894)
### Security
+ Add SELinux profiles to the rpm package (#15832)
- Fix various issues with AppArmor profiles provided in the deb package
(#14609)
- Add AppArmor policy that prevents writing to /proc (#15571)
## 1.8.3 (2015-10-12)
### Distribution
- Fix layer IDs lead to local graph poisoning (CVE-2014-8178)
- Fix manifest validation and parsing logic errors allow pull-by-digest validation bypass (CVE-2014-8179)
+ Add `--disable-legacy-registry` to prevent a daemon from using a v1 registry
## 1.8.2 (2015-09-10)
### Distribution
### Distribution:
- Fixes rare edge case of handling GNU LongLink and LongName entries.
- Avoid buffering to tempfile when pushing to registry V2.
- Fix ^C on docker pull.
- Fix docker pull issues on client disconnection.
- Fix issue that caused the daemon to panic when loggers weren't configured properly.
- Fix goroutine leak pulling images from registry V2.
### Runtime
### Runtime:
- Fix a bug mounting cgroups for docker daemons running inside docker containers.
- Initialize log configuration properly.
@@ -559,11 +19,11 @@ by another client (#15489)
- Handle `-q` flag in `docker ps` properly when there is a default format.
### Networking
### Networking:
- Fix several corner cases with netlink.
### Contrib
### Contrib:
- Fix several issues with bash completion.
@@ -571,7 +31,7 @@ by another client (#15489)
### Distribution
* Fix a bug where pushing multiple tags would result in invalid images
- Fix a bug where pushing multiple tags would result in invalid images
## 1.8.0 (2015-08-11)
@@ -609,7 +69,7 @@ by another client (#15489)
+ Add logging driver for GELF via UDP
+ Allow to copy files from host to containers with `docker cp`
+ Promote volume drivers from experimental to master
+ Add rollover options to json-file log driver, and --log-driver-opts flag
+ Add rollover log driver, and --log-driver-opts flag
+ Add memory swappiness tuning options
* Remove cgroup read-only flag when privileged
* Make /proc, /sys, & /dev readonly for readonly containers
@@ -630,7 +90,7 @@ by another client (#15489)
### Bug fixes
- Add missing 'Names' field to /containers/json API output
- Make `docker rmi` of dangling images safe while pulling
- Make `docker rmi --dangling` safe when pulling
- Devicemapper: Change default basesize to 100G
- Go Scheduler issue with sync.Mutex and gcc
- Fix issue where Search API endpoint would panic due to empty AuthConfig
@@ -789,7 +249,7 @@ by another client (#15489)
+ Docker daemon has full IPv6 support
+ The `docker run` command can take the `--pid=host` flag to use the host PID namespace, which makes it possible for example to debug host processes using containerized debugging tools
+ The `docker run` command can take the `--read-only` flag to make the containers root filesystem mounted as readonly, which can be used in combination with volumes to force a containers processes to only write to locations that will be persisted
+ Container total memory usage can be limited for `docker run` using the `--memory-swap` flag
+ Container total memory usage can be limited for `docker run` using the `memory-swap` flag
* Major stability improvements for devicemapper storage driver
* Better integration with host system: containers will reflect changes to the host's `/etc/resolv.conf` file when restarted
* Better integration with host system: per-container iptable rules are moved to the DOCKER chain
@@ -808,7 +268,7 @@ by another client (#15489)
#### Notable Features since 1.3.0
+ Set key=value labels to the daemon (displayed in `docker info`), applied with
new `-label` daemon flag
+ Add support for `ENV` in Dockerfile of the form:
+ Add support for `ENV` in Dockerfile of the form:
`ENV name=value name2=value2...`
+ New Overlayfs Storage Driver
+ `docker info` now returns an `ID` and `Name` field
@@ -1270,7 +730,7 @@ by another client (#15489)
- Fix broken images API for version less than 1.7
- Use the right encoding for all API endpoints which return JSON
- Move remote api client to api/
- Queue calls to the API using generic socket wait
- Queue calls to the API using generic socket wait
#### Runtime
@@ -1350,7 +810,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
- Do not add hostname when networking is disabled
* Return most recent image from the cache by date
- Return all errors from docker wait
* Add Content-Type Header "application/json" to GET /version and /info responses
* Add Content-Type Header "application/json" to GET /version and /info responses
#### Other
@@ -1378,7 +838,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
#### Runtime
- Only get the image's rootfs when we need to calculate the image size
- Correctly handle unmapping UDP ports
- Correctly handle unmapping UDP ports
* Make CopyFileWithTar use a pipe instead of a buffer to save memory on docker build
- Fix login message to say pull instead of push
- Fix "docker load" help by removing "SOURCE" prompt and mentioning STDIN
@@ -1686,7 +1146,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
+ Add -rm to docker run for removing a container on exit
- Remove error messages which are not actually errors
- Fix `docker rm` with volumes
- Fix some error cases where an HTTP body might not be closed
- Fix some error cases where a HTTP body might not be closed
- Fix panic with wrong dockercfg file
- Fix the attach behavior with -i
* Record termination time in state.
@@ -2020,7 +1480,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
+ Containers can expose public UDP ports (eg, '-p 123/udp')
+ Optionally specify an exact public port (eg. '-p 80:4500')
* 'docker login' supports additional options
- Don't save a container`s hostname when committing an image.
- Dont save a container`s hostname when committing an image.
#### Registry

View File

@@ -2,7 +2,7 @@
Want to hack on Docker? Awesome! We have a contributor's guide that explains
[setting up a Docker development environment and the contribution
process](https://docs.docker.com/opensource/project/who-written-for/).
process](https://docs.docker.com/project/who-written-for/).
![Contributors guide](docs/static_files/contributors.png)
@@ -25,7 +25,7 @@ The Docker maintainers take security seriously. If you discover a security
issue, please bring it to their attention right away!
Please **DO NOT** file a public issue, instead send your report privately to
[security@docker.com](mailto:security@docker.com).
[security@docker.com](mailto:security@docker.com),
Security reports are greatly appreciated and we will publicly thank you for it.
We also like to send gifts&mdash;if you're into Docker schwag, make sure to let
@@ -41,22 +41,18 @@ and will thank you for it!
Check that [our issue database](https://github.com/docker/docker/issues)
doesn't already include that problem or suggestion before submitting an issue.
If you find a match, you can use the "subscribe" button to get notified on
updates. Do *not* leave random "+1" or "I have this too" comments, as they
only clutter the discussion, and don't help resolving it. However, if you
have ways to reproduce the issue or have additional information that may help
resolving the issue, please leave a comment.
If you find a match, add a quick "+1" or "I have this problem too." Doing this
helps prioritize the most common problems and requests.
When reporting issues, always include:
When reporting issues, please include your host OS (Ubuntu 12.04, Fedora 19,
etc). Please include:
* The output of `uname -a`.
* The output of `docker version`.
* The output of `docker info`.
* The output of `docker -D info`.
Also include the steps required to reproduce the problem if possible and
Please also include the steps required to reproduce the problem if possible and
applicable. This information will help us review and fix your issue faster.
When sending lengthy log-files, consider posting them as a gist (https://gist.github.com).
Don't forget to remove sensitive data from your logfiles before posting (you can
replace those parts with "REDACTED").
**Issue Report Template**:
@@ -112,7 +108,7 @@ anybody starts working on it.
We are always thrilled to receive pull requests. We do our best to process them
quickly. If your pull request is not accepted on the first try,
don't get discouraged! Our contributor's guide explains [the review process we
use for simple changes](https://docs.docker.com/opensource/workflow/make-a-contribution/).
use for simple changes](https://docs.docker.com/project/make-a-contribution/).
### Design and cleanup proposals
@@ -120,8 +116,8 @@ You can propose new designs for existing Docker features. You can also design
entirely new features. We really appreciate contributors who want to refactor or
otherwise cleanup our project. For information on making these types of
contributions, see [the advanced contribution
section](https://docs.docker.com/opensource/workflow/advanced-contributing/) in
the contributors guide.
section](https://docs.docker.com/project/advanced-contributing/) in the
contributors guide.
We try hard to keep Docker lean and focused. Docker can't do everything for
everybody. This means that we might decide against incorporating a new feature.
@@ -142,7 +138,7 @@ However, there might be a way to implement that feature *on top of* Docker.
IRC is a rich chat protocol but it can overwhelm new users. You can search
<a href="https://botbot.me/freenode/docker/#" target="_blank">our chat archives</a>.
</p>
Read our <a href="https://docs.docker.com/opensource/get-help/#irc-quickstart" target="_blank">IRC quickstart guide</a> for an easy way to get started.
Read our <a href="https://docs.docker.com/project/get-help/#irc-quickstart" target="_blank">IRC quickstart guide</a> for an easy way to get started.
</td>
</tr>
<tr>
@@ -152,10 +148,8 @@ However, there might be a way to implement that feature *on top of* Docker.
<a href="https://groups.google.com/forum/#!forum/docker-user" target="_blank">Docker-user</a>
is for people using Docker containers.
The <a href="https://groups.google.com/forum/#!forum/docker-dev" target="_blank">docker-dev</a>
group is for contributors and other people contributing to the Docker project.
You can join them without a google account by sending an email to
<a href="mailto:docker-dev+subscribe@googlegroups.com">docker-dev+subscribe@googlegroups.com</a>.
After receiving the join-request message, you can simply reply to that to confirm the subscribtion.
group is for contributors and other people contributing to the Docker
project.
</td>
</tr>
<tr>
@@ -169,7 +163,7 @@ However, there might be a way to implement that feature *on top of* Docker.
<tr>
<td>Stack Overflow</td>
<td>
Stack Overflow has over 17000 Docker questions listed. We regularly
Stack Overflow has over 7000K Docker questions listed. We regularly
monitor <a href="https://stackoverflow.com/search?tab=newest&q=docker" target="_blank">Docker questions</a>
and so do many other knowledgeable Docker users.
</td>
@@ -189,14 +183,14 @@ Fork the repository and make changes on your fork in a feature branch:
Submit unit tests for your changes. Go has a great test framework built in; use
it! Take a look at existing tests for inspiration. [Run the full test
suite](https://docs.docker.com/opensource/project/test-and-docs/) on your branch before
suite](https://docs.docker.com/project/test-and-docs/) on your branch before
submitting a pull request.
Update the documentation when creating or modifying features. Test your
documentation changes for clarity, concision, and correctness, as well as a
clean documentation build. See our contributors guide for [our style
guide](https://docs.docker.com/opensource/doc-style) and instructions on [building
the documentation](https://docs.docker.com/opensource/project/test-and-docs/#build-and-test-the-documentation).
guide](https://docs.docker.com/project/doc-style) and instructions on [building
the documentation](https://docs.docker.com/project/test-and-docs/#build-and-test-the-documentation).
Write clean code. Universally formatted code promotes ease of writing, reading,
and maintenance. Always run `gofmt -s -w file.go` on each changed file before
@@ -230,7 +224,7 @@ high majority of submissions should have a single commit, so if in doubt: squash
down to one.
After every commit, [make sure the test suite passes]
(https://docs.docker.com/opensource/project/test-and-docs/). Include documentation
((https://docs.docker.com/project/test-and-docs/)). Include documentation
changes in the same pull request so that a revert would remove all traces of
the feature or fix.
@@ -314,15 +308,26 @@ format right away, but please do adjust your processes for future contributions.
### How can I become a maintainer?
The procedures for adding new maintainers are explained in the
global [MAINTAINERS](https://github.com/docker/opensource/blob/master/MAINTAINERS)
file in the [https://github.com/docker/opensource/](https://github.com/docker/opensource/)
repository.
* Step 1: Learn the component inside out
* Step 2: Make yourself useful by contributing code, bug fixes, support etc.
* Step 3: Volunteer on the IRC channel (#docker at Freenode)
* Step 4: Propose yourself at a scheduled docker meeting in #docker-dev
Don't forget: being a maintainer is a time investment. Make sure you
will have time to make yourself available. You don't have to be a
maintainer to make a difference on the project!
### IRC meetings
There are two monthly meetings taking place on #docker-dev IRC to accommodate all
timezones. Anybody can propose a topic for discussion prior to the meeting.
If you feel the conversation is going off-topic, feel free to point it out.
For the exact dates and times, have a look at [the irc-minutes
repo](https://github.com/docker/irc-minutes). The minutes also contain all the
notes from previous meetings.
## Docker community guidelines
We want to keep the Docker community awesome, growing and collaborative. We need
@@ -347,12 +352,6 @@ guidelines for the community as a whole:
to an email you are potentially sending to a large number of people. Please
consider this before you update. Also remember that nobody likes spam.
* Don't send email to the maintainers: There's no need to send email to the
maintainers to ask them to investigate an issue or to take a look at a
pull request. Instead of sending an email, GitHub mentions should be
used to ping maintainers to review a pull request, a proposal or an
issue.
### Guideline violations — 3 strikes method
The point of this section is not to find opportunities to punish people, but we

View File

@@ -23,127 +23,93 @@
# the case. Therefore, you don't have to disable it anymore.
#
FROM debian:jessie
FROM ubuntu:14.04
MAINTAINER Tianon Gravi <admwiggin@gmail.com> (@tianon)
# add zfs ppa
RUN apt-key adv --keyserver hkp://p80.pool.sks-keyservers.net:80 --recv-keys E871F18B51E0147C77796AC81196BA81F6B0FC61 \
|| apt-key adv --keyserver hkp://pgp.mit.edu:80 --recv-keys E871F18B51E0147C77796AC81196BA81F6B0FC61
RUN echo deb http://ppa.launchpad.net/zfs-native/stable/ubuntu trusty main > /etc/apt/sources.list.d/zfs.list
# add llvm repo
RUN apt-key adv --keyserver hkp://p80.pool.sks-keyservers.net:80 --recv-keys 6084F3CF814B57C1CF12EFD515CF4D18AF4F7421 \
|| apt-key adv --keyserver hkp://pgp.mit.edu:80 --recv-keys 6084F3CF814B57C1CF12EFD515CF4D18AF4F7421
RUN echo deb http://llvm.org/apt/jessie/ llvm-toolchain-jessie-3.8 main > /etc/apt/sources.list.d/llvm.list
# allow replacing httpredir mirror
ARG APT_MIRROR=httpredir.debian.org
RUN sed -i s/httpredir.debian.org/$APT_MIRROR/g /etc/apt/sources.list
RUN apt-key adv --keyserver hkp://p80.pool.sks-keyservers.net:80 --recv-keys E871F18B51E0147C77796AC81196BA81F6B0FC61
RUN echo deb http://ppa.launchpad.net/zfs-native/stable/ubuntu trusty main > /etc/apt/sources.list.d/zfs.list
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
apt-utils \
aufs-tools \
automake \
bash-completion \
bsdmainutils \
btrfs-tools \
build-essential \
clang-3.8 \
createrepo \
curl \
dpkg-sig \
gcc-mingw-w64 \
git \
iptables \
jq \
libapparmor-dev \
libcap-dev \
libltdl-dev \
libsqlite3-dev \
libsystemd-journal-dev \
libtool \
mercurial \
net-tools \
pkg-config \
python-dev \
parallel \
python-mock \
python-pip \
python-websocket \
reprepro \
ruby1.9.1 \
ruby1.9.1-dev \
s3cmd=1.1.0* \
ubuntu-zfs \
xfsprogs \
libzfs-dev \
tar \
zip \
--no-install-recommends \
&& pip install awscli==1.10.15 \
&& ln -snf /usr/bin/clang-3.8 /usr/local/bin/clang \
&& ln -snf /usr/bin/clang++-3.8 /usr/local/bin/clang++
--no-install-recommends
# Get lvm2 source for compiling statically
ENV LVM2_VERSION 2.02.103
RUN mkdir -p /usr/local/lvm2 \
&& curl -fsSL "https://mirrors.kernel.org/sourceware/lvm2/LVM2.${LVM2_VERSION}.tgz" \
| tar -xzC /usr/local/lvm2 --strip-components=1
RUN git clone -b v2_02_103 https://git.fedorahosted.org/git/lvm2.git /usr/local/lvm2
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure \
--build="$(gcc -print-multiarch)" \
--enable-static_link \
&& ./configure --enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# Configure the container for OSX cross compilation
ENV OSX_SDK MacOSX10.11.sdk
ENV OSX_CROSS_COMMIT 8aa9b71a394905e6c5f4b59e2b97b87a004658a4
RUN set -x \
&& export OSXCROSS_PATH="/osxcross" \
&& git clone https://github.com/tpoechtrager/osxcross.git $OSXCROSS_PATH \
&& ( cd $OSXCROSS_PATH && git checkout -q $OSX_CROSS_COMMIT) \
&& curl -sSL https://s3.dockerproject.org/darwin/v2/${OSX_SDK}.tar.xz -o "${OSXCROSS_PATH}/tarballs/${OSX_SDK}.tar.xz" \
&& UNATTENDED=yes OSX_VERSION_MIN=10.6 ${OSXCROSS_PATH}/build.sh
ENV PATH /osxcross/target/bin:$PATH
# install seccomp: the version shipped in trusty is too old
ENV SECCOMP_VERSION 2.3.0
RUN set -x \
&& export SECCOMP_PATH="$(mktemp -d)" \
&& curl -fsSL "https://github.com/seccomp/libseccomp/releases/download/v${SECCOMP_VERSION}/libseccomp-${SECCOMP_VERSION}.tar.gz" \
| tar -xzC "$SECCOMP_PATH" --strip-components=1 \
&& ( \
cd "$SECCOMP_PATH" \
&& ./configure --prefix=/usr/local \
&& make \
&& make install \
&& ldconfig \
) \
&& rm -rf "$SECCOMP_PATH"
# Install lxc
ENV LXC_VERSION 1.1.2
RUN mkdir -p /usr/src/lxc \
&& curl -sSL https://linuxcontainers.org/downloads/lxc/lxc-${LXC_VERSION}.tar.gz | tar -v -C /usr/src/lxc/ -xz --strip-components=1
RUN cd /usr/src/lxc \
&& ./configure \
&& make \
&& make install \
&& ldconfig
# Install Go
# IMPORTANT: If the version of Go is updated, the Windows to Linux CI machines
# will need updating, to avoid errors. Ping #docker-maintainers on IRC
# with a heads-up.
ENV GO_VERSION 1.5.4
RUN curl -fsSL "https://storage.googleapis.com/golang/go${GO_VERSION}.linux-amd64.tar.gz" \
| tar -xzC /usr/local
ENV GO_VERSION 1.4.2
RUN curl -sSL https://golang.org/dl/go${GO_VERSION}.src.tar.gz | tar -v -C /usr/local -xz \
&& mkdir -p /go/bin
ENV PATH /go/bin:/usr/local/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
RUN cd /usr/local/go/src && ./make.bash --no-clean 2>&1
# Compile Go for cross compilation
ENV DOCKER_CROSSPLATFORMS \
linux/386 linux/arm \
darwin/amd64 \
darwin/amd64 darwin/386 \
freebsd/amd64 freebsd/386 freebsd/arm \
windows/amd64 windows/386
# (set an explicit GOARM of 5 for maximum compatibility)
ENV GOARM 5
RUN cd /usr/local/go/src \
&& set -x \
&& for platform in $DOCKER_CROSSPLATFORMS; do \
GOOS=${platform%/*} \
GOARCH=${platform##*/} \
./make.bash --no-clean 2>&1; \
done
# This has been commented out and kept as reference because we don't support compiling with older Go anymore.
# ENV GOFMT_VERSION 1.3.3
# RUN curl -sSL https://storage.googleapis.com/golang/go${GOFMT_VERSION}.$(go env GOOS)-$(go env GOARCH).tar.gz | tar -C /go/bin -xz --strip-components=2 go/bin/gofmt
ENV GO_TOOLS_COMMIT 823804e1ae08dbb14eb807afc7db9993bc9e3cc3
# Update this sha when we upgrade to go 1.5.0
ENV GO_TOOLS_COMMIT 069d2f3bcb68257b627205f0486d6cc69a231ff9
# Grab Go's cover tool for dead-simple code coverage testing
# Grab Go's vet tool for examining go code to find suspicious constructs
# and help prevent errors that the compiler might not catch
@@ -152,47 +118,46 @@ RUN git clone https://github.com/golang/tools.git /go/src/golang.org/x/tools \
&& go install -v golang.org/x/tools/cmd/cover \
&& go install -v golang.org/x/tools/cmd/vet
# Grab Go's lint tool
ENV GO_LINT_COMMIT 32a87160691b3c96046c0c678fe57c5bef761456
ENV GO_LINT_COMMIT f42f5c1c440621302702cb0741e9d2ca547ae80f
RUN git clone https://github.com/golang/lint.git /go/src/github.com/golang/lint \
&& (cd /go/src/github.com/golang/lint && git checkout -q $GO_LINT_COMMIT) \
&& go install -v github.com/golang/lint/golint
# Install two versions of the registry. The first is an older version that
# only supports schema1 manifests. The second is a newer version that supports
# both. This allows integration-cli tests to cover push/pull with both schema1
# and schema2 manifests.
ENV REGISTRY_COMMIT_SCHEMA1 ec87e9b6971d831f0eff752ddb54fb64693e51cd
ENV REGISTRY_COMMIT 47a064d4195a9b56133891bbb13620c3ac83a827
# TODO replace FPM with some very minimal debhelper stuff
RUN gem install --no-rdoc --no-ri fpm --version 1.3.2
# Install registry
ENV REGISTRY_COMMIT ec87e9b6971d831f0eff752ddb54fb64693e51cd
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/distribution.git "$GOPATH/src/github.com/docker/distribution" \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT_SCHEMA1") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2-schema1 github.com/docker/distribution/cmd/registry \
&& rm -rf "$GOPATH"
# Install notary server
ENV NOTARY_VERSION docker-v1.11-3
ENV NOTARY_COMMIT 8e8122eb5528f621afcd4e2854c47302f17392f7
RUN set -x \
&& export GO15VENDOREXPERIMENT=1 \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/notary.git "$GOPATH/src/github.com/docker/notary" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_VERSION") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/notary-server github.com/docker/notary/cmd/notary-server \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary github.com/docker/notary/cmd/notary \
&& rm -rf "$GOPATH"
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
ENV DOCKER_PY_COMMIT 8a87001d09852058f08a807ab6e8491d57ca1e88
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT \
&& pip install -r test-requirements.txt
&& git checkout -q $DOCKER_PY_COMMIT
# Setup s3cmd config
RUN { \
echo '[default]'; \
echo 'access_key=$AWS_ACCESS_KEY'; \
echo 'secret_key=$AWS_SECRET_KEY'; \
} > ~/.s3cfg
# Set user.email so crosbymichael's in-container merge commits go smoothly
RUN git config --global user.email 'docker-dummy@example.com'
@@ -203,7 +168,7 @@ RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor pkcs11 seccomp selinux
ENV DOCKER_BUILDTAGS apparmor selinux
# Let us use a .bashrc file
RUN ln -sfv $PWD/.bashrc ~/.bashrc
@@ -212,19 +177,18 @@ RUN ln -sfv $PWD/.bashrc ~/.bashrc
RUN ln -sv $PWD/contrib/completion/bash/docker /etc/bash_completion.d/docker
# Get useful and necessary Hub images so we can "docker load" locally instead of pulling
COPY contrib/download-frozen-image-v2.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image-v2.sh /docker-frozen-images \
buildpack-deps:jessie@sha256:25785f89240fbcdd8a74bdaf30dd5599a9523882c6dfc567f2e9ef7cf6f79db6 \
busybox:latest@sha256:e4f93f6ed15a0cdd342f5aae387886fba0ab98af0a102da6276eaf24d6e6ade0 \
debian:jessie@sha256:f968f10b4b523737e253a97eac59b0d1420b5c19b69928d35801a6373ffe330e \
hello-world:latest@sha256:8be990ef2aeb16dbcb9271ddfe2610fa6658d13f6dfb8bc72074cc1ca36966a7
COPY contrib/download-frozen-image.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image.sh /docker-frozen-images \
busybox:latest@8c2e06607696bd4afb3d03b687e361cc43cf8ec1a4a725bc96e39f05ba97dd55 \
hello-world:frozen@91c95931e552b11604fea91c2f537284149ec32fff0f700a4769cfd31d7696ae \
jess/unshare@5c9f6ea50341a2a8eb6677527f2bdedbf331ae894a41714fda770fb130f3314d
# see also "hack/make/.ensure-frozen-images" (which needs to be updated any time this list is)
# Download man page generator
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b v1.0.4 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone --depth 1 -b v1.4 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& git clone -b v1.0.3 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone -b v1.2 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& go get -v -d github.com/cpuguy83/go-md2man \
&& go build -v -o /usr/local/bin/go-md2man github.com/cpuguy83/go-md2man \
&& rm -rf "$GOPATH"
@@ -239,35 +203,12 @@ RUN set -x \
&& rm -rf "$GOPATH"
# Build/install the tool for embedding resources in Windows binaries
ENV RSRC_COMMIT ba14da1f827188454a4591717fff29999010887f
ENV RSRC_COMMIT e48dbf1b7fc464a9e85fcec450dddf80816b76e0
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/akavel/rsrc.git "$GOPATH/src/github.com/akavel/rsrc" \
&& (cd "$GOPATH/src/github.com/akavel/rsrc" && git checkout -q "$RSRC_COMMIT") \
&& go build -v -o /usr/local/bin/rsrc github.com/akavel/rsrc \
&& rm -rf "$GOPATH"
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
&& git clone https://github.com/akavel/rsrc.git /go/src/github.com/akavel/rsrc \
&& cd /go/src/github.com/akavel/rsrc \
&& git checkout -q $RSRC_COMMIT \
&& go install -v
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]

View File

@@ -1,209 +0,0 @@
# This file describes the standard way to build Docker on aarch64, using docker
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time.
# docker build -t docker -f Dockerfile.aarch64 .
#
# # Mount your source in an interactive container for quick testing:
# docker run -v `pwd`:/go/src/github.com/docker/docker --privileged -i -t docker bash
#
# # Run the test suite:
# docker run --privileged docker hack/make.sh test
#
# Note: AppArmor used to mess with privileged mode, but this is no longer
# the case. Therefore, you don't have to disable it anymore.
#
FROM aarch64/ubuntu:wily
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
aufs-tools \
automake \
bash-completion \
btrfs-tools \
build-essential \
createrepo \
curl \
dpkg-sig \
g++ \
gcc \
git \
iptables \
jq \
libapparmor-dev \
libc6-dev \
libcap-dev \
libsqlite3-dev \
libsystemd-dev \
mercurial \
net-tools \
parallel \
pkg-config \
python-dev \
python-mock \
python-pip \
python-websocket \
gccgo \
--no-install-recommends
# Install armhf loader to use armv6 binaries on armv8
RUN dpkg --add-architecture armhf \
&& apt-get update \
&& apt-get install -y libc6:armhf
# Get lvm2 source for compiling statically
ENV LVM2_VERSION 2.02.103
RUN mkdir -p /usr/local/lvm2 \
&& curl -fsSL "https://mirrors.kernel.org/sourceware/lvm2/LVM2.${LVM2_VERSION}.tgz" \
| tar -xzC /usr/local/lvm2 --strip-components=1
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# fix platform enablement in lvm2 to support aarch64 properly
RUN set -e \
&& for f in config.guess config.sub; do \
curl -fsSL -o "/usr/local/lvm2/autoconf/$f" "http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=$f;hb=HEAD"; \
done
# "arch.c:78:2: error: #error the arch code needs to know about your machine type"
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure \
--build="$(gcc -print-multiarch)" \
--enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# install seccomp: the version shipped in trusty is too old
ENV SECCOMP_VERSION 2.3.0
RUN set -x \
&& export SECCOMP_PATH="$(mktemp -d)" \
&& curl -fsSL "https://github.com/seccomp/libseccomp/releases/download/v${SECCOMP_VERSION}/libseccomp-${SECCOMP_VERSION}.tar.gz" \
| tar -xzC "$SECCOMP_PATH" --strip-components=1 \
&& ( \
cd "$SECCOMP_PATH" \
&& ./configure --prefix=/usr/local \
&& make \
&& make install \
&& ldconfig \
) \
&& rm -rf "$SECCOMP_PATH"
# Install Go
# We don't have official binary tarballs for ARM64, eigher for Go or bootstrap,
# so we use the official armv6 released binaries as a GOROOT_BOOTSTRAP, and
# build Go from source code.
ENV GO_VERSION 1.5.4
RUN mkdir /usr/src/go && curl -fsSL https://storage.googleapis.com/golang/go${GO_VERSION}.src.tar.gz | tar -v -C /usr/src/go -xz --strip-components=1 \
&& cd /usr/src/go/src \
&& GOOS=linux GOARCH=arm64 GOROOT_BOOTSTRAP="$(go env GOROOT)" ./make.bash
ENV PATH /usr/src/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
# Only install one version of the registry, because old version which support
# schema1 manifests is not working on ARM64, we should skip integration-cli
# tests for schema1 manifests on ARM64.
ENV REGISTRY_COMMIT 47a064d4195a9b56133891bbb13620c3ac83a827
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/distribution.git "$GOPATH/src/github.com/docker/distribution" \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& rm -rf "$GOPATH"
# Install notary server
ENV NOTARY_VERSION docker-v1.11-3
RUN set -x \
&& export GO15VENDOREXPERIMENT=1 \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/notary.git "$GOPATH/src/github.com/docker/notary" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_VERSION") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary-server github.com/docker/notary/cmd/notary-server \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary github.com/docker/notary/cmd/notary \
&& rm -rf "$GOPATH"
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT \
&& pip install -r test-requirements.txt
# Set user.email so crosbymichael's in-container merge commits go smoothly
RUN git config --global user.email 'docker-dummy@example.com'
# Add an unprivileged user to be used for tests which need it
RUN groupadd -r docker
RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor pkcs11 seccomp selinux
# Let us use a .bashrc file
RUN ln -sfv $PWD/.bashrc ~/.bashrc
# Register Docker's bash completion.
RUN ln -sv $PWD/contrib/completion/bash/docker /etc/bash_completion.d/docker
# Get useful and necessary Hub images so we can "docker load" locally instead of pulling
COPY contrib/download-frozen-image-v2.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image-v2.sh /docker-frozen-images \
aarch64/buildpack-deps:jessie@sha256:6aa1d6910791b7ac78265fd0798e5abd6cb3f27ae992f6f960f6c303ec9535f2 \
aarch64/busybox:latest@sha256:b23a6a37cf269dff6e46d2473b6e227afa42b037e6d23435f1d2bc40fc8c2828 \
aarch64/debian:jessie@sha256:4be74a41a7c70ebe887b634b11ffe516cf4fcd56864a54941e56bb49883c3170 \
aarch64/hello-world:latest@sha256:65a4a158587b307bb02db4de41b836addb0c35175bdc801367b1ac1ddeb9afda
# see also "hack/make/.ensure-frozen-images" (which needs to be updated any time this list is)
# Download man page generator
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b v1.0.4 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone --depth 1 -b v1.4 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& go get -v -d github.com/cpuguy83/go-md2man \
&& go build -v -o /usr/local/bin/go-md2man github.com/cpuguy83/go-md2man \
&& rm -rf "$GOPATH"
# Download toml validator
ENV TOMLV_COMMIT 9baf8a8a9f2ed20a8e54160840c492f937eeaf9a
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/BurntSushi/toml.git "$GOPATH/src/github.com/BurntSushi/toml" \
&& (cd "$GOPATH/src/github.com/BurntSushi/toml" && git checkout -q "$TOMLV_COMMIT") \
&& go build -v -o /usr/local/bin/tomlv github.com/BurntSushi/toml/cmd/tomlv \
&& rm -rf "$GOPATH"
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker

View File

@@ -1,227 +0,0 @@
# This file describes the standard way to build Docker on ARMv7, using docker
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time.
# docker build -t docker -f Dockerfile.armhf .
#
# # Mount your source in an interactive container for quick testing:
# docker run -v `pwd`:/go/src/github.com/docker/docker --privileged -i -t docker bash
#
# # Run the test suite:
# docker run --privileged docker hack/make.sh test
#
# Note: AppArmor used to mess with privileged mode, but this is no longer
# the case. Therefore, you don't have to disable it anymore.
#
FROM armhf/debian:jessie
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
aufs-tools \
automake \
bash-completion \
btrfs-tools \
build-essential \
createrepo \
curl \
dpkg-sig \
git \
iptables \
jq \
net-tools \
libapparmor-dev \
libcap-dev \
libltdl-dev \
libsqlite3-dev \
libsystemd-journal-dev \
libtool \
mercurial \
pkg-config \
python-dev \
python-mock \
python-pip \
python-websocket \
xfsprogs \
tar \
--no-install-recommends
# Get lvm2 source for compiling statically
ENV LVM2_VERSION 2.02.103
RUN mkdir -p /usr/local/lvm2 \
&& curl -fsSL "https://mirrors.kernel.org/sourceware/lvm2/LVM2.${LVM2_VERSION}.tgz" \
| tar -xzC /usr/local/lvm2 --strip-components=1
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure \
--build="$(gcc -print-multiarch)" \
--enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# Install Go
# TODO Update to 1.5.4 once available, or build from source, as these builds
# are marked "end of life", see http://dave.cheney.net/unofficial-arm-tarballs
ENV GO_VERSION 1.5.3
RUN curl -fsSL "http://dave.cheney.net/paste/go${GO_VERSION}.linux-arm.tar.gz" \
| tar -xzC /usr/local
ENV PATH /go/bin:/usr/local/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
# we're building for armhf, which is ARMv7, so let's be explicit about that
ENV GOARCH arm
ENV GOARM 7
# This has been commented out and kept as reference because we don't support compiling with older Go anymore.
# ENV GOFMT_VERSION 1.3.3
# RUN curl -sSL https://storage.googleapis.com/golang/go${GOFMT_VERSION}.$(go env GOOS)-$(go env GOARCH).tar.gz | tar -C /go/bin -xz --strip-components=2 go/bin/gofmt
ENV GO_TOOLS_COMMIT 823804e1ae08dbb14eb807afc7db9993bc9e3cc3
# Grab Go's cover tool for dead-simple code coverage testing
# Grab Go's vet tool for examining go code to find suspicious constructs
# and help prevent errors that the compiler might not catch
RUN git clone https://github.com/golang/tools.git /go/src/golang.org/x/tools \
&& (cd /go/src/golang.org/x/tools && git checkout -q $GO_TOOLS_COMMIT) \
&& go install -v golang.org/x/tools/cmd/cover \
&& go install -v golang.org/x/tools/cmd/vet
# Grab Go's lint tool
ENV GO_LINT_COMMIT 32a87160691b3c96046c0c678fe57c5bef761456
RUN git clone https://github.com/golang/lint.git /go/src/github.com/golang/lint \
&& (cd /go/src/github.com/golang/lint && git checkout -q $GO_LINT_COMMIT) \
&& go install -v github.com/golang/lint/golint
# install seccomp: the version shipped in trusty is too old
ENV SECCOMP_VERSION 2.3.0
RUN set -x \
&& export SECCOMP_PATH="$(mktemp -d)" \
&& curl -fsSL "https://github.com/seccomp/libseccomp/releases/download/v${SECCOMP_VERSION}/libseccomp-${SECCOMP_VERSION}.tar.gz" \
| tar -xzC "$SECCOMP_PATH" --strip-components=1 \
&& ( \
cd "$SECCOMP_PATH" \
&& ./configure --prefix=/usr/local \
&& make \
&& make install \
&& ldconfig \
) \
&& rm -rf "$SECCOMP_PATH"
# Install two versions of the registry. The first is an older version that
# only supports schema1 manifests. The second is a newer version that supports
# both. This allows integration-cli tests to cover push/pull with both schema1
# and schema2 manifests.
ENV REGISTRY_COMMIT_SCHEMA1 ec87e9b6971d831f0eff752ddb54fb64693e51cd
ENV REGISTRY_COMMIT cb08de17d74bef86ce6c5abe8b240e282f5750be
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/distribution.git "$GOPATH/src/github.com/docker/distribution" \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT_SCHEMA1") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2-schema1 github.com/docker/distribution/cmd/registry \
&& rm -rf "$GOPATH"
# Install notary server
ENV NOTARY_VERSION docker-v1.11-3
RUN set -x \
&& export GO15VENDOREXPERIMENT=1 \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/notary.git "$GOPATH/src/github.com/docker/notary" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_VERSION") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary-server github.com/docker/notary/cmd/notary-server \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary github.com/docker/notary/cmd/notary \
&& rm -rf "$GOPATH"
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT \
&& pip install -r test-requirements.txt
# Set user.email so crosbymichael's in-container merge commits go smoothly
RUN git config --global user.email 'docker-dummy@example.com'
# Add an unprivileged user to be used for tests which need it
RUN groupadd -r docker
RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor pkcs11 seccomp selinux
# Let us use a .bashrc file
RUN ln -sfv $PWD/.bashrc ~/.bashrc
# Register Docker's bash completion.
RUN ln -sv $PWD/contrib/completion/bash/docker /etc/bash_completion.d/docker
# Get useful and necessary Hub images so we can "docker load" locally instead of pulling
COPY contrib/download-frozen-image-v2.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image-v2.sh /docker-frozen-images \
armhf/buildpack-deps:jessie@sha256:ca6cce8e5bf5c952129889b5cc15cd6aa8d995d77e55e3749bbaadae50e476cb \
armhf/busybox:latest@sha256:d98a7343ac750ffe387e3d514f8521ba69846c216778919b01414b8617cfb3d4 \
armhf/debian:jessie@sha256:4a2187483f04a84f9830910fe3581d69b3c985cc045d9f01d8e2f3795b28107b \
armhf/hello-world:latest@sha256:161dcecea0225975b2ad5f768058212c1e0d39e8211098666ffa1ac74cfb7791
# see also "hack/make/.ensure-frozen-images" (which needs to be updated any time this list is)
# Download man page generator
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b v1.0.4 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone --depth 1 -b v1.4 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& go get -v -d github.com/cpuguy83/go-md2man \
&& go build -v -o /usr/local/bin/go-md2man github.com/cpuguy83/go-md2man \
&& rm -rf "$GOPATH"
# Download toml validator
ENV TOMLV_COMMIT 9baf8a8a9f2ed20a8e54160840c492f937eeaf9a
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/BurntSushi/toml.git "$GOPATH/src/github.com/BurntSushi/toml" \
&& (cd "$GOPATH/src/github.com/BurntSushi/toml" && git checkout -q "$TOMLV_COMMIT") \
&& go build -v -o /usr/local/bin/tomlv github.com/BurntSushi/toml/cmd/tomlv \
&& rm -rf "$GOPATH"
# Build/install the tool for embedding resources in Windows binaries
ENV RSRC_VERSION v2
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b "$RSRC_VERSION" https://github.com/akavel/rsrc.git "$GOPATH/src/github.com/akavel/rsrc" \
&& go build -v -o /usr/local/bin/rsrc github.com/akavel/rsrc \
&& rm -rf "$GOPATH"
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker

View File

@@ -1,102 +0,0 @@
# This file describes the standard way to build Docker, using docker
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time.
# docker build -t docker -f Dockerfile.gccgo .
#
FROM gcc:5.3
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
aufs-tools \
btrfs-tools \
build-essential \
curl \
git \
iptables \
jq \
net-tools \
libapparmor-dev \
libcap-dev \
libsqlite3-dev \
mercurial \
net-tools \
parallel \
python-dev \
python-mock \
python-pip \
python-websocket \
--no-install-recommends
# Get lvm2 source for compiling statically
RUN git clone -b v2_02_103 https://git.fedorahosted.org/git/lvm2.git /usr/local/lvm2
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure --enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# install seccomp: the version shipped in jessie is too old
ENV SECCOMP_VERSION v2.3.0
RUN set -x \
&& export SECCOMP_PATH=$(mktemp -d) \
&& git clone https://github.com/seccomp/libseccomp.git "$SECCOMP_PATH" \
&& ( \
cd "$SECCOMP_PATH" \
&& git checkout "$SECCOMP_VERSION" \
&& ./autogen.sh \
&& ./configure --prefix=/usr \
&& make \
&& make install \
) \
&& rm -rf "$SECCOMP_PATH"
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT
# Add an unprivileged user to be used for tests which need it
RUN groupadd -r docker
RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor seccomp selinux
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker

View File

@@ -1,227 +0,0 @@
# This file describes the standard way to build Docker on ppc64le, using docker
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time.
# docker build -t docker -f Dockerfile.ppc64le .
#
# # Mount your source in an interactive container for quick testing:
# docker run -v `pwd`:/go/src/github.com/docker/docker --privileged -i -t docker bash
#
# # Run the test suite:
# docker run --privileged docker hack/make.sh test
#
# Note: AppArmor used to mess with privileged mode, but this is no longer
# the case. Therefore, you don't have to disable it anymore.
#
FROM ppc64le/gcc:5.3
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
aufs-tools \
automake \
bash-completion \
btrfs-tools \
build-essential \
createrepo \
curl \
dpkg-sig \
git \
iptables \
jq \
net-tools \
libapparmor-dev \
libcap-dev \
libltdl-dev \
libsqlite3-dev \
libsystemd-journal-dev \
libtool \
mercurial \
pkg-config \
python-dev \
python-mock \
python-pip \
python-websocket \
xfsprogs \
tar \
--no-install-recommends
# Get lvm2 source for compiling statically
ENV LVM2_VERSION 2.02.103
RUN mkdir -p /usr/local/lvm2 \
&& curl -fsSL "https://mirrors.kernel.org/sourceware/lvm2/LVM2.${LVM2_VERSION}.tgz" \
| tar -xzC /usr/local/lvm2 --strip-components=1
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# fix platform enablement in lvm2 to support ppc64le properly
RUN set -e \
&& for f in config.guess config.sub; do \
curl -fsSL -o "/usr/local/lvm2/autoconf/$f" "http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=$f;hb=HEAD"; \
done
# "arch.c:78:2: error: #error the arch code needs to know about your machine type"
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure \
--build="$(gcc -print-multiarch)" \
--enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# TODO install Go, using gccgo as GOROOT_BOOTSTRAP (Go 1.5+ supports ppc64le properly)
# possibly a ppc64le/golang image?
## BUILD GOLANG
ENV GO_VERSION 1.5.4
ENV GO_DOWNLOAD_URL https://golang.org/dl/go${GO_VERSION}.src.tar.gz
ENV GO_DOWNLOAD_SHA256 002acabce7ddc140d0d55891f9d4fcfbdd806b9332fb8b110c91bc91afb0bc93
ENV GOROOT_BOOTSTRAP /usr/local
RUN curl -fsSL "$GO_DOWNLOAD_URL" -o golang.tar.gz \
&& echo "$GO_DOWNLOAD_SHA256 golang.tar.gz" | sha256sum -c - \
&& tar -C /usr/src -xzf golang.tar.gz \
&& rm golang.tar.gz \
&& cd /usr/src/go/src && ./make.bash 2>&1
ENV GOROOT_BOOTSTRAP /usr/src/
ENV PATH /usr/src/go/bin/:/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
# This has been commented out and kept as reference because we don't support compiling with older Go anymore.
# ENV GOFMT_VERSION 1.3.3
# RUN curl -sSL https://storage.googleapis.com/golang/go${GOFMT_VERSION}.$(go env GOOS)-$(go env GOARCH).tar.gz | tar -C /go/bin -xz --strip-components=2 go/bin/gofmt
ENV GO_TOOLS_COMMIT 823804e1ae08dbb14eb807afc7db9993bc9e3cc3
# Grab Go's cover tool for dead-simple code coverage testing
# Grab Go's vet tool for examining go code to find suspicious constructs
# and help prevent errors that the compiler might not catch
RUN git clone https://github.com/golang/tools.git /go/src/golang.org/x/tools \
&& (cd /go/src/golang.org/x/tools && git checkout -q $GO_TOOLS_COMMIT) \
&& go install -v golang.org/x/tools/cmd/cover \
&& go install -v golang.org/x/tools/cmd/vet
# Grab Go's lint tool
ENV GO_LINT_COMMIT 32a87160691b3c96046c0c678fe57c5bef761456
RUN git clone https://github.com/golang/lint.git /go/src/github.com/golang/lint \
&& (cd /go/src/github.com/golang/lint && git checkout -q $GO_LINT_COMMIT) \
&& go install -v github.com/golang/lint/golint
# Install two versions of the registry. The first is an older version that
# only supports schema1 manifests. The second is a newer version that supports
# both. This allows integration-cli tests to cover push/pull with both schema1
# and schema2 manifests.
ENV REGISTRY_COMMIT_SCHEMA1 ec87e9b6971d831f0eff752ddb54fb64693e51cd
ENV REGISTRY_COMMIT 47a064d4195a9b56133891bbb13620c3ac83a827
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/distribution.git "$GOPATH/src/github.com/docker/distribution" \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT_SCHEMA1") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2-schema1 github.com/docker/distribution/cmd/registry \
&& rm -rf "$GOPATH"
# Install notary and notary-server
ENV NOTARY_VERSION docker-v1.11-3
RUN set -x \
&& export GO15VENDOREXPERIMENT=1 \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/notary.git "$GOPATH/src/github.com/docker/notary" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_VERSION") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary-server github.com/docker/notary/cmd/notary-server \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary github.com/docker/notary/cmd/notary \
&& rm -rf "$GOPATH"
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT \
&& pip install -r test-requirements.txt
# Set user.email so crosbymichael's in-container merge commits go smoothly
RUN git config --global user.email 'docker-dummy@example.com'
# Add an unprivileged user to be used for tests which need it
RUN groupadd -r docker
RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor pkcs11 selinux
# Let us use a .bashrc file
RUN ln -sfv $PWD/.bashrc ~/.bashrc
# Register Docker's bash completion.
RUN ln -sv $PWD/contrib/completion/bash/docker /etc/bash_completion.d/docker
# Get useful and necessary Hub images so we can "docker load" locally instead of pulling
COPY contrib/download-frozen-image-v2.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image-v2.sh /docker-frozen-images \
ppc64le/buildpack-deps:jessie@sha256:902bfe4ef1389f94d143d64516dd50a2de75bca2e66d4a44b1d73f63ddf05dda \
ppc64le/busybox:latest@sha256:38bb82085248d5a3c24bd7a5dc146f2f2c191e189da0441f1c2ca560e3fc6f1b \
ppc64le/debian:jessie@sha256:412845f51b6ab662afba71bc7a716e20fdb9b84f185d180d4c7504f8a75c4f91 \
ppc64le/hello-world:latest@sha256:186a40a9a02ca26df0b6c8acdfb8ac2f3ae6678996a838f977e57fac9d963974
# see also "hack/make/.ensure-frozen-images" (which needs to be updated any time this list is)
# Download man page generator
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b v1.0.4 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone --depth 1 -b v1.4 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& go get -v -d github.com/cpuguy83/go-md2man \
&& go build -v -o /usr/local/bin/go-md2man github.com/cpuguy83/go-md2man \
&& rm -rf "$GOPATH"
# Download toml validator
ENV TOMLV_COMMIT 9baf8a8a9f2ed20a8e54160840c492f937eeaf9a
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/BurntSushi/toml.git "$GOPATH/src/github.com/BurntSushi/toml" \
&& (cd "$GOPATH/src/github.com/BurntSushi/toml" && git checkout -q "$TOMLV_COMMIT") \
&& go build -v -o /usr/local/bin/tomlv github.com/BurntSushi/toml/cmd/tomlv \
&& rm -rf "$GOPATH"
# Build/install the tool for embedding resources in Windows binaries
ENV RSRC_VERSION v2
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b "$RSRC_VERSION" https://github.com/akavel/rsrc.git "$GOPATH/src/github.com/akavel/rsrc" \
&& go build -v -o /usr/local/bin/rsrc github.com/akavel/rsrc \
&& rm -rf "$GOPATH"
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker

View File

@@ -1,206 +0,0 @@
# This file describes the standard way to build Docker on s390x, using docker
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time.
# docker build -t docker -f Dockerfile.s390x .
#
# # Mount your source in an interactive container for quick testing:
# docker run -v `pwd`:/go/src/github.com/docker/docker --privileged -i -t docker bash
#
# # Run the test suite:
# docker run --privileged docker hack/make.sh test
#
# Note: AppArmor used to mess with privileged mode, but this is no longer
# the case. Therefore, you don't have to disable it anymore.
#
FROM s390x/gcc:5.3
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
aufs-tools \
automake \
bash-completion \
btrfs-tools \
build-essential \
createrepo \
curl \
dpkg-sig \
git \
iptables \
jq \
net-tools \
libapparmor-dev \
libcap-dev \
libltdl-dev \
libsqlite3-dev \
libsystemd-journal-dev \
libtool \
mercurial \
pkg-config \
python-dev \
python-mock \
python-pip \
python-websocket \
xfsprogs \
tar \
--no-install-recommends
# Get lvm2 source for compiling statically
ENV LVM2_VERSION 2.02.103
RUN mkdir -p /usr/local/lvm2 \
&& curl -fsSL "https://mirrors.kernel.org/sourceware/lvm2/LVM2.${LVM2_VERSION}.tgz" \
| tar -xzC /usr/local/lvm2 --strip-components=1
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# fix platform enablement in lvm2 to support s390x properly
RUN set -e \
&& for f in config.guess config.sub; do \
curl -fsSL -o "/usr/local/lvm2/autoconf/$f" "http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=$f;hb=HEAD"; \
done
# "arch.c:78:2: error: #error the arch code needs to know about your machine type"
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure \
--build="$(gcc -print-multiarch)" \
--enable-static_link \
&& make device-mapper \
&& make install_device-mapper
# see https://git.fedorahosted.org/cgit/lvm2.git/tree/INSTALL
# Note: Go comes from the base image (gccgo, specifically)
# We can't compile Go proper because s390x isn't an officially supported architecture yet.
ENV PATH /go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
# This has been commented out and kept as reference because we don't support compiling with older Go anymore.
# ENV GOFMT_VERSION 1.3.3
# RUN curl -sSL https://storage.googleapis.com/golang/go${GOFMT_VERSION}.$(go env GOOS)-$(go env GOARCH).tar.gz | tar -C /go/bin -xz --strip-components=2 go/bin/gofmt
# TODO update this sha when we upgrade to Go 1.5+
ENV GO_TOOLS_COMMIT 069d2f3bcb68257b627205f0486d6cc69a231ff9
# Grab Go's cover tool for dead-simple code coverage testing
# Grab Go's vet tool for examining go code to find suspicious constructs
# and help prevent errors that the compiler might not catch
RUN git clone https://github.com/golang/tools.git /go/src/golang.org/x/tools \
&& (cd /go/src/golang.org/x/tools && git checkout -q $GO_TOOLS_COMMIT) \
&& go install -v golang.org/x/tools/cmd/cover \
&& go install -v golang.org/x/tools/cmd/vet
# Grab Go's lint tool
ENV GO_LINT_COMMIT f42f5c1c440621302702cb0741e9d2ca547ae80f
RUN git clone https://github.com/golang/lint.git /go/src/github.com/golang/lint \
&& (cd /go/src/github.com/golang/lint && git checkout -q $GO_LINT_COMMIT) \
&& go install -v github.com/golang/lint/golint
# Install registry
ENV REGISTRY_COMMIT ec87e9b6971d831f0eff752ddb54fb64693e51cd
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/distribution.git "$GOPATH/src/github.com/docker/distribution" \
&& (cd "$GOPATH/src/github.com/docker/distribution" && git checkout -q "$REGISTRY_COMMIT") \
&& GOPATH="$GOPATH/src/github.com/docker/distribution/Godeps/_workspace:$GOPATH" \
go build -o /usr/local/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& rm -rf "$GOPATH"
# Install notary server
ENV NOTARY_VERSION docker-v1.11-3
RUN set -x \
&& export GO15VENDOREXPERIMENT=1 \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/docker/notary.git "$GOPATH/src/github.com/docker/notary" \
&& (cd "$GOPATH/src/github.com/docker/notary" && git checkout -q "$NOTARY_VERSION") \
&& GOPATH="$GOPATH/src/github.com/docker/notary/vendor:$GOPATH" \
go build -o /usr/local/bin/notary-server github.com/docker/notary/cmd/notary-server \
&& rm -rf "$GOPATH"
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT e2878cbcc3a7eef99917adc1be252800b0e41ece
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT \
&& pip install -r test-requirements.txt
# Set user.email so crosbymichael's in-container merge commits go smoothly
RUN git config --global user.email 'docker-dummy@example.com'
# Add an unprivileged user to be used for tests which need it
RUN groupadd -r docker
RUN useradd --create-home --gid docker unprivilegeduser
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/docker/docker
ENV DOCKER_BUILDTAGS apparmor pkcs11 selinux
# Let us use a .bashrc file
RUN ln -sfv $PWD/.bashrc ~/.bashrc
# Register Docker's bash completion.
RUN ln -sv $PWD/contrib/completion/bash/docker /etc/bash_completion.d/docker
# Get useful and necessary Hub images so we can "docker load" locally instead of pulling
COPY contrib/download-frozen-image-v2.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image-v2.sh /docker-frozen-images \
s390x/buildpack-deps:jessie@sha256:4d1381224acaca6c4bfe3604de3af6972083a8558a99672cb6989c7541780099 \
s390x/busybox:latest@sha256:dd61522c983884a66ed72d60301925889028c6d2d5e0220a8fe1d9b4c6a4f01b \
s390x/debian:jessie@sha256:b74c863400909eff3c5e196cac9bfd1f6333ce47aae6a38398d87d5875da170a \
s390x/hello-world:latest@sha256:780d80b3a7677c3788c0d5cd9168281320c8d4a6d9183892d8ee5cdd610f5699
# see also "hack/make/.ensure-frozen-images" (which needs to be updated any time this list is)
# Download man page generator
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b v1.0.4 https://github.com/cpuguy83/go-md2man.git "$GOPATH/src/github.com/cpuguy83/go-md2man" \
&& git clone --depth 1 -b v1.4 https://github.com/russross/blackfriday.git "$GOPATH/src/github.com/russross/blackfriday" \
&& go get -v -d github.com/cpuguy83/go-md2man \
&& go build -v -o /usr/local/bin/go-md2man github.com/cpuguy83/go-md2man \
&& rm -rf "$GOPATH"
# Download toml validator
ENV TOMLV_COMMIT 9baf8a8a9f2ed20a8e54160840c492f937eeaf9a
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone https://github.com/BurntSushi/toml.git "$GOPATH/src/github.com/BurntSushi/toml" \
&& (cd "$GOPATH/src/github.com/BurntSushi/toml" && git checkout -q "$TOMLV_COMMIT") \
&& go build -v -o /usr/local/bin/tomlv github.com/BurntSushi/toml/cmd/tomlv \
&& rm -rf "$GOPATH"
# Build/install the tool for embedding resources in Windows binaries
ENV RSRC_VERSION v2
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone --depth 1 -b "$RSRC_VERSION" https://github.com/akavel/rsrc.git "$GOPATH/src/github.com/akavel/rsrc" \
&& go build -v -o /usr/local/bin/rsrc github.com/akavel/rsrc \
&& rm -rf "$GOPATH"
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker

View File

@@ -23,34 +23,12 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
e2fsprogs \
iptables \
procps \
xfsprogs \
xz-utils \
\
aufs-tools \
lxc \
&& rm -rf /var/lib/apt/lists/*
# Install runc
ENV RUNC_COMMIT e87436998478d222be209707503c27f6f91be0c5
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/opencontainers/runc.git "$GOPATH/src/github.com/opencontainers/runc" \
&& cd "$GOPATH/src/github.com/opencontainers/runc" \
&& git checkout -q "$RUNC_COMMIT" \
&& make static BUILDTAGS="seccomp apparmor selinux" \
&& cp runc /usr/local/bin/docker-runc
# Install containerd
ENV CONTAINERD_COMMIT d2f03861c91edaafdcb3961461bf82ae83785ed7
RUN set -x \
&& export GOPATH="$(mktemp -d)" \
&& git clone git://github.com/docker/containerd.git "$GOPATH/src/github.com/docker/containerd" \
&& cd "$GOPATH/src/github.com/docker/containerd" \
&& git checkout -q "$CONTAINERD_COMMIT" \
&& make static \
&& cp bin/containerd /usr/local/bin/docker-containerd \
&& cp bin/containerd-shim /usr/local/bin/docker-containerd-shim \
&& cp bin/ctr /usr/local/bin/docker-containerd-ctr
ENV AUTO_GOPATH 1
WORKDIR /usr/src/docker
COPY . /usr/src/docker

View File

@@ -1,101 +0,0 @@
# This file describes the standard way to build Docker, using a docker container on Windows
# Server 2016
#
# Usage:
#
# # Assemble the full dev environment. This is slow the first time. Run this from
# # a directory containing the sources you are validating. For example from
# # c:\go\src\github.com\docker\docker
#
# docker build -t docker -f Dockerfile.windows .
#
#
# # Build docker in a container. Run the following from a Windows cmd command prommpt,
# # replacing c:\built with the directory you want the binaries to be placed on the
# # host system.
#
# docker run --rm -v "c:\built:c:\target" docker sh -c 'cd /c/go/src/github.com/docker/docker; hack/make.sh binary; ec=$?; if [ $ec -eq 0 ]; then robocopy /c/go/src/github.com/docker/docker/bundles/$(cat VERSION)/binary /c/target/binary; fi; exit $ec'
#
# Important notes:
# ---------------
#
# 'Start-Sleep' is a deliberate workaround for a current problem on containers in Windows
# Server 2016. It ensures that the network is up and available for when the command is
# network related. This bug is being tracked internally at Microsoft and exists in TP4.
# Generally sleep 1 or 2 is probably enough, but making it 5 to make the build file
# as bullet proof as possible. This isn't a big deal as this only runs the first time.
#
# The cygwin posix utilities from GIT aren't usable interactively as at January 2016. This
# is because they require a console window which isn't present in a container in Windows.
# See the example at the top of this file. Do NOT use -it in that docker run!!!
#
# Don't try to use a volume for passing the source through. The cygwin posix utilities will
# balk at reparse points. Again, see the example at the top of this file on how use a volume
# to get the built binary out of the container.
#
# The steps are minimised dramatically to improve performance (TP4 is slow on commit)
FROM windowsservercore
# Environment variable notes:
# - GO_VERSION must consistent with 'Dockerfile' used by Linux'.
# - FROM_DOCKERFILE is used for detection of building within a container.
ENV GO_VERSION=1.5.4 \
GIT_LOCATION=https://github.com/git-for-windows/git/releases/download/v2.7.2.windows.1/Git-2.7.2-64-bit.exe \
RSRC_COMMIT=ba14da1f827188454a4591717fff29999010887f \
GOPATH=C:/go;C:/go/src/github.com/docker/docker/vendor \
FROM_DOCKERFILE=1
WORKDIR c:/
# Everything downloaded/installed in one go (better performance, esp on TP4)
RUN \
setx /M Path "c:\git\cmd;c:\git\bin;c:\git\usr\bin;%Path%;c:\gcc\bin;c:\go\bin" && \
setx GOROOT "c:\go" && \
powershell -command \
$ErrorActionPreference = 'Stop'; \
Start-Sleep -Seconds 5; \
Function Download-File([string] $source, [string] $target) { \
$wc = New-Object net.webclient; $wc.Downloadfile($source, $target) \
} \
\
Write-Host INFO: Downloading git...; \
Download-File %GIT_LOCATION% gitsetup.exe; \
\
Write-Host INFO: Downloading go...; \
Download-File https://storage.googleapis.com/golang/go%GO_VERSION%.windows-amd64.msi go.msi; \
\
Write-Host INFO: Downloading compiler 1 of 3...; \
Download-File https://raw.githubusercontent.com/jhowardmsft/docker-tdmgcc/master/gcc.zip gcc.zip; \
\
Write-Host INFO: Downloading compiler 2 of 3...; \
Download-File https://raw.githubusercontent.com/jhowardmsft/docker-tdmgcc/master/runtime.zip runtime.zip; \
\
Write-Host INFO: Downloading compiler 3 of 3...; \
Download-File https://raw.githubusercontent.com/jhowardmsft/docker-tdmgcc/master/binutils.zip binutils.zip; \
\
Write-Host INFO: Installing git...; \
Start-Process gitsetup.exe -ArgumentList '/VERYSILENT /SUPPRESSMSGBOXES /CLOSEAPPLICATIONS /DIR=c:\git\' -Wait; \
\
Write-Host INFO: Installing go..."; \
Start-Process msiexec -ArgumentList '-i go.msi -quiet' -Wait; \
\
Write-Host INFO: Unzipping compiler...; \
c:\git\usr\bin\unzip.exe -q -o gcc.zip -d /c/gcc; \
c:\git\usr\bin\unzip.exe -q -o runtime.zip -d /c/gcc; \
c:\git\usr\bin\unzip.exe -q -o binutils.zip -d /c/gcc"; \
\
Write-Host INFO: Removing interim files; \
Remove-Item *.zip; \
Remove-Item go.msi; \
Remove-Item gitsetup.exe; \
\
Write-Host INFO: Cloning and installing RSRC; \
c:\git\bin\git.exe clone https://github.com/akavel/rsrc.git c:\go\src\github.com\akavel\rsrc; \
cd \go\src\github.com\akavel\rsrc; c:\git\bin\git.exe checkout -q %RSRC_COMMIT%; c:\go\bin\go.exe install -v; \
\
Write-Host INFO: Completed
# Prepare for building
COPY . /go/src/github.com/docker/docker

View File

@@ -176,7 +176,7 @@
END OF TERMS AND CONDITIONS
Copyright 2013-2016 Docker, Inc.
Copyright 2013-2015 Docker, Inc.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.

View File

@@ -1,16 +1,253 @@
# Docker maintainers file
#
# This file describes who runs the docker/docker project and how.
# This is a living document - if you see something out of date or missing, speak up!
# This file describes who runs the Docker project and how.
# This is a living document - if you see something out of date or missing,
# speak up!
#
# It is structured to be consumable by both humans and programs.
# To extract its contents programmatically, use any TOML-compliant
# parser.
#
# This file is compiled into the MAINTAINERS file in docker/opensource.
#
[Rules]
[Rules.maintainers]
title = "What is a maintainer?"
text = """
There are different types of maintainers, with different responsibilities, but
all maintainers have 3 things in common:
1) They share responsibility in the project's success.
2) They have made a long-term, recurring time investment to improve the project.
3) They spend that time doing whatever needs to be done, not necessarily what
is the most interesting or fun.
Maintainers are often under-appreciated, because their work is harder to appreciate.
It's easy to appreciate a really cool and technically advanced feature. It's harder
to appreciate the absence of bugs, the slow but steady improvement in stability,
or the reliability of a release process. But those things distinguish a good
project from a great one.
"""
[Rules.bdfl]
title = "The Benevolent dictator for life (BDFL)"
text = """
Docker follows the timeless, highly efficient and totally unfair system
known as [Benevolent dictator for
life](https://en.wikipedia.org/wiki/Benevolent_Dictator_for_Life), with
yours truly, Solomon Hykes, in the role of BDFL. This means that all
decisions are made, by default, by Solomon. Since making every decision
myself would be highly un-scalable, in practice decisions are spread
across multiple maintainers.
Ideally, the BDFL role is like the Queen of England: awesome crown, but not
an actual operational role day-to-day. The real job of a BDFL is to NEVER GO AWAY.
Every other rule can change, perhaps drastically so, but the BDFL will always
be there, preserving the philosophy and principles of the project, and keeping
ultimate authority over its fate. This gives us great flexibility in experimenting
with various governance models, knowing that we can always press the "reset" button
without fear of fragmentation or deadlock. See the US congress for a counter-example.
BDFL daily routine:
* Is the project governance stuck in a deadlock or irreversibly fragmented?
* If yes: refactor the project governance
* Are there issues or conflicts escalated by core?
* If yes: resolve them
* Go back to polishing that crown.
"""
[Rules.decisions]
title = "How are decisions made?"
text = """
Short answer: EVERYTHING IS A PULL REQUEST.
Docker is an open-source project with an open design philosophy. This
means that the repository is the source of truth for EVERY aspect of the
project, including its philosophy, design, road map, and APIs. *If it's
part of the project, it's in the repo. If it's in the repo, it's part of
the project.*
As a result, all decisions can be expressed as changes to the
repository. An implementation change is a change to the source code. An
API change is a change to the API specification. A philosophy change is
a change to the philosophy manifesto, and so on.
All decisions affecting Docker, big and small, follow the same 3 steps:
* Step 1: Open a pull request. Anyone can do this.
* Step 2: Discuss the pull request. Anyone can do this.
* Step 3: Merge or refuse the pull request. Who does this depends on the nature
of the pull request and which areas of the project it affects. See *review flow*
for details.
Because Docker is such a large and active project, it's important for everyone to know
who is responsible for deciding what. That is determined by a precise set of rules.
* For every *decision* in the project, the rules should designate, in a deterministic way,
who should *decide*.
* For every *problem* in the project, the rules should designate, in a deterministic way,
who should be responsible for *fixing* it.
* For every *question* in the project, the rules should designate, in a deterministic way,
who should be expected to have the *answer*.
"""
[Rules.review]
title = "Review flow"
text = """
Pull requests should be processed according to the following flow:
* For each subsystem affected by the change, the maintainers of the subsystem must approve or refuse it.
It is the responsibility of the subsystem maintainers to process patches affecting them in a timely
manner.
* If the change affects areas of the code which are not part of a subsystem,
or if subsystem maintainers are unable to reach a timely decision, it must be approved by
the core maintainers.
* If the change affects the UI or public APIs, or if it represents a major change in architecture,
the architects must approve or refuse it.
* If the change affects the operations of the project, it must be approved or rejected by
the relevant operators.
* If the change affects the governance, philosophy, goals or principles of the project,
it must be approved by BDFL.
"""
[Rules.DCO]
title = "Helping contributors with the DCO"
text = """
The [DCO or `Sign your work`](
https://github.com/docker/docker/blob/master/CONTRIBUTING.md#sign-your-work)
requirement is not intended as a roadblock or speed bump.
Some Docker contributors are not as familiar with `git`, or have used a web based
editor, and thus asking them to `git commit --amend -s` is not the best way forward.
In this case, maintainers can update the commits based on clause (c) of the DCO. The
most trivial way for a contributor to allow the maintainer to do this, is to add
a DCO signature in a Pull Requests's comment, or a maintainer can simply note that
the change is sufficiently trivial that it does not substantivly change the existing
contribution - i.e., a spelling change.
When you add someone's DCO, please also add your own to keep a log.
"""
[Rules.holiday]
title = "I'm a maintainer, and I'm going on holiday"
text = """
Please let your co-maintainers and other contributors know by raising a pull
request that comments out your `MAINTAINERS` file entry using a `#`.
"""
[Rules."no direct push"]
title = "I'm a maintainer. Should I make pull requests too?"
text = """
Yes. Nobody should ever push to master directly. All changes should be
made through a pull request.
"""
[Rules.meta]
title = "How is this process changed?"
text = "Just like everything else: by making a pull request :)"
# Current project organization
[Org]
bdfl = "shykes"
# The chief architect is responsible for the overall integrity of the technical architecture
# across all subsystems, and the consistency of APIs and UI.
#
# Changes to UI, public APIs and overall architecture (for example a plugin system) must
# be approved by the chief architect.
"Chief Architect" = "shykes"
# The Chief Operator is responsible for the day-to-day operations of the project including:
# - facilitating communications amongst all the contributors;
# - tracking release schedules;
# - managing the relationship with downstream distributions and upstream dependencies;
# - helping new contributors to get involved and become successful contributors and maintainers
#
# The role is also responsible for managing and measuring the success of the overall project
# and ensuring it is governed properly working in concert with the Docker Governance Advisory Board (DGAB).
"Chief Operator" = "spf13"
[Org.Operators]
# The operators make sure the trains run on time. They are responsible for overall operations
# of the project. This includes facilitating communication between all the participants; helping
# newcomers get involved and become successful contributors and maintainers; tracking the schedule
# of releases; managing the relationship with downstream distributions and upstream dependencies;
# define measures of success for the project and measure progress; Devise and implement tools and
# processes which make contributors and maintainers happier and more efficient.
[Org.Operators.security]
people = [
"erw",
"diogomonica",
"nathanmccauley"
]
[Org.Operators."monthly meetings"]
people = [
"sven",
"tianon"
]
[Org.Operators.infrastructure]
people = [
"jfrazelle",
"crosbymichael"
]
[Org.Operators.community]
people = [
"theadactyl"
]
# The chief maintainer is responsible for all aspects of quality for the project including
# code reviews, usability, stability, security, performance, etc.
# The most important function of the chief maintainer is to lead by example. On the first
# day of a new maintainer, the best advice should be "follow the C.M.'s example and you'll
# be fine".
"Chief Maintainer" = "crosbymichael"
# The community manager is responsible for serving the project community, including users,
# contributors and partners. This involves:
# - facilitating communication between maintainers, contributors and users
# - organizing contributor and maintainer events
# - helping new contributors get involved
# - anything the project community needs to be successful
#
# The community manager is a point of contact for any contributor who has questions, concerns
# or feedback about project operations.
"Community Manager" = "theadactyl"
[Org."Core maintainers"]
# The Core maintainers are the ghostbusters of the project: when there's a problem others
@@ -24,39 +261,151 @@
# For each release (including minor releases), a "release captain" is assigned from the
# pool of core maintainers. Rotation is encouraged across all maintainers, to ensure
# the release process is clear and up-to-date.
#
# It is common for core maintainers to "branch out" to join or start a subsystem.
people = [
"aaronlehmann",
"calavera",
"coolljt0725",
"cpuguy83",
"crosbymichael",
"duglin",
"erikh",
"estesp",
"icecrime",
"jhowardmsft",
"jfrazelle",
"lk4d4",
"mhbauer",
"runcom",
"tianon",
"tibor",
"tonistiigi",
"unclejack",
"vbatts",
"vdemeester"
"vieux",
"vishh"
]
[Org."Docs maintainers"]
# TODO Describe the docs maintainers role.
[Org.Subsystems]
people = [
"jamtur01",
"moxiegirl",
"sven",
"thajeztah"
]
# As the project grows, it gets separated into well-defined subsystems. Each subsystem
# has a dedicated group of maintainers, which are dedicated to that subsytem and responsible
# for its quality.
# This "cellular division" is the primary mechanism for scaling maintenance of the project as it grows.
#
# The maintainers of each subsytem are responsible for:
#
# 1. Exposing a clear road map for improving their subsystem.
# 2. Deliver prompt feedback and decisions on pull requests affecting their subsystem.
# 3. Be available to anyone with questions, bug reports, criticism etc.
# on their component. This includes IRC, GitHub requests and the mailing
# list.
# 4. Make sure their subsystem respects the philosophy, design and
# road map of the project.
#
# #### How to review patches to your subsystem
#
# Accepting pull requests:
#
# - If the pull request appears to be ready to merge, give it a `LGTM`, which
# stands for "Looks Good To Me".
# - If the pull request has some small problems that need to be changed, make
# a comment adressing the issues.
# - If the changes needed to a PR are small, you can add a "LGTM once the
# following comments are adressed..." this will reduce needless back and
# forth.
# - If the PR only needs a few changes before being merged, any MAINTAINER can
# make a replacement PR that incorporates the existing commits and fixes the
# problems before a fast track merge.
#
# Closing pull requests:
#
# - If a PR appears to be abandoned, after having attempted to contact the
# original contributor, then a replacement PR may be made. Once the
# replacement PR is made, any contributor may close the original one.
# - If you are not sure if the pull request implements a good feature or you
# do not understand the purpose of the PR, ask the contributor to provide
# more documentation. If the contributor is not able to adequately explain
# the purpose of the PR, the PR may be closed by any MAINTAINER.
# - If a MAINTAINER feels that the pull request is sufficiently architecturally
# flawed, or if the pull request needs significantly more design discussion
# before being considered, the MAINTAINER should close the pull request with
# a short explanation of what discussion still needs to be had. It is
# important not to leave such pull requests open, as this will waste both the
# MAINTAINER's time and the contributor's time. It is not good to string a
# contributor on for weeks or months, having them make many changes to a PR
# that will eventually be rejected.
[Org.Subsystems.Documentation]
people = [
"fredlf",
"james",
"moxiegirl",
"thaJeztah",
"jamtur01",
"spf13",
"sven"
]
[Org.Subsystems.libcontainer]
people = [
"crosbymichael",
"jnagal",
"lk4d4",
"mpatel",
"vmarmol"
]
[Org.Subsystems.registry]
people = [
"dmcg",
"dmp42",
"jlhawn",
"samalba",
"sday",
"vbatts"
]
[Org.Subsystems."build tools"]
people = [
"shykes",
"tianon"
]
[Org.Subsystem."remote api"]
people = [
"vieux"
]
[Org.Subsystem.swarm]
people = [
"aluzzardi",
"vieux"
]
[Org.Subsystem.machine]
people = [
"bfirsh",
"ehazlett"
]
[Org.Subsystem.compose]
people = [
"aanand"
]
[Org.Subsystem.builder]
people = [
"duglin",
"erikh",
"tibor"
]
[Org.Curators]
@@ -70,41 +419,10 @@
# - close an issue or pull request when it's an exact duplicate
# - close an issue or pull request when it's inappropriate or off-topic
people = [
"programmerq",
"thajeztah"
]
people = [
"thajeztah"
]
[Org.Alumni]
# This list contains maintainers that are no longer active on the project.
# It is thanks to these people that the project has become what it is today.
# Thank you!
people = [
# As a maintainer, Erik was responsible for the "builder", and
# started the first designs for the new networking model in
# Docker. Erik is now working on all kinds of plugins for Docker
# (https://github.com/contiv) and various open source projects
# in his own repository https://github.com/erikh. You may
# still stumble into him in our issue tracker, or on IRC.
"erikh",
# Victor is one of the earliest contributors to Docker, having worked on the
# project when it was still "dotCloud" in April 2013. He's been responsible
# for multiple releases (https://github.com/docker/docker/pulls?q=is%3Apr+bump+in%3Atitle+author%3Avieux),
# and up until today (2015), our number 2 contributor. Although he's no longer
# a maintainer for the Docker "Engine", he's still actively involved in other
# Docker projects, and most likely can be found in the Docker Swarm repository,
# for which he's a core maintainer.
"vieux",
# Vishnu became a maintainer to help out on the daemon codebase and
# libcontainer integration. He's currently involved in the
# Open Containers Initiative, working on the specifications,
# besides his work on cAdvisor and Kubernetes for Google.
"vishh"
]
[people]
@@ -114,21 +432,26 @@
# ADD YOURSELF HERE IN ALPHABETICAL ORDER
[people.aaronlehmann]
Name = "Aaron Lehmann"
Email = "aaron.lehmann@docker.com"
GitHub = "aaronlehmann"
[people.aanand]
Name = "Aanand Prasad"
Email = "aanand@docker.com"
GitHub = "aanand"
[people.aluzzardi]
Name = "Andrea Luzzardi"
Email = "aluzzardi@docker.com"
GitHub = "aluzzardi"
[people.bfirsh]
Name = "Ben Firshman"
Email = "ben@firshman.co.uk"
GitHub = "bfirsh"
[people.calavera]
Name = "David Calavera"
Email = "david.calavera@gmail.com"
GitHub = "calavera"
[people.coolljt0725]
Name = "Lei Jitang"
Email = "leijitang@huawei.com"
GitHub = "coolljt0725"
[people.cpuguy83]
Name = "Brian Goff"
Email = "cpuguy83@gmail.com"
@@ -139,71 +462,101 @@
Email = "crosbymichael@gmail.com"
GitHub = "crosbymichael"
[people.diogomonica]
Name = "Diogo Monica"
Email = "diogo@docker.com"
GitHub = "diogomonica"
[people.duglin]
Name = "Doug Davis"
Email = "dug@us.ibm.com"
GitHub = "duglin"
[people.dmcg]
Name = "Derek McGowan"
Email = "derek@docker.com"
Github = "dmcgowan"
[people.dmp42]
Name = "Olivier Gambier"
Email = "olivier@docker.com"
Github = "dmp42"
[people.ehazlett]
Name = "Evan Hazlett"
Email = "ejhazlett@gmail.com"
GitHub = "ehazlett"
[people.erikh]
Name = "Erik Hollensbe"
Email = "erik@docker.com"
GitHub = "erikh"
[people.erw]
Name = "Eric Windisch"
Email = "eric@windisch.us"
GitHub = "ewindisch"
[people.estesp]
Name = "Phil Estes"
Email = "estesp@linux.vnet.ibm.com"
GitHub = "estesp"
[people.fredlf]
Name = "Fred Lifton"
Email = "fred.lifton@docker.com"
GitHub = "fredlf"
[people.icecrime]
Name = "Arnaud Porterie"
Email = "arnaud@docker.com"
GitHub = "icecrime"
[people.jamtur01]
Name = "James Turnbull"
Email = "james@lovedthanlost.net"
GitHub = "jamtur01"
[people.jhowardmsft]
Name = "John Howard"
Email = "jhoward@microsoft.com"
GitHub = "jhowardmsft"
[people.jfrazelle]
Name = "Jessie Frazelle"
Email = "jess@linux.com"
Email = "j@docker.com"
GitHub = "jfrazelle"
[people.jlhawn]
Name = "Josh Hawn"
Email = "josh.hawn@docker.com"
Github = "jlhawn"
[people.lk4d4]
Name = "Alexander Morozov"
Email = "lk4d4@docker.com"
GitHub = "lk4d4"
[people.mhbauer]
Name = "Morgan Bauer"
Email = "mbauer@us.ibm.com"
GitHub = "mhbauer"
[people.moxiegirl]
Name = "Mary Anthony"
Email = "mary.anthony@docker.com"
GitHub = "moxiegirl"
[people.programmerq]
Name = "Jeff Anderson"
Email = "jeff@docker.com"
GitHub = "programmerq"
[people.nathanmccauley]
Name = "Nathan McCauley"
Email = "nathan.mccauley@docker.com"
GitHub = "nathanmccauley"
[people.runcom]
Name = "Antonio Murdaca"
Email = "runcom@redhat.com"
Email = "me@runcom.ninja"
GitHub = "runcom"
[people.sday]
Name = "Stephen Day"
Email = "stephen.day@docker.com"
Github = "stevvooe"
[people.shykes]
Name = "Solomon Hykes"
Email = "solomon@docker.com"
GitHub = "shykes"
[people.spf13]
Name = "Steve Francia"
Email = "steve.francia@gmail.com"
GitHub = "spf13"
[people.sven]
Name = "Sven Dowideit"
Email = "SvenDowideit@home.org.au"
@@ -214,6 +567,11 @@
Email = "github@gone.nl"
GitHub = "thaJeztah"
[people.theadactyl]
Name = "Thea Lamkin"
Email = "thea@docker.com"
GitHub = "theadactyl"
[people.tianon]
Name = "Tianon Gravi"
Email = "admwiggin@gmail.com"
@@ -224,31 +582,36 @@
Email = "tibor@docker.com"
GitHub = "tiborvass"
[people.tonistiigi]
Name = "Tõnis Tiigi"
Email = "tonis@docker.com"
GitHub = "tonistiigi"
[people.unclejack]
Name = "Cristian Staretu"
Email = "cristian.staretu@gmail.com"
GitHub = "unclejack"
[people.vbatts]
Name = "Vincent Batts"
Email = "vbatts@redhat.com"
GitHub = "vbatts"
[people.vdemeester]
Name = "Vincent Demeester"
Email = "vincent@sbr.pm"
GitHub = "vdemeester"
[people.vieux]
Name = "Victor Vieux"
Email = "vieux@docker.com"
GitHub = "vieux"
[people.vmarmol]
Name = "Victor Marmol"
Email = "vmarmol@google.com"
GitHub = "vmarmol"
[people.jnagal]
Name = "Rohit Jnagal"
Email = "jnagal@google.com"
GitHub = "rjnagal"
[people.mpatel]
Name = "Mrunal Patel"
Email = "mpatel@redhat.com"
GitHub = "mrunalp"
[people.unclejack]
Name = "Cristian Staretu"
Email = "cristian.staretu@gmail.com"
GitHub = "unclejack"
[people.vishh]
Name = "Vishnu Kannan"
Email = "vishnuk@google.com"

View File

@@ -1,23 +1,15 @@
.PHONY: all binary build cross default docs docs-build docs-shell shell test test-docker-py test-integration-cli test-unit validate
# get OS/Arch of docker engine
DOCKER_OSARCH := $(shell bash -c 'source hack/make/.detect-daemon-osarch && echo $${DOCKER_ENGINE_OSARCH:-$$DOCKER_CLIENT_OSARCH}')
DOCKERFILE := $(shell bash -c 'source hack/make/.detect-daemon-osarch && echo $${DOCKERFILE}')
.PHONY: all binary build cross default docs docs-build docs-shell shell test test-unit test-integration-cli test-docker-py validate
# env vars passed through directly to Docker's build scripts
# to allow things like `make DOCKER_CLIENTONLY=1 binary` easily
# `docs/sources/contributing/devenvironment.md ` and `project/PACKAGERS.md` have some limited documentation of some of these
DOCKER_ENVS := \
-e BUILDFLAGS \
-e KEEPBUNDLE \
-e DOCKER_BUILD_GOGC \
-e DOCKER_BUILD_PKGS \
-e DOCKER_CLIENTONLY \
-e DOCKER_DEBUG \
-e DOCKER_EXECDRIVER \
-e DOCKER_EXPERIMENTAL \
-e DOCKER_GRAPHDRIVER \
-e DOCKER_INCREMENTAL_BINARY \
-e DOCKER_REMAP_ROOT \
-e DOCKER_STORAGE_OPTS \
-e DOCKER_USERLANDPROXY \
-e TESTDIRS \
@@ -31,26 +23,17 @@ DOCKER_ENVS := \
BIND_DIR := $(if $(BINDDIR),$(BINDDIR),$(if $(DOCKER_HOST),,bundles))
DOCKER_MOUNT := $(if $(BIND_DIR),-v "$(CURDIR)/$(BIND_DIR):/go/src/github.com/docker/docker/$(BIND_DIR)")
# This allows the test suite to be able to run without worrying about the underlying fs used by the container running the daemon (e.g. aufs-on-aufs), so long as the host running the container is running a supported fs.
# The volume will be cleaned up when the container is removed due to `--rm`.
# Note that `BIND_DIR` will already be set to `bundles` if `DOCKER_HOST` is not set (see above BIND_DIR line), in such case this will do nothing since `DOCKER_MOUNT` will already be set.
DOCKER_MOUNT := $(if $(DOCKER_MOUNT),$(DOCKER_MOUNT),-v "/go/src/github.com/docker/docker/bundles")
GIT_BRANCH := $(shell git rev-parse --abbrev-ref HEAD 2>/dev/null)
DOCKER_IMAGE := docker-dev$(if $(GIT_BRANCH),:$(GIT_BRANCH))
DOCKER_DOCS_IMAGE := docker-docs$(if $(GIT_BRANCH),:$(GIT_BRANCH))
DOCKER_FLAGS := docker run --rm -i --privileged $(DOCKER_ENVS) $(DOCKER_MOUNT)
DOCKER_RUN_DOCKER := docker run --rm -it --privileged $(DOCKER_ENVS) $(DOCKER_MOUNT) "$(DOCKER_IMAGE)"
# if this session isn't interactive, then we don't want to allocate a
# TTY, which would fail, but if it is interactive, we do want to attach
# so that the user can send e.g. ^C through.
INTERACTIVE := $(shell [ -t 0 ] && echo 1 || echo 0)
ifeq ($(INTERACTIVE), 1)
DOCKER_FLAGS += -t
endif
DOCKER_RUN_DOCS := docker run --rm -it $(DOCS_MOUNT) -e AWS_S3_BUCKET -e NOCACHE
DOCKER_RUN_DOCKER := $(DOCKER_FLAGS) "$(DOCKER_IMAGE)"
# for some docs workarounds (see below in "docs-build" target)
GITCOMMIT := $(shell git rev-parse --short HEAD 2>/dev/null)
default: binary
@@ -60,47 +43,38 @@ all: build
binary: build
$(DOCKER_RUN_DOCKER) hack/make.sh binary
build: bundles
docker build ${DOCKER_BUILD_ARGS} -t "$(DOCKER_IMAGE)" -f "$(DOCKERFILE)" .
bundles:
mkdir bundles
cross: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary binary cross
win: build
$(DOCKER_RUN_DOCKER) hack/make.sh win
tgz: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary binary cross tgz
$(DOCKER_RUN_DOCKER) hack/make.sh binary cross
deb: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary build-deb
docs:
$(MAKE) -C docs docs
gccgo: build
$(DOCKER_RUN_DOCKER) hack/make.sh gccgo
$(DOCKER_RUN_DOCKER) hack/make.sh binary build-deb
rpm: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary build-rpm
shell: build
$(DOCKER_RUN_DOCKER) bash
$(DOCKER_RUN_DOCKER) hack/make.sh binary build-rpm
test: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary cross test-unit test-integration-cli test-docker-py
test-docker-py: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary test-docker-py
test-integration-cli: build
$(DOCKER_RUN_DOCKER) hack/make.sh dynbinary test-integration-cli
$(DOCKER_RUN_DOCKER) hack/make.sh binary cross test-unit test-integration-cli test-docker-py
test-unit: build
$(DOCKER_RUN_DOCKER) hack/make.sh test-unit
test-integration-cli: build
$(DOCKER_RUN_DOCKER) hack/make.sh binary test-integration-cli
test-docker-py: build
$(DOCKER_RUN_DOCKER) hack/make.sh binary test-docker-py
validate: build
$(DOCKER_RUN_DOCKER) hack/make.sh validate-dco validate-default-seccomp validate-gofmt validate-pkg validate-lint validate-test validate-toml validate-vet validate-vendor
$(DOCKER_RUN_DOCKER) hack/make.sh validate-dco validate-gofmt validate-pkg validate-lint validate-test validate-toml validate-vet
shell: build
$(DOCKER_RUN_DOCKER) bash
build: bundles
docker build -t "$(DOCKER_IMAGE)" .
bundles:
mkdir bundles
docs:
$(MAKE) -C docs docs

2
NOTICE
View File

@@ -1,5 +1,5 @@
Docker
Copyright 2012-2016 Docker, Inc.
Copyright 2012-2015 Docker, Inc.
This product includes software developed at Docker, Inc. (https://www.docker.com).

View File

@@ -1,12 +1,12 @@
Docker: the container engine [![Release](https://img.shields.io/github/release/docker/docker.svg)](https://github.com/docker/docker/releases/latest)
============================
Docker: the Linux container engine
==================================
Docker is an open source project to pack, ship and run any application
as a lightweight container.
Docker containers are both *hardware-agnostic* and *platform-agnostic*.
This means they can run anywhere, from your laptop to the largest
cloud compute instance and everything in between - and they don't require
EC2 compute instance and everything in between - and they don't require
you to use a particular language, framework or packaging system. That
makes them great building blocks for deploying and scaling web apps,
databases, and backend services without depending on a particular stack
@@ -18,7 +18,7 @@ It benefits directly from the experience accumulated over several years
of large-scale operation and support of hundreds of thousands of
applications and databases.
![](docs/static_files/docker-logo-compressed.png "Docker")
![Docker L](docs/static_files/docker-logo-compressed.png "Docker")
## Security Disclosure
@@ -143,14 +143,15 @@ as they can be built by running a Unix command in a container.
Getting started
===============
Docker can be installed either on your computer for building applications or
on servers for running them. To get started, [check out the installation
instructions in the
documentation](https://docs.docker.com/engine/installation/).
Docker can be installed on your local machine as well as servers - both
bare metal and virtualized. It is available as a binary on most modern
Linux systems, or as a VM on Windows, Mac and other systems.
We also offer an [interactive tutorial](https://www.docker.com/tryit/)
for quickly learning the basics of using Docker.
For up-to-date install instructions, see the [Docs](https://docs.docker.com).
Usage examples
==============
@@ -158,7 +159,7 @@ Docker can be used to run short-lived commands, long-running daemons
(app servers, databases, etc.), interactive shell sessions, etc.
You can find a [list of real-world
examples](https://docs.docker.com/engine/examples/) in the
examples](https://docs.docker.com/examples/) in the
documentation.
Under the hood
@@ -167,23 +168,22 @@ Under the hood
Under the hood, Docker is built on the following components:
* The
[cgroups](https://www.kernel.org/doc/Documentation/cgroup-v1/cgroups.txt)
[cgroups](https://www.kernel.org/doc/Documentation/cgroups/cgroups.txt)
and
[namespaces](http://man7.org/linux/man-pages/man7/namespaces.7.html)
capabilities of the Linux kernel
* The [Go](https://golang.org) programming language
* The [Docker Image Specification](https://github.com/docker/docker/blob/master/image/spec/v1.md)
* The [Libcontainer Specification](https://github.com/opencontainers/runc/blob/master/libcontainer/SPEC.md)
* The [Libcontainer Specification](https://github.com/docker/libcontainer/blob/master/SPEC.md)
Contributing to Docker [![GoDoc](https://godoc.org/github.com/docker/docker?status.svg)](https://godoc.org/github.com/docker/docker)
Contributing to Docker
======================
| **Master** (Linux) | **Experimental** (linux) | **Windows** | **FreeBSD** |
|------------------|----------------------|---------|---------|
| [![Jenkins Build Status](https://jenkins.dockerproject.org/view/Docker/job/Docker%20Master/badge/icon)](https://jenkins.dockerproject.org/view/Docker/job/Docker%20Master/) | [![Jenkins Build Status](https://jenkins.dockerproject.org/view/Docker/job/Docker%20Master%20%28experimental%29/badge/icon)](https://jenkins.dockerproject.org/view/Docker/job/Docker%20Master%20%28experimental%29/) | [![Build Status](http://jenkins.dockerproject.org/job/Docker%20Master%20(windows)/badge/icon)](http://jenkins.dockerproject.org/job/Docker%20Master%20(windows)/) | [![Build Status](http://jenkins.dockerproject.org/job/Docker%20Master%20(freebsd)/badge/icon)](http://jenkins.dockerproject.org/job/Docker%20Master%20(freebsd)/) |
[![GoDoc](https://godoc.org/github.com/docker/docker?status.svg)](https://godoc.org/github.com/docker/docker)
[![Jenkins Build Status](https://jenkins.dockerproject.org/job/Docker%20Master/badge/icon)](https://jenkins.dockerproject.org/job/Docker%20Master/)
Want to hack on Docker? Awesome! We have [instructions to help you get
started contributing code or documentation](https://docs.docker.com/opensource/project/who-written-for/).
started contributing code or documentation](https://docs.docker.com/project/who-written-for/).
These instructions are probably not perfect, please let us know if anything
feels wrong or incomplete. Better yet, submit a PR and improve them yourself.
@@ -216,7 +216,7 @@ We are always open to suggestions on process improvements, and are always lookin
<td>Internet&nbsp;Relay&nbsp;Chat&nbsp;(IRC)</td>
<td>
<p>
IRC is a direct line to our most knowledgeable Docker users; we have
IRC a direct line to our most knowledgeable Docker users; we have
both the <code>#docker</code> and <code>#docker-dev</code> group on
<strong>irc.freenode.net</strong>.
IRC is a rich chat protocol but it can overwhelm new users. You can search
@@ -234,8 +234,6 @@ We are always open to suggestions on process improvements, and are always lookin
The <a href="https://groups.google.com/forum/#!forum/docker-dev" target="_blank">docker-dev</a>
group is for contributors and other people contributing to the Docker
project.
You can join them without an google account by sending an email to e.g. "docker-user+subscribe@googlegroups.com".
After receiving the join-request message, you can simply reply to that to confirm the subscribtion.
</td>
</tr>
<tr>
@@ -249,7 +247,7 @@ We are always open to suggestions on process improvements, and are always lookin
<tr>
<td>Stack Overflow</td>
<td>
Stack Overflow has over 7000 Docker questions listed. We regularly
Stack Overflow has over 7000K Docker questions listed. We regularly
monitor <a href="https://stackoverflow.com/search?tab=newest&q=docker" target="_blank">Docker questions</a>
and so do many other knowledgeable Docker users.
</td>
@@ -290,12 +288,8 @@ for a container-centric world
system
* [Docker Compose](https://github.com/docker/compose) (formerly Fig):
Define and run multi-container apps
* [Kitematic](https://github.com/docker/kitematic): The easiest way to use
* [Kitematic](https://github.com/kitematic/kitematic): The easiest way to use
Docker on Mac and Windows
If you know of another project underway that should be listed here, please help
us keep this list up-to-date by submitting a PR.
Awesome-Docker
==============
You can find more projects, tools and articles related to Docker on the [awesome-docker list](https://github.com/veggiemonk/awesome-docker). Add your project there.

View File

@@ -33,108 +33,151 @@ won't be accepting pull requests adding or removing items from this file.
# 1. Features and refactoring
## 1.1 Runtime improvements
## 1.1 Security
We recently introduced [`runC`](https://runc.io) as a standalone low-level tool for container
execution. The initial goal was to integrate runC as a replacement in the Engine for the traditional
default libcontainer `execdriver`, but the Engine internals were not ready for this.
Security is a top objective for the Docker Engine. The most notable items we intend to provide in
the near future are:
As runC continued evolving, and the OCI specification along with it, we created
[`containerd`](https://containerd.tools/), a daemon to control and monitor multiple `runC`. This is
the new target for Engine integration, as it can entirely replace the whole `execdriver`
architecture, and container monitoring along with it.
- Trusted distribution of images: the effort is driven by the [distribution](https://github.com/docker/distribution)
group but will have significant impact on the Engine
- [User namespaces](https://github.com/docker/docker/pull/12648)
- [Seccomp support](https://github.com/docker/libcontainer/pull/613)
Docker Engine will rely on a long-running `containerd` companion daemon for all container execution
related operations. This could open the door in the future for Engine restarts without interrupting
running containers.
## 1.2 Plumbing project
## 1.2 Plugins improvements
We define a plumbing tool as a standalone piece of software usable and meaningful on its own. In
the current state of the Docker Engine, most subsystems provide independent functionalities (such
the builder, pushing and pulling images, running applications in a containerized environment, etc)
but all are coupled in a single binary. We want to offer the users to flexibility to use only the
pieces they need, and we will also gain in maintainability by splitting the project among multiple
repositories.
Docker Engine 1.7.0 introduced plugin support, initially for the use cases of volumes and networks
extensions. The plugin infrastructure was kept minimal as we were collecting use cases and real
world feedback before optimizing for any particular workflow.
As it currently stands, the rough design outlines is to have:
- Low level plumbing tools, each dealing with one responsibility (e.g., [runC](https://runc.io))
- Docker subsystems services, each exposing an elementary concept over an API, and relying on one or
multiple lower level plumbing tools for their implementation (e.g., network management)
- Docker Engine to expose higher level actions (e.g., create a container with volume `V` and network
`N`), while still providing pass-through access to the individual subsystems.
In the future, we'd like plugins to become first class citizens, and encourage an ecosystem of
plugins. This implies in particular making it trivially easy to distribute plugins as containers
through any Registry instance, as well as solving the commonly heard pain points of plugins needing
to be treated as somewhat special (being active at all time, started before any other user
containers, and not as easily dismissed).
The architectural details are still being worked on, but one thing we know for sure is that we need
to technically decouple the pieces.
## 1.3 Internal decoupling
### 1.2.1 Runtime
A lot of work has been done in trying to decouple the Docker Engine's internals. In particular, the
API implementation has been refactored and ongoing work is happening to move the code to a separate
repository ([`docker/engine-api`](https://github.com/docker/engine-api)), and the Builder side of
the daemon is now [fully independent](https://github.com/docker/docker/tree/master/builder) while
still residing in the same repository.
A Runtime tool already exists today in the form of [runC](https://github.com/opencontainers/runc).
We intend to modify the Engine to directly call out to a binary implementing the Open Containers
Specification such as runC rather than relying on libcontainer to set the container runtime up.
We are exploring ways to go further with that decoupling, capitalizing on the work introduced by the
runtime renovation and plugins improvement efforts. Indeed, the combination of `containerd` support
with the concept of "special" containers opens the door for bootstrapping more Engine internals
using the same facilities.
This plan will deprecate the existing [`execdriver`](https://github.com/docker/docker/tree/master/daemon/execdriver)
as different runtime backends will be implemented as separated binaries instead of being compiled
into the Engine.
## 1.4 Cluster capable Engine
### 1.2.2 Builder
The community has been pushing for a more cluster capable Docker Engine, and a huge effort was spent
adding features such as multihost networking, and node discovery down at the Engine level. Yet, the
Engine is currently incapable of taking scheduling decisions alone, and continues relying on Swarm
for that.
The Builder (i.e., the ability to build an image from a Dockerfile) is already nicely decoupled,
but would benefit from being entirely separated from the Engine, and rely on the standard Engine
API for its operations.
We plan to complete this effort and make Engine fully cluster capable. Multiple instances of the
Docker Engine being already capable of discovering each other and establish overlay networking for
their container to communicate, the next step is for a given Engine to gain ability to dispatch work
to another node in the cluster. This will be introduced in a backward compatible way, such that a
`docker run` invocation on a particular node remains fully deterministic.
### 1.2.3 Distribution
Distribution already has a [dedicated repository](https://github.com/docker/distribution) which
holds the implementation for Registry v2 and client libraries. We could imagine going further by
having the Engine call out to a binary providing image distribution related functionalities.
There are two short term goals related to image distribution. The first is stabilize and simplify
the push/pull code. Following that is the conversion to the more secure Registry V2 protocol.
### 1.2.4 Networking
Most of networking related code was already decoupled today in [libnetwork](https://github.com/docker/libnetwork).
As with other ingredients, we might want to take it a step further and make it a meaningful utility
that the Engine would call out to instead of a library.
## 1.3 Plugins
An initiative around plugins started with Docker 1.7.0, with the goal of allowing for out of
process extensibility of some Docker functionalities, starting with volumes and networking. The
approach is to provide specific extension points rather than generic hooking facilities. We also
deliberately keep the extensions API the simplest possible, expanding as we discover valid use
cases that cannot be implemented.
At the time of writing:
- Plugin support is merged as an experimental feature: real world use cases and user feedback will
help us refine the UX to make the feature more user friendly.
- There are no immediate plans to expand on the number of pluggable subsystems.
- Golang 1.5 might add language support for [plugins](https://docs.google.com/document/d/1nr-TQHw_er6GOQRsF6T43GGhFDelrAP0NqSS_00RgZQ)
which we consider supporting as an alternative to JSON/HTTP.
## 1.4 Volume management
Volumes are not a first class citizen in the Engine today: we would like better volume management,
similar to the way network are managed in the new [CNM](https://github.com/docker/docker/issues/9983).
## 1.5 Better API implementation
The current Engine API is insufficiently typed, versioned, and ultimately hard to maintain. We
also suffer from the lack of a common implementation with [Swarm](https://github.com/docker/swarm).
## 1.6 Checkpoint/restore
Support for checkpoint/restore was [merged](https://github.com/docker/libcontainer/pull/479) in
[libcontainer](https://github.com/docker/libcontainer) and made available through [runC](https://runc.io):
we intend to take advantage of it in the Engine.
# 2 Frozen features
## 2.1 Docker exec
We won't accept patches expanding the surface of `docker exec`, which we intend to keep as a
*debugging* feature, as well as being strongly dependent on the Runtime ingredient effort.
*debugging* feature, as well as being strongly dependent on the the Runtime ingredient effort.
## 2.2 Dockerfile syntax
The Dockerfile syntax as we know it is simple, and has proven successful in supporting all our
The Dockerfile syntax as we know it is simple, and has proven succesful in supporting all our
[official images](https://github.com/docker-library/official-images). Although this is *not* a
definitive move, we temporarily won't accept more patches to the Dockerfile syntax for several
reasons:
- Long term impact of syntax changes is a sensitive matter that require an amount of attention the
volume of Engine codebase and activity today doesn't allow us to provide.
- Allowing the Builder to be implemented as a separate utility consuming the Engine's API will
open the door for many possibilities, such as offering alternate syntaxes or DSL for existing
languages without cluttering the Engine's codebase.
- A standalone Builder will also offer the opportunity for a better dedicated group of maintainers
to own the Dockerfile syntax and decide collectively on the direction to give it.
- Our experience with official images tend to show that no new instruction or syntax expansion is
*strictly* necessary for the majority of use cases, and although we are aware many things are
still lacking for many, we cannot make it a priority yet for the above reasons.
- Long term impact of syntax changes is a sensitive matter that require an amount of attention
the volume of Engine codebase and activity today doesn't allow us to provide.
- Allowing the Builder to be implemented as a separate utility consuming the Engine's API will
open the door for many possibilities, such as offering alternate syntaxes or DSL for existing
languages without cluttering the Engine's codebase.
- A standalone Builder will also offer the opportunity for a better dedicated group of maintainers
to own the Dockerfile syntax and decide collectively on the direction to give it.
- Our experience with official images tend to show that no new instruction or syntax expansion is
*strictly* necessary for the majority of use cases, and although we are aware many things are still
lacking for many, we cannot make it a priority yet for the above reasons.
Again, this is not about saying that the Dockerfile syntax is done, it's about making choices about
what we want to do first!
## 2.3 Remote Registry Operations
A large amount of work is ongoing in the area of image distribution and provenance. This includes
moving to the V2 Registry API and heavily refactoring the code that powers these features. The
desired result is more secure, reliable and easier to use image distribution.
A large amount of work is ongoing in the area of image distribution and
provenance. This includes moving to the V2 Registry API and heavily
refactoring the code that powers these features. The desired result is more
secure, reliable and easier to use image distribution.
Part of the problem with this part of the code base is the lack of a stable and flexible interface.
If new features are added that access the registry without solidifying these interfaces, achieving
feature parity will continue to be elusive. While we get a handle on this situation, we are imposing
a moratorium on new code that accesses the Registry API in commands that don't already make remote
calls.
Part of the problem with this part of the code base is the lack of a stable
and flexible interface. If new features are added that access the registry
without solidifying these interfaces, achieving feature parity will continue
to be elusive. While we get a handle on this situation, we are imposing a
moratorium on new code that accesses the Registry API in commands that don't
already make remote calls.
Currently, only the following commands cause interaction with a remote registry:
Currently, only the following commands cause interaction with a remote
registry:
- push
- pull
- run
- build
- search
- login
- push
- pull
- run
- build
- search
- login
In the interest of stabilizing the registry access model during this ongoing work, we are not
accepting additions to other commands that will cause remote interaction with the Registry API. This
moratorium will lift when the goals of the distribution project have been met.
In the interest of stabilizing the registry access model during this ongoing
work, we are not accepting additions to other commands that will cause remote
interaction with the Registry API. This moratorium will lift when the goals of
the distribution project have been met.

View File

@@ -1,45 +0,0 @@
# Vendoring policies
This document outlines recommended Vendoring policies for Docker repositories.
(Example, libnetwork is a Docker repo and logrus is not.)
## Vendoring using tags
Commit ID based vendoring provides little/no information about the updates
vendored. To fix this, vendors will now require that repositories use annotated
tags along with commit ids to snapshot commits. Annotated tags by themselves
are not sufficient, since the same tag can be force updated to reference
different commits.
Each tag should:
- Follow Semantic Versioning rules (refer to section on "Semantic Versioning")
- Have a corresponding entry in the change tracking document.
Each repo should:
- Have a change tracking document between tags/releases. Ex: CHANGELOG.md,
github releases file.
The goal here is for consuming repos to be able to use the tag version and
changelog updates to determine whether the vendoring will cause any breaking or
backward incompatible changes. This also means that repos can specify having
dependency on a package of a specific version or greater up to the next major
release, without encountering breaking changes.
## Semantic Versioning
Annotated version tags should follow Schema Versioning policies.
According to http://semver.org:
"Given a version number MAJOR.MINOR.PATCH, increment the:
MAJOR version when you make incompatible API changes,
MINOR version when you add functionality in a backwards-compatible manner, and
PATCH version when you make backwards-compatible bug fixes.
Additional labels for pre-release and build metadata are available as extensions
to the MAJOR.MINOR.PATCH format."
## Vendoring cadence
In order to avoid huge vendoring changes, it is recommended to have a regular
cadence for vendoring updates. eg. monthly.
## Pre-merge vendoring tests
All related repos will be vendored into docker/docker.
CI on docker/docker should catch any breaking changes involving multiple repos.

View File

@@ -1 +1 @@
1.11.0
1.8.2-rc1

19
api/api_unit_test.go Normal file
View File

@@ -0,0 +1,19 @@
package api
import (
"testing"
)
func TestJsonContentType(t *testing.T) {
if !MatchesContentType("application/json", "application/json") {
t.Fail()
}
if !MatchesContentType("application/json; charset=utf-8", "application/json") {
t.Fail()
}
if MatchesContentType("dockerapplication/json", "application/json") {
t.Fail()
}
}

View File

@@ -1,103 +1,78 @@
package client
import (
"encoding/json"
"fmt"
"io"
"golang.org/x/net/context"
"net/url"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/signal"
"github.com/docker/engine-api/types"
)
// CmdAttach attaches to a running container.
//
// Usage: docker attach [OPTIONS] CONTAINER
func (cli *DockerCli) CmdAttach(args ...string) error {
cmd := Cli.Subcmd("attach", []string{"CONTAINER"}, Cli.DockerCommands["attach"].Description, true)
noStdin := cmd.Bool([]string{"-no-stdin"}, false, "Do not attach STDIN")
proxy := cmd.Bool([]string{"-sig-proxy"}, true, "Proxy all received signals to the process")
detachKeys := cmd.String([]string{"-detach-keys"}, "", "Override the key sequence for detaching a container")
cmd := Cli.Subcmd("attach", []string{"CONTAINER"}, "Attach to a running container", true)
noStdin := cmd.Bool([]string{"#nostdin", "-no-stdin"}, false, "Do not attach STDIN")
proxy := cmd.Bool([]string{"#sig-proxy", "-sig-proxy"}, true, "Proxy all received signals to the process")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
c, err := cli.client.ContainerInspect(context.Background(), cmd.Arg(0))
serverResp, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/json", nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
var c types.ContainerJSON
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return err
}
if !c.State.Running {
return fmt.Errorf("You cannot attach to a stopped container, start it first")
}
if c.State.Paused {
return fmt.Errorf("You cannot attach to a paused container, unpause it first")
}
if err := cli.CheckTtyInput(!*noStdin, c.Config.Tty); err != nil {
return err
}
if *detachKeys != "" {
cli.configFile.DetachKeys = *detachKeys
}
options := types.ContainerAttachOptions{
ContainerID: cmd.Arg(0),
Stream: true,
Stdin: !*noStdin && c.Config.OpenStdin,
Stdout: true,
Stderr: true,
DetachKeys: cli.configFile.DetachKeys,
}
var in io.ReadCloser
if options.Stdin {
in = cli.in
}
if *proxy && !c.Config.Tty {
sigc := cli.forwardAllSignals(options.ContainerID)
defer signal.StopCatch(sigc)
}
resp, err := cli.client.ContainerAttach(context.Background(), options)
if err != nil {
return err
}
defer resp.Close()
if in != nil && c.Config.Tty {
if err := cli.setRawTerminal(); err != nil {
return err
}
defer cli.restoreTerminal(in)
}
if c.Config.Tty && cli.isTerminalOut {
height, width := cli.getTtySize()
// To handle the case where a user repeatedly attaches/detaches without resizing their
// terminal, the only way to get the shell prompt to display for attaches 2+ is to artificially
// resize it, then go back to normal. Without this, every attach after the first will
// require the user to manually resize or hit enter.
cli.resizeTtyTo(cmd.Arg(0), height+1, width+1, false)
// After the above resizing occurs, the call to monitorTtySize below will handle resetting back
// to the actual size.
if err := cli.monitorTtySize(cmd.Arg(0), false); err != nil {
logrus.Debugf("Error monitoring TTY size: %s", err)
}
}
if err := cli.holdHijackedConnection(c.Config.Tty, in, cli.out, cli.err, resp); err != nil {
var in io.ReadCloser
v := url.Values{}
v.Set("stream", "1")
if !*noStdin && c.Config.OpenStdin {
v.Set("stdin", "1")
in = cli.in
}
v.Set("stdout", "1")
v.Set("stderr", "1")
if *proxy && !c.Config.Tty {
sigc := cli.forwardAllSignals(cmd.Arg(0))
defer signal.StopCatch(sigc)
}
if err := cli.hijack("POST", "/containers/"+cmd.Arg(0)+"/attach?"+v.Encode(), c.Config.Tty, in, cli.out, cli.err, nil, nil); err != nil {
return err
}
_, status, err := getExitCode(cli, options.ContainerID)
_, status, err := getExitCode(cli, cmd.Arg(0))
if err != nil {
return err
}

View File

@@ -3,36 +3,45 @@ package client
import (
"archive/tar"
"bufio"
"bytes"
"encoding/base64"
"encoding/json"
"fmt"
"io"
"io/ioutil"
"net/http"
"net/url"
"os"
"os/exec"
"path"
"path/filepath"
"regexp"
"runtime"
"golang.org/x/net/context"
"strconv"
"strings"
"github.com/docker/docker/api"
"github.com/docker/docker/builder"
"github.com/docker/docker/builder/dockerignore"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/graph/tags"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/archive"
"github.com/docker/docker/pkg/fileutils"
"github.com/docker/docker/pkg/httputils"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/progress"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/progressreader"
"github.com/docker/docker/pkg/streamformatter"
"github.com/docker/docker/pkg/symlink"
"github.com/docker/docker/pkg/ulimit"
"github.com/docker/docker/pkg/units"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/reference"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/container"
"github.com/docker/go-units"
"github.com/docker/docker/registry"
"github.com/docker/docker/utils"
)
type translatorFunc func(reference.NamedTagged) (reference.Canonical, error)
const (
tarHeaderSize = 512
)
// CmdBuild builds a new image from the source code at a given path.
//
@@ -40,33 +49,25 @@ type translatorFunc func(reference.NamedTagged) (reference.Canonical, error)
//
// Usage: docker build [OPTIONS] PATH | URL | -
func (cli *DockerCli) CmdBuild(args ...string) error {
cmd := Cli.Subcmd("build", []string{"PATH | URL | -"}, Cli.DockerCommands["build"].Description, true)
flTags := opts.NewListOpts(validateTag)
cmd.Var(&flTags, []string{"t", "-tag"}, "Name and optionally a tag in the 'name:tag' format")
suppressOutput := cmd.Bool([]string{"q", "-quiet"}, false, "Suppress the build output and print image ID on success")
noCache := cmd.Bool([]string{"-no-cache"}, false, "Do not use cache when building the image")
rm := cmd.Bool([]string{"-rm"}, true, "Remove intermediate containers after a successful build")
cmd := Cli.Subcmd("build", []string{"PATH | URL | -"}, "Build a new image from the source code at PATH", true)
tag := cmd.String([]string{"t", "-tag"}, "", "Repository name (and optionally a tag) for the image")
suppressOutput := cmd.Bool([]string{"q", "-quiet"}, false, "Suppress the verbose output generated by the containers")
noCache := cmd.Bool([]string{"#no-cache", "-no-cache"}, false, "Do not use cache when building the image")
rm := cmd.Bool([]string{"#rm", "-rm"}, true, "Remove intermediate containers after a successful build")
forceRm := cmd.Bool([]string{"-force-rm"}, false, "Always remove intermediate containers")
pull := cmd.Bool([]string{"-pull"}, false, "Always attempt to pull a newer version of the image")
dockerfileName := cmd.String([]string{"f", "-file"}, "", "Name of the Dockerfile (Default is 'PATH/Dockerfile')")
flMemoryString := cmd.String([]string{"m", "-memory"}, "", "Memory limit")
flMemorySwap := cmd.String([]string{"-memory-swap"}, "", "Swap limit equal to memory plus swap: '-1' to enable unlimited swap")
flShmSize := cmd.String([]string{"-shm-size"}, "", "Size of /dev/shm, default value is 64MB")
flCPUShares := cmd.Int64([]string{"#c", "-cpu-shares"}, 0, "CPU shares (relative weight)")
flCPUPeriod := cmd.Int64([]string{"-cpu-period"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) period")
flCPUQuota := cmd.Int64([]string{"-cpu-quota"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) quota")
flMemorySwap := cmd.String([]string{"-memory-swap"}, "", "Total memory (memory + swap), '-1' to disable swap")
flCPUShares := cmd.Int64([]string{"c", "-cpu-shares"}, 0, "CPU shares (relative weight)")
flCpuPeriod := cmd.Int64([]string{"-cpu-period"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) period")
flCpuQuota := cmd.Int64([]string{"-cpu-quota"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) quota")
flCPUSetCpus := cmd.String([]string{"-cpuset-cpus"}, "", "CPUs in which to allow execution (0-3, 0,1)")
flCPUSetMems := cmd.String([]string{"-cpuset-mems"}, "", "MEMs in which to allow execution (0-3, 0,1)")
flCgroupParent := cmd.String([]string{"-cgroup-parent"}, "", "Optional parent cgroup for the container")
flBuildArg := opts.NewListOpts(runconfigopts.ValidateEnv)
cmd.Var(&flBuildArg, []string{"-build-arg"}, "Set build-time variables")
isolation := cmd.String([]string{"-isolation"}, "", "Container isolation technology")
flLabels := opts.NewListOpts(nil)
cmd.Var(&flLabels, []string{"-label"}, "Set metadata for an image")
ulimits := make(map[string]*units.Ulimit)
flUlimits := runconfigopts.NewUlimitOpt(&ulimits)
ulimits := make(map[string]*ulimit.Ulimit)
flUlimits := opts.NewUlimitOpt(&ulimits)
cmd.Var(flUlimits, []string{"-ulimit"}, "Ulimit options")
cmd.Require(flag.Exact, 1)
@@ -77,42 +78,34 @@ func (cli *DockerCli) CmdBuild(args ...string) error {
cmd.ParseFlags(args, true)
var (
ctx io.ReadCloser
err error
context io.ReadCloser
isRemote bool
err error
)
_, err = exec.LookPath("git")
hasGit := err == nil
specifiedContext := cmd.Arg(0)
var (
contextDir string
tempDir string
relDockerfile string
progBuff io.Writer
buildBuff io.Writer
)
progBuff = cli.out
buildBuff = cli.out
if *suppressOutput {
progBuff = bytes.NewBuffer(nil)
buildBuff = bytes.NewBuffer(nil)
}
switch {
case specifiedContext == "-":
ctx, relDockerfile, err = builder.GetContextFromReader(cli.in, *dockerfileName)
case urlutil.IsGitURL(specifiedContext):
tempDir, relDockerfile, err = builder.GetContextFromGitURL(specifiedContext, *dockerfileName)
tempDir, relDockerfile, err = getContextFromReader(cli.in, *dockerfileName)
case urlutil.IsGitURL(specifiedContext) && hasGit:
tempDir, relDockerfile, err = getContextFromGitURL(specifiedContext, *dockerfileName)
case urlutil.IsURL(specifiedContext):
ctx, relDockerfile, err = builder.GetContextFromURL(progBuff, specifiedContext, *dockerfileName)
tempDir, relDockerfile, err = getContextFromURL(cli.out, specifiedContext, *dockerfileName)
default:
contextDir, relDockerfile, err = builder.GetContextFromLocalDir(specifiedContext, *dockerfileName)
contextDir, relDockerfile, err = getContextFromLocalDir(specifiedContext, *dockerfileName)
}
if err != nil {
if *suppressOutput && urlutil.IsURL(specifiedContext) {
fmt.Fprintln(cli.err, progBuff)
}
return fmt.Errorf("unable to prepare context: %s", err)
}
@@ -121,65 +114,69 @@ func (cli *DockerCli) CmdBuild(args ...string) error {
contextDir = tempDir
}
if ctx == nil {
// And canonicalize dockerfile name to a platform-independent one
relDockerfile, err = archive.CanonicalTarNameForPath(relDockerfile)
if err != nil {
return fmt.Errorf("cannot canonicalize dockerfile path %s: %v", relDockerfile, err)
}
// Resolve the FROM lines in the Dockerfile to trusted digest references
// using Notary. On a successful build, we must tag the resolved digests
// to the original name specified in the Dockerfile.
newDockerfile, resolvedTags, err := rewriteDockerfileFrom(filepath.Join(contextDir, relDockerfile), cli.trustedReference)
if err != nil {
return fmt.Errorf("unable to process Dockerfile: %v", err)
}
defer newDockerfile.Close()
f, err := os.Open(filepath.Join(contextDir, ".dockerignore"))
if err != nil && !os.IsNotExist(err) {
return err
}
var excludes []string
if err == nil {
excludes, err = dockerignore.ReadAll(f)
if err != nil {
return err
}
}
if err := builder.ValidateContextDirectory(contextDir, excludes); err != nil {
return fmt.Errorf("Error checking context: '%s'.", err)
}
// If .dockerignore mentions .dockerignore or the Dockerfile
// then make sure we send both files over to the daemon
// because Dockerfile is, obviously, needed no matter what, and
// .dockerignore is needed to know if either one needs to be
// removed. The daemon will remove them for us, if needed, after it
// parses the Dockerfile. Ignore errors here, as they will have been
// caught by validateContextDirectory above.
var includes = []string{"."}
keepThem1, _ := fileutils.Matches(".dockerignore", excludes)
keepThem2, _ := fileutils.Matches(relDockerfile, excludes)
if keepThem1 || keepThem2 {
includes = append(includes, ".dockerignore", relDockerfile)
}
ctx, err = archive.TarWithOptions(contextDir, &archive.TarOptions{
Compression: archive.Uncompressed,
ExcludePatterns: excludes,
IncludeFiles: includes,
})
if err != nil {
return err
}
// And canonicalize dockerfile name to a platform-independent one
relDockerfile, err = archive.CanonicalTarNameForPath(relDockerfile)
if err != nil {
return fmt.Errorf("cannot canonicalize dockerfile path %s: %v", relDockerfile, err)
}
var resolvedTags []*resolvedTag
if isTrusted() {
// Wrap the tar archive to replace the Dockerfile entry with the rewritten
// Dockerfile which uses trusted pulls.
ctx = replaceDockerfileTarWrapper(ctx, relDockerfile, cli.trustedReference, &resolvedTags)
var includes = []string{"."}
excludes, err := utils.ReadDockerIgnore(path.Join(contextDir, ".dockerignore"))
if err != nil {
return err
}
if err := utils.ValidateContextDirectory(contextDir, excludes); err != nil {
return fmt.Errorf("Error checking context: '%s'.", err)
}
// If .dockerignore mentions .dockerignore or the Dockerfile
// then make sure we send both files over to the daemon
// because Dockerfile is, obviously, needed no matter what, and
// .dockerignore is needed to know if either one needs to be
// removed. The deamon will remove them for us, if needed, after it
// parses the Dockerfile. Ignore errors here, as they will have been
// caught by ValidateContextDirectory above.
keepThem1, _ := fileutils.Matches(".dockerignore", excludes)
keepThem2, _ := fileutils.Matches(relDockerfile, excludes)
if keepThem1 || keepThem2 {
includes = append(includes, ".dockerignore", relDockerfile)
}
context, err = archive.TarWithOptions(contextDir, &archive.TarOptions{
Compression: archive.Uncompressed,
ExcludePatterns: excludes,
IncludeFiles: includes,
})
if err != nil {
return err
}
// Wrap the tar archive to replace the Dockerfile entry with the rewritten
// Dockerfile which uses trusted pulls.
context = replaceDockerfileTarWrapper(context, newDockerfile, relDockerfile)
// Setup an upload progress bar
progressOutput := streamformatter.NewStreamFormatter().NewProgressOutput(progBuff, true)
var body io.Reader = progress.NewProgressReader(ctx, progressOutput, 0, "", "Sending build context to Docker daemon")
// FIXME: ProgressReader shouldn't be this annoying to use
sf := streamformatter.NewStreamFormatter()
var body io.Reader = progressreader.New(progressreader.Config{
In: context,
Out: cli.out,
Formatter: sf,
NewLines: true,
ID: "",
Action: "Sending build context to Docker daemon",
})
var memory int64
if *flMemoryString != "" {
@@ -202,153 +199,396 @@ func (cli *DockerCli) CmdBuild(args ...string) error {
memorySwap = parsedMemorySwap
}
}
// Send the build context
v := &url.Values{}
var shmSize int64
if *flShmSize != "" {
shmSize, err = units.RAMInBytes(*flShmSize)
if err != nil {
//Check if the given image name can be resolved
if *tag != "" {
repository, tag := parsers.ParseRepositoryTag(*tag)
if err := registry.ValidateRepositoryName(repository); err != nil {
return err
}
if len(tag) > 0 {
if err := tags.ValidateTagName(tag); err != nil {
return err
}
}
}
options := types.ImageBuildOptions{
Context: body,
Memory: memory,
MemorySwap: memorySwap,
Tags: flTags.GetAll(),
SuppressOutput: *suppressOutput,
NoCache: *noCache,
Remove: *rm,
ForceRemove: *forceRm,
PullParent: *pull,
Isolation: container.Isolation(*isolation),
CPUSetCPUs: *flCPUSetCpus,
CPUSetMems: *flCPUSetMems,
CPUShares: *flCPUShares,
CPUQuota: *flCPUQuota,
CPUPeriod: *flCPUPeriod,
CgroupParent: *flCgroupParent,
Dockerfile: relDockerfile,
ShmSize: shmSize,
Ulimits: flUlimits.GetList(),
BuildArgs: runconfigopts.ConvertKVStringsToMap(flBuildArg.GetAll()),
AuthConfigs: cli.retrieveAuthConfigs(),
Labels: runconfigopts.ConvertKVStringsToMap(flLabels.GetAll()),
v.Set("t", *tag)
if *suppressOutput {
v.Set("q", "1")
}
if isRemote {
v.Set("remote", cmd.Arg(0))
}
if *noCache {
v.Set("nocache", "1")
}
if *rm {
v.Set("rm", "1")
} else {
v.Set("rm", "0")
}
response, err := cli.client.ImageBuild(context.Background(), options)
if *forceRm {
v.Set("forcerm", "1")
}
if *pull {
v.Set("pull", "1")
}
v.Set("cpusetcpus", *flCPUSetCpus)
v.Set("cpusetmems", *flCPUSetMems)
v.Set("cpushares", strconv.FormatInt(*flCPUShares, 10))
v.Set("cpuquota", strconv.FormatInt(*flCpuQuota, 10))
v.Set("cpuperiod", strconv.FormatInt(*flCpuPeriod, 10))
v.Set("memory", strconv.FormatInt(memory, 10))
v.Set("memswap", strconv.FormatInt(memorySwap, 10))
v.Set("cgroupparent", *flCgroupParent)
v.Set("dockerfile", relDockerfile)
ulimitsVar := flUlimits.GetList()
ulimitsJson, err := json.Marshal(ulimitsVar)
if err != nil {
return err
}
defer response.Body.Close()
v.Set("ulimits", string(ulimitsJson))
err = jsonmessage.DisplayJSONMessagesStream(response.Body, buildBuff, cli.outFd, cli.isTerminalOut, nil)
headers := http.Header(make(map[string][]string))
buf, err := json.Marshal(cli.configFile.AuthConfigs)
if err != nil {
if jerr, ok := err.(*jsonmessage.JSONError); ok {
// If no error code is set, default to 1
if jerr.Code == 0 {
jerr.Code = 1
return err
}
headers.Add("X-Registry-Config", base64.URLEncoding.EncodeToString(buf))
headers.Set("Content-Type", "application/tar")
sopts := &streamOpts{
rawTerminal: true,
in: body,
out: cli.out,
headers: headers,
}
serverResp, err := cli.stream("POST", fmt.Sprintf("/build?%s", v.Encode()), sopts)
// Windows: show error message about modified file permissions.
if runtime.GOOS == "windows" {
h, err := httputils.ParseServerHeader(serverResp.header.Get("Server"))
if err == nil {
if h.OS != "windows" {
fmt.Fprintln(cli.err, `SECURITY WARNING: You are building a Docker image from Windows against a non-Windows Docker host. All files and directories added to build context will have '-rwxr-xr-x' permissions. It is recommended to double check and reset permissions for sensitive files and directories.`)
}
if *suppressOutput {
fmt.Fprintf(cli.err, "%s%s", progBuff, buildBuff)
}
return Cli.StatusError{Status: jerr.Message, StatusCode: jerr.Code}
}
}
// Windows: show error message about modified file permissions if the
// daemon isn't running Windows.
if response.OSType != "windows" && runtime.GOOS == "windows" {
fmt.Fprintln(cli.err, `SECURITY WARNING: You are building a Docker image from Windows against a non-Windows Docker host. All files and directories added to build context will have '-rwxr-xr-x' permissions. It is recommended to double check and reset permissions for sensitive files and directories.`)
if jerr, ok := err.(*jsonmessage.JSONError); ok {
// If no error code is set, default to 1
if jerr.Code == 0 {
jerr.Code = 1
}
return Cli.StatusError{Status: jerr.Message, StatusCode: jerr.Code}
}
// Everything worked so if -q was provided the output from the daemon
// should be just the image ID and we'll print that to stdout.
if *suppressOutput {
fmt.Fprintf(cli.out, "%s", buildBuff)
if err != nil {
return err
}
if isTrusted() {
// Since the build was successful, now we must tag any of the resolved
// images from the above Dockerfile rewrite.
for _, resolved := range resolvedTags {
if err := cli.tagTrusted(resolved.digestRef, resolved.tagRef); err != nil {
return err
}
// Since the build was successful, now we must tag any of the resolved
// images from the above Dockerfile rewrite.
for _, resolved := range resolvedTags {
if err := cli.tagTrusted(resolved.repoInfo, resolved.digestRef, resolved.tagRef); err != nil {
return err
}
}
return nil
}
// validateTag checks if the given image name can be resolved.
func validateTag(rawRepo string) (string, error) {
_, err := reference.ParseNamed(rawRepo)
if err != nil {
return "", err
// getDockerfileRelPath uses the given context directory for a `docker build`
// and returns the absolute path to the context directory, the relative path of
// the dockerfile in that context directory, and a non-nil error on success.
func getDockerfileRelPath(givenContextDir, givenDockerfile string) (absContextDir, relDockerfile string, err error) {
if absContextDir, err = filepath.Abs(givenContextDir); err != nil {
return "", "", fmt.Errorf("unable to get absolute context directory: %v", err)
}
return rawRepo, nil
// The context dir might be a symbolic link, so follow it to the actual
// target directory.
absContextDir, err = filepath.EvalSymlinks(absContextDir)
if err != nil {
return "", "", fmt.Errorf("unable to evaluate symlinks in context path: %v", err)
}
stat, err := os.Lstat(absContextDir)
if err != nil {
return "", "", fmt.Errorf("unable to stat context directory %q: %v", absContextDir, err)
}
if !stat.IsDir() {
return "", "", fmt.Errorf("context must be a directory: %s", absContextDir)
}
absDockerfile := givenDockerfile
if absDockerfile == "" {
// No -f/--file was specified so use the default relative to the
// context directory.
absDockerfile = filepath.Join(absContextDir, api.DefaultDockerfileName)
// Just to be nice ;-) look for 'dockerfile' too but only
// use it if we found it, otherwise ignore this check
if _, err = os.Lstat(absDockerfile); os.IsNotExist(err) {
altPath := filepath.Join(absContextDir, strings.ToLower(api.DefaultDockerfileName))
if _, err = os.Lstat(altPath); err == nil {
absDockerfile = altPath
}
}
}
// If not already an absolute path, the Dockerfile path should be joined to
// the base directory.
if !filepath.IsAbs(absDockerfile) {
absDockerfile = filepath.Join(absContextDir, absDockerfile)
}
// Verify that 'filename' is within the build context
absDockerfile, err = symlink.FollowSymlinkInScope(absDockerfile, absContextDir)
if err != nil {
return "", "", fmt.Errorf("The Dockerfile (%s) must be within the build context (%s)", givenDockerfile, givenContextDir)
}
if _, err := os.Lstat(absDockerfile); err != nil {
if os.IsNotExist(err) {
return "", "", fmt.Errorf("Cannot locate Dockerfile: absDockerfile: %q", absDockerfile)
}
return "", "", fmt.Errorf("unable to stat Dockerfile: %v", err)
}
if relDockerfile, err = filepath.Rel(absContextDir, absDockerfile); err != nil {
return "", "", fmt.Errorf("unable to get relative Dockerfile path: %v", err)
}
return absContextDir, relDockerfile, nil
}
// writeToFile copies from the given reader and writes it to a file with the
// given filename.
func writeToFile(r io.Reader, filename string) error {
file, err := os.OpenFile(filename, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, os.FileMode(0600))
if err != nil {
return fmt.Errorf("unable to create file: %v", err)
}
defer file.Close()
if _, err := io.Copy(file, r); err != nil {
return fmt.Errorf("unable to write file: %v", err)
}
return nil
}
// getContextFromReader will read the contents of the given reader as either a
// Dockerfile or tar archive to be extracted to a temporary directory used as
// the context directory. Returns the absolute path to the temporary context
// directory, the relative path of the dockerfile in that context directory,
// and a non-nil error on success.
func getContextFromReader(r io.Reader, dockerfileName string) (absContextDir, relDockerfile string, err error) {
buf := bufio.NewReader(r)
magic, err := buf.Peek(tarHeaderSize)
if err != nil && err != io.EOF {
return "", "", fmt.Errorf("failed to peek context header from STDIN: %v", err)
}
if absContextDir, err = ioutil.TempDir("", "docker-build-context-"); err != nil {
return "", "", fmt.Errorf("unbale to create temporary context directory: %v", err)
}
defer func(d string) {
if err != nil {
os.RemoveAll(d)
}
}(absContextDir)
if !archive.IsArchive(magic) { // Input should be read as a Dockerfile.
// -f option has no meaning when we're reading it from stdin,
// so just use our default Dockerfile name
relDockerfile = api.DefaultDockerfileName
return absContextDir, relDockerfile, writeToFile(buf, filepath.Join(absContextDir, relDockerfile))
}
if err := archive.Untar(buf, absContextDir, nil); err != nil {
return "", "", fmt.Errorf("unable to extract stdin to temporary context direcotry: %v", err)
}
return getDockerfileRelPath(absContextDir, dockerfileName)
}
// getContextFromGitURL uses a Git URL as context for a `docker build`. The
// git repo is cloned into a temporary directory used as the context directory.
// Returns the absolute path to the temporary context directory, the relative
// path of the dockerfile in that context directory, and a non-nil error on
// success.
func getContextFromGitURL(gitURL, dockerfileName string) (absContextDir, relDockerfile string, err error) {
if absContextDir, err = utils.GitClone(gitURL); err != nil {
return "", "", fmt.Errorf("unable to 'git clone' to temporary context directory: %v", err)
}
return getDockerfileRelPath(absContextDir, dockerfileName)
}
// getContextFromURL uses a remote URL as context for a `docker build`. The
// remote resource is downloaded as either a Dockerfile or a context tar
// archive and stored in a temporary directory used as the context directory.
// Returns the absolute path to the temporary context directory, the relative
// path of the dockerfile in that context directory, and a non-nil error on
// success.
func getContextFromURL(out io.Writer, remoteURL, dockerfileName string) (absContextDir, relDockerfile string, err error) {
response, err := httputils.Download(remoteURL)
if err != nil {
return "", "", fmt.Errorf("unable to download remote context %s: %v", remoteURL, err)
}
defer response.Body.Close()
// Pass the response body through a progress reader.
progReader := &progressreader.Config{
In: response.Body,
Out: out,
Formatter: streamformatter.NewStreamFormatter(),
Size: int(response.ContentLength),
NewLines: true,
ID: "",
Action: fmt.Sprintf("Downloading build context from remote url: %s", remoteURL),
}
return getContextFromReader(progReader, dockerfileName)
}
// getContextFromLocalDir uses the given local directory as context for a
// `docker build`. Returns the absolute path to the local context directory,
// the relative path of the dockerfile in that context directory, and a non-nil
// error on success.
func getContextFromLocalDir(localDir, dockerfileName string) (absContextDir, relDockerfile string, err error) {
// When using a local context directory, when the Dockerfile is specified
// with the `-f/--file` option then it is considered relative to the
// current directory and not the context directory.
if dockerfileName != "" {
if dockerfileName, err = filepath.Abs(dockerfileName); err != nil {
return "", "", fmt.Errorf("unable to get absolute path to Dockerfile: %v", err)
}
}
return getDockerfileRelPath(localDir, dockerfileName)
}
var dockerfileFromLinePattern = regexp.MustCompile(`(?i)^[\s]*FROM[ \f\r\t\v]+(?P<image>[^ \f\r\t\v\n#]+)`)
type trustedDockerfile struct {
*os.File
size int64
}
func (td *trustedDockerfile) Close() error {
td.File.Close()
return os.Remove(td.File.Name())
}
// resolvedTag records the repository, tag, and resolved digest reference
// from a Dockerfile rewrite.
type resolvedTag struct {
digestRef reference.Canonical
tagRef reference.NamedTagged
repoInfo *registry.RepositoryInfo
digestRef, tagRef registry.Reference
}
// rewriteDockerfileFrom rewrites the given Dockerfile by resolving images in
// "FROM <image>" instructions to a digest reference. `translator` is a
// function that takes a repository name and tag reference and returns a
// trusted digest reference.
func rewriteDockerfileFrom(dockerfile io.Reader, translator translatorFunc) (newDockerfile []byte, resolvedTags []*resolvedTag, err error) {
func rewriteDockerfileFrom(dockerfileName string, translator func(string, registry.Reference) (registry.Reference, error)) (newDockerfile *trustedDockerfile, resolvedTags []*resolvedTag, err error) {
dockerfile, err := os.Open(dockerfileName)
if err != nil {
return nil, nil, fmt.Errorf("unable to open Dockerfile: %v", err)
}
defer dockerfile.Close()
scanner := bufio.NewScanner(dockerfile)
buf := bytes.NewBuffer(nil)
// Make a tempfile to store the rewritten Dockerfile.
tempFile, err := ioutil.TempFile("", "trusted-dockerfile-")
if err != nil {
return nil, nil, fmt.Errorf("unable to make temporary trusted Dockerfile: %v", err)
}
trustedFile := &trustedDockerfile{
File: tempFile,
}
defer func() {
if err != nil {
// Close the tempfile if there was an error during Notary lookups.
// Otherwise the caller should close it.
trustedFile.Close()
}
}()
// Scan the lines of the Dockerfile, looking for a "FROM" line.
for scanner.Scan() {
line := scanner.Text()
matches := dockerfileFromLinePattern.FindStringSubmatch(line)
if matches != nil && matches[1] != api.NoBaseImageSpecifier {
if matches != nil && matches[1] != "scratch" {
// Replace the line with a resolved "FROM repo@digest"
ref, err := reference.ParseNamed(matches[1])
if err != nil {
return nil, nil, err
repo, tag := parsers.ParseRepositoryTag(matches[1])
if tag == "" {
tag = tags.DEFAULTTAG
}
ref = reference.WithDefaultTag(ref)
if ref, ok := ref.(reference.NamedTagged); ok && isTrusted() {
trustedRef, err := translator(ref)
repoInfo, err := registry.ParseRepositoryInfo(repo)
if err != nil {
return nil, nil, fmt.Errorf("unable to parse repository info: %v", err)
}
ref := registry.ParseReference(tag)
if !ref.HasDigest() && isTrusted() {
trustedRef, err := translator(repo, ref)
if err != nil {
return nil, nil, err
}
line = dockerfileFromLinePattern.ReplaceAllLiteralString(line, fmt.Sprintf("FROM %s", trustedRef.String()))
line = dockerfileFromLinePattern.ReplaceAllLiteralString(line, fmt.Sprintf("FROM %s", trustedRef.ImageName(repo)))
resolvedTags = append(resolvedTags, &resolvedTag{
repoInfo: repoInfo,
digestRef: trustedRef,
tagRef: ref,
})
}
}
_, err := fmt.Fprintln(buf, line)
n, err := fmt.Fprintln(tempFile, line)
if err != nil {
return nil, nil, err
}
trustedFile.size += int64(n)
}
return buf.Bytes(), resolvedTags, scanner.Err()
tempFile.Seek(0, os.SEEK_SET)
return trustedFile, resolvedTags, scanner.Err()
}
// replaceDockerfileTarWrapper wraps the given input tar archive stream and
// replaces the entry with the given Dockerfile name with the contents of the
// new Dockerfile. Returns a new tar archive stream with the replaced
// Dockerfile.
func replaceDockerfileTarWrapper(inputTarStream io.ReadCloser, dockerfileName string, translator translatorFunc, resolvedTags *[]*resolvedTag) io.ReadCloser {
func replaceDockerfileTarWrapper(inputTarStream io.ReadCloser, newDockerfile *trustedDockerfile, dockerfileName string) io.ReadCloser {
pipeReader, pipeWriter := io.Pipe()
go func() {
tarReader := tar.NewReader(inputTarStream)
tarWriter := tar.NewWriter(pipeWriter)
@@ -369,18 +609,13 @@ func replaceDockerfileTarWrapper(inputTarStream io.ReadCloser, dockerfileName st
}
var content io.Reader = tarReader
if hdr.Name == dockerfileName {
// This entry is the Dockerfile. Since the tar archive was
// generated from a directory on the local filesystem, the
// Dockerfile will only appear once in the archive.
var newDockerfile []byte
newDockerfile, *resolvedTags, err = rewriteDockerfileFrom(content, translator)
if err != nil {
pipeWriter.CloseWithError(err)
return
}
hdr.Size = int64(len(newDockerfile))
content = bytes.NewBuffer(newDockerfile)
hdr.Size = newDockerfile.size
content = newDockerfile
}
if err := tarWriter.WriteHeader(hdr); err != nil {

View File

@@ -1,23 +1,21 @@
package client
import (
"crypto/tls"
"errors"
"fmt"
"io"
"net/http"
"net/url"
"os"
"runtime"
"strings"
"github.com/docker/docker/api"
"github.com/docker/docker/cli"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/cliconfig/credentials"
"github.com/docker/docker/dockerversion"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/sockets"
"github.com/docker/docker/pkg/term"
"github.com/docker/engine-api/client"
"github.com/docker/go-connections/sockets"
"github.com/docker/go-connections/tlsconfig"
"github.com/docker/docker/pkg/tlsconfig"
)
// DockerCli represents the docker command line client.
@@ -26,6 +24,13 @@ type DockerCli struct {
// initializing closure
init func() error
// proto holds the client protocol i.e. unix.
proto string
// addr holds the client address.
addr string
// basePath holds the path to prepend to the requests
basePath string
// configFile has the client configuration file
configFile *cliconfig.ConfigFile
// in holds the input stream and closer (io.ReadCloser) for the client.
@@ -36,22 +41,23 @@ type DockerCli struct {
err io.Writer
// keyFile holds the key file as a string.
keyFile string
// tlsConfig holds the TLS configuration for the client, and will
// set the scheme to https in NewDockerCli if present.
tlsConfig *tls.Config
// scheme holds the scheme of the client i.e. https.
scheme string
// inFd holds the file descriptor of the client's STDIN (if valid).
inFd uintptr
// outFd holds file descriptor of the client's STDOUT (if valid).
outFd uintptr
// isTerminalIn indicates whether the client's STDIN is a TTY
isTerminalIn bool
// isTerminalOut indicates whether the client's STDOUT is a TTY
// isTerminalOut dindicates whether the client's STDOUT is a TTY
isTerminalOut bool
// client is the http client that performs all API operations
client client.APIClient
// state holds the terminal state
state *term.State
// transport holds the client transport instance.
transport *http.Transport
}
// Initialize calls the init function that will setup the configuration for the client
// such as the TLS, tcp and other parameters used to run the client.
func (cli *DockerCli) Initialize() error {
if cli.init == nil {
return nil
@@ -71,43 +77,10 @@ func (cli *DockerCli) CheckTtyInput(attachStdin, ttyMode bool) error {
return nil
}
// PsFormat returns the format string specified in the configuration.
// String contains columns and format specification, for example {{ID}}\t{{Name}}.
func (cli *DockerCli) PsFormat() string {
return cli.configFile.PsFormat
}
// ImagesFormat returns the format string specified in the configuration.
// String contains columns and format specification, for example {{ID}}\t{{Name}}.
func (cli *DockerCli) ImagesFormat() string {
return cli.configFile.ImagesFormat
}
func (cli *DockerCli) setRawTerminal() error {
if cli.isTerminalIn && os.Getenv("NORAW") == "" {
state, err := term.SetRawTerminal(cli.inFd)
if err != nil {
return err
}
cli.state = state
}
return nil
}
func (cli *DockerCli) restoreTerminal(in io.Closer) error {
if cli.state != nil {
term.RestoreTerminal(cli.inFd, cli.state)
}
// WARNING: DO NOT REMOVE THE OS CHECK !!!
// For some reason this Close call blocks on darwin..
// As the client exists right after, simply discard the close
// until we find a better solution.
if in != nil && runtime.GOOS != "darwin" {
return in.Close()
}
return nil
}
// NewDockerCli returns a DockerCli instance with IO output and error streams set by in, out and err.
// The key file, protocol (i.e. unix) and address are passed in as strings, along with the tls.Config. If the tls.Config
// is set the client scheme will be set to https.
@@ -122,42 +95,47 @@ func NewDockerCli(in io.ReadCloser, out, err io.Writer, clientFlags *cli.ClientF
cli.init = func() error {
clientFlags.PostParse()
configFile, e := cliconfig.Load(cliconfig.ConfigDir())
if e != nil {
fmt.Fprintf(cli.err, "WARNING: Error loading config file:%v\n", e)
}
if !configFile.ContainsAuth() {
credentials.DetectDefaultStore(configFile)
}
cli.configFile = configFile
host, err := getServerHost(clientFlags.Common.Hosts, clientFlags.Common.TLSOptions)
if err != nil {
return err
hosts := clientFlags.Common.Hosts
switch len(hosts) {
case 0:
defaultHost := os.Getenv("DOCKER_HOST")
if defaultHost == "" {
defaultHost = opts.DefaultHost
}
defaultHost, err := opts.ValidateHost(defaultHost)
if err != nil {
return err
}
hosts = []string{defaultHost}
case 1:
// only accept one host to talk to
default:
return errors.New("Please specify only one -H")
}
customHeaders := cli.configFile.HTTPHeaders
if customHeaders == nil {
customHeaders = map[string]string{}
}
customHeaders["User-Agent"] = clientUserAgent()
protoAddrParts := strings.SplitN(hosts[0], "://", 2)
cli.proto, cli.addr = protoAddrParts[0], protoAddrParts[1]
verStr := api.DefaultVersion.String()
if tmpStr := os.Getenv("DOCKER_API_VERSION"); tmpStr != "" {
verStr = tmpStr
if cli.proto == "tcp" {
// error is checked in pkg/parsers already
parsed, _ := url.Parse("tcp://" + cli.addr)
cli.addr = parsed.Host
cli.basePath = parsed.Path
}
httpClient, err := newHTTPClient(host, clientFlags.Common.TLSOptions)
if err != nil {
return err
if clientFlags.Common.TLSOptions != nil {
cli.scheme = "https"
var e error
cli.tlsConfig, e = tlsconfig.Client(*clientFlags.Common.TLSOptions)
if e != nil {
return e
}
} else {
cli.scheme = "http"
}
client, err := client.NewClient(host, verStr, httpClient, customHeaders)
if err != nil {
return err
}
cli.client = client
if cli.in != nil {
cli.inFd, cli.isTerminalIn = term.GetFdInfo(cli.in)
}
@@ -165,51 +143,20 @@ func NewDockerCli(in io.ReadCloser, out, err io.Writer, clientFlags *cli.ClientF
cli.outFd, cli.isTerminalOut = term.GetFdInfo(cli.out)
}
// The transport is created here for reuse during the client session.
cli.transport = &http.Transport{
TLSClientConfig: cli.tlsConfig,
}
sockets.ConfigureTCPTransport(cli.transport, cli.proto, cli.addr)
configFile, e := cliconfig.Load(cliconfig.ConfigDir())
if e != nil {
fmt.Fprintf(cli.err, "WARNING: Error loading config file:%v\n", e)
}
cli.configFile = configFile
return nil
}
return cli
}
func getServerHost(hosts []string, tlsOptions *tlsconfig.Options) (host string, err error) {
switch len(hosts) {
case 0:
host = os.Getenv("DOCKER_HOST")
case 1:
host = hosts[0]
default:
return "", errors.New("Please specify only one -H")
}
host, err = opts.ParseHost(tlsOptions != nil, host)
return
}
func newHTTPClient(host string, tlsOptions *tlsconfig.Options) (*http.Client, error) {
if tlsOptions == nil {
// let the api client configure the default transport.
return nil, nil
}
config, err := tlsconfig.Client(*tlsOptions)
if err != nil {
return nil, err
}
tr := &http.Transport{
TLSClientConfig: config,
}
proto, addr, _, err := client.ParseHost(host)
if err != nil {
return nil, err
}
sockets.ConfigureTransport(tr, proto, addr)
return &http.Client{
Transport: tr,
}, nil
}
func clientUserAgent() string {
return "Docker-Client/" + dockerversion.Version + " (" + runtime.GOOS + ")"
}

View File

@@ -2,81 +2,80 @@ package client
import (
"encoding/json"
"errors"
"fmt"
"net/url"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/reference"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/container"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
"github.com/docker/docker/runconfig"
)
// CmdCommit creates a new image from a container's changes.
//
// Usage: docker commit [OPTIONS] CONTAINER [REPOSITORY[:TAG]]
func (cli *DockerCli) CmdCommit(args ...string) error {
cmd := Cli.Subcmd("commit", []string{"CONTAINER [REPOSITORY[:TAG]]"}, Cli.DockerCommands["commit"].Description, true)
cmd := Cli.Subcmd("commit", []string{"CONTAINER [REPOSITORY[:TAG]]"}, "Create a new image from a container's changes", true)
flPause := cmd.Bool([]string{"p", "-pause"}, true, "Pause container during commit")
flComment := cmd.String([]string{"m", "-message"}, "", "Commit message")
flAuthor := cmd.String([]string{"a", "-author"}, "", "Author (e.g., \"John Hannibal Smith <hannibal@a-team.com>\")")
flAuthor := cmd.String([]string{"a", "#author", "-author"}, "", "Author (e.g., \"John Hannibal Smith <hannibal@a-team.com>\")")
flChanges := opts.NewListOpts(nil)
cmd.Var(&flChanges, []string{"c", "-change"}, "Apply Dockerfile instruction to the created image")
// FIXME: --run is deprecated, it will be replaced with inline Dockerfile commands.
flConfig := cmd.String([]string{"#-run"}, "", "This option is deprecated and will be removed in a future version in favor of inline Dockerfile-compatible commands")
flConfig := cmd.String([]string{"#run", "#-run"}, "", "This option is deprecated and will be removed in a future version in favor of inline Dockerfile-compatible commands")
cmd.Require(flag.Max, 2)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var (
name = cmd.Arg(0)
repositoryAndTag = cmd.Arg(1)
repositoryName string
tag string
name = cmd.Arg(0)
repository, tag = parsers.ParseRepositoryTag(cmd.Arg(1))
)
//Check if the given image name can be resolved
if repositoryAndTag != "" {
ref, err := reference.ParseNamed(repositoryAndTag)
if err != nil {
if repository != "" {
if err := registry.ValidateRepositoryName(repository); err != nil {
return err
}
repositoryName = ref.Name()
switch x := ref.(type) {
case reference.Canonical:
return errors.New("cannot commit to digest reference")
case reference.NamedTagged:
tag = x.Tag()
}
}
var config *container.Config
v := url.Values{}
v.Set("container", name)
v.Set("repo", repository)
v.Set("tag", tag)
v.Set("comment", *flComment)
v.Set("author", *flAuthor)
for _, change := range flChanges.GetAll() {
v.Add("changes", change)
}
if *flPause != true {
v.Set("pause", "0")
}
var (
config *runconfig.Config
response types.ContainerCommitResponse
)
if *flConfig != "" {
config = &container.Config{}
config = &runconfig.Config{}
if err := json.Unmarshal([]byte(*flConfig), config); err != nil {
return err
}
}
options := types.ContainerCommitOptions{
ContainerID: name,
RepositoryName: repositoryName,
Tag: tag,
Comment: *flComment,
Author: *flAuthor,
Changes: flChanges.GetAll(),
Pause: *flPause,
Config: config,
serverResp, err := cli.call("POST", "/commit?"+v.Encode(), config, nil)
if err != nil {
return err
}
response, err := cli.client.ContainerCommit(context.Background(), options)
if err != nil {
defer serverResp.body.Close()
if err := json.NewDecoder(serverResp.body).Decode(&response); err != nil {
return err
}

View File

@@ -1,19 +1,20 @@
package client
import (
"encoding/base64"
"encoding/json"
"fmt"
"io"
"net/http"
"net/url"
"os"
"path/filepath"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/archive"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/system"
"github.com/docker/engine-api/types"
)
type copyDirection int
@@ -24,29 +25,25 @@ const (
acrossContainers = fromContainer | toContainer
)
type cpConfig struct {
followLink bool
}
// CmdCp copies files/folders to or from a path in a container.
//
// When copying from a container, if DEST_PATH is '-' the data is written as a
// When copying from a container, if LOCALPATH is '-' the data is written as a
// tar archive file to STDOUT.
//
// When copying to a container, if SRC_PATH is '-' the data is read as a tar
// archive file from STDIN, and the destination CONTAINER:DEST_PATH, must specify
// When copying to a container, if LOCALPATH is '-' the data is read as a tar
// archive file from STDIN, and the destination CONTAINER:PATH, must specify
// a directory.
//
// Usage:
// docker cp CONTAINER:SRC_PATH DEST_PATH|-
// docker cp SRC_PATH|- CONTAINER:DEST_PATH
// docker cp CONTAINER:PATH LOCALPATH|-
// docker cp LOCALPATH|- CONTAINER:PATH
func (cli *DockerCli) CmdCp(args ...string) error {
cmd := Cli.Subcmd(
"cp",
[]string{"CONTAINER:SRC_PATH DEST_PATH|-", "SRC_PATH|- CONTAINER:DEST_PATH"},
[]string{"CONTAINER:PATH LOCALPATH|-", "LOCALPATH|- CONTAINER:PATH"},
strings.Join([]string{
Cli.DockerCommands["cp"].Description,
"\nUse '-' as the source to read a tar archive from stdin\n",
"Copy files/folders between a container and your host.\n",
"Use '-' as the source to read a tar archive from stdin\n",
"and extract it to a directory destination in a container.\n",
"Use '-' as the destination to stream a tar archive of a\n",
"container source to stdout.",
@@ -54,8 +51,6 @@ func (cli *DockerCli) CmdCp(args ...string) error {
true,
)
followLink := cmd.Bool([]string{"L", "-follow-link"}, false, "Always follow symbol link in SRC_PATH")
cmd.Require(flag.Exact, 2)
cmd.ParseFlags(args, true)
@@ -77,15 +72,11 @@ func (cli *DockerCli) CmdCp(args ...string) error {
direction |= toContainer
}
cpParam := &cpConfig{
followLink: *followLink,
}
switch direction {
case fromContainer:
return cli.copyFromContainer(srcContainer, srcPath, dstPath, cpParam)
return cli.copyFromContainer(srcContainer, srcPath, dstPath)
case toContainer:
return cli.copyToContainer(srcPath, dstContainer, dstPath, cpParam)
return cli.copyToContainer(srcPath, dstContainer, dstPath)
case acrossContainers:
// Copying between containers isn't supported.
return fmt.Errorf("copying between containers is not supported")
@@ -110,7 +101,7 @@ func (cli *DockerCli) CmdCp(args ...string) error {
// client, a `:` could be part of an absolute Windows path, in which case it
// is immediately proceeded by a backslash.
func splitCpArg(arg string) (container, path string) {
if system.IsAbs(arg) {
if filepath.IsAbs(arg) {
// Explicit local absolute path, e.g., `C:\foo` or `/foo`.
return "", arg
}
@@ -127,7 +118,38 @@ func splitCpArg(arg string) (container, path string) {
}
func (cli *DockerCli) statContainerPath(containerName, path string) (types.ContainerPathStat, error) {
return cli.client.ContainerStatPath(context.Background(), containerName, path)
var stat types.ContainerPathStat
query := make(url.Values, 1)
query.Set("path", filepath.ToSlash(path)) // Normalize the paths used in the API.
urlStr := fmt.Sprintf("/containers/%s/archive?%s", containerName, query.Encode())
response, err := cli.call("HEAD", urlStr, nil, nil)
if err != nil {
return stat, err
}
defer response.body.Close()
if response.statusCode != http.StatusOK {
return stat, fmt.Errorf("unexpected status code from daemon: %d", response.statusCode)
}
return getContainerPathStatFromHeader(response.header)
}
func getContainerPathStatFromHeader(header http.Header) (types.ContainerPathStat, error) {
var stat types.ContainerPathStat
encodedStat := header.Get("X-Docker-Container-Path-Stat")
statDecoder := base64.NewDecoder(base64.StdEncoding, strings.NewReader(encodedStat))
err := json.NewDecoder(statDecoder).Decode(&stat)
if err != nil {
err = fmt.Errorf("unable to decode container path stat header: %s", err)
}
return stat, err
}
func resolveLocalPath(localPath string) (absPath string, err error) {
@@ -138,7 +160,7 @@ func resolveLocalPath(localPath string) (absPath string, err error) {
return archive.PreserveTrailingDotOrSeparator(absPath, localPath), nil
}
func (cli *DockerCli) copyFromContainer(srcContainer, srcPath, dstPath string, cpParam *cpConfig) (err error) {
func (cli *DockerCli) copyFromContainer(srcContainer, srcPath, dstPath string) (err error) {
if dstPath != "-" {
// Get an absolute destination path.
dstPath, err = resolveLocalPath(dstPath)
@@ -147,59 +169,53 @@ func (cli *DockerCli) copyFromContainer(srcContainer, srcPath, dstPath string, c
}
}
// if client requests to follow symbol link, then must decide target file to be copied
var rebaseName string
if cpParam.followLink {
srcStat, err := cli.statContainerPath(srcContainer, srcPath)
query := make(url.Values, 1)
query.Set("path", filepath.ToSlash(srcPath)) // Normalize the paths used in the API.
// If the destination is a symbolic link, we should follow it.
if err == nil && srcStat.Mode&os.ModeSymlink != 0 {
linkTarget := srcStat.LinkTarget
if !system.IsAbs(linkTarget) {
// Join with the parent directory.
srcParent, _ := archive.SplitPathDirEntry(srcPath)
linkTarget = filepath.Join(srcParent, linkTarget)
}
urlStr := fmt.Sprintf("/containers/%s/archive?%s", srcContainer, query.Encode())
linkTarget, rebaseName = archive.GetRebaseName(srcPath, linkTarget)
srcPath = linkTarget
}
}
content, stat, err := cli.client.CopyFromContainer(context.Background(), srcContainer, srcPath)
response, err := cli.call("GET", urlStr, nil, nil)
if err != nil {
return err
}
defer content.Close()
defer response.body.Close()
if response.statusCode != http.StatusOK {
return fmt.Errorf("unexpected status code from daemon: %d", response.statusCode)
}
if dstPath == "-" {
// Send the response to STDOUT.
_, err = io.Copy(os.Stdout, content)
_, err = io.Copy(os.Stdout, response.body)
return err
}
// In order to get the copy behavior right, we need to know information
// about both the source and the destination. The response headers include
// stat info about the source that we can use in deciding exactly how to
// copy it locally. Along with the stat info about the local destination,
// we have everything we need to handle the multiple possibilities there
// can be when copying a file/dir from one location to another file/dir.
stat, err := getContainerPathStatFromHeader(response.header)
if err != nil {
return fmt.Errorf("unable to get resource stat from response: %s", err)
}
// Prepare source copy info.
srcInfo := archive.CopyInfo{
Path: srcPath,
Exists: true,
IsDir: stat.Mode.IsDir(),
RebaseName: rebaseName,
Path: srcPath,
Exists: true,
IsDir: stat.Mode.IsDir(),
}
preArchive := content
if len(srcInfo.RebaseName) != 0 {
_, srcBase := archive.SplitPathDirEntry(srcInfo.Path)
preArchive = archive.RebaseArchiveEntries(content, srcBase, srcInfo.RebaseName)
}
// See comments in the implementation of `archive.CopyTo` for exactly what
// goes into deciding how and whether the source archive needs to be
// altered for the correct copy behavior.
return archive.CopyTo(preArchive, srcInfo, dstPath)
return archive.CopyTo(response.body, srcInfo, dstPath)
}
func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpParam *cpConfig) (err error) {
func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string) (err error) {
if srcPath != "-" {
// Get an absolute source path.
srcPath, err = resolveLocalPath(srcPath)
@@ -220,7 +236,7 @@ func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpP
// If the destination is a symbolic link, we should evaluate it.
if err == nil && dstStat.Mode&os.ModeSymlink != 0 {
linkTarget := dstStat.LinkTarget
if !system.IsAbs(linkTarget) {
if !filepath.IsAbs(linkTarget) {
// Join with the parent directory.
dstParent, _ := archive.SplitPathDirEntry(dstPath)
linkTarget = filepath.Join(dstParent, linkTarget)
@@ -233,7 +249,7 @@ func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpP
// Ignore any error and assume that the parent directory of the destination
// path exists, in which case the copy may still succeed. If there is any
// type of conflict (e.g., non-directory overwriting an existing directory
// or vice versa) the extraction will fail. If the destination simply did
// or vice versia) the extraction will fail. If the destination simply did
// not exist, but the parent directory does, the extraction will still
// succeed.
if err == nil {
@@ -254,7 +270,7 @@ func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpP
}
} else {
// Prepare source copy info.
srcInfo, err := archive.CopyInfoSourcePath(srcPath, cpParam.followLink)
srcInfo, err := archive.CopyInfoSourcePath(srcPath)
if err != nil {
return err
}
@@ -268,7 +284,7 @@ func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpP
// With the stat info about the local source as well as the
// destination, we have enough information to know whether we need to
// alter the archive that we upload so that when the server extracts
// it to the specified directory in the container we get the desired
// it to the specified directory in the container we get the disired
// copy behavior.
// See comments in the implementation of `archive.PrepareArchiveCopy`
@@ -287,12 +303,22 @@ func (cli *DockerCli) copyToContainer(srcPath, dstContainer, dstPath string, cpP
content = preparedArchive
}
options := types.CopyToContainerOptions{
ContainerID: dstContainer,
Path: resolvedDstPath,
Content: content,
AllowOverwriteDirWithFile: false,
query := make(url.Values, 2)
query.Set("path", filepath.ToSlash(resolvedDstPath)) // Normalize the paths used in the API.
// Do not allow for an existing directory to be overwritten by a non-directory and vice versa.
query.Set("noOverwriteDirNonDir", "true")
urlStr := fmt.Sprintf("/containers/%s/archive?%s", dstContainer, query.Encode())
response, err := cli.stream("PUT", urlStr, &streamOpts{in: content})
if err != nil {
return err
}
defer response.body.Close()
if response.statusCode != http.StatusOK {
return fmt.Errorf("unexpected status code from daemon: %d", response.statusCode)
}
return cli.client.CopyToContainer(context.Background(), options)
return nil
}

View File

@@ -1,21 +1,20 @@
package client
import (
"encoding/base64"
"encoding/json"
"fmt"
"io"
"net/url"
"os"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/jsonmessage"
"github.com/docker/docker/reference"
"github.com/docker/docker/graph/tags"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/container"
networktypes "github.com/docker/engine-api/types/network"
"github.com/docker/docker/runconfig"
)
func (cli *DockerCli) pullImage(image string) error {
@@ -23,44 +22,40 @@ func (cli *DockerCli) pullImage(image string) error {
}
func (cli *DockerCli) pullImageCustomOut(image string, out io.Writer) error {
ref, err := reference.ParseNamed(image)
if err != nil {
return err
}
var tag string
switch x := reference.WithDefaultTag(ref).(type) {
case reference.Canonical:
tag = x.Digest().String()
case reference.NamedTagged:
tag = x.Tag()
v := url.Values{}
repos, tag := parsers.ParseRepositoryTag(image)
// pull only the image tagged 'latest' if no tag was specified
if tag == "" {
tag = tags.DEFAULTTAG
}
v.Set("fromImage", repos)
v.Set("tag", tag)
// Resolve the Repository name from fqn to RepositoryInfo
repoInfo, err := registry.ParseRepositoryInfo(ref)
repoInfo, err := registry.ParseRepositoryInfo(repos)
if err != nil {
return err
}
authConfig := cli.resolveAuthConfig(repoInfo.Index)
encodedAuth, err := encodeAuthToBase64(authConfig)
// Resolve the Auth config relevant for this server
authConfig := registry.ResolveAuthConfig(cli.configFile, repoInfo.Index)
buf, err := json.Marshal(authConfig)
if err != nil {
return err
}
options := types.ImageCreateOptions{
Parent: ref.Name(),
Tag: tag,
RegistryAuth: encodedAuth,
registryAuthHeader := []string{
base64.URLEncoding.EncodeToString(buf),
}
responseBody, err := cli.client.ImageCreate(context.Background(), options)
if err != nil {
sopts := &streamOpts{
rawTerminal: true,
out: out,
headers: map[string][]string{"X-Registry-Auth": registryAuthHeader},
}
if _, err := cli.stream("POST", "/images/create?"+v.Encode(), sopts); err != nil {
return err
}
defer responseBody.Close()
return jsonmessage.DisplayJSONMessagesStream(responseBody, out, cli.outFd, cli.isTerminalOut, nil)
return nil
}
type cidFile struct {
@@ -82,7 +77,14 @@ func newCIDFile(path string) (*cidFile, error) {
return &cidFile{path: path, file: f}, nil
}
func (cli *DockerCli) createContainer(config *container.Config, hostConfig *container.HostConfig, networkingConfig *networktypes.NetworkingConfig, cidfile, name string) (*types.ContainerCreateResponse, error) {
func (cli *DockerCli) createContainer(config *runconfig.Config, hostConfig *runconfig.HostConfig, cidfile, name string) (*types.ContainerCreateResponse, error) {
containerValues := url.Values{}
if name != "" {
containerValues.Set("name", name)
}
mergedConfig := runconfig.MergeConfigs(config, hostConfig)
var containerIDFile *cidFile
if cidfile != "" {
var err error
@@ -92,52 +94,56 @@ func (cli *DockerCli) createContainer(config *container.Config, hostConfig *cont
defer containerIDFile.Close()
}
var trustedRef reference.Canonical
_, ref, err := reference.ParseIDOrReference(config.Image)
if err != nil {
return nil, err
repo, tag := parsers.ParseRepositoryTag(config.Image)
if tag == "" {
tag = tags.DEFAULTTAG
}
if ref != nil {
ref = reference.WithDefaultTag(ref)
if ref, ok := ref.(reference.NamedTagged); ok && isTrusted() {
var err error
trustedRef, err = cli.trustedReference(ref)
if err != nil {
return nil, err
}
config.Image = trustedRef.String()
ref := registry.ParseReference(tag)
var trustedRef registry.Reference
if isTrusted() && !ref.HasDigest() {
var err error
trustedRef, err = cli.trustedReference(repo, ref)
if err != nil {
return nil, err
}
config.Image = trustedRef.ImageName(repo)
}
//create the container
response, err := cli.client.ContainerCreate(context.Background(), config, hostConfig, networkingConfig, name)
serverResp, err := cli.call("POST", "/containers/create?"+containerValues.Encode(), mergedConfig, nil)
//if image not found try to pull it
if err != nil {
if client.IsErrImageNotFound(err) && ref != nil {
fmt.Fprintf(cli.err, "Unable to find image '%s' locally\n", ref.String())
if serverResp.statusCode == 404 && strings.Contains(err.Error(), config.Image) {
fmt.Fprintf(cli.err, "Unable to find image '%s' locally\n", ref.ImageName(repo))
// we don't want to write to stdout anything apart from container.ID
if err = cli.pullImageCustomOut(config.Image, cli.err); err != nil {
return nil, err
}
if ref, ok := ref.(reference.NamedTagged); ok && trustedRef != nil {
if err := cli.tagTrusted(trustedRef, ref); err != nil {
return nil, err
}
}
// Retry
var retryErr error
response, retryErr = cli.client.ContainerCreate(context.Background(), config, hostConfig, networkingConfig, name)
if retryErr != nil {
return nil, retryErr
}
} else {
// we don't want to write to stdout anything apart from container.ID
if err = cli.pullImageCustomOut(config.Image, cli.err); err != nil {
return nil, err
}
if trustedRef != nil && !ref.HasDigest() {
repoInfo, err := registry.ParseRepositoryInfo(repo)
if err != nil {
return nil, err
}
if err := cli.tagTrusted(repoInfo, trustedRef, ref); err != nil {
return nil, err
}
}
// Retry
if serverResp, err = cli.call("POST", "/containers/create?"+containerValues.Encode(), mergedConfig, nil); err != nil {
return nil, err
}
} else if err != nil {
return nil, err
}
defer serverResp.body.Close()
var response types.ContainerCreateResponse
if err := json.NewDecoder(serverResp.body).Decode(&response); err != nil {
return nil, err
}
for _, warning := range response.Warnings {
fmt.Fprintf(cli.err, "WARNING: %s\n", warning)
}
@@ -153,7 +159,7 @@ func (cli *DockerCli) createContainer(config *container.Config, hostConfig *cont
//
// Usage: docker create [OPTIONS] IMAGE [COMMAND] [ARG...]
func (cli *DockerCli) CmdCreate(args ...string) error {
cmd := Cli.Subcmd("create", []string{"IMAGE [COMMAND] [ARG...]"}, Cli.DockerCommands["create"].Description, true)
cmd := Cli.Subcmd("create", []string{"IMAGE [COMMAND] [ARG...]"}, "Create a new container", true)
addTrustedFlags(cmd, true)
// These are flags not stored in Config/HostConfig
@@ -161,8 +167,7 @@ func (cli *DockerCli) CmdCreate(args ...string) error {
flName = cmd.String([]string{"-name"}, "", "Assign a name to the container")
)
config, hostConfig, networkingConfig, cmd, err := runconfigopts.Parse(cmd, args)
config, hostConfig, cmd, err := runconfig.Parse(cmd, args)
if err != nil {
cmd.ReportError(err.Error(), true)
os.Exit(1)
@@ -171,7 +176,7 @@ func (cli *DockerCli) CmdCreate(args ...string) error {
cmd.Usage()
return nil
}
response, err := cli.createContainer(config, hostConfig, networkingConfig, hostConfig.ContainerIDFile, *flName)
response, err := cli.createContainer(config, hostConfig, hostConfig.ContainerIDFile, *flName)
if err != nil {
return err
}

View File

@@ -1,10 +1,10 @@
package client
import (
"encoding/json"
"fmt"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/archive"
flag "github.com/docker/docker/pkg/mflag"
@@ -18,7 +18,7 @@ import (
//
// Usage: docker diff CONTAINER
func (cli *DockerCli) CmdDiff(args ...string) error {
cmd := Cli.Subcmd("diff", []string{"CONTAINER"}, Cli.DockerCommands["diff"].Description, true)
cmd := Cli.Subcmd("diff", []string{"CONTAINER"}, "Inspect changes on a container's filesystem", true)
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
@@ -27,11 +27,18 @@ func (cli *DockerCli) CmdDiff(args ...string) error {
return fmt.Errorf("Container name cannot be empty")
}
changes, err := cli.client.ContainerDiff(context.Background(), cmd.Arg(0))
serverResp, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/changes", nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
changes := []types.ContainerChange{}
if err := json.NewDecoder(serverResp.body).Decode(&changes); err != nil {
return err
}
for _, change := range changes {
var kind string
switch change.Kind {

View File

@@ -1,32 +1,22 @@
package client
import (
"encoding/json"
"fmt"
"io"
"sort"
"strings"
"sync"
"net/url"
"time"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/jsonlog"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/engine-api/types"
eventtypes "github.com/docker/engine-api/types/events"
"github.com/docker/engine-api/types/filters"
"github.com/docker/docker/pkg/parsers/filters"
"github.com/docker/docker/pkg/timeutils"
)
// CmdEvents prints a live stream of real time events from the server.
//
// Usage: docker events [OPTIONS]
func (cli *DockerCli) CmdEvents(args ...string) error {
cmd := Cli.Subcmd("events", nil, Cli.DockerCommands["events"].Description, true)
since := cmd.String([]string{"-since"}, "", "Show all events created since timestamp")
cmd := Cli.Subcmd("events", nil, "Get real time events from the server", true)
since := cmd.String([]string{"#since", "-since"}, "", "Show all events created since timestamp")
until := cmd.String([]string{"-until"}, "", "Stream events until this timestamp")
flFilter := opts.NewListOpts(nil)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Filter output based on conditions provided")
@@ -34,7 +24,10 @@ func (cli *DockerCli) CmdEvents(args ...string) error {
cmd.ParseFlags(args, true)
eventFilterArgs := filters.NewArgs()
var (
v = url.Values{}
eventFilterArgs = filters.Args{}
)
// Consolidate all filter flags, and sanity check them early.
// They'll get process in the daemon/server.
@@ -45,102 +38,26 @@ func (cli *DockerCli) CmdEvents(args ...string) error {
return err
}
}
options := types.EventsOptions{
Since: *since,
Until: *until,
Filters: eventFilterArgs,
ref := time.Now()
if *since != "" {
v.Set("since", timeutils.GetTimestamp(*since, ref))
}
responseBody, err := cli.client.Events(context.Background(), options)
if err != nil {
return err
if *until != "" {
v.Set("until", timeutils.GetTimestamp(*until, ref))
}
defer responseBody.Close()
return streamEvents(responseBody, cli.out)
}
// streamEvents decodes prints the incoming events in the provided output.
func streamEvents(input io.Reader, output io.Writer) error {
return decodeEvents(input, func(event eventtypes.Message, err error) error {
if len(eventFilterArgs) > 0 {
filterJSON, err := filters.ToParam(eventFilterArgs)
if err != nil {
return err
}
printOutput(event, output)
return nil
})
}
type eventProcessor func(event eventtypes.Message, err error) error
func decodeEvents(input io.Reader, ep eventProcessor) error {
dec := json.NewDecoder(input)
for {
var event eventtypes.Message
err := dec.Decode(&event)
if err != nil && err == io.EOF {
break
}
if procErr := ep(event, err); procErr != nil {
return procErr
}
v.Set("filters", filterJSON)
}
sopts := &streamOpts{
rawTerminal: true,
out: cli.out,
}
if _, err := cli.stream("GET", "/events?"+v.Encode(), sopts); err != nil {
return err
}
return nil
}
// printOutput prints all types of event information.
// Each output includes the event type, actor id, name and action.
// Actor attributes are printed at the end if the actor has any.
func printOutput(event eventtypes.Message, output io.Writer) {
if event.TimeNano != 0 {
fmt.Fprintf(output, "%s ", time.Unix(0, event.TimeNano).Format(jsonlog.RFC3339NanoFixed))
} else if event.Time != 0 {
fmt.Fprintf(output, "%s ", time.Unix(event.Time, 0).Format(jsonlog.RFC3339NanoFixed))
}
fmt.Fprintf(output, "%s %s %s", event.Type, event.Action, event.Actor.ID)
if len(event.Actor.Attributes) > 0 {
var attrs []string
var keys []string
for k := range event.Actor.Attributes {
keys = append(keys, k)
}
sort.Strings(keys)
for _, k := range keys {
v := event.Actor.Attributes[k]
attrs = append(attrs, fmt.Sprintf("%s=%s", k, v))
}
fmt.Fprintf(output, " (%s)", strings.Join(attrs, ", "))
}
fmt.Fprint(output, "\n")
}
type eventHandler struct {
handlers map[string]func(eventtypes.Message)
mu sync.Mutex
}
func (w *eventHandler) Handle(action string, h func(eventtypes.Message)) {
w.mu.Lock()
w.handlers[action] = h
w.mu.Unlock()
}
// Watch ranges over the passed in event chan and processes the events based on the
// handlers created for a given action.
// To stop watching, close the event chan.
func (w *eventHandler) Watch(c <-chan eventtypes.Message) {
for e := range c {
w.mu.Lock()
h, exists := w.handlers[e.Action]
w.mu.Unlock()
if !exists {
continue
}
logrus.Debugf("event handler: received event: %v", e)
go h(e)
}
}

View File

@@ -1,61 +1,60 @@
package client
import (
"encoding/json"
"fmt"
"io"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/promise"
"github.com/docker/engine-api/types"
"github.com/docker/docker/runconfig"
)
// CmdExec runs a command in a running container.
//
// Usage: docker exec [OPTIONS] CONTAINER COMMAND [ARG...]
func (cli *DockerCli) CmdExec(args ...string) error {
cmd := Cli.Subcmd("exec", []string{"CONTAINER COMMAND [ARG...]"}, Cli.DockerCommands["exec"].Description, true)
detachKeys := cmd.String([]string{"-detach-keys"}, "", "Override the key sequence for detaching a container")
cmd := Cli.Subcmd("exec", []string{"CONTAINER COMMAND [ARG...]"}, "Run a command in a running container", true)
execConfig, err := ParseExec(cmd, args)
execConfig, err := runconfig.ParseExec(cmd, args)
// just in case the ParseExec does not exit
if execConfig.Container == "" || err != nil {
return Cli.StatusError{StatusCode: 1}
}
if *detachKeys != "" {
cli.configFile.DetachKeys = *detachKeys
}
// Send client escape keys
execConfig.DetachKeys = cli.configFile.DetachKeys
response, err := cli.client.ContainerExecCreate(context.Background(), *execConfig)
serverResp, err := cli.call("POST", "/containers/"+execConfig.Container+"/exec", execConfig, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
var response types.ContainerExecCreateResponse
if err := json.NewDecoder(serverResp.body).Decode(&response); err != nil {
return err
}
execID := response.ID
if execID == "" {
fmt.Fprintf(cli.out, "exec ID empty")
return nil
}
//Temp struct for execStart so that we don't need to transfer all the execConfig
execStartCheck := &types.ExecStartCheck{
Detach: execConfig.Detach,
Tty: execConfig.Tty,
}
if !execConfig.Detach {
if err := cli.CheckTtyInput(execConfig.AttachStdin, execConfig.Tty); err != nil {
return err
}
} else {
execStartCheck := types.ExecStartCheck{
Detach: execConfig.Detach,
Tty: execConfig.Tty,
}
if err := cli.client.ContainerExecStart(context.Background(), execID, execStartCheck); err != nil {
if _, _, err := readBody(cli.call("POST", "/exec/"+execID+"/start", execStartCheck, nil)); err != nil {
return err
}
// For now don't print this - wait for when we support exec wait()
@@ -67,9 +66,18 @@ func (cli *DockerCli) CmdExec(args ...string) error {
var (
out, stderr io.Writer
in io.ReadCloser
hijacked = make(chan io.Closer)
errCh chan error
)
// Block the return until the chan gets closed
defer func() {
logrus.Debugf("End of CmdExec(), Waiting for hijack to finish.")
if _, ok := <-hijacked; ok {
fmt.Fprintln(cli.err, "Hijack did not finish (chan still open)")
}
}()
if execConfig.AttachStdin {
in = cli.in
}
@@ -83,21 +91,24 @@ func (cli *DockerCli) CmdExec(args ...string) error {
stderr = cli.err
}
}
errCh = promise.Go(func() error {
return cli.hijack("POST", "/exec/"+execID+"/start", execConfig.Tty, in, out, stderr, hijacked, execConfig)
})
resp, err := cli.client.ContainerExecAttach(context.Background(), execID, *execConfig)
if err != nil {
return err
}
defer resp.Close()
if in != nil && execConfig.Tty {
if err := cli.setRawTerminal(); err != nil {
// Acknowledge the hijack before starting
select {
case closer := <-hijacked:
// Make sure that hijack gets closed when returning. (result
// in closing hijack chan and freeing server's goroutines.
if closer != nil {
defer closer.Close()
}
case err := <-errCh:
if err != nil {
logrus.Debugf("Error hijack: %s", err)
return err
}
defer cli.restoreTerminal(in)
}
errCh = promise.Go(func() error {
return cli.holdHijackedConnection(execConfig.Tty, in, out, stderr, resp)
})
if execConfig.Tty && cli.isTerminalIn {
if err := cli.monitorTtySize(execID, true); err != nil {
@@ -121,46 +132,3 @@ func (cli *DockerCli) CmdExec(args ...string) error {
return nil
}
// ParseExec parses the specified args for the specified command and generates
// an ExecConfig from it.
// If the minimal number of specified args is not right or if specified args are
// not valid, it will return an error.
func ParseExec(cmd *flag.FlagSet, args []string) (*types.ExecConfig, error) {
var (
flStdin = cmd.Bool([]string{"i", "-interactive"}, false, "Keep STDIN open even if not attached")
flTty = cmd.Bool([]string{"t", "-tty"}, false, "Allocate a pseudo-TTY")
flDetach = cmd.Bool([]string{"d", "-detach"}, false, "Detached mode: run command in the background")
flUser = cmd.String([]string{"u", "-user"}, "", "Username or UID (format: <name|uid>[:<group|gid>])")
flPrivileged = cmd.Bool([]string{"-privileged"}, false, "Give extended privileges to the command")
execCmd []string
container string
)
cmd.Require(flag.Min, 2)
if err := cmd.ParseFlags(args, true); err != nil {
return nil, err
}
container = cmd.Arg(0)
parsedArgs := cmd.Args()
execCmd = parsedArgs[1:]
execConfig := &types.ExecConfig{
User: *flUser,
Privileged: *flPrivileged,
Tty: *flTty,
Cmd: execCmd,
Container: container,
Detach: *flDetach,
}
// If -d is not set, attach to everything by default
if !*flDetach {
execConfig.AttachStdout = true
execConfig.AttachStderr = true
if *flStdin {
execConfig.AttachStdin = true
}
}
return execConfig, nil
}

View File

@@ -3,8 +3,7 @@ package client
import (
"errors"
"io"
"golang.org/x/net/context"
"os"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -16,27 +15,33 @@ import (
//
// Usage: docker export [OPTIONS] CONTAINER
func (cli *DockerCli) CmdExport(args ...string) error {
cmd := Cli.Subcmd("export", []string{"CONTAINER"}, Cli.DockerCommands["export"].Description, true)
cmd := Cli.Subcmd("export", []string{"CONTAINER"}, "Export the contents of a container's filesystem as a tar archive", true)
outfile := cmd.String([]string{"o", "-output"}, "", "Write to a file, instead of STDOUT")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
if *outfile == "" && cli.isTerminalOut {
var (
output io.Writer = cli.out
err error
)
if *outfile != "" {
output, err = os.Create(*outfile)
if err != nil {
return err
}
} else if cli.isTerminalOut {
return errors.New("Cowardly refusing to save to a terminal. Use the -o flag or redirect.")
}
responseBody, err := cli.client.ContainerExport(context.Background(), cmd.Arg(0))
if err != nil {
return err
image := cmd.Arg(0)
sopts := &streamOpts{
rawTerminal: true,
out: output,
}
defer responseBody.Close()
if *outfile == "" {
_, err := io.Copy(cli.out, responseBody)
if _, err := cli.stream("GET", "/containers/"+image+"/export", sopts); err != nil {
return err
}
return copyToFile(*outfile, responseBody)
return nil
}

View File

@@ -1,192 +0,0 @@
package formatter
import (
"reflect"
"strings"
"testing"
"time"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/engine-api/types"
)
func TestContainerPsContext(t *testing.T) {
containerID := stringid.GenerateRandomID()
unix := time.Now().Add(-65 * time.Second).Unix()
var ctx containerContext
cases := []struct {
container types.Container
trunc bool
expValue string
expHeader string
call func() string
}{
{types.Container{ID: containerID}, true, stringid.TruncateID(containerID), containerIDHeader, ctx.ID},
{types.Container{ID: containerID}, false, containerID, containerIDHeader, ctx.ID},
{types.Container{Names: []string{"/foobar_baz"}}, true, "foobar_baz", namesHeader, ctx.Names},
{types.Container{Image: "ubuntu"}, true, "ubuntu", imageHeader, ctx.Image},
{types.Container{Image: "verylongimagename"}, true, "verylongimagename", imageHeader, ctx.Image},
{types.Container{Image: "verylongimagename"}, false, "verylongimagename", imageHeader, ctx.Image},
{types.Container{
Image: "a5a665ff33eced1e0803148700880edab4",
ImageID: "a5a665ff33eced1e0803148700880edab4269067ed77e27737a708d0d293fbf5",
},
true,
"a5a665ff33ec",
imageHeader,
ctx.Image,
},
{types.Container{
Image: "a5a665ff33eced1e0803148700880edab4",
ImageID: "a5a665ff33eced1e0803148700880edab4269067ed77e27737a708d0d293fbf5",
},
false,
"a5a665ff33eced1e0803148700880edab4",
imageHeader,
ctx.Image,
},
{types.Container{Image: ""}, true, "<no image>", imageHeader, ctx.Image},
{types.Container{Command: "sh -c 'ls -la'"}, true, `"sh -c 'ls -la'"`, commandHeader, ctx.Command},
{types.Container{Created: unix}, true, time.Unix(unix, 0).String(), createdAtHeader, ctx.CreatedAt},
{types.Container{Ports: []types.Port{{PrivatePort: 8080, PublicPort: 8080, Type: "tcp"}}}, true, "8080/tcp", portsHeader, ctx.Ports},
{types.Container{Status: "RUNNING"}, true, "RUNNING", statusHeader, ctx.Status},
{types.Container{SizeRw: 10}, true, "10 B", sizeHeader, ctx.Size},
{types.Container{SizeRw: 10, SizeRootFs: 20}, true, "10 B (virtual 20 B)", sizeHeader, ctx.Size},
{types.Container{}, true, "", labelsHeader, ctx.Labels},
{types.Container{Labels: map[string]string{"cpu": "6", "storage": "ssd"}}, true, "cpu=6,storage=ssd", labelsHeader, ctx.Labels},
{types.Container{Created: unix}, true, "About a minute", runningForHeader, ctx.RunningFor},
}
for _, c := range cases {
ctx = containerContext{c: c.container, trunc: c.trunc}
v := c.call()
if strings.Contains(v, ",") {
compareMultipleValues(t, v, c.expValue)
} else if v != c.expValue {
t.Fatalf("Expected %s, was %s\n", c.expValue, v)
}
h := ctx.fullHeader()
if h != c.expHeader {
t.Fatalf("Expected %s, was %s\n", c.expHeader, h)
}
}
c1 := types.Container{Labels: map[string]string{"com.docker.swarm.swarm-id": "33", "com.docker.swarm.node_name": "ubuntu"}}
ctx = containerContext{c: c1, trunc: true}
sid := ctx.Label("com.docker.swarm.swarm-id")
node := ctx.Label("com.docker.swarm.node_name")
if sid != "33" {
t.Fatalf("Expected 33, was %s\n", sid)
}
if node != "ubuntu" {
t.Fatalf("Expected ubuntu, was %s\n", node)
}
h := ctx.fullHeader()
if h != "SWARM ID\tNODE NAME" {
t.Fatalf("Expected %s, was %s\n", "SWARM ID\tNODE NAME", h)
}
c2 := types.Container{}
ctx = containerContext{c: c2, trunc: true}
label := ctx.Label("anything.really")
if label != "" {
t.Fatalf("Expected an empty string, was %s", label)
}
ctx = containerContext{c: c2, trunc: true}
fullHeader := ctx.fullHeader()
if fullHeader != "" {
t.Fatalf("Expected fullHeader to be empty, was %s", fullHeader)
}
}
func TestImagesContext(t *testing.T) {
imageID := stringid.GenerateRandomID()
unix := time.Now().Unix()
var ctx imageContext
cases := []struct {
imageCtx imageContext
expValue string
expHeader string
call func() string
}{
{imageContext{
i: types.Image{ID: imageID},
trunc: true,
}, stringid.TruncateID(imageID), imageIDHeader, ctx.ID},
{imageContext{
i: types.Image{ID: imageID},
trunc: false,
}, imageID, imageIDHeader, ctx.ID},
{imageContext{
i: types.Image{Size: 10},
trunc: true,
}, "10 B", sizeHeader, ctx.Size},
{imageContext{
i: types.Image{Created: unix},
trunc: true,
}, time.Unix(unix, 0).String(), createdAtHeader, ctx.CreatedAt},
// FIXME
// {imageContext{
// i: types.Image{Created: unix},
// trunc: true,
// }, units.HumanDuration(time.Unix(unix, 0)), createdSinceHeader, ctx.CreatedSince},
{imageContext{
i: types.Image{},
repo: "busybox",
}, "busybox", repositoryHeader, ctx.Repository},
{imageContext{
i: types.Image{},
tag: "latest",
}, "latest", tagHeader, ctx.Tag},
{imageContext{
i: types.Image{},
digest: "sha256:d149ab53f8718e987c3a3024bb8aa0e2caadf6c0328f1d9d850b2a2a67f2819a",
}, "sha256:d149ab53f8718e987c3a3024bb8aa0e2caadf6c0328f1d9d850b2a2a67f2819a", digestHeader, ctx.Digest},
}
for _, c := range cases {
ctx = c.imageCtx
v := c.call()
if strings.Contains(v, ",") {
compareMultipleValues(t, v, c.expValue)
} else if v != c.expValue {
t.Fatalf("Expected %s, was %s\n", c.expValue, v)
}
h := ctx.fullHeader()
if h != c.expHeader {
t.Fatalf("Expected %s, was %s\n", c.expHeader, h)
}
}
}
func compareMultipleValues(t *testing.T, value, expected string) {
// comma-separated values means probably a map input, which won't
// be guaranteed to have the same order as our expected value
// We'll create maps and use reflect.DeepEquals to check instead:
entriesMap := make(map[string]string)
expMap := make(map[string]string)
entries := strings.Split(value, ",")
expectedEntries := strings.Split(expected, ",")
for _, entry := range entries {
keyval := strings.Split(entry, "=")
entriesMap[keyval[0]] = keyval[1]
}
for _, expected := range expectedEntries {
keyval := strings.Split(expected, "=")
expMap[keyval[0]] = keyval[1]
}
if !reflect.DeepEqual(expMap, entriesMap) {
t.Fatalf("Expected entries: %v, got: %v", expected, value)
}
}

View File

@@ -1,255 +0,0 @@
package formatter
import (
"bytes"
"fmt"
"io"
"strings"
"text/tabwriter"
"text/template"
"github.com/docker/docker/reference"
"github.com/docker/docker/utils/templates"
"github.com/docker/engine-api/types"
)
const (
tableFormatKey = "table"
rawFormatKey = "raw"
defaultContainerTableFormat = "table {{.ID}}\t{{.Image}}\t{{.Command}}\t{{.RunningFor}} ago\t{{.Status}}\t{{.Ports}}\t{{.Names}}"
defaultImageTableFormat = "table {{.Repository}}\t{{.Tag}}\t{{.ID}}\t{{.CreatedSince}} ago\t{{.Size}}"
defaultImageTableFormatWithDigest = "table {{.Repository}}\t{{.Tag}}\t{{.Digest}}\t{{.ID}}\t{{.CreatedSince}} ago\t{{.Size}}"
defaultQuietFormat = "{{.ID}}"
)
// Context contains information required by the formatter to print the output as desired.
type Context struct {
// Output is the output stream to which the formatted string is written.
Output io.Writer
// Format is used to choose raw, table or custom format for the output.
Format string
// Quiet when set to true will simply print minimal information.
Quiet bool
// Trunc when set to true will truncate the output of certain fields such as Container ID.
Trunc bool
// internal element
table bool
finalFormat string
header string
buffer *bytes.Buffer
}
func (c *Context) preformat() {
c.finalFormat = c.Format
if strings.HasPrefix(c.Format, tableKey) {
c.table = true
c.finalFormat = c.finalFormat[len(tableKey):]
}
c.finalFormat = strings.Trim(c.finalFormat, " ")
r := strings.NewReplacer(`\t`, "\t", `\n`, "\n")
c.finalFormat = r.Replace(c.finalFormat)
}
func (c *Context) parseFormat() (*template.Template, error) {
tmpl, err := templates.Parse(c.finalFormat)
if err != nil {
c.buffer.WriteString(fmt.Sprintf("Template parsing error: %v\n", err))
c.buffer.WriteTo(c.Output)
}
return tmpl, err
}
func (c *Context) postformat(tmpl *template.Template, subContext subContext) {
if c.table {
if len(c.header) == 0 {
// if we still don't have a header, we didn't have any containers so we need to fake it to get the right headers from the template
tmpl.Execute(bytes.NewBufferString(""), subContext)
c.header = subContext.fullHeader()
}
t := tabwriter.NewWriter(c.Output, 20, 1, 3, ' ', 0)
t.Write([]byte(c.header))
t.Write([]byte("\n"))
c.buffer.WriteTo(t)
t.Flush()
} else {
c.buffer.WriteTo(c.Output)
}
}
func (c *Context) contextFormat(tmpl *template.Template, subContext subContext) error {
if err := tmpl.Execute(c.buffer, subContext); err != nil {
c.buffer = bytes.NewBufferString(fmt.Sprintf("Template parsing error: %v\n", err))
c.buffer.WriteTo(c.Output)
return err
}
if c.table && len(c.header) == 0 {
c.header = subContext.fullHeader()
}
c.buffer.WriteString("\n")
return nil
}
// ContainerContext contains container specific information required by the formater, encapsulate a Context struct.
type ContainerContext struct {
Context
// Size when set to true will display the size of the output.
Size bool
// Containers
Containers []types.Container
}
// ImageContext contains image specific information required by the formater, encapsulate a Context struct.
type ImageContext struct {
Context
Digest bool
// Images
Images []types.Image
}
func (ctx ContainerContext) Write() {
switch ctx.Format {
case tableFormatKey:
ctx.Format = defaultContainerTableFormat
if ctx.Quiet {
ctx.Format = defaultQuietFormat
}
case rawFormatKey:
if ctx.Quiet {
ctx.Format = `container_id: {{.ID}}`
} else {
ctx.Format = `container_id: {{.ID}}
image: {{.Image}}
command: {{.Command}}
created_at: {{.CreatedAt}}
status: {{.Status}}
names: {{.Names}}
labels: {{.Labels}}
ports: {{.Ports}}
`
if ctx.Size {
ctx.Format += `size: {{.Size}}
`
}
}
}
ctx.buffer = bytes.NewBufferString("")
ctx.preformat()
if ctx.table && ctx.Size {
ctx.finalFormat += "\t{{.Size}}"
}
tmpl, err := ctx.parseFormat()
if err != nil {
return
}
for _, container := range ctx.Containers {
containerCtx := &containerContext{
trunc: ctx.Trunc,
c: container,
}
err = ctx.contextFormat(tmpl, containerCtx)
if err != nil {
return
}
}
ctx.postformat(tmpl, &containerContext{})
}
func (ctx ImageContext) Write() {
switch ctx.Format {
case tableFormatKey:
ctx.Format = defaultImageTableFormat
if ctx.Digest {
ctx.Format = defaultImageTableFormatWithDigest
}
if ctx.Quiet {
ctx.Format = defaultQuietFormat
}
case rawFormatKey:
if ctx.Quiet {
ctx.Format = `image_id: {{.ID}}`
} else {
if ctx.Digest {
ctx.Format = `repository: {{ .Repository }}
tag: {{.Tag}}
digest: {{.Digest}}
image_id: {{.ID}}
created_at: {{.CreatedAt}}
virtual_size: {{.Size}}
`
} else {
ctx.Format = `repository: {{ .Repository }}
tag: {{.Tag}}
image_id: {{.ID}}
created_at: {{.CreatedAt}}
virtual_size: {{.Size}}
`
}
}
}
ctx.buffer = bytes.NewBufferString("")
ctx.preformat()
if ctx.table && ctx.Digest && !strings.Contains(ctx.Format, "{{.Digest}}") {
ctx.finalFormat += "\t{{.Digest}}"
}
tmpl, err := ctx.parseFormat()
if err != nil {
return
}
for _, image := range ctx.Images {
repoTags := image.RepoTags
repoDigests := image.RepoDigests
if len(repoTags) == 1 && repoTags[0] == "<none>:<none>" && len(repoDigests) == 1 && repoDigests[0] == "<none>@<none>" {
// dangling image - clear out either repoTags or repoDigests so we only show it once below
repoDigests = []string{}
}
// combine the tags and digests lists
tagsAndDigests := append(repoTags, repoDigests...)
for _, repoAndRef := range tagsAndDigests {
repo := "<none>"
tag := "<none>"
digest := "<none>"
if !strings.HasPrefix(repoAndRef, "<none>") {
ref, err := reference.ParseNamed(repoAndRef)
if err != nil {
continue
}
repo = ref.Name()
switch x := ref.(type) {
case reference.Canonical:
digest = x.Digest().String()
case reference.NamedTagged:
tag = x.Tag()
}
}
imageCtx := &imageContext{
trunc: ctx.Trunc,
i: image,
repo: repo,
tag: tag,
digest: digest,
}
err = ctx.contextFormat(tmpl, imageCtx)
if err != nil {
return
}
}
}
ctx.postformat(tmpl, &imageContext{})
}

View File

@@ -1,535 +0,0 @@
package formatter
import (
"bytes"
"fmt"
"testing"
"time"
"github.com/docker/engine-api/types"
)
func TestContainerContextWrite(t *testing.T) {
unixTime := time.Now().AddDate(0, 0, -1).Unix()
expectedTime := time.Unix(unixTime, 0).String()
contexts := []struct {
context ContainerContext
expected string
}{
// Errors
{
ContainerContext{
Context: Context{
Format: "{{InvalidFunction}}",
},
},
`Template parsing error: template: :1: function "InvalidFunction" not defined
`,
},
{
ContainerContext{
Context: Context{
Format: "{{nil}}",
},
},
`Template parsing error: template: :1:2: executing "" at <nil>: nil is not a command
`,
},
// Table Format
{
ContainerContext{
Context: Context{
Format: "table",
},
},
`CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
containerID1 ubuntu "" 24 hours ago foobar_baz
containerID2 ubuntu "" 24 hours ago foobar_bar
`,
},
{
ContainerContext{
Context: Context{
Format: "table {{.Image}}",
},
},
"IMAGE\nubuntu\nubuntu\n",
},
{
ContainerContext{
Context: Context{
Format: "table {{.Image}}",
},
Size: true,
},
"IMAGE SIZE\nubuntu 0 B\nubuntu 0 B\n",
},
{
ContainerContext{
Context: Context{
Format: "table {{.Image}}",
Quiet: true,
},
},
"IMAGE\nubuntu\nubuntu\n",
},
{
ContainerContext{
Context: Context{
Format: "table",
Quiet: true,
},
},
"containerID1\ncontainerID2\n",
},
// Raw Format
{
ContainerContext{
Context: Context{
Format: "raw",
},
},
fmt.Sprintf(`container_id: containerID1
image: ubuntu
command: ""
created_at: %s
status:
names: foobar_baz
labels:
ports:
container_id: containerID2
image: ubuntu
command: ""
created_at: %s
status:
names: foobar_bar
labels:
ports:
`, expectedTime, expectedTime),
},
{
ContainerContext{
Context: Context{
Format: "raw",
},
Size: true,
},
fmt.Sprintf(`container_id: containerID1
image: ubuntu
command: ""
created_at: %s
status:
names: foobar_baz
labels:
ports:
size: 0 B
container_id: containerID2
image: ubuntu
command: ""
created_at: %s
status:
names: foobar_bar
labels:
ports:
size: 0 B
`, expectedTime, expectedTime),
},
{
ContainerContext{
Context: Context{
Format: "raw",
Quiet: true,
},
},
"container_id: containerID1\ncontainer_id: containerID2\n",
},
// Custom Format
{
ContainerContext{
Context: Context{
Format: "{{.Image}}",
},
},
"ubuntu\nubuntu\n",
},
{
ContainerContext{
Context: Context{
Format: "{{.Image}}",
},
Size: true,
},
"ubuntu\nubuntu\n",
},
}
for _, context := range contexts {
containers := []types.Container{
{ID: "containerID1", Names: []string{"/foobar_baz"}, Image: "ubuntu", Created: unixTime},
{ID: "containerID2", Names: []string{"/foobar_bar"}, Image: "ubuntu", Created: unixTime},
}
out := bytes.NewBufferString("")
context.context.Output = out
context.context.Containers = containers
context.context.Write()
actual := out.String()
if actual != context.expected {
t.Fatalf("Expected \n%s, got \n%s", context.expected, actual)
}
// Clean buffer
out.Reset()
}
}
func TestContainerContextWriteWithNoContainers(t *testing.T) {
out := bytes.NewBufferString("")
containers := []types.Container{}
contexts := []struct {
context ContainerContext
expected string
}{
{
ContainerContext{
Context: Context{
Format: "{{.Image}}",
Output: out,
},
},
"",
},
{
ContainerContext{
Context: Context{
Format: "table {{.Image}}",
Output: out,
},
},
"IMAGE\n",
},
{
ContainerContext{
Context: Context{
Format: "{{.Image}}",
Output: out,
},
Size: true,
},
"",
},
{
ContainerContext{
Context: Context{
Format: "table {{.Image}}",
Output: out,
},
Size: true,
},
"IMAGE SIZE\n",
},
}
for _, context := range contexts {
context.context.Containers = containers
context.context.Write()
actual := out.String()
if actual != context.expected {
t.Fatalf("Expected \n%s, got \n%s", context.expected, actual)
}
// Clean buffer
out.Reset()
}
}
func TestImageContextWrite(t *testing.T) {
unixTime := time.Now().AddDate(0, 0, -1).Unix()
expectedTime := time.Unix(unixTime, 0).String()
contexts := []struct {
context ImageContext
expected string
}{
// Errors
{
ImageContext{
Context: Context{
Format: "{{InvalidFunction}}",
},
},
`Template parsing error: template: :1: function "InvalidFunction" not defined
`,
},
{
ImageContext{
Context: Context{
Format: "{{nil}}",
},
},
`Template parsing error: template: :1:2: executing "" at <nil>: nil is not a command
`,
},
// Table Format
{
ImageContext{
Context: Context{
Format: "table",
},
},
`REPOSITORY TAG IMAGE ID CREATED SIZE
image tag1 imageID1 24 hours ago 0 B
image <none> imageID1 24 hours ago 0 B
image tag2 imageID2 24 hours ago 0 B
<none> <none> imageID3 24 hours ago 0 B
`,
},
{
ImageContext{
Context: Context{
Format: "table {{.Repository}}",
},
},
"REPOSITORY\nimage\nimage\nimage\n<none>\n",
},
{
ImageContext{
Context: Context{
Format: "table {{.Repository}}",
},
Digest: true,
},
`REPOSITORY DIGEST
image <none>
image sha256:cbbf2f9a99b47fc460d422812b6a5adff7dfee951d8fa2e4a98caa0382cfbdbf
image <none>
<none> <none>
`,
},
{
ImageContext{
Context: Context{
Format: "table {{.Repository}}",
Quiet: true,
},
},
"REPOSITORY\nimage\nimage\nimage\n<none>\n",
},
{
ImageContext{
Context: Context{
Format: "table",
Quiet: true,
},
},
"imageID1\nimageID1\nimageID2\nimageID3\n",
},
{
ImageContext{
Context: Context{
Format: "table",
Quiet: false,
},
Digest: true,
},
`REPOSITORY TAG DIGEST IMAGE ID CREATED SIZE
image tag1 <none> imageID1 24 hours ago 0 B
image <none> sha256:cbbf2f9a99b47fc460d422812b6a5adff7dfee951d8fa2e4a98caa0382cfbdbf imageID1 24 hours ago 0 B
image tag2 <none> imageID2 24 hours ago 0 B
<none> <none> <none> imageID3 24 hours ago 0 B
`,
},
{
ImageContext{
Context: Context{
Format: "table",
Quiet: true,
},
Digest: true,
},
"imageID1\nimageID1\nimageID2\nimageID3\n",
},
// Raw Format
{
ImageContext{
Context: Context{
Format: "raw",
},
},
fmt.Sprintf(`repository: image
tag: tag1
image_id: imageID1
created_at: %s
virtual_size: 0 B
repository: image
tag: <none>
image_id: imageID1
created_at: %s
virtual_size: 0 B
repository: image
tag: tag2
image_id: imageID2
created_at: %s
virtual_size: 0 B
repository: <none>
tag: <none>
image_id: imageID3
created_at: %s
virtual_size: 0 B
`, expectedTime, expectedTime, expectedTime, expectedTime),
},
{
ImageContext{
Context: Context{
Format: "raw",
},
Digest: true,
},
fmt.Sprintf(`repository: image
tag: tag1
digest: <none>
image_id: imageID1
created_at: %s
virtual_size: 0 B
repository: image
tag: <none>
digest: sha256:cbbf2f9a99b47fc460d422812b6a5adff7dfee951d8fa2e4a98caa0382cfbdbf
image_id: imageID1
created_at: %s
virtual_size: 0 B
repository: image
tag: tag2
digest: <none>
image_id: imageID2
created_at: %s
virtual_size: 0 B
repository: <none>
tag: <none>
digest: <none>
image_id: imageID3
created_at: %s
virtual_size: 0 B
`, expectedTime, expectedTime, expectedTime, expectedTime),
},
{
ImageContext{
Context: Context{
Format: "raw",
Quiet: true,
},
},
`image_id: imageID1
image_id: imageID1
image_id: imageID2
image_id: imageID3
`,
},
// Custom Format
{
ImageContext{
Context: Context{
Format: "{{.Repository}}",
},
},
"image\nimage\nimage\n<none>\n",
},
{
ImageContext{
Context: Context{
Format: "{{.Repository}}",
},
Digest: true,
},
"image\nimage\nimage\n<none>\n",
},
}
for _, context := range contexts {
images := []types.Image{
{ID: "imageID1", RepoTags: []string{"image:tag1"}, RepoDigests: []string{"image@sha256:cbbf2f9a99b47fc460d422812b6a5adff7dfee951d8fa2e4a98caa0382cfbdbf"}, Created: unixTime},
{ID: "imageID2", RepoTags: []string{"image:tag2"}, Created: unixTime},
{ID: "imageID3", RepoTags: []string{"<none>:<none>"}, RepoDigests: []string{"<none>@<none>"}, Created: unixTime},
}
out := bytes.NewBufferString("")
context.context.Output = out
context.context.Images = images
context.context.Write()
actual := out.String()
if actual != context.expected {
t.Fatalf("Expected \n%s, got \n%s", context.expected, actual)
}
// Clean buffer
out.Reset()
}
}
func TestImageContextWriteWithNoImage(t *testing.T) {
out := bytes.NewBufferString("")
images := []types.Image{}
contexts := []struct {
context ImageContext
expected string
}{
{
ImageContext{
Context: Context{
Format: "{{.Repository}}",
Output: out,
},
},
"",
},
{
ImageContext{
Context: Context{
Format: "table {{.Repository}}",
Output: out,
},
},
"REPOSITORY\n",
},
{
ImageContext{
Context: Context{
Format: "{{.Repository}}",
Output: out,
},
Digest: true,
},
"",
},
{
ImageContext{
Context: Context{
Format: "table {{.Repository}}",
Output: out,
},
Digest: true,
},
"REPOSITORY DIGEST\n",
},
}
for _, context := range contexts {
context.context.Images = images
context.context.Write()
actual := out.String()
if actual != context.expected {
t.Fatalf("Expected \n%s, got \n%s", context.expected, actual)
}
// Clean buffer
out.Reset()
}
}

View File

@@ -1,56 +1,257 @@
package client
import (
"crypto/tls"
"errors"
"fmt"
"io"
"net"
"net/http"
"net/http/httputil"
"os"
"runtime"
"strings"
"time"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api"
"github.com/docker/docker/autogen/dockerversion"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/stdcopy"
"github.com/docker/engine-api/types"
"github.com/docker/docker/pkg/term"
)
func (cli *DockerCli) holdHijackedConnection(tty bool, inputStream io.ReadCloser, outputStream, errorStream io.Writer, resp types.HijackedResponse) error {
var err error
receiveStdout := make(chan error, 1)
if outputStream != nil || errorStream != nil {
go func() {
// When TTY is ON, use regular copy
if tty && outputStream != nil {
_, err = io.Copy(outputStream, resp.Reader)
} else {
_, err = stdcopy.StdCopy(outputStream, errorStream, resp.Reader)
}
logrus.Debugf("[hijack] End of stdout")
receiveStdout <- err
}()
type tlsClientCon struct {
*tls.Conn
rawConn net.Conn
}
func (c *tlsClientCon) CloseWrite() error {
// Go standard tls.Conn doesn't provide the CloseWrite() method so we do it
// on its underlying connection.
if cwc, ok := c.rawConn.(interface {
CloseWrite() error
}); ok {
return cwc.CloseWrite()
}
return nil
}
func tlsDial(network, addr string, config *tls.Config) (net.Conn, error) {
return tlsDialWithDialer(new(net.Dialer), network, addr, config)
}
// We need to copy Go's implementation of tls.Dial (pkg/cryptor/tls/tls.go) in
// order to return our custom tlsClientCon struct which holds both the tls.Conn
// object _and_ its underlying raw connection. The rationale for this is that
// we need to be able to close the write end of the connection when attaching,
// which tls.Conn does not provide.
func tlsDialWithDialer(dialer *net.Dialer, network, addr string, config *tls.Config) (net.Conn, error) {
// We want the Timeout and Deadline values from dialer to cover the
// whole process: TCP connection and TLS handshake. This means that we
// also need to start our own timers now.
timeout := dialer.Timeout
if !dialer.Deadline.IsZero() {
deadlineTimeout := dialer.Deadline.Sub(time.Now())
if timeout == 0 || deadlineTimeout < timeout {
timeout = deadlineTimeout
}
}
stdinDone := make(chan struct{})
go func() {
if inputStream != nil {
io.Copy(resp.Conn, inputStream)
var errChannel chan error
if timeout != 0 {
errChannel = make(chan error, 2)
time.AfterFunc(timeout, func() {
errChannel <- errors.New("")
})
}
rawConn, err := dialer.Dial(network, addr)
if err != nil {
return nil, err
}
// When we set up a TCP connection for hijack, there could be long periods
// of inactivity (a long running command with no output) that in certain
// network setups may cause ECONNTIMEOUT, leaving the client in an unknown
// state. Setting TCP KeepAlive on the socket connection will prohibit
// ECONNTIMEOUT unless the socket connection truly is broken
if tcpConn, ok := rawConn.(*net.TCPConn); ok {
tcpConn.SetKeepAlive(true)
tcpConn.SetKeepAlivePeriod(30 * time.Second)
}
colonPos := strings.LastIndex(addr, ":")
if colonPos == -1 {
colonPos = len(addr)
}
hostname := addr[:colonPos]
// If no ServerName is set, infer the ServerName
// from the hostname we're connecting to.
if config.ServerName == "" {
// Make a copy to avoid polluting argument or default.
c := *config
c.ServerName = hostname
config = &c
}
conn := tls.Client(rawConn, config)
if timeout == 0 {
err = conn.Handshake()
} else {
go func() {
errChannel <- conn.Handshake()
}()
err = <-errChannel
}
if err != nil {
rawConn.Close()
return nil, err
}
// This is Docker difference with standard's crypto/tls package: returned a
// wrapper which holds both the TLS and raw connections.
return &tlsClientCon{conn, rawConn}, nil
}
func (cli *DockerCli) dial() (net.Conn, error) {
if cli.tlsConfig != nil && cli.proto != "unix" {
// Notice this isn't Go standard's tls.Dial function
return tlsDial(cli.proto, cli.addr, cli.tlsConfig)
}
return net.Dial(cli.proto, cli.addr)
}
func (cli *DockerCli) hijack(method, path string, setRawTerminal bool, in io.ReadCloser, stdout, stderr io.Writer, started chan io.Closer, data interface{}) error {
defer func() {
if started != nil {
close(started)
}
}()
params, err := cli.encodeData(data)
if err != nil {
return err
}
req, err := http.NewRequest(method, fmt.Sprintf("%s/v%s%s", cli.basePath, api.Version, path), params)
if err != nil {
return err
}
// Add CLI Config's HTTP Headers BEFORE we set the Docker headers
// then the user can't change OUR headers
for k, v := range cli.configFile.HTTPHeaders {
req.Header.Set(k, v)
}
req.Header.Set("User-Agent", "Docker-Client/"+dockerversion.VERSION+" ("+runtime.GOOS+")")
req.Header.Set("Content-Type", "text/plain")
req.Header.Set("Connection", "Upgrade")
req.Header.Set("Upgrade", "tcp")
req.Host = cli.addr
dial, err := cli.dial()
// When we set up a TCP connection for hijack, there could be long periods
// of inactivity (a long running command with no output) that in certain
// network setups may cause ECONNTIMEOUT, leaving the client in an unknown
// state. Setting TCP KeepAlive on the socket connection will prohibit
// ECONNTIMEOUT unless the socket connection truly is broken
if tcpConn, ok := dial.(*net.TCPConn); ok {
tcpConn.SetKeepAlive(true)
tcpConn.SetKeepAlivePeriod(30 * time.Second)
}
if err != nil {
if strings.Contains(err.Error(), "connection refused") {
return fmt.Errorf("Cannot connect to the Docker daemon. Is 'docker -d' running on this host?")
}
return err
}
clientconn := httputil.NewClientConn(dial, nil)
defer clientconn.Close()
// Server hijacks the connection, error 'connection closed' expected
clientconn.Do(req)
rwc, br := clientconn.Hijack()
defer rwc.Close()
if started != nil {
started <- rwc
}
var receiveStdout chan error
var oldState *term.State
if in != nil && setRawTerminal && cli.isTerminalIn && os.Getenv("NORAW") == "" {
oldState, err = term.SetRawTerminal(cli.inFd)
if err != nil {
return err
}
defer term.RestoreTerminal(cli.inFd, oldState)
}
if stdout != nil || stderr != nil {
receiveStdout = promise.Go(func() (err error) {
defer func() {
if in != nil {
if setRawTerminal && cli.isTerminalIn {
term.RestoreTerminal(cli.inFd, oldState)
}
// For some reason this Close call blocks on darwin..
// As the client exists right after, simply discard the close
// until we find a better solution.
if runtime.GOOS != "darwin" {
in.Close()
}
}
}()
// When TTY is ON, use regular copy
if setRawTerminal && stdout != nil {
_, err = io.Copy(stdout, br)
} else {
_, err = stdcopy.StdCopy(stdout, stderr, br)
}
logrus.Debugf("[hijack] End of stdout")
return err
})
}
sendStdin := promise.Go(func() error {
if in != nil {
io.Copy(rwc, in)
logrus.Debugf("[hijack] End of stdin")
}
if err := resp.CloseWrite(); err != nil {
logrus.Debugf("Couldn't send EOF: %s", err)
if conn, ok := rwc.(interface {
CloseWrite() error
}); ok {
if err := conn.CloseWrite(); err != nil {
logrus.Debugf("Couldn't send EOF: %s", err)
}
}
close(stdinDone)
}()
// Discard errors due to pipe interruption
return nil
})
select {
case err := <-receiveStdout:
if err != nil {
if stdout != nil || stderr != nil {
if err := <-receiveStdout; err != nil {
logrus.Debugf("Error receiveStdout: %s", err)
return err
}
case <-stdinDone:
if outputStream != nil || errorStream != nil {
if err := <-receiveStdout; err != nil {
logrus.Debugf("Error receiveStdout: %s", err)
return err
}
}
}
if !cli.isTerminalIn {
if err := <-sendStdin; err != nil {
logrus.Debugf("Error sendStdin: %s", err)
return err
}
}
return nil
}

View File

@@ -1,75 +1,76 @@
package client
import (
"encoding/json"
"fmt"
"strconv"
"strings"
"text/tabwriter"
"time"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/stringutils"
"github.com/docker/go-units"
"github.com/docker/docker/pkg/units"
)
// CmdHistory shows the history of an image.
//
// Usage: docker history [OPTIONS] IMAGE
func (cli *DockerCli) CmdHistory(args ...string) error {
cmd := Cli.Subcmd("history", []string{"IMAGE"}, Cli.DockerCommands["history"].Description, true)
cmd := Cli.Subcmd("history", []string{"IMAGE"}, "Show the history of an image", true)
human := cmd.Bool([]string{"H", "-human"}, true, "Print sizes and dates in human readable format")
quiet := cmd.Bool([]string{"q", "-quiet"}, false, "Only show numeric IDs")
noTrunc := cmd.Bool([]string{"-no-trunc"}, false, "Don't truncate output")
noTrunc := cmd.Bool([]string{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
history, err := cli.client.ImageHistory(context.Background(), cmd.Arg(0))
serverResp, err := cli.call("GET", "/images/"+cmd.Arg(0)+"/history", nil, nil)
if err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
defer serverResp.body.Close()
if *quiet {
for _, entry := range history {
if *noTrunc {
fmt.Fprintf(w, "%s\n", entry.ID)
} else {
fmt.Fprintf(w, "%s\n", stringid.TruncateID(entry.ID))
}
}
w.Flush()
return nil
history := []types.ImageHistory{}
if err := json.NewDecoder(serverResp.body).Decode(&history); err != nil {
return err
}
var imageID string
var createdBy string
var created string
var size string
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
if !*quiet {
fmt.Fprintln(w, "IMAGE\tCREATED\tCREATED BY\tSIZE\tCOMMENT")
}
fmt.Fprintln(w, "IMAGE\tCREATED\tCREATED BY\tSIZE\tCOMMENT")
for _, entry := range history {
imageID = entry.ID
createdBy = strings.Replace(entry.CreatedBy, "\t", " ", -1)
if *noTrunc == false {
createdBy = stringutils.Truncate(createdBy, 45)
imageID = stringid.TruncateID(entry.ID)
}
if *human {
created = units.HumanDuration(time.Now().UTC().Sub(time.Unix(entry.Created, 0))) + " ago"
size = units.HumanSize(float64(entry.Size))
if *noTrunc {
fmt.Fprintf(w, entry.ID)
} else {
created = time.Unix(entry.Created, 0).Format(time.RFC3339)
size = strconv.FormatInt(entry.Size, 10)
fmt.Fprintf(w, stringid.TruncateID(entry.ID))
}
if !*quiet {
if *human {
fmt.Fprintf(w, "\t%s ago\t", units.HumanDuration(time.Now().UTC().Sub(time.Unix(entry.Created, 0))))
} else {
fmt.Fprintf(w, "\t%s\t", time.Unix(entry.Created, 0).Format(time.RFC3339))
}
fmt.Fprintf(w, "%s\t%s\t%s\t%s\t%s\n", imageID, created, createdBy, size, entry.Comment)
if *noTrunc {
fmt.Fprintf(w, "%s\t", entry.CreatedBy)
} else {
fmt.Fprintf(w, "%s\t", stringutils.Truncate(entry.CreatedBy, 45))
}
if *human {
fmt.Fprintf(w, "%s\t", units.HumanSize(float64(entry.Size)))
} else {
fmt.Fprintf(w, "%d\t", entry.Size)
}
fmt.Fprintf(w, "%s", entry.Comment)
}
fmt.Fprintf(w, "\n")
}
w.Flush()
return nil

View File

@@ -1,26 +1,32 @@
package client
import (
"golang.org/x/net/context"
"encoding/json"
"fmt"
"net/url"
"text/tabwriter"
"time"
"github.com/docker/docker/api/client/formatter"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/filters"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/parsers/filters"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/units"
"github.com/docker/docker/utils"
)
// CmdImages lists the images in a specified repository, or all top-level images if no repository is specified.
//
// Usage: docker images [OPTIONS] [REPOSITORY]
func (cli *DockerCli) CmdImages(args ...string) error {
cmd := Cli.Subcmd("images", []string{"[REPOSITORY[:TAG]]"}, Cli.DockerCommands["images"].Description, true)
cmd := Cli.Subcmd("images", []string{"[REPOSITORY]"}, "List images", true)
quiet := cmd.Bool([]string{"q", "-quiet"}, false, "Only show numeric IDs")
all := cmd.Bool([]string{"a", "-all"}, false, "Show all images (default hides intermediate images)")
noTrunc := cmd.Bool([]string{"-no-trunc"}, false, "Don't truncate output")
noTrunc := cmd.Bool([]string{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
showDigests := cmd.Bool([]string{"-digests"}, false, "Show digests")
format := cmd.String([]string{"-format"}, "", "Pretty-print images using a Go template")
flFilter := opts.NewListOpts(nil)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Filter output based on conditions provided")
@@ -30,7 +36,7 @@ func (cli *DockerCli) CmdImages(args ...string) error {
// Consolidate all filter flags, and sanity check them early.
// They'll get process in the daemon/server.
imageFilterArgs := filters.NewArgs()
imageFilterArgs := filters.Args{}
for _, f := range flFilter.GetAll() {
var err error
imageFilterArgs, err = filters.ParseFlag(f, imageFilterArgs)
@@ -39,43 +45,86 @@ func (cli *DockerCli) CmdImages(args ...string) error {
}
}
var matchName string
matchName := cmd.Arg(0)
v := url.Values{}
if len(imageFilterArgs) > 0 {
filterJSON, err := filters.ToParam(imageFilterArgs)
if err != nil {
return err
}
v.Set("filters", filterJSON)
}
if cmd.NArg() == 1 {
matchName = cmd.Arg(0)
// FIXME rename this parameter, to not be confused with the filters flag
v.Set("filter", matchName)
}
if *all {
v.Set("all", "1")
}
options := types.ImageListOptions{
MatchName: matchName,
All: *all,
Filters: imageFilterArgs,
}
images, err := cli.client.ImageList(context.Background(), options)
serverResp, err := cli.call("GET", "/images/json?"+v.Encode(), nil, nil)
if err != nil {
return err
}
f := *format
if len(f) == 0 {
if len(cli.ImagesFormat()) > 0 && !*quiet {
f = cli.ImagesFormat()
defer serverResp.body.Close()
images := []types.Image{}
if err := json.NewDecoder(serverResp.body).Decode(&images); err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
if !*quiet {
if *showDigests {
fmt.Fprintln(w, "REPOSITORY\tTAG\tDIGEST\tIMAGE ID\tCREATED\tVIRTUAL SIZE")
} else {
f = "table"
fmt.Fprintln(w, "REPOSITORY\tTAG\tIMAGE ID\tCREATED\tVIRTUAL SIZE")
}
}
imagesCtx := formatter.ImageContext{
Context: formatter.Context{
Output: cli.out,
Format: f,
Quiet: *quiet,
Trunc: !*noTrunc,
},
Digest: *showDigests,
Images: images,
for _, image := range images {
ID := image.ID
if !*noTrunc {
ID = stringid.TruncateID(ID)
}
repoTags := image.RepoTags
repoDigests := image.RepoDigests
if len(repoTags) == 1 && repoTags[0] == "<none>:<none>" && len(repoDigests) == 1 && repoDigests[0] == "<none>@<none>" {
// dangling image - clear out either repoTags or repoDigsts so we only show it once below
repoDigests = []string{}
}
// combine the tags and digests lists
tagsAndDigests := append(repoTags, repoDigests...)
for _, repoAndRef := range tagsAndDigests {
repo, ref := parsers.ParseRepositoryTag(repoAndRef)
// default tag and digest to none - if there's a value, it'll be set below
tag := "<none>"
digest := "<none>"
if utils.DigestReference(ref) {
digest = ref
} else {
tag = ref
}
if !*quiet {
if *showDigests {
fmt.Fprintf(w, "%s\t%s\t%s\t%s\t%s ago\t%s\n", repo, tag, digest, ID, units.HumanDuration(time.Now().UTC().Sub(time.Unix(int64(image.Created), 0))), units.HumanSize(float64(image.VirtualSize)))
} else {
fmt.Fprintf(w, "%s\t%s\t%s\t%s ago\t%s\n", repo, tag, ID, units.HumanDuration(time.Now().UTC().Sub(time.Unix(int64(image.Created), 0))), units.HumanSize(float64(image.VirtualSize)))
}
} else {
fmt.Fprintln(w, ID)
}
}
}
imagesCtx.Write()
if !*quiet {
w.Flush()
}
return nil
}

View File

@@ -3,17 +3,15 @@ package client
import (
"fmt"
"io"
"net/url"
"os"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/reference"
"github.com/docker/engine-api/types"
"github.com/docker/docker/registry"
)
// CmdImport creates an empty filesystem image, imports the contents of the tarball into the image, and optionally tags the image.
@@ -22,61 +20,58 @@ import (
//
// Usage: docker import [OPTIONS] file|URL|- [REPOSITORY[:TAG]]
func (cli *DockerCli) CmdImport(args ...string) error {
cmd := Cli.Subcmd("import", []string{"file|URL|- [REPOSITORY[:TAG]]"}, Cli.DockerCommands["import"].Description, true)
cmd := Cli.Subcmd("import", []string{"file|URL|- [REPOSITORY[:TAG]]"}, "Create an empty filesystem image and import the contents of the\ntarball (.tar, .tar.gz, .tgz, .bzip, .tar.xz, .txz) into it, then\noptionally tag it.", true)
flChanges := opts.NewListOpts(nil)
cmd.Var(&flChanges, []string{"c", "-change"}, "Apply Dockerfile instruction to the created image")
message := cmd.String([]string{"m", "-message"}, "", "Set commit message for imported image")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var (
in io.Reader
tag string
v = url.Values{}
src = cmd.Arg(0)
srcName = src
repository = cmd.Arg(1)
changes = flChanges.GetAll()
)
v.Set("fromSrc", src)
v.Set("repo", repository)
for _, change := range flChanges.GetAll() {
v.Add("changes", change)
}
if cmd.NArg() == 3 {
fmt.Fprintf(cli.err, "[DEPRECATED] The format 'file|URL|- [REPOSITORY [TAG]]' has been deprecated. Please use file|URL|- [REPOSITORY[:TAG]]\n")
tag = cmd.Arg(2)
v.Set("tag", cmd.Arg(2))
}
if repository != "" {
//Check if the given image name can be resolved
if _, err := reference.ParseNamed(repository); err != nil {
repo, _ := parsers.ParseRepositoryTag(repository)
if err := registry.ValidateRepositoryName(repo); err != nil {
return err
}
}
var in io.Reader
if src == "-" {
in = cli.in
} else if !urlutil.IsURL(src) {
srcName = "-"
v.Set("fromSrc", "-")
file, err := os.Open(src)
if err != nil {
return err
}
defer file.Close()
in = file
}
options := types.ImageImportOptions{
Source: in,
SourceName: srcName,
RepositoryName: repository,
Message: *message,
Tag: tag,
Changes: changes,
sopts := &streamOpts{
rawTerminal: true,
in: in,
out: cli.out,
}
responseBody, err := cli.client.ImageImport(context.Background(), options)
if err != nil {
return err
}
defer responseBody.Close()
return jsonmessage.DisplayJSONMessagesStream(responseBody, cli.out, cli.outFd, cli.isTerminalOut, nil)
_, err := cli.stream("POST", "/images/create?"+v.Encode(), sopts)
return err
}

View File

@@ -1,138 +1,95 @@
package client
import (
"encoding/json"
"fmt"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/httputils"
"github.com/docker/docker/pkg/ioutils"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/utils"
"github.com/docker/go-units"
"github.com/docker/docker/pkg/units"
)
// CmdInfo displays system-wide information.
//
// Usage: docker info
func (cli *DockerCli) CmdInfo(args ...string) error {
cmd := Cli.Subcmd("info", nil, Cli.DockerCommands["info"].Description, true)
cmd := Cli.Subcmd("info", nil, "Display system-wide information", true)
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
info, err := cli.client.Info(context.Background())
serverResp, err := cli.call("GET", "/info", nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
info := &types.Info{}
if err := json.NewDecoder(serverResp.body).Decode(info); err != nil {
return fmt.Errorf("Error reading remote info: %v", err)
}
fmt.Fprintf(cli.out, "Containers: %d\n", info.Containers)
fmt.Fprintf(cli.out, " Running: %d\n", info.ContainersRunning)
fmt.Fprintf(cli.out, " Paused: %d\n", info.ContainersPaused)
fmt.Fprintf(cli.out, " Stopped: %d\n", info.ContainersStopped)
fmt.Fprintf(cli.out, "Images: %d\n", info.Images)
ioutils.FprintfIfNotEmpty(cli.out, "Server Version: %s\n", info.ServerVersion)
ioutils.FprintfIfNotEmpty(cli.out, "Storage Driver: %s\n", info.Driver)
if info.DriverStatus != nil {
for _, pair := range info.DriverStatus {
fmt.Fprintf(cli.out, " %s: %s\n", pair[0], pair[1])
// print a warning if devicemapper is using a loopback file
if pair[0] == "Data loop file" {
fmt.Fprintln(cli.err, " WARNING: Usage of loopback devices is strongly discouraged for production use. Either use `--storage-opt dm.thinpooldev` or use `--storage-opt dm.no_warn_on_loop_devices=true` to suppress this warning.")
}
}
}
if info.SystemStatus != nil {
for _, pair := range info.SystemStatus {
fmt.Fprintf(cli.out, "%s: %s\n", pair[0], pair[1])
}
}
ioutils.FprintfIfNotEmpty(cli.out, "Execution Driver: %s\n", info.ExecutionDriver)
ioutils.FprintfIfNotEmpty(cli.out, "Logging Driver: %s\n", info.LoggingDriver)
ioutils.FprintfIfNotEmpty(cli.out, "Cgroup Driver: %s\n", info.CgroupDriver)
fmt.Fprintf(cli.out, "Plugins: \n")
fmt.Fprintf(cli.out, " Volume:")
fmt.Fprintf(cli.out, " %s", strings.Join(info.Plugins.Volume, " "))
fmt.Fprintf(cli.out, "\n")
fmt.Fprintf(cli.out, " Network:")
fmt.Fprintf(cli.out, " %s", strings.Join(info.Plugins.Network, " "))
fmt.Fprintf(cli.out, "\n")
if len(info.Plugins.Authorization) != 0 {
fmt.Fprintf(cli.out, " Authorization:")
fmt.Fprintf(cli.out, " %s", strings.Join(info.Plugins.Authorization, " "))
fmt.Fprintf(cli.out, "\n")
}
ioutils.FprintfIfNotEmpty(cli.out, "Kernel Version: %s\n", info.KernelVersion)
ioutils.FprintfIfNotEmpty(cli.out, "Operating System: %s\n", info.OperatingSystem)
ioutils.FprintfIfNotEmpty(cli.out, "OSType: %s\n", info.OSType)
ioutils.FprintfIfNotEmpty(cli.out, "Architecture: %s\n", info.Architecture)
fmt.Fprintf(cli.out, "CPUs: %d\n", info.NCPU)
fmt.Fprintf(cli.out, "Total Memory: %s\n", units.BytesSize(float64(info.MemTotal)))
ioutils.FprintfIfNotEmpty(cli.out, "Name: %s\n", info.Name)
ioutils.FprintfIfNotEmpty(cli.out, "ID: %s\n", info.ID)
fmt.Fprintf(cli.out, "Docker Root Dir: %s\n", info.DockerRootDir)
fmt.Fprintf(cli.out, "Debug mode (client): %v\n", utils.IsDebugEnabled())
fmt.Fprintf(cli.out, "Debug mode (server): %v\n", info.Debug)
if info.Debug {
fmt.Fprintf(cli.out, " File Descriptors: %d\n", info.NFd)
fmt.Fprintf(cli.out, " Goroutines: %d\n", info.NGoroutines)
fmt.Fprintf(cli.out, " System Time: %s\n", info.SystemTime)
fmt.Fprintf(cli.out, " EventsListeners: %d\n", info.NEventsListener)
fmt.Fprintf(cli.out, "Debug mode (server): %v\n", info.Debug)
fmt.Fprintf(cli.out, "File Descriptors: %d\n", info.NFd)
fmt.Fprintf(cli.out, "Goroutines: %d\n", info.NGoroutines)
fmt.Fprintf(cli.out, "System Time: %s\n", info.SystemTime)
fmt.Fprintf(cli.out, "EventsListeners: %d\n", info.NEventsListener)
fmt.Fprintf(cli.out, "Init SHA1: %s\n", info.InitSha1)
fmt.Fprintf(cli.out, "Init Path: %s\n", info.InitPath)
fmt.Fprintf(cli.out, "Docker Root Dir: %s\n", info.DockerRootDir)
}
ioutils.FprintfIfNotEmpty(cli.out, "Http Proxy: %s\n", info.HTTPProxy)
ioutils.FprintfIfNotEmpty(cli.out, "Https Proxy: %s\n", info.HTTPSProxy)
ioutils.FprintfIfNotEmpty(cli.out, "Http Proxy: %s\n", info.HttpProxy)
ioutils.FprintfIfNotEmpty(cli.out, "Https Proxy: %s\n", info.HttpsProxy)
ioutils.FprintfIfNotEmpty(cli.out, "No Proxy: %s\n", info.NoProxy)
if info.IndexServerAddress != "" {
u := cli.configFile.AuthConfigs[info.IndexServerAddress].Username
if len(u) > 0 {
fmt.Fprintf(cli.out, "Username: %v\n", u)
fmt.Fprintf(cli.out, "Registry: %v\n", info.IndexServerAddress)
}
fmt.Fprintf(cli.out, "Registry: %v\n", info.IndexServerAddress)
}
// Only output these warnings if the server does not support these features
if info.OSType != "windows" {
if !info.MemoryLimit {
fmt.Fprintln(cli.err, "WARNING: No memory limit support")
}
if !info.SwapLimit {
fmt.Fprintln(cli.err, "WARNING: No swap limit support")
}
if !info.KernelMemory {
fmt.Fprintln(cli.err, "WARNING: No kernel memory limit support")
}
if !info.OomKillDisable {
fmt.Fprintln(cli.err, "WARNING: No oom kill disable support")
}
if !info.CPUCfsQuota {
fmt.Fprintln(cli.err, "WARNING: No cpu cfs quota support")
}
if !info.CPUCfsPeriod {
fmt.Fprintln(cli.err, "WARNING: No cpu cfs period support")
}
if !info.CPUShares {
fmt.Fprintln(cli.err, "WARNING: No cpu shares support")
}
if !info.CPUSet {
fmt.Fprintln(cli.err, "WARNING: No cpuset support")
}
if !info.IPv4Forwarding {
fmt.Fprintln(cli.err, "WARNING: IPv4 forwarding is disabled")
}
if !info.BridgeNfIptables {
fmt.Fprintln(cli.err, "WARNING: bridge-nf-call-iptables is disabled")
}
if !info.BridgeNfIP6tables {
fmt.Fprintln(cli.err, "WARNING: bridge-nf-call-ip6tables is disabled")
// Only output these warnings if the server supports these features
if h, err := httputils.ParseServerHeader(serverResp.header.Get("Server")); err == nil {
if h.OS != "windows" {
if !info.MemoryLimit {
fmt.Fprintf(cli.err, "WARNING: No memory limit support\n")
}
if !info.SwapLimit {
fmt.Fprintf(cli.err, "WARNING: No swap limit support\n")
}
if !info.IPv4Forwarding {
fmt.Fprintf(cli.err, "WARNING: IPv4 forwarding is disabled.\n")
}
if !info.BridgeNfIptables {
fmt.Fprintf(cli.err, "WARNING: bridge-nf-call-iptables is disabled\n")
}
if !info.BridgeNfIp6tables {
fmt.Fprintf(cli.err, "WARNING: bridge-nf-call-ip6tables is disabled\n")
}
}
}
@@ -143,13 +100,9 @@ func (cli *DockerCli) CmdInfo(args ...string) error {
}
}
ioutils.FprintfIfTrue(cli.out, "Experimental: %v\n", info.ExperimentalBuild)
if info.ClusterStore != "" {
fmt.Fprintf(cli.out, "Cluster store: %s\n", info.ClusterStore)
if info.ExperimentalBuild {
fmt.Fprintf(cli.out, "Experimental: true\n")
}
if info.ClusterAdvertise != "" {
fmt.Fprintf(cli.out, "Cluster advertise: %s\n", info.ClusterAdvertise)
}
return nil
}

View File

@@ -1,127 +1,157 @@
package client
import (
"bytes"
"encoding/json"
"fmt"
"io"
"strings"
"text/template"
"golang.org/x/net/context"
"github.com/docker/docker/api/client/inspect"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/utils/templates"
"github.com/docker/engine-api/client"
)
var funcMap = template.FuncMap{
"json": func(v interface{}) string {
a, _ := json.Marshal(v)
return string(a)
},
}
// CmdInspect displays low-level information on one or more containers or images.
//
// Usage: docker inspect [OPTIONS] CONTAINER|IMAGE [CONTAINER|IMAGE...]
func (cli *DockerCli) CmdInspect(args ...string) error {
cmd := Cli.Subcmd("inspect", []string{"CONTAINER|IMAGE [CONTAINER|IMAGE...]"}, Cli.DockerCommands["inspect"].Description, true)
tmplStr := cmd.String([]string{"f", "-format"}, "", "Format the output using the given go template")
cmd := Cli.Subcmd("inspect", []string{"CONTAINER|IMAGE [CONTAINER|IMAGE...]"}, "Return low-level information on a container or image", true)
tmplStr := cmd.String([]string{"f", "#format", "-format"}, "", "Format the output using the given go template")
inspectType := cmd.String([]string{"-type"}, "", "Return JSON for specified type, (e.g image or container)")
size := cmd.Bool([]string{"s", "-size"}, false, "Display total file sizes if the type is container")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var tmpl *template.Template
var err error
var obj []byte
if *tmplStr != "" {
if tmpl, err = template.New("").Funcs(funcMap).Parse(*tmplStr); err != nil {
return Cli.StatusError{StatusCode: 64,
Status: "Template parsing error: " + err.Error()}
}
}
if *inspectType != "" && *inspectType != "container" && *inspectType != "image" {
return fmt.Errorf("%q is not a valid value for --type", *inspectType)
}
var elementSearcher inspectSearcher
switch *inspectType {
case "container":
elementSearcher = cli.inspectContainers(*size)
case "image":
elementSearcher = cli.inspectImages(*size)
default:
elementSearcher = cli.inspectAll(*size)
}
indented := new(bytes.Buffer)
indented.WriteString("[\n")
status := 0
isImage := false
return cli.inspectElements(*tmplStr, cmd.Args(), elementSearcher)
}
for _, name := range cmd.Args() {
func (cli *DockerCli) inspectContainers(getSize bool) inspectSearcher {
return func(ref string) (interface{}, []byte, error) {
return cli.client.ContainerInspectWithRaw(context.Background(), ref, getSize)
}
}
func (cli *DockerCli) inspectImages(getSize bool) inspectSearcher {
return func(ref string) (interface{}, []byte, error) {
return cli.client.ImageInspectWithRaw(context.Background(), ref, getSize)
}
}
func (cli *DockerCli) inspectAll(getSize bool) inspectSearcher {
return func(ref string) (interface{}, []byte, error) {
c, rawContainer, err := cli.client.ContainerInspectWithRaw(context.Background(), ref, getSize)
if err != nil {
// Search for image with that id if a container doesn't exist.
if client.IsErrContainerNotFound(err) {
i, rawImage, err := cli.client.ImageInspectWithRaw(context.Background(), ref, getSize)
if err != nil {
if client.IsErrImageNotFound(err) {
return nil, nil, fmt.Errorf("Error: No such image or container: %s", ref)
}
return nil, nil, err
if *inspectType == "" || *inspectType == "container" {
obj, _, err = readBody(cli.call("GET", "/containers/"+name+"/json", nil, nil))
if err != nil && *inspectType == "container" {
if strings.Contains(err.Error(), "No such") {
fmt.Fprintf(cli.err, "Error: No such container: %s\n", name)
} else {
fmt.Fprintf(cli.err, "%s", err)
}
return i, rawImage, err
status = 1
continue
}
return nil, nil, err
}
return c, rawContainer, err
}
}
type inspectSearcher func(ref string) (interface{}, []byte, error)
func (cli *DockerCli) inspectElements(tmplStr string, references []string, searchByReference inspectSearcher) error {
elementInspector, err := cli.newInspectorWithTemplate(tmplStr)
if err != nil {
return Cli.StatusError{StatusCode: 64, Status: err.Error()}
}
var inspectErr error
for _, ref := range references {
element, raw, err := searchByReference(ref)
if err != nil {
inspectErr = err
break
}
if err := elementInspector.Inspect(element, raw); err != nil {
inspectErr = err
break
if obj == nil && (*inspectType == "" || *inspectType == "image") {
obj, _, err = readBody(cli.call("GET", "/images/"+name+"/json", nil, nil))
isImage = true
if err != nil {
if strings.Contains(err.Error(), "No such") {
if *inspectType == "" {
fmt.Fprintf(cli.err, "Error: No such image or container: %s\n", name)
} else {
fmt.Fprintf(cli.err, "Error: No such image: %s\n", name)
}
} else {
fmt.Fprintf(cli.err, "%s", err)
}
status = 1
continue
}
}
if tmpl == nil {
if err := json.Indent(indented, obj, "", " "); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
continue
}
} else {
rdr := bytes.NewReader(obj)
dec := json.NewDecoder(rdr)
if isImage {
inspPtr := types.ImageInspect{}
if err := dec.Decode(&inspPtr); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
continue
}
if err := tmpl.Execute(cli.out, inspPtr); err != nil {
rdr.Seek(0, 0)
var raw interface{}
if err := dec.Decode(&raw); err != nil {
return err
}
if err = tmpl.Execute(cli.out, raw); err != nil {
return err
}
}
} else {
inspPtr := types.ContainerJSON{}
if err := dec.Decode(&inspPtr); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
continue
}
if err := tmpl.Execute(cli.out, inspPtr); err != nil {
rdr.Seek(0, 0)
var raw interface{}
if err := dec.Decode(&raw); err != nil {
return err
}
if err = tmpl.Execute(cli.out, raw); err != nil {
return err
}
}
}
cli.out.Write([]byte{'\n'})
}
indented.WriteString(",")
}
if indented.Len() > 1 {
// Remove trailing ','
indented.Truncate(indented.Len() - 1)
}
indented.WriteString("]\n")
if tmpl == nil {
// Note that we will always write "[]" when "-f" isn't specified,
// to make sure the output would always be array, see
// https://github.com/docker/docker/pull/9500#issuecomment-65846734
if _, err := io.Copy(cli.out, indented); err != nil {
return err
}
}
if err := elementInspector.Flush(); err != nil {
cli.inspectErrorStatus(err)
}
if status := cli.inspectErrorStatus(inspectErr); status != 0 {
if status != 0 {
return Cli.StatusError{StatusCode: status}
}
return nil
}
func (cli *DockerCli) inspectErrorStatus(err error) (status int) {
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
}
return
}
func (cli *DockerCli) newInspectorWithTemplate(tmplStr string) (inspect.Inspector, error) {
elementInspector := inspect.NewIndentedInspector(cli.out)
if tmplStr != "" {
tmpl, err := templates.Parse(tmplStr)
if err != nil {
return nil, fmt.Errorf("Template parsing error: %s", err)
}
elementInspector = inspect.NewTemplateInspector(cli.out, tmpl)
}
return elementInspector, nil
}

View File

@@ -1,119 +0,0 @@
package inspect
import (
"bytes"
"encoding/json"
"fmt"
"io"
"text/template"
)
// Inspector defines an interface to implement to process elements
type Inspector interface {
Inspect(typedElement interface{}, rawElement []byte) error
Flush() error
}
// TemplateInspector uses a text template to inspect elements.
type TemplateInspector struct {
outputStream io.Writer
buffer *bytes.Buffer
tmpl *template.Template
}
// NewTemplateInspector creates a new inspector with a template.
func NewTemplateInspector(outputStream io.Writer, tmpl *template.Template) Inspector {
return &TemplateInspector{
outputStream: outputStream,
buffer: new(bytes.Buffer),
tmpl: tmpl,
}
}
// Inspect executes the inspect template.
// It decodes the raw element into a map if the initial execution fails.
// This allows docker cli to parse inspect structs injected with Swarm fields.
func (i *TemplateInspector) Inspect(typedElement interface{}, rawElement []byte) error {
buffer := new(bytes.Buffer)
if err := i.tmpl.Execute(buffer, typedElement); err != nil {
if rawElement == nil {
return fmt.Errorf("Template parsing error: %v", err)
}
return i.tryRawInspectFallback(rawElement, err)
}
i.buffer.Write(buffer.Bytes())
i.buffer.WriteByte('\n')
return nil
}
// Flush write the result of inspecting all elements into the output stream.
func (i *TemplateInspector) Flush() error {
if i.buffer.Len() == 0 {
_, err := io.WriteString(i.outputStream, "\n")
return err
}
_, err := io.Copy(i.outputStream, i.buffer)
return err
}
// IndentedInspector uses a buffer to stop the indented representation of an element.
type IndentedInspector struct {
outputStream io.Writer
elements []interface{}
rawElements [][]byte
}
// NewIndentedInspector generates a new IndentedInspector.
func NewIndentedInspector(outputStream io.Writer) Inspector {
return &IndentedInspector{
outputStream: outputStream,
}
}
// Inspect writes the raw element with an indented json format.
func (i *IndentedInspector) Inspect(typedElement interface{}, rawElement []byte) error {
if rawElement != nil {
i.rawElements = append(i.rawElements, rawElement)
} else {
i.elements = append(i.elements, typedElement)
}
return nil
}
// Flush write the result of inspecting all elements into the output stream.
func (i *IndentedInspector) Flush() error {
if len(i.elements) == 0 && len(i.rawElements) == 0 {
_, err := io.WriteString(i.outputStream, "[]\n")
return err
}
var buffer io.Reader
if len(i.rawElements) > 0 {
bytesBuffer := new(bytes.Buffer)
bytesBuffer.WriteString("[")
for idx, r := range i.rawElements {
bytesBuffer.Write(r)
if idx < len(i.rawElements)-1 {
bytesBuffer.WriteString(",")
}
}
bytesBuffer.WriteString("]")
indented := new(bytes.Buffer)
if err := json.Indent(indented, bytesBuffer.Bytes(), "", " "); err != nil {
return err
}
buffer = indented
} else {
b, err := json.MarshalIndent(i.elements, "", " ")
if err != nil {
return err
}
buffer = bytes.NewReader(b)
}
if _, err := io.Copy(i.outputStream, buffer); err != nil {
return err
}
_, err := io.WriteString(i.outputStream, "\n")
return err
}

View File

@@ -1,40 +0,0 @@
// +build !go1.5
package inspect
import (
"bytes"
"encoding/json"
"fmt"
"strings"
)
// tryeRawInspectFallback executes the inspect template with a raw interface.
// This allows docker cli to parse inspect structs injected with Swarm fields.
// Unfortunately, go 1.4 doesn't fail executing invalid templates when the input is an interface.
// It doesn't allow to modify this behavior either, sending <no value> messages to the output.
// We assume that the template is invalid when there is a <no value>, if the template was valid
// we'd get <nil> or "" values. In that case we fail with the original error raised executing the
// template with the typed input.
func (i *TemplateInspector) tryRawInspectFallback(rawElement []byte, originalErr error) error {
var raw interface{}
buffer := new(bytes.Buffer)
rdr := bytes.NewReader(rawElement)
dec := json.NewDecoder(rdr)
if rawErr := dec.Decode(&raw); rawErr != nil {
return fmt.Errorf("unable to read inspect data: %v", rawErr)
}
if rawErr := i.tmpl.Execute(buffer, raw); rawErr != nil {
return fmt.Errorf("Template parsing error: %v", rawErr)
}
if strings.Contains(buffer.String(), "<no value>") {
return fmt.Errorf("Template parsing error: %v", originalErr)
}
i.buffer.Write(buffer.Bytes())
i.buffer.WriteByte('\n')
return nil
}

View File

@@ -1,29 +0,0 @@
// +build go1.5
package inspect
import (
"bytes"
"encoding/json"
"fmt"
)
func (i *TemplateInspector) tryRawInspectFallback(rawElement []byte, _ error) error {
var raw interface{}
buffer := new(bytes.Buffer)
rdr := bytes.NewReader(rawElement)
dec := json.NewDecoder(rdr)
if rawErr := dec.Decode(&raw); rawErr != nil {
return fmt.Errorf("unable to read inspect data: %v", rawErr)
}
tmplMissingKey := i.tmpl.Option("missingkey=error")
if rawErr := tmplMissingKey.Execute(buffer, raw); rawErr != nil {
return fmt.Errorf("Template parsing error: %v", rawErr)
}
i.buffer.Write(buffer.Bytes())
i.buffer.WriteByte('\n')
return nil
}

View File

@@ -1,221 +0,0 @@
package inspect
import (
"bytes"
"strings"
"testing"
"github.com/docker/docker/utils/templates"
)
type testElement struct {
DNS string `json:"Dns"`
}
func TestTemplateInspectorDefault(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.DNS}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
if err := i.Inspect(testElement{"0.0.0.0"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
if b.String() != "0.0.0.0\n" {
t.Fatalf("Expected `0.0.0.0\\n`, got `%s`", b.String())
}
}
func TestTemplateInspectorEmpty(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.DNS}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
if err := i.Flush(); err != nil {
t.Fatal(err)
}
if b.String() != "\n" {
t.Fatalf("Expected `\\n`, got `%s`", b.String())
}
}
func TestTemplateInspectorTemplateError(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.Foo}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
err = i.Inspect(testElement{"0.0.0.0"}, nil)
if err == nil {
t.Fatal("Expected error got nil")
}
if !strings.HasPrefix(err.Error(), "Template parsing error") {
t.Fatalf("Expected template error, got %v", err)
}
}
func TestTemplateInspectorRawFallback(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.Dns}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
if err := i.Inspect(testElement{"0.0.0.0"}, []byte(`{"Dns": "0.0.0.0"}`)); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
if b.String() != "0.0.0.0\n" {
t.Fatalf("Expected `0.0.0.0\\n`, got `%s`", b.String())
}
}
func TestTemplateInspectorRawFallbackError(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.Dns}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
err = i.Inspect(testElement{"0.0.0.0"}, []byte(`{"Foo": "0.0.0.0"}`))
if err == nil {
t.Fatal("Expected error got nil")
}
if !strings.HasPrefix(err.Error(), "Template parsing error") {
t.Fatalf("Expected template error, got %v", err)
}
}
func TestTemplateInspectorMultiple(t *testing.T) {
b := new(bytes.Buffer)
tmpl, err := templates.Parse("{{.DNS}}")
if err != nil {
t.Fatal(err)
}
i := NewTemplateInspector(b, tmpl)
if err := i.Inspect(testElement{"0.0.0.0"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Inspect(testElement{"1.1.1.1"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
if b.String() != "0.0.0.0\n1.1.1.1\n" {
t.Fatalf("Expected `0.0.0.0\\n1.1.1.1\\n`, got `%s`", b.String())
}
}
func TestIndentedInspectorDefault(t *testing.T) {
b := new(bytes.Buffer)
i := NewIndentedInspector(b)
if err := i.Inspect(testElement{"0.0.0.0"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
expected := `[
{
"Dns": "0.0.0.0"
}
]
`
if b.String() != expected {
t.Fatalf("Expected `%s`, got `%s`", expected, b.String())
}
}
func TestIndentedInspectorMultiple(t *testing.T) {
b := new(bytes.Buffer)
i := NewIndentedInspector(b)
if err := i.Inspect(testElement{"0.0.0.0"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Inspect(testElement{"1.1.1.1"}, nil); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
expected := `[
{
"Dns": "0.0.0.0"
},
{
"Dns": "1.1.1.1"
}
]
`
if b.String() != expected {
t.Fatalf("Expected `%s`, got `%s`", expected, b.String())
}
}
func TestIndentedInspectorEmpty(t *testing.T) {
b := new(bytes.Buffer)
i := NewIndentedInspector(b)
if err := i.Flush(); err != nil {
t.Fatal(err)
}
expected := "[]\n"
if b.String() != expected {
t.Fatalf("Expected `%s`, got `%s`", expected, b.String())
}
}
func TestIndentedInspectorRawElements(t *testing.T) {
b := new(bytes.Buffer)
i := NewIndentedInspector(b)
if err := i.Inspect(testElement{"0.0.0.0"}, []byte(`{"Dns": "0.0.0.0", "Node": "0"}`)); err != nil {
t.Fatal(err)
}
if err := i.Inspect(testElement{"1.1.1.1"}, []byte(`{"Dns": "1.1.1.1", "Node": "1"}`)); err != nil {
t.Fatal(err)
}
if err := i.Flush(); err != nil {
t.Fatal(err)
}
expected := `[
{
"Dns": "0.0.0.0",
"Node": "0"
},
{
"Dns": "1.1.1.1",
"Node": "1"
}
]
`
if b.String() != expected {
t.Fatalf("Expected `%s`, got `%s`", expected, b.String())
}
}

View File

@@ -2,9 +2,6 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -14,22 +11,23 @@ import (
//
// Usage: docker kill [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdKill(args ...string) error {
cmd := Cli.Subcmd("kill", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["kill"].Description, true)
cmd := Cli.Subcmd("kill", []string{"CONTAINER [CONTAINER...]"}, "Kill a running container using SIGKILL or a specified signal", true)
signal := cmd.String([]string{"s", "-signal"}, "KILL", "Signal to send to the container")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
var errNames []string
for _, name := range cmd.Args() {
if err := cli.client.ContainerKill(context.Background(), name, *signal); err != nil {
errs = append(errs, err.Error())
if _, _, err := readBody(cli.call("POST", fmt.Sprintf("/containers/%s/kill?signal=%s", name, *signal), nil, nil)); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to kill containers: %v", errNames)
}
return nil
}

View File

@@ -4,10 +4,7 @@ import (
"io"
"os"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
)
@@ -17,34 +14,29 @@ import (
//
// Usage: docker load [OPTIONS]
func (cli *DockerCli) CmdLoad(args ...string) error {
cmd := Cli.Subcmd("load", nil, Cli.DockerCommands["load"].Description, true)
cmd := Cli.Subcmd("load", nil, "Load an image from a tar archive or STDIN", true)
infile := cmd.String([]string{"i", "-input"}, "", "Read from a tar archive file, instead of STDIN")
quiet := cmd.Bool([]string{"q", "-quiet"}, false, "Suppress the load output")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
var input io.Reader = cli.in
var (
input io.Reader = cli.in
err error
)
if *infile != "" {
file, err := os.Open(*infile)
input, err = os.Open(*infile)
if err != nil {
return err
}
defer file.Close()
input = file
}
if !cli.isTerminalOut {
*quiet = true
sopts := &streamOpts{
rawTerminal: true,
in: input,
out: cli.out,
}
response, err := cli.client.ImageLoad(context.Background(), input, *quiet)
if err != nil {
if _, err := cli.stream("POST", "/images/load", sopts); err != nil {
return err
}
defer response.Body.Close()
if response.Body != nil && response.JSON {
return jsonmessage.DisplayJSONMessagesStream(response.Body, cli.out, cli.outFd, cli.isTerminalOut, nil)
}
_, err = io.Copy(cli.out, response.Body)
return err
return nil
}

View File

@@ -2,176 +2,146 @@ package client
import (
"bufio"
"encoding/json"
"fmt"
"io"
"os"
"runtime"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/cliconfig/credentials"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/term"
"github.com/docker/engine-api/types"
"github.com/docker/docker/registry"
)
// CmdLogin logs in a user to a Docker registry service.
// CmdLogin logs in or registers a user to a Docker registry service.
//
// If no server is specified, the user will be logged into or registered to the registry's index server.
//
// Usage: docker login SERVER
func (cli *DockerCli) CmdLogin(args ...string) error {
cmd := Cli.Subcmd("login", []string{"[SERVER]"}, Cli.DockerCommands["login"].Description+".\nIf no server is specified, the default is defined by the daemon.", true)
cmd := Cli.Subcmd("login", []string{"[SERVER]"}, "Register or log in to a Docker registry server, if no server is\nspecified \""+registry.IndexServer+"\" is the default.", true)
cmd.Require(flag.Max, 1)
flUser := cmd.String([]string{"u", "-username"}, "", "Username")
flPassword := cmd.String([]string{"p", "-password"}, "", "Password")
var username, password, email string
// Deprecated in 1.11: Should be removed in docker 1.13
cmd.String([]string{"#e", "#-email"}, "", "Email")
cmd.StringVar(&username, []string{"u", "-username"}, "", "Username")
cmd.StringVar(&password, []string{"p", "-password"}, "", "Password")
cmd.StringVar(&email, []string{"e", "-email"}, "", "Email")
cmd.ParseFlags(args, true)
// On Windows, force the use of the regular OS stdin stream. Fixes #14336/#14210
if runtime.GOOS == "windows" {
cli.in = os.Stdin
}
var serverAddress string
var isDefaultRegistry bool
serverAddress := registry.IndexServer
if len(cmd.Args()) > 0 {
serverAddress = cmd.Arg(0)
}
promptDefault := func(prompt string, configDefault string) {
if configDefault == "" {
fmt.Fprintf(cli.out, "%s: ", prompt)
} else {
fmt.Fprintf(cli.out, "%s (%s): ", prompt, configDefault)
}
}
readInput := func(in io.Reader, out io.Writer) string {
reader := bufio.NewReader(in)
line, _, err := reader.ReadLine()
if err != nil {
fmt.Fprintln(out, err.Error())
os.Exit(1)
}
return string(line)
}
authconfig, ok := cli.configFile.AuthConfigs[serverAddress]
if !ok {
authconfig = cliconfig.AuthConfig{}
}
if username == "" {
promptDefault("Username", authconfig.Username)
username = readInput(cli.in, cli.out)
username = strings.Trim(username, " ")
if username == "" {
username = authconfig.Username
}
}
// Assume that a different username means they may not want to use
// the password or email from the config file, so prompt them
if username != authconfig.Username {
if password == "" {
oldState, err := term.SaveState(cli.inFd)
if err != nil {
return err
}
fmt.Fprintf(cli.out, "Password: ")
term.DisableEcho(cli.inFd, oldState)
password = readInput(cli.in, cli.out)
fmt.Fprint(cli.out, "\n")
term.RestoreTerminal(cli.inFd, oldState)
if password == "" {
return fmt.Errorf("Error : Password Required")
}
}
if email == "" {
promptDefault("Email", authconfig.Email)
email = readInput(cli.in, cli.out)
if email == "" {
email = authconfig.Email
}
}
} else {
serverAddress = cli.electAuthServer()
isDefaultRegistry = true
// However, if they don't override the username use the
// password or email from the cmd line if specified. IOW, allow
// then to change/override them. And if not specified, just
// use what's in the config file
if password == "" {
password = authconfig.Password
}
if email == "" {
email = authconfig.Email
}
}
authconfig.Username = username
authconfig.Password = password
authconfig.Email = email
authconfig.ServerAddress = serverAddress
cli.configFile.AuthConfigs[serverAddress] = authconfig
authConfig, err := cli.configureAuth(*flUser, *flPassword, serverAddress, isDefaultRegistry)
serverResp, err := cli.call("POST", "/auth", cli.configFile.AuthConfigs[serverAddress], nil)
if serverResp.statusCode == 401 {
delete(cli.configFile.AuthConfigs, serverAddress)
if err2 := cli.configFile.Save(); err2 != nil {
fmt.Fprintf(cli.out, "WARNING: could not save config file: %v\n", err2)
}
return err
}
if err != nil {
return err
}
response, err := cli.client.RegistryLogin(context.Background(), authConfig)
if err != nil {
defer serverResp.body.Close()
var response types.AuthResponse
if err := json.NewDecoder(serverResp.body).Decode(&response); err != nil {
// Upon error, remove entry
delete(cli.configFile.AuthConfigs, serverAddress)
return err
}
if response.IdentityToken != "" {
authConfig.Password = ""
authConfig.IdentityToken = response.IdentityToken
}
if err := storeCredentials(cli.configFile, authConfig); err != nil {
return fmt.Errorf("Error saving credentials: %v", err)
if err := cli.configFile.Save(); err != nil {
return fmt.Errorf("Error saving config file: %v", err)
}
fmt.Fprintf(cli.out, "WARNING: login credentials saved in %s\n", cli.configFile.Filename())
if response.Status != "" {
fmt.Fprintln(cli.out, response.Status)
fmt.Fprintf(cli.out, "%s\n", response.Status)
}
return nil
}
func (cli *DockerCli) promptWithDefault(prompt string, configDefault string) {
if configDefault == "" {
fmt.Fprintf(cli.out, "%s: ", prompt)
} else {
fmt.Fprintf(cli.out, "%s (%s): ", prompt, configDefault)
}
}
func (cli *DockerCli) configureAuth(flUser, flPassword, serverAddress string, isDefaultRegistry bool) (types.AuthConfig, error) {
authconfig, err := getCredentials(cli.configFile, serverAddress)
if err != nil {
return authconfig, err
}
authconfig.Username = strings.TrimSpace(authconfig.Username)
if flUser = strings.TrimSpace(flUser); flUser == "" {
if isDefaultRegistry {
// if this is a defauly registry (docker hub), then display the following message.
fmt.Fprintln(cli.out, "Login with your Docker ID to push and pull images from Docker Hub. If you don't have a Docker ID, head over to https://hub.docker.com to create one.")
}
cli.promptWithDefault("Username", authconfig.Username)
flUser = readInput(cli.in, cli.out)
flUser = strings.TrimSpace(flUser)
if flUser == "" {
flUser = authconfig.Username
}
}
if flUser == "" {
return authconfig, fmt.Errorf("Error: Non-null Username Required")
}
if flPassword == "" {
oldState, err := term.SaveState(cli.inFd)
if err != nil {
return authconfig, err
}
fmt.Fprintf(cli.out, "Password: ")
term.DisableEcho(cli.inFd, oldState)
flPassword = readInput(cli.in, cli.out)
fmt.Fprint(cli.out, "\n")
term.RestoreTerminal(cli.inFd, oldState)
if flPassword == "" {
return authconfig, fmt.Errorf("Error: Password Required")
}
}
authconfig.Username = flUser
authconfig.Password = flPassword
authconfig.ServerAddress = serverAddress
authconfig.IdentityToken = ""
return authconfig, nil
}
func readInput(in io.Reader, out io.Writer) string {
reader := bufio.NewReader(in)
line, _, err := reader.ReadLine()
if err != nil {
fmt.Fprintln(out, err.Error())
os.Exit(1)
}
return string(line)
}
// getCredentials loads the user credentials from a credentials store.
// The store is determined by the config file settings.
func getCredentials(c *cliconfig.ConfigFile, serverAddress string) (types.AuthConfig, error) {
s := loadCredentialsStore(c)
return s.Get(serverAddress)
}
func getAllCredentials(c *cliconfig.ConfigFile) (map[string]types.AuthConfig, error) {
s := loadCredentialsStore(c)
return s.GetAll()
}
// storeCredentials saves the user credentials in a credentials store.
// The store is determined by the config file settings.
func storeCredentials(c *cliconfig.ConfigFile, auth types.AuthConfig) error {
s := loadCredentialsStore(c)
return s.Store(auth)
}
// eraseCredentials removes the user credentials from a credentials store.
// The store is determined by the config file settings.
func eraseCredentials(c *cliconfig.ConfigFile, serverAddress string) error {
s := loadCredentialsStore(c)
return s.Erase(serverAddress)
}
// loadCredentialsStore initializes a new credentials store based
// in the settings provided in the configuration file.
func loadCredentialsStore(c *cliconfig.ConfigFile) credentials.Store {
if c.CredentialsStore != "" {
return credentials.NewNativeStore(c)
}
return credentials.NewFileStore(c)
}

View File

@@ -5,6 +5,7 @@ import (
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/registry"
)
// CmdLogout logs a user out from a Docker registry.
@@ -13,29 +14,25 @@ import (
//
// Usage: docker logout [SERVER]
func (cli *DockerCli) CmdLogout(args ...string) error {
cmd := Cli.Subcmd("logout", []string{"[SERVER]"}, Cli.DockerCommands["logout"].Description+".\nIf no server is specified, the default is defined by the daemon.", true)
cmd := Cli.Subcmd("logout", []string{"[SERVER]"}, "Log out from a Docker registry, if no server is\nspecified \""+registry.IndexServer+"\" is the default.", true)
cmd.Require(flag.Max, 1)
cmd.ParseFlags(args, true)
var serverAddress string
serverAddress := registry.IndexServer
if len(cmd.Args()) > 0 {
serverAddress = cmd.Arg(0)
} else {
serverAddress = cli.electAuthServer()
}
// check if we're logged in based on the records in the config file
// which means it couldn't have user/pass cause they may be in the creds store
if _, ok := cli.configFile.AuthConfigs[serverAddress]; !ok {
fmt.Fprintf(cli.out, "Not logged in to %s\n", serverAddress)
return nil
}
} else {
fmt.Fprintf(cli.out, "Remove login credentials for %s\n", serverAddress)
delete(cli.configFile.AuthConfigs, serverAddress)
fmt.Fprintf(cli.out, "Remove login credentials for %s\n", serverAddress)
if err := eraseCredentials(cli.configFile, serverAddress); err != nil {
fmt.Fprintf(cli.out, "WARNING: could not erase credentials: %v\n", err)
if err := cli.configFile.Save(); err != nil {
return fmt.Errorf("Failed to save docker config: %v", err)
}
}
return nil
}

View File

@@ -1,27 +1,22 @@
package client
import (
"encoding/json"
"fmt"
"io"
"golang.org/x/net/context"
"net/url"
"time"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/stdcopy"
"github.com/docker/engine-api/types"
"github.com/docker/docker/pkg/timeutils"
)
var validDrivers = map[string]bool{
"json-file": true,
"journald": true,
}
// CmdLogs fetches the logs of a given container.
//
// docker logs [OPTIONS] CONTAINER
func (cli *DockerCli) CmdLogs(args ...string) error {
cmd := Cli.Subcmd("logs", []string{"CONTAINER"}, Cli.DockerCommands["logs"].Description, true)
cmd := Cli.Subcmd("logs", []string{"CONTAINER"}, "Fetch the logs of a container", true)
follow := cmd.Bool([]string{"f", "-follow"}, false, "Follow log output")
since := cmd.String([]string{"-since"}, "", "Show logs since timestamp")
times := cmd.Bool([]string{"t", "-timestamps"}, false, "Show timestamps")
@@ -32,34 +27,43 @@ func (cli *DockerCli) CmdLogs(args ...string) error {
name := cmd.Arg(0)
c, err := cli.client.ContainerInspect(context.Background(), name)
serverResp, err := cli.call("GET", "/containers/"+name+"/json", nil, nil)
if err != nil {
return err
}
if !validDrivers[c.HostConfig.LogConfig.Type] {
return fmt.Errorf("\"logs\" command is supported only for \"json-file\" and \"journald\" logging drivers (got: %s)", c.HostConfig.LogConfig.Type)
}
options := types.ContainerLogsOptions{
ContainerID: name,
ShowStdout: true,
ShowStderr: true,
Since: *since,
Timestamps: *times,
Follow: *follow,
Tail: *tail,
}
responseBody, err := cli.client.ContainerLogs(context.Background(), options)
if err != nil {
var c types.ContainerJSON
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return err
}
defer responseBody.Close()
if c.Config.Tty {
_, err = io.Copy(cli.out, responseBody)
} else {
_, err = stdcopy.StdCopy(cli.out, cli.err, responseBody)
if logType := c.HostConfig.LogConfig.Type; logType != "json-file" {
return fmt.Errorf("\"logs\" command is supported only for \"json-file\" logging driver (got: %s)", logType)
}
v := url.Values{}
v.Set("stdout", "1")
v.Set("stderr", "1")
if *since != "" {
v.Set("since", timeutils.GetTimestamp(*since, time.Now()))
}
if *times {
v.Set("timestamps", "1")
}
if *follow {
v.Set("follow", "1")
}
v.Set("tail", *tail)
sopts := &streamOpts{
rawTerminal: c.Config.Tty,
out: cli.out,
err: cli.err,
}
_, err = cli.stream("GET", "/containers/"+name+"/logs?"+v.Encode(), sopts)
return err
}

View File

@@ -1,392 +1,15 @@
// +build experimental
package client
import (
"fmt"
"net"
"sort"
"strings"
"text/tabwriter"
"os"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/stringid"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/filters"
"github.com/docker/engine-api/types/network"
nwclient "github.com/docker/libnetwork/client"
)
// CmdNetwork is the parent subcommand for all network commands
//
// Usage: docker network <COMMAND> [OPTIONS]
func (cli *DockerCli) CmdNetwork(args ...string) error {
cmd := Cli.Subcmd("network", []string{"COMMAND [OPTIONS]"}, networkUsage(), false)
cmd.Require(flag.Min, 1)
err := cmd.ParseFlags(args, true)
cmd.Usage()
return err
}
// CmdNetworkCreate creates a new network with a given name
//
// Usage: docker network create [OPTIONS] <NETWORK-NAME>
func (cli *DockerCli) CmdNetworkCreate(args ...string) error {
cmd := Cli.Subcmd("network create", []string{"NETWORK-NAME"}, "Creates a new network with a name specified by the user", false)
flDriver := cmd.String([]string{"d", "-driver"}, "bridge", "Driver to manage the Network")
flOpts := opts.NewMapOpts(nil, nil)
flIpamDriver := cmd.String([]string{"-ipam-driver"}, "default", "IP Address Management Driver")
flIpamSubnet := opts.NewListOpts(nil)
flIpamIPRange := opts.NewListOpts(nil)
flIpamGateway := opts.NewListOpts(nil)
flIpamAux := opts.NewMapOpts(nil, nil)
flIpamOpt := opts.NewMapOpts(nil, nil)
flLabels := opts.NewListOpts(nil)
cmd.Var(&flIpamSubnet, []string{"-subnet"}, "subnet in CIDR format that represents a network segment")
cmd.Var(&flIpamIPRange, []string{"-ip-range"}, "allocate container ip from a sub-range")
cmd.Var(&flIpamGateway, []string{"-gateway"}, "ipv4 or ipv6 Gateway for the master subnet")
cmd.Var(flIpamAux, []string{"-aux-address"}, "auxiliary ipv4 or ipv6 addresses used by Network driver")
cmd.Var(flOpts, []string{"o", "-opt"}, "set driver specific options")
cmd.Var(flIpamOpt, []string{"-ipam-opt"}, "set IPAM driver specific options")
cmd.Var(&flLabels, []string{"-label"}, "set metadata on a network")
flInternal := cmd.Bool([]string{"-internal"}, false, "restricts external access to the network")
flIPv6 := cmd.Bool([]string{"-ipv6"}, false, "enable IPv6 networking")
cmd.Require(flag.Exact, 1)
err := cmd.ParseFlags(args, true)
if err != nil {
return err
}
// Set the default driver to "" if the user didn't set the value.
// That way we can know whether it was user input or not.
driver := *flDriver
if !cmd.IsSet("-driver") && !cmd.IsSet("d") {
driver = ""
}
ipamCfg, err := consolidateIpam(flIpamSubnet.GetAll(), flIpamIPRange.GetAll(), flIpamGateway.GetAll(), flIpamAux.GetAll())
if err != nil {
return err
}
// Construct network create request body
nc := types.NetworkCreate{
Name: cmd.Arg(0),
Driver: driver,
IPAM: network.IPAM{Driver: *flIpamDriver, Config: ipamCfg, Options: flIpamOpt.GetAll()},
Options: flOpts.GetAll(),
CheckDuplicate: true,
Internal: *flInternal,
EnableIPv6: *flIPv6,
Labels: runconfigopts.ConvertKVStringsToMap(flLabels.GetAll()),
}
resp, err := cli.client.NetworkCreate(context.Background(), nc)
if err != nil {
return err
}
fmt.Fprintf(cli.out, "%s\n", resp.ID)
return nil
}
// CmdNetworkRm deletes one or more networks
//
// Usage: docker network rm NETWORK-NAME|NETWORK-ID [NETWORK-NAME|NETWORK-ID...]
func (cli *DockerCli) CmdNetworkRm(args ...string) error {
cmd := Cli.Subcmd("network rm", []string{"NETWORK [NETWORK...]"}, "Deletes one or more networks", false)
cmd.Require(flag.Min, 1)
if err := cmd.ParseFlags(args, true); err != nil {
return err
}
status := 0
for _, net := range cmd.Args() {
if err := cli.client.NetworkRemove(context.Background(), net); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
continue
}
}
if status != 0 {
return Cli.StatusError{StatusCode: status}
}
return nil
}
// CmdNetworkConnect connects a container to a network
//
// Usage: docker network connect [OPTIONS] <NETWORK> <CONTAINER>
func (cli *DockerCli) CmdNetworkConnect(args ...string) error {
cmd := Cli.Subcmd("network connect", []string{"NETWORK CONTAINER"}, "Connects a container to a network", false)
flIPAddress := cmd.String([]string{"-ip"}, "", "IP Address")
flIPv6Address := cmd.String([]string{"-ip6"}, "", "IPv6 Address")
flLinks := opts.NewListOpts(runconfigopts.ValidateLink)
cmd.Var(&flLinks, []string{"-link"}, "Add link to another container")
flAliases := opts.NewListOpts(nil)
cmd.Var(&flAliases, []string{"-alias"}, "Add network-scoped alias for the container")
cmd.Require(flag.Min, 2)
if err := cmd.ParseFlags(args, true); err != nil {
return err
}
epConfig := &network.EndpointSettings{
IPAMConfig: &network.EndpointIPAMConfig{
IPv4Address: *flIPAddress,
IPv6Address: *flIPv6Address,
},
Links: flLinks.GetAll(),
Aliases: flAliases.GetAll(),
}
return cli.client.NetworkConnect(context.Background(), cmd.Arg(0), cmd.Arg(1), epConfig)
}
// CmdNetworkDisconnect disconnects a container from a network
//
// Usage: docker network disconnect <NETWORK> <CONTAINER>
func (cli *DockerCli) CmdNetworkDisconnect(args ...string) error {
cmd := Cli.Subcmd("network disconnect", []string{"NETWORK CONTAINER"}, "Disconnects container from a network", false)
force := cmd.Bool([]string{"f", "-force"}, false, "Force the container to disconnect from a network")
cmd.Require(flag.Exact, 2)
if err := cmd.ParseFlags(args, true); err != nil {
return err
}
return cli.client.NetworkDisconnect(context.Background(), cmd.Arg(0), cmd.Arg(1), *force)
}
// CmdNetworkLs lists all the networks managed by docker daemon
//
// Usage: docker network ls [OPTIONS]
func (cli *DockerCli) CmdNetworkLs(args ...string) error {
cmd := Cli.Subcmd("network ls", nil, "Lists networks", true)
quiet := cmd.Bool([]string{"q", "-quiet"}, false, "Only display numeric IDs")
noTrunc := cmd.Bool([]string{"-no-trunc"}, false, "Do not truncate the output")
flFilter := opts.NewListOpts(nil)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Filter output based on conditions provided")
cmd.Require(flag.Exact, 0)
err := cmd.ParseFlags(args, true)
if err != nil {
return err
}
// Consolidate all filter flags, and sanity check them early.
// They'll get process after get response from server.
netFilterArgs := filters.NewArgs()
for _, f := range flFilter.GetAll() {
if netFilterArgs, err = filters.ParseFlag(f, netFilterArgs); err != nil {
return err
}
}
options := types.NetworkListOptions{
Filters: netFilterArgs,
}
networkResources, err := cli.client.NetworkList(context.Background(), options)
if err != nil {
return err
}
wr := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
// unless quiet (-q) is specified, print field titles
if !*quiet {
fmt.Fprintln(wr, "NETWORK ID\tNAME\tDRIVER")
}
sort.Sort(byNetworkName(networkResources))
for _, networkResource := range networkResources {
ID := networkResource.ID
netName := networkResource.Name
if !*noTrunc {
ID = stringid.TruncateID(ID)
}
if *quiet {
fmt.Fprintln(wr, ID)
continue
}
driver := networkResource.Driver
fmt.Fprintf(wr, "%s\t%s\t%s\t",
ID,
netName,
driver)
fmt.Fprint(wr, "\n")
}
wr.Flush()
return nil
}
type byNetworkName []types.NetworkResource
func (r byNetworkName) Len() int { return len(r) }
func (r byNetworkName) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r byNetworkName) Less(i, j int) bool { return r[i].Name < r[j].Name }
// CmdNetworkInspect inspects the network object for more details
//
// Usage: docker network inspect [OPTIONS] <NETWORK> [NETWORK...]
func (cli *DockerCli) CmdNetworkInspect(args ...string) error {
cmd := Cli.Subcmd("network inspect", []string{"NETWORK [NETWORK...]"}, "Displays detailed information on one or more networks", false)
tmplStr := cmd.String([]string{"f", "-format"}, "", "Format the output using the given go template")
cmd.Require(flag.Min, 1)
if err := cmd.ParseFlags(args, true); err != nil {
return err
}
inspectSearcher := func(name string) (interface{}, []byte, error) {
i, err := cli.client.NetworkInspect(context.Background(), name)
return i, nil, err
}
return cli.inspectElements(*tmplStr, cmd.Args(), inspectSearcher)
}
// Consolidates the ipam configuration as a group from different related configurations
// user can configure network with multiple non-overlapping subnets and hence it is
// possible to correlate the various related parameters and consolidate them.
// consoidateIpam consolidates subnets, ip-ranges, gateways and auxiliary addresses into
// structured ipam data.
func consolidateIpam(subnets, ranges, gateways []string, auxaddrs map[string]string) ([]network.IPAMConfig, error) {
if len(subnets) < len(ranges) || len(subnets) < len(gateways) {
return nil, fmt.Errorf("every ip-range or gateway must have a corresponding subnet")
}
iData := map[string]*network.IPAMConfig{}
// Populate non-overlapping subnets into consolidation map
for _, s := range subnets {
for k := range iData {
ok1, err := subnetMatches(s, k)
if err != nil {
return nil, err
}
ok2, err := subnetMatches(k, s)
if err != nil {
return nil, err
}
if ok1 || ok2 {
return nil, fmt.Errorf("multiple overlapping subnet configuration is not supported")
}
}
iData[s] = &network.IPAMConfig{Subnet: s, AuxAddress: map[string]string{}}
}
// Validate and add valid ip ranges
for _, r := range ranges {
match := false
for _, s := range subnets {
ok, err := subnetMatches(s, r)
if err != nil {
return nil, err
}
if !ok {
continue
}
if iData[s].IPRange != "" {
return nil, fmt.Errorf("cannot configure multiple ranges (%s, %s) on the same subnet (%s)", r, iData[s].IPRange, s)
}
d := iData[s]
d.IPRange = r
match = true
}
if !match {
return nil, fmt.Errorf("no matching subnet for range %s", r)
}
}
// Validate and add valid gateways
for _, g := range gateways {
match := false
for _, s := range subnets {
ok, err := subnetMatches(s, g)
if err != nil {
return nil, err
}
if !ok {
continue
}
if iData[s].Gateway != "" {
return nil, fmt.Errorf("cannot configure multiple gateways (%s, %s) for the same subnet (%s)", g, iData[s].Gateway, s)
}
d := iData[s]
d.Gateway = g
match = true
}
if !match {
return nil, fmt.Errorf("no matching subnet for gateway %s", g)
}
}
// Validate and add aux-addresses
for key, aa := range auxaddrs {
match := false
for _, s := range subnets {
ok, err := subnetMatches(s, aa)
if err != nil {
return nil, err
}
if !ok {
continue
}
iData[s].AuxAddress[key] = aa
match = true
}
if !match {
return nil, fmt.Errorf("no matching subnet for aux-address %s", aa)
}
}
idl := []network.IPAMConfig{}
for _, v := range iData {
idl = append(idl, *v)
}
return idl, nil
}
func subnetMatches(subnet, data string) (bool, error) {
var (
ip net.IP
)
_, s, err := net.ParseCIDR(subnet)
if err != nil {
return false, fmt.Errorf("Invalid subnet %s : %v", s, err)
}
if strings.Contains(data, "/") {
ip, _, err = net.ParseCIDR(data)
if err != nil {
return false, fmt.Errorf("Invalid cidr %s : %v", data, err)
}
} else {
ip = net.ParseIP(data)
}
return s.Contains(ip), nil
}
func networkUsage() string {
networkCommands := map[string]string{
"create": "Create a network",
"connect": "Connect container to a network",
"disconnect": "Disconnect container from a network",
"inspect": "Display detailed network information",
"ls": "List all networks",
"rm": "Remove a network",
}
help := "Commands:\n"
for cmd, description := range networkCommands {
help += fmt.Sprintf(" %-25.25s%s\n", cmd, description)
}
help += fmt.Sprintf("\nRun 'docker network COMMAND --help' for more information on a command.")
return help
nCli := nwclient.NewNetworkCli(cli.out, cli.err, nwclient.CallFunc(cli.callWrapper))
args = append([]string{"network"}, args...)
return nCli.Cmd(os.Args[0], args...)
}

View File

@@ -2,9 +2,6 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -14,21 +11,22 @@ import (
//
// Usage: docker pause CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdPause(args ...string) error {
cmd := Cli.Subcmd("pause", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["pause"].Description, true)
cmd := Cli.Subcmd("pause", []string{"CONTAINER [CONTAINER...]"}, "Pause all processes within a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
var errNames []string
for _, name := range cmd.Args() {
if err := cli.client.ContainerPause(context.Background(), name); err != nil {
errs = append(errs, err.Error())
if _, _, err := readBody(cli.call("POST", fmt.Sprintf("/containers/%s/pause", name), nil, nil)); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to pause containers: %v", errNames)
}
return nil
}

View File

@@ -1,14 +1,13 @@
package client
import (
"encoding/json"
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/go-connections/nat"
"github.com/docker/docker/pkg/nat"
)
// CmdPort lists port mappings for a container.
@@ -16,16 +15,28 @@ import (
//
// Usage: docker port CONTAINER [PRIVATE_PORT[/PROTO]]
func (cli *DockerCli) CmdPort(args ...string) error {
cmd := Cli.Subcmd("port", []string{"CONTAINER [PRIVATE_PORT[/PROTO]]"}, Cli.DockerCommands["port"].Description, true)
cmd := Cli.Subcmd("port", []string{"CONTAINER [PRIVATE_PORT[/PROTO]]"}, "List port mappings for the CONTAINER, or lookup the public-facing port that\nis NAT-ed to the PRIVATE_PORT", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
c, err := cli.client.ContainerInspect(context.Background(), cmd.Arg(0))
serverResp, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/json", nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
var c struct {
NetworkSettings struct {
Ports nat.PortMap
}
}
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return err
}
if cmd.NArg() == 2 {
var (
port = cmd.Arg(1)

View File

@@ -1,14 +1,16 @@
package client
import (
"golang.org/x/net/context"
"encoding/json"
"net/url"
"strconv"
"github.com/docker/docker/api/client/formatter"
"github.com/docker/docker/api/client/ps"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/filters"
"github.com/docker/docker/pkg/parsers/filters"
)
// CmdPs outputs a list of Docker containers.
@@ -18,17 +20,18 @@ func (cli *DockerCli) CmdPs(args ...string) error {
var (
err error
psFilterArgs = filters.NewArgs()
psFilterArgs = filters.Args{}
v = url.Values{}
cmd = Cli.Subcmd("ps", nil, Cli.DockerCommands["ps"].Description, true)
cmd = Cli.Subcmd("ps", nil, "List containers", true)
quiet = cmd.Bool([]string{"q", "-quiet"}, false, "Only display numeric IDs")
size = cmd.Bool([]string{"s", "-size"}, false, "Display total file sizes")
all = cmd.Bool([]string{"a", "-all"}, false, "Show all containers (default shows just running)")
noTrunc = cmd.Bool([]string{"-no-trunc"}, false, "Don't truncate output")
nLatest = cmd.Bool([]string{"l", "-latest"}, false, "Show the latest created container (includes all states)")
since = cmd.String([]string{"#-since"}, "", "Show containers created since Id or Name (includes all states)")
before = cmd.String([]string{"#-before"}, "", "Only show containers created before Id or Name")
last = cmd.Int([]string{"n"}, -1, "Show n last created containers (includes all states)")
noTrunc = cmd.Bool([]string{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
nLatest = cmd.Bool([]string{"l", "-latest"}, false, "Show the latest created container, include non-running")
since = cmd.String([]string{"#sinceId", "#-since-id", "-since"}, "", "Show created since Id or Name, include non-running")
before = cmd.String([]string{"#beforeId", "#-before-id", "-before"}, "", "Show only container created before Id or Name")
last = cmd.Int([]string{"n"}, -1, "Show n last created containers, include non-running")
format = cmd.String([]string{"-format"}, "", "Pretty-print containers using a Go template")
flFilter = opts.NewListOpts(nil)
)
@@ -41,6 +44,26 @@ func (cli *DockerCli) CmdPs(args ...string) error {
*last = 1
}
if *all {
v.Set("all", "1")
}
if *last != -1 {
v.Set("limit", strconv.Itoa(*last))
}
if *since != "" {
v.Set("since", *since)
}
if *before != "" {
v.Set("before", *before)
}
if *size {
v.Set("size", "1")
}
// Consolidate all filter flags, and sanity check them.
// They'll get processed in the daemon/server.
for _, f := range flFilter.GetAll() {
@@ -49,20 +72,27 @@ func (cli *DockerCli) CmdPs(args ...string) error {
}
}
options := types.ContainerListOptions{
All: *all,
Limit: *last,
Since: *since,
Before: *before,
Size: *size,
Filter: psFilterArgs,
if len(psFilterArgs) > 0 {
filterJSON, err := filters.ToParam(psFilterArgs)
if err != nil {
return err
}
v.Set("filters", filterJSON)
}
containers, err := cli.client.ContainerList(context.Background(), options)
serverResp, err := cli.call("GET", "/containers/json?"+v.Encode(), nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
containers := []types.Container{}
if err := json.NewDecoder(serverResp.body).Decode(&containers); err != nil {
return err
}
f := *format
if len(f) == 0 {
if len(cli.PsFormat()) > 0 && !*quiet {
@@ -72,18 +102,15 @@ func (cli *DockerCli) CmdPs(args ...string) error {
}
}
psCtx := formatter.ContainerContext{
Context: formatter.Context{
Output: cli.out,
Format: f,
Quiet: *quiet,
Trunc: !*noTrunc,
},
Size: *size,
Containers: containers,
psCtx := ps.Context{
Output: cli.out,
Format: f,
Quiet: *quiet,
Size: *size,
Trunc: !*noTrunc,
}
psCtx.Write()
ps.Format(psCtx, containers)
return nil
}

View File

@@ -1,47 +1,44 @@
package formatter
package ps
import (
"bytes"
"fmt"
"strconv"
"strings"
"text/tabwriter"
"text/template"
"time"
"github.com/docker/docker/api"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/stringutils"
"github.com/docker/engine-api/types"
"github.com/docker/go-units"
"github.com/docker/docker/pkg/units"
)
const (
tableKey = "table"
containerIDHeader = "CONTAINER ID"
imageHeader = "IMAGE"
namesHeader = "NAMES"
commandHeader = "COMMAND"
createdSinceHeader = "CREATED"
createdAtHeader = "CREATED AT"
runningForHeader = "CREATED"
statusHeader = "STATUS"
portsHeader = "PORTS"
sizeHeader = "SIZE"
labelsHeader = "LABELS"
imageIDHeader = "IMAGE ID"
repositoryHeader = "REPOSITORY"
tagHeader = "TAG"
digestHeader = "DIGEST"
mountsHeader = "MOUNTS"
idHeader = "CONTAINER ID"
imageHeader = "IMAGE"
namesHeader = "NAMES"
commandHeader = "COMMAND"
createdAtHeader = "CREATED AT"
runningForHeader = "CREATED"
statusHeader = "STATUS"
portsHeader = "PORTS"
sizeHeader = "SIZE"
labelsHeader = "LABELS"
)
type containerContext struct {
baseSubContext
trunc bool
c types.Container
trunc bool
header []string
c types.Container
}
func (c *containerContext) ID() string {
c.addHeader(containerIDHeader)
c.addHeader(idHeader)
if c.trunc {
return stringid.TruncateID(c.c.ID)
}
@@ -67,11 +64,6 @@ func (c *containerContext) Image() string {
if c.c.Image == "" {
return "<no image>"
}
if c.trunc {
if trunc := stringid.TruncateID(c.c.ImageID); trunc == stringid.TruncateID(c.c.Image) {
return trunc
}
}
return c.c.Image
}
@@ -143,96 +135,82 @@ func (c *containerContext) Label(name string) string {
return c.c.Labels[name]
}
func (c *containerContext) Mounts() string {
c.addHeader(mountsHeader)
var name string
var mounts []string
for _, m := range c.c.Mounts {
if m.Name == "" {
name = m.Source
} else {
name = m.Name
}
if c.trunc {
name = stringutils.Truncate(name, 15)
}
mounts = append(mounts, name)
}
return strings.Join(mounts, ",")
}
type imageContext struct {
baseSubContext
trunc bool
i types.Image
repo string
tag string
digest string
}
func (c *imageContext) ID() string {
c.addHeader(imageIDHeader)
if c.trunc {
return stringid.TruncateID(c.i.ID)
}
return c.i.ID
}
func (c *imageContext) Repository() string {
c.addHeader(repositoryHeader)
return c.repo
}
func (c *imageContext) Tag() string {
c.addHeader(tagHeader)
return c.tag
}
func (c *imageContext) Digest() string {
c.addHeader(digestHeader)
return c.digest
}
func (c *imageContext) CreatedSince() string {
c.addHeader(createdSinceHeader)
createdAt := time.Unix(int64(c.i.Created), 0)
return units.HumanDuration(time.Now().UTC().Sub(createdAt))
}
func (c *imageContext) CreatedAt() string {
c.addHeader(createdAtHeader)
return time.Unix(int64(c.i.Created), 0).String()
}
func (c *imageContext) Size() string {
c.addHeader(sizeHeader)
return units.HumanSize(float64(c.i.Size))
}
type subContext interface {
fullHeader() string
addHeader(header string)
}
type baseSubContext struct {
header []string
}
func (c *baseSubContext) fullHeader() string {
func (c *containerContext) fullHeader() string {
if c.header == nil {
return ""
}
return strings.Join(c.header, "\t")
}
func (c *baseSubContext) addHeader(header string) {
func (c *containerContext) addHeader(header string) {
if c.header == nil {
c.header = []string{}
}
c.header = append(c.header, strings.ToUpper(header))
}
func customFormat(ctx Context, containers []types.Container) {
var (
table bool
header string
format = ctx.Format
buffer = bytes.NewBufferString("")
)
if strings.HasPrefix(ctx.Format, tableKey) {
table = true
format = format[len(tableKey):]
}
format = strings.Trim(format, " ")
r := strings.NewReplacer(`\t`, "\t", `\n`, "\n")
format = r.Replace(format)
if table && ctx.Size {
format += "\t{{.Size}}"
}
tmpl, err := template.New("").Parse(format)
if err != nil {
buffer.WriteString(fmt.Sprintf("Template parsing error: %v\n", err))
buffer.WriteTo(ctx.Output)
return
}
for _, container := range containers {
containerCtx := &containerContext{
trunc: ctx.Trunc,
c: container,
}
if err := tmpl.Execute(buffer, containerCtx); err != nil {
buffer = bytes.NewBufferString(fmt.Sprintf("Template parsing error: %v\n", err))
buffer.WriteTo(ctx.Output)
return
}
if table && len(header) == 0 {
header = containerCtx.fullHeader()
}
buffer.WriteString("\n")
}
if table {
if len(header) == 0 {
// if we still don't have a header, we didn't have any containers so we need to fake it to get the right headers from the template
containerCtx := &containerContext{}
tmpl.Execute(bytes.NewBufferString(""), containerCtx)
header = containerCtx.fullHeader()
}
t := tabwriter.NewWriter(ctx.Output, 20, 1, 3, ' ', 0)
t.Write([]byte(header))
t.Write([]byte("\n"))
buffer.WriteTo(t)
t.Flush()
} else {
buffer.WriteTo(ctx.Output)
}
}
func stripNamePrefix(ss []string) []string {
for i, s := range ss {
ss[i] = s[1:]

View File

@@ -0,0 +1,102 @@
package ps
import (
"bytes"
"reflect"
"strings"
"testing"
"time"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/stringid"
)
func TestContainerPsContext(t *testing.T) {
containerId := stringid.GenerateRandomID()
unix := time.Now().Unix()
var ctx containerContext
cases := []struct {
container types.Container
trunc bool
expValue string
expHeader string
call func() string
}{
{types.Container{ID: containerId}, true, stringid.TruncateID(containerId), idHeader, ctx.ID},
{types.Container{Names: []string{"/foobar_baz"}}, true, "foobar_baz", namesHeader, ctx.Names},
{types.Container{Image: "ubuntu"}, true, "ubuntu", imageHeader, ctx.Image},
{types.Container{Image: ""}, true, "<no image>", imageHeader, ctx.Image},
{types.Container{Command: "sh -c 'ls -la'"}, true, `"sh -c 'ls -la'"`, commandHeader, ctx.Command},
{types.Container{Created: int(unix)}, true, time.Unix(unix, 0).String(), createdAtHeader, ctx.CreatedAt},
{types.Container{Ports: []types.Port{{PrivatePort: 8080, PublicPort: 8080, Type: "tcp"}}}, true, "8080/tcp", portsHeader, ctx.Ports},
{types.Container{Status: "RUNNING"}, true, "RUNNING", statusHeader, ctx.Status},
{types.Container{SizeRw: 10}, true, "10 B", sizeHeader, ctx.Size},
{types.Container{SizeRw: 10, SizeRootFs: 20}, true, "10 B (virtual 20 B)", sizeHeader, ctx.Size},
{types.Container{Labels: map[string]string{"cpu": "6", "storage": "ssd"}}, true, "cpu=6,storage=ssd", labelsHeader, ctx.Labels},
}
for _, c := range cases {
ctx = containerContext{c: c.container, trunc: c.trunc}
v := c.call()
if strings.Contains(v, ",") {
// comma-separated values means probably a map input, which won't
// be guaranteed to have the same order as our expected value
// We'll create maps and use reflect.DeepEquals to check instead:
entriesMap := make(map[string]string)
expMap := make(map[string]string)
entries := strings.Split(v, ",")
expectedEntries := strings.Split(c.expValue, ",")
for _, entry := range entries {
keyval := strings.Split(entry, "=")
entriesMap[keyval[0]] = keyval[1]
}
for _, expected := range expectedEntries {
keyval := strings.Split(expected, "=")
expMap[keyval[0]] = keyval[1]
}
if !reflect.DeepEqual(expMap, entriesMap) {
t.Fatalf("Expected entries: %v, got: %v", c.expValue, v)
}
} else if v != c.expValue {
t.Fatalf("Expected %s, was %s\n", c.expValue, v)
}
h := ctx.fullHeader()
if h != c.expHeader {
t.Fatalf("Expected %s, was %s\n", c.expHeader, h)
}
}
c := types.Container{Labels: map[string]string{"com.docker.swarm.swarm-id": "33", "com.docker.swarm.node_name": "ubuntu"}}
ctx = containerContext{c: c, trunc: true}
sid := ctx.Label("com.docker.swarm.swarm-id")
node := ctx.Label("com.docker.swarm.node_name")
if sid != "33" {
t.Fatalf("Expected 33, was %s\n", sid)
}
if node != "ubuntu" {
t.Fatalf("Expected ubuntu, was %s\n", node)
}
h := ctx.fullHeader()
if h != "SWARM ID\tNODE NAME" {
t.Fatalf("Expected %s, was %s\n", "SWARM ID\tNODE NAME", h)
}
}
func TestContainerPsFormatError(t *testing.T) {
out := bytes.NewBufferString("")
ctx := Context{
Format: "{{InvalidFunction}}",
Output: out,
}
customFormat(ctx, make([]types.Container, 0))
if out.String() != "Template parsing error: template: :1: function \"InvalidFunction\" not defined\n" {
t.Fatalf("Expected format error, got `%v`\n", out.String())
}
}

View File

@@ -0,0 +1,65 @@
package ps
import (
"io"
"github.com/docker/docker/api/types"
)
const (
tableFormatKey = "table"
rawFormatKey = "raw"
defaultTableFormat = "table {{.ID}}\t{{.Image}}\t{{.Command}}\t{{.RunningFor}} ago\t{{.Status}}\t{{.Ports}}\t{{.Names}}"
defaultQuietFormat = "{{.ID}}"
)
type Context struct {
Output io.Writer
Format string
Size bool
Quiet bool
Trunc bool
}
func Format(ctx Context, containers []types.Container) {
switch ctx.Format {
case tableFormatKey:
tableFormat(ctx, containers)
case rawFormatKey:
rawFormat(ctx, containers)
default:
customFormat(ctx, containers)
}
}
func rawFormat(ctx Context, containers []types.Container) {
if ctx.Quiet {
ctx.Format = `container_id: {{.ID}}`
} else {
ctx.Format = `container_id: {{.ID}}
image: {{.Image}}
command: {{.Command}}
created_at: {{.CreatedAt}}
status: {{.Status}}
names: {{.Names}}
labels: {{.Labels}}
ports: {{.Ports}}
`
if ctx.Size {
ctx.Format += `size: {{.Size}}
`
}
}
customFormat(ctx, containers)
}
func tableFormat(ctx Context, containers []types.Container) {
ctx.Format = defaultTableFormat
if ctx.Quiet {
ctx.Format = defaultQuietFormat
}
customFormat(ctx, containers)
}

View File

@@ -1,25 +1,21 @@
package client
import (
"errors"
"fmt"
"golang.org/x/net/context"
"net/url"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/jsonmessage"
"github.com/docker/docker/graph/tags"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/reference"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
"github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
)
// CmdPull pulls an image or a repository from the registry.
//
// Usage: docker pull [OPTIONS] IMAGENAME[:TAG|@DIGEST]
func (cli *DockerCli) CmdPull(args ...string) error {
cmd := Cli.Subcmd("pull", []string{"NAME[:TAG|@DIGEST]"}, Cli.DockerCommands["pull"].Description, true)
cmd := Cli.Subcmd("pull", []string{"NAME[:TAG|@DIGEST]"}, "Pull an image or a repository from a registry", true)
allTags := cmd.Bool([]string{"a", "-all-tags"}, false, "Download all tagged images in the repository")
addTrustedFlags(cmd, true)
cmd.Require(flag.Exact, 1)
@@ -27,63 +23,31 @@ func (cli *DockerCli) CmdPull(args ...string) error {
cmd.ParseFlags(args, true)
remote := cmd.Arg(0)
distributionRef, err := reference.ParseNamed(remote)
if err != nil {
return err
}
if *allTags && !reference.IsNameOnly(distributionRef) {
return errors.New("tag can't be used with --all-tags/-a")
}
if !*allTags && reference.IsNameOnly(distributionRef) {
distributionRef = reference.WithDefaultTag(distributionRef)
fmt.Fprintf(cli.out, "Using default tag: %s\n", reference.DefaultTag)
}
var tag string
switch x := distributionRef.(type) {
case reference.Canonical:
tag = x.Digest().String()
case reference.NamedTagged:
tag = x.Tag()
taglessRemote, tag := parsers.ParseRepositoryTag(remote)
if tag == "" && !*allTags {
tag = tags.DEFAULTTAG
fmt.Fprintf(cli.out, "Using default tag: %s\n", tag)
} else if tag != "" && *allTags {
return fmt.Errorf("tag can't be used with --all-tags/-a")
}
ref := registry.ParseReference(tag)
// Resolve the Repository name from fqn to RepositoryInfo
repoInfo, err := registry.ParseRepositoryInfo(distributionRef)
repoInfo, err := registry.ParseRepositoryInfo(taglessRemote)
if err != nil {
return err
}
authConfig := cli.resolveAuthConfig(repoInfo.Index)
requestPrivilege := cli.registryAuthenticationPrivilegedFunc(repoInfo.Index, "pull")
if isTrusted() && !ref.HasDigest() {
// Check if tag is digest
return cli.trustedPull(repoInfo, ref, authConfig, requestPrivilege)
authConfig := registry.ResolveAuthConfig(cli.configFile, repoInfo.Index)
return cli.trustedPull(repoInfo, ref, authConfig)
}
return cli.imagePullPrivileged(authConfig, distributionRef.String(), "", requestPrivilege)
}
func (cli *DockerCli) imagePullPrivileged(authConfig types.AuthConfig, imageID, tag string, requestPrivilege client.RequestPrivilegeFunc) error {
encodedAuth, err := encodeAuthToBase64(authConfig)
if err != nil {
return err
}
options := types.ImagePullOptions{
ImageID: imageID,
Tag: tag,
RegistryAuth: encodedAuth,
}
responseBody, err := cli.client.ImagePull(context.Background(), options, requestPrivilege)
if err != nil {
return err
}
defer responseBody.Close()
return jsonmessage.DisplayJSONMessagesStream(responseBody, cli.out, cli.outFd, cli.isTerminalOut, nil)
v := url.Values{}
v.Set("fromImage", ref.ImageName(taglessRemote))
_, _, err = cli.clientRequestAttemptLogin("POST", "/images/create?"+v.Encode(), nil, cli.out, repoInfo.Index, "pull")
return err
}

View File

@@ -1,76 +1,53 @@
package client
import (
"errors"
"io"
"golang.org/x/net/context"
"fmt"
"net/url"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/reference"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
"github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
)
// CmdPush pushes an image or repository to the registry.
//
// Usage: docker push NAME[:TAG]
func (cli *DockerCli) CmdPush(args ...string) error {
cmd := Cli.Subcmd("push", []string{"NAME[:TAG]"}, Cli.DockerCommands["push"].Description, true)
cmd := Cli.Subcmd("push", []string{"NAME[:TAG]"}, "Push an image or a repository to a registry", true)
addTrustedFlags(cmd, false)
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
ref, err := reference.ParseNamed(cmd.Arg(0))
if err != nil {
return err
}
var tag string
switch x := ref.(type) {
case reference.Canonical:
return errors.New("cannot push a digest reference")
case reference.NamedTagged:
tag = x.Tag()
}
remote, tag := parsers.ParseRepositoryTag(cmd.Arg(0))
// Resolve the Repository name from fqn to RepositoryInfo
repoInfo, err := registry.ParseRepositoryInfo(ref)
repoInfo, err := registry.ParseRepositoryInfo(remote)
if err != nil {
return err
}
// Resolve the Auth config relevant for this server
authConfig := cli.resolveAuthConfig(repoInfo.Index)
authConfig := registry.ResolveAuthConfig(cli.configFile, repoInfo.Index)
// If we're not using a custom registry, we know the restrictions
// applied to repository names and can warn the user in advance.
// Custom repositories can have different rules, and we must also
// allow pushing by image ID.
if repoInfo.Official {
username := authConfig.Username
if username == "" {
username = "<user>"
}
return fmt.Errorf("You cannot push a \"root\" repository. Please rename your repository to <user>/<repo> (ex: %s/%s)", username, repoInfo.LocalName)
}
requestPrivilege := cli.registryAuthenticationPrivilegedFunc(repoInfo.Index, "push")
if isTrusted() {
return cli.trustedPush(repoInfo, tag, authConfig, requestPrivilege)
return cli.trustedPush(repoInfo, tag, authConfig)
}
responseBody, err := cli.imagePushPrivileged(authConfig, ref.Name(), tag, requestPrivilege)
if err != nil {
return err
}
v := url.Values{}
v.Set("tag", tag)
defer responseBody.Close()
return jsonmessage.DisplayJSONMessagesStream(responseBody, cli.out, cli.outFd, cli.isTerminalOut, nil)
}
func (cli *DockerCli) imagePushPrivileged(authConfig types.AuthConfig, imageID, tag string, requestPrivilege client.RequestPrivilegeFunc) (io.ReadCloser, error) {
encodedAuth, err := encodeAuthToBase64(authConfig)
if err != nil {
return nil, err
}
options := types.ImagePushOptions{
ImageID: imageID,
Tag: tag,
RegistryAuth: encodedAuth,
}
return cli.client.ImagePush(context.Background(), options, requestPrivilege)
_, _, err = cli.clientRequestAttemptLogin("POST", "/images/"+remote+"/push?"+v.Encode(), nil, cli.out, repoInfo.Index, "push")
return err
}

View File

@@ -2,9 +2,6 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -14,19 +11,15 @@ import (
//
// Usage: docker rename OLD_NAME NEW_NAME
func (cli *DockerCli) CmdRename(args ...string) error {
cmd := Cli.Subcmd("rename", []string{"OLD_NAME NEW_NAME"}, Cli.DockerCommands["rename"].Description, true)
cmd := Cli.Subcmd("rename", []string{"OLD_NAME NEW_NAME"}, "Rename a container", true)
cmd.Require(flag.Exact, 2)
cmd.ParseFlags(args, true)
oldName := strings.TrimSpace(cmd.Arg(0))
newName := strings.TrimSpace(cmd.Arg(1))
oldName := cmd.Arg(0)
newName := cmd.Arg(1)
if oldName == "" || newName == "" {
return fmt.Errorf("Error: Neither old nor new names may be empty")
}
if err := cli.client.ContainerRename(context.Background(), oldName, newName); err != nil {
if _, _, err := readBody(cli.call("POST", fmt.Sprintf("/containers/%s/rename?name=%s", oldName, newName), nil, nil)); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
return fmt.Errorf("Error: failed to rename container named %s", oldName)
}

View File

@@ -2,34 +2,38 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
"net/url"
"strconv"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdRestart restarts one or more containers.
// CmdRestart restarts one or more running containers.
//
// Usage: docker restart [OPTIONS] CONTAINER [CONTAINER...]
// Usage: docker stop [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdRestart(args ...string) error {
cmd := Cli.Subcmd("restart", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["restart"].Description, true)
cmd := Cli.Subcmd("restart", []string{"CONTAINER [CONTAINER...]"}, "Restart a running container", true)
nSeconds := cmd.Int([]string{"t", "-time"}, 10, "Seconds to wait for stop before killing the container")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
v := url.Values{}
v.Set("t", strconv.Itoa(*nSeconds))
var errNames []string
for _, name := range cmd.Args() {
if err := cli.client.ContainerRestart(context.Background(), name, *nSeconds); err != nil {
errs = append(errs, err.Error())
_, _, err := readBody(cli.call("POST", "/containers/"+name+"/restart?"+v.Encode(), nil, nil))
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to restart containers: %v", errNames)
}
return nil
}

View File

@@ -2,55 +2,54 @@ package client
import (
"fmt"
"net/url"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/engine-api/types"
)
// CmdRm removes one or more containers.
//
// Usage: docker rm [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdRm(args ...string) error {
cmd := Cli.Subcmd("rm", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["rm"].Description, true)
cmd := Cli.Subcmd("rm", []string{"CONTAINER [CONTAINER...]"}, "Remove one or more containers", true)
v := cmd.Bool([]string{"v", "-volumes"}, false, "Remove the volumes associated with the container")
link := cmd.Bool([]string{"l", "-link"}, false, "Remove the specified link")
link := cmd.Bool([]string{"l", "#link", "-link"}, false, "Remove the specified link")
force := cmd.Bool([]string{"f", "-force"}, false, "Force the removal of a running container (uses SIGKILL)")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
val := url.Values{}
if *v {
val.Set("v", "1")
}
if *link {
val.Set("link", "1")
}
if *force {
val.Set("force", "1")
}
var errNames []string
for _, name := range cmd.Args() {
if name == "" {
return fmt.Errorf("Container name cannot be empty")
}
name = strings.Trim(name, "/")
if err := cli.removeContainer(name, *v, *link, *force); err != nil {
errs = append(errs, err.Error())
_, _, err := readBody(cli.call("DELETE", "/containers/"+name+"?"+val.Encode(), nil, nil))
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
}
return nil
}
func (cli *DockerCli) removeContainer(containerID string, removeVolumes, removeLinks, force bool) error {
options := types.ContainerRemoveOptions{
ContainerID: containerID,
RemoveVolumes: removeVolumes,
RemoveLinks: removeLinks,
Force: force,
}
if err := cli.client.ContainerRemove(context.Background(), options); err != nil {
return err
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to remove containers: %v", errNames)
}
return nil
}

View File

@@ -1,22 +1,20 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/engine-api/types"
)
// CmdRmi removes all images with the specified name(s).
//
// Usage: docker rmi [OPTIONS] IMAGE [IMAGE...]
func (cli *DockerCli) CmdRmi(args ...string) error {
cmd := Cli.Subcmd("rmi", []string{"IMAGE [IMAGE...]"}, Cli.DockerCommands["rmi"].Description, true)
cmd := Cli.Subcmd("rmi", []string{"IMAGE [IMAGE...]"}, "Remove one or more images", true)
force := cmd.Bool([]string{"f", "-force"}, false, "Force removal of the image")
noprune := cmd.Bool([]string{"-no-prune"}, false, "Do not delete untagged parents")
cmd.Require(flag.Min, 1)
@@ -31,18 +29,22 @@ func (cli *DockerCli) CmdRmi(args ...string) error {
v.Set("noprune", "1")
}
var errs []string
var errNames []string
for _, name := range cmd.Args() {
options := types.ImageRemoveOptions{
ImageID: name,
Force: *force,
PruneChildren: !*noprune,
}
dels, err := cli.client.ImageRemove(context.Background(), options)
serverResp, err := cli.call("DELETE", "/images/"+name+"?"+v.Encode(), nil, nil)
if err != nil {
errs = append(errs, err.Error())
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
defer serverResp.body.Close()
dels := []types.ImageDelete{}
if err := json.NewDecoder(serverResp.body).Decode(&dels); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
continue
}
for _, del := range dels {
if del.Deleted != "" {
fmt.Fprintf(cli.out, "Deleted: %s\n", del.Deleted)
@@ -52,8 +54,8 @@ func (cli *DockerCli) CmdRmi(args ...string) error {
}
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to remove images: %v", errNames)
}
return nil
}

View File

@@ -3,27 +3,19 @@ package client
import (
"fmt"
"io"
"net/url"
"os"
"runtime"
"strings"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/signal"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types"
"github.com/docker/docker/runconfig"
"github.com/docker/libnetwork/resolvconf/dns"
)
const (
errCmdNotFound = "not found or does not exist."
errCmdCouldNotBeInvoked = "could not be invoked."
)
func (cid *cidFile) Close() error {
cid.file.Close()
@@ -44,29 +36,11 @@ func (cid *cidFile) Write(id string) error {
return nil
}
// if container start fails with 'command not found' error, return 127
// if container start fails with 'command cannot be invoked' error, return 126
// return 125 for generic docker daemon failures
func runStartContainerErr(err error) error {
trimmedErr := strings.Trim(err.Error(), "Error response from daemon: ")
statusError := Cli.StatusError{StatusCode: 125}
if strings.HasPrefix(trimmedErr, "Container command") {
if strings.Contains(trimmedErr, errCmdNotFound) {
statusError = Cli.StatusError{StatusCode: 127}
} else if strings.Contains(trimmedErr, errCmdCouldNotBeInvoked) {
statusError = Cli.StatusError{StatusCode: 126}
}
}
return statusError
}
// CmdRun runs a command in a new container.
//
// Usage: docker run [OPTIONS] IMAGE [COMMAND] [ARG...]
func (cli *DockerCli) CmdRun(args ...string) error {
cmd := Cli.Subcmd("run", []string{"IMAGE [COMMAND] [ARG...]"}, Cli.DockerCommands["run"].Description, true)
cmd := Cli.Subcmd("run", []string{"IMAGE [COMMAND] [ARG...]"}, "Run a command in a new container", true)
addTrustedFlags(cmd, true)
// These are flags not stored in Config/HostConfig
@@ -75,7 +49,6 @@ func (cli *DockerCli) CmdRun(args ...string) error {
flDetach = cmd.Bool([]string{"d", "-detach"}, false, "Run container in background and print container ID")
flSigProxy = cmd.Bool([]string{"-sig-proxy"}, true, "Proxy received signals to the process")
flName = cmd.String([]string{"-name"}, "", "Assign a name to the container")
flDetachKeys = cmd.String([]string{"-detach-keys"}, "", "Override the key sequence for detaching a container")
flAttach *opts.ListOpts
ErrConflictAttachDetach = fmt.Errorf("Conflicting options: -a and -d")
@@ -83,23 +56,18 @@ func (cli *DockerCli) CmdRun(args ...string) error {
ErrConflictDetachAutoRemove = fmt.Errorf("Conflicting options: --rm and -d")
)
config, hostConfig, networkingConfig, cmd, err := runconfigopts.Parse(cmd, args)
config, hostConfig, cmd, err := runconfig.Parse(cmd, args)
// just in case the Parse does not exit
if err != nil {
cmd.ReportError(err.Error(), true)
os.Exit(125)
os.Exit(1)
}
if hostConfig.OomKillDisable != nil && *hostConfig.OomKillDisable && hostConfig.Memory == 0 {
fmt.Fprintf(cli.err, "WARNING: Disabling the OOM killer on containers without setting a '-m/--memory' limit may be dangerous.\n")
}
if len(hostConfig.DNS) > 0 {
if len(hostConfig.Dns) > 0 {
// check the DNS settings passed via --dns against
// localhost regexp to warn if they are trying to
// set a DNS to a localhost address
for _, dnsIP := range hostConfig.DNS {
for _, dnsIP := range hostConfig.Dns {
if dns.IsLocalhost(dnsIP) {
fmt.Fprintf(cli.err, "WARNING: Localhost DNS setting (--dns=%s) may fail in containers.\n", dnsIP)
break
@@ -111,8 +79,6 @@ func (cli *DockerCli) CmdRun(args ...string) error {
return nil
}
config.ArgsEscaped = false
if !*flDetach {
if err := cli.CheckTtyInput(config.AttachStdin, config.Tty); err != nil {
return err
@@ -147,10 +113,9 @@ func (cli *DockerCli) CmdRun(args ...string) error {
hostConfig.ConsoleSize[0], hostConfig.ConsoleSize[1] = cli.getTtySize()
}
createResponse, err := cli.createContainer(config, hostConfig, networkingConfig, hostConfig.ContainerIDFile, *flName)
createResponse, err := cli.createContainer(config, hostConfig, hostConfig.ContainerIDFile, *flName)
if err != nil {
cmd.ReportError(err.Error(), true)
return runStartContainerErr(err)
return err
}
if sigProxy {
sigc := cli.forwardAllSignals(createResponse.ID)
@@ -171,66 +136,71 @@ func (cli *DockerCli) CmdRun(args ...string) error {
if *flAutoRemove && (hostConfig.RestartPolicy.IsAlways() || hostConfig.RestartPolicy.IsOnFailure()) {
return ErrConflictRestartPolicyAndAutoRemove
}
// We need to instantiate the chan because the select needs it. It can
// be closed but can't be uninitialized.
hijacked := make(chan io.Closer)
// Block the return until the chan gets closed
defer func() {
logrus.Debugf("End of CmdRun(), Waiting for hijack to finish.")
if _, ok := <-hijacked; ok {
fmt.Fprintln(cli.err, "Hijack did not finish (chan still open)")
}
}()
if config.AttachStdin || config.AttachStdout || config.AttachStderr {
var (
out, stderr io.Writer
in io.ReadCloser
v = url.Values{}
)
v.Set("stream", "1")
if config.AttachStdin {
v.Set("stdin", "1")
in = cli.in
}
if config.AttachStdout {
v.Set("stdout", "1")
out = cli.out
}
if config.AttachStderr {
v.Set("stderr", "1")
if config.Tty {
stderr = cli.out
} else {
stderr = cli.err
}
}
if *flDetachKeys != "" {
cli.configFile.DetachKeys = *flDetachKeys
errCh = promise.Go(func() error {
return cli.hijack("POST", "/containers/"+createResponse.ID+"/attach?"+v.Encode(), config.Tty, in, out, stderr, hijacked, nil)
})
} else {
close(hijacked)
}
// Acknowledge the hijack before starting
select {
case closer := <-hijacked:
// Make sure that the hijack gets closed when returning (results
// in closing the hijack chan and freeing server's goroutines)
if closer != nil {
defer closer.Close()
}
options := types.ContainerAttachOptions{
ContainerID: createResponse.ID,
Stream: true,
Stdin: config.AttachStdin,
Stdout: config.AttachStdout,
Stderr: config.AttachStderr,
DetachKeys: cli.configFile.DetachKeys,
}
resp, err := cli.client.ContainerAttach(context.Background(), options)
case err := <-errCh:
if err != nil {
logrus.Debugf("Error hijack: %s", err)
return err
}
if in != nil && config.Tty {
if err := cli.setRawTerminal(); err != nil {
return err
}
defer cli.restoreTerminal(in)
}
errCh = promise.Go(func() error {
return cli.holdHijackedConnection(config.Tty, in, out, stderr, resp)
})
}
if *flAutoRemove {
defer func() {
if err := cli.removeContainer(createResponse.ID, true, false, false); err != nil {
fmt.Fprintf(cli.err, "%v\n", err)
defer func() {
if *flAutoRemove {
if _, _, err = readBody(cli.call("DELETE", "/containers/"+createResponse.ID+"?v=1", nil, nil)); err != nil {
fmt.Fprintf(cli.err, "Error deleting container: %s\n", err)
}
}()
}
}
}()
//start the container
if err := cli.client.ContainerStart(context.Background(), createResponse.ID); err != nil {
cmd.ReportError(err.Error(), false)
return runStartContainerErr(err)
if _, _, err = readBody(cli.call("POST", "/containers/"+createResponse.ID+"/start", nil, nil)); err != nil {
return err
}
if (config.AttachStdin || config.AttachStdout || config.AttachStderr) && config.Tty && cli.isTerminalOut {
@@ -259,8 +229,8 @@ func (cli *DockerCli) CmdRun(args ...string) error {
if *flAutoRemove {
// Autoremove: wait for the container to finish, retrieve
// the exit code and remove the container
if status, err = cli.client.ContainerWait(context.Background(), createResponse.ID); err != nil {
return runStartContainerErr(err)
if _, _, err := readBody(cli.call("POST", "/containers/"+createResponse.ID+"/wait", nil, nil)); err != nil {
return err
}
if _, status, err = getExitCode(cli, createResponse.ID); err != nil {
return err
@@ -269,7 +239,7 @@ func (cli *DockerCli) CmdRun(args ...string) error {
// No Autoremove: Simply retrieve the exit code
if !config.Tty {
// In non-TTY mode, we can't detach, so we must wait for container exit
if status, err = cli.client.ContainerWait(context.Background(), createResponse.ID); err != nil {
if status, err = waitForExit(cli, createResponse.ID); err != nil {
return err
}
} else {

View File

@@ -3,8 +3,8 @@ package client
import (
"errors"
"io"
"golang.org/x/net/context"
"net/url"
"os"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -16,27 +16,43 @@ import (
//
// Usage: docker save [OPTIONS] IMAGE [IMAGE...]
func (cli *DockerCli) CmdSave(args ...string) error {
cmd := Cli.Subcmd("save", []string{"IMAGE [IMAGE...]"}, Cli.DockerCommands["save"].Description+" (streamed to STDOUT by default)", true)
outfile := cmd.String([]string{"o", "-output"}, "", "Write to a file, instead of STDOUT")
cmd := Cli.Subcmd("save", []string{"IMAGE [IMAGE...]"}, "Save an image(s) to a tar archive (streamed to STDOUT by default)", true)
outfile := cmd.String([]string{"o", "-output"}, "", "Write to an file, instead of STDOUT")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
if *outfile == "" && cli.isTerminalOut {
var (
output io.Writer = cli.out
err error
)
if *outfile != "" {
output, err = os.Create(*outfile)
if err != nil {
return err
}
} else if cli.isTerminalOut {
return errors.New("Cowardly refusing to save to a terminal. Use the -o flag or redirect.")
}
responseBody, err := cli.client.ImageSave(context.Background(), cmd.Args())
if err != nil {
return err
}
defer responseBody.Close()
if *outfile == "" {
_, err := io.Copy(cli.out, responseBody)
return err
sopts := &streamOpts{
rawTerminal: true,
out: output,
}
return copyToFile(*outfile, responseBody)
if len(cmd.Args()) == 1 {
image := cmd.Arg(0)
if _, err := cli.stream("GET", "/images/"+image+"/get", sopts); err != nil {
return err
}
} else {
v := url.Values{}
for _, arg := range cmd.Args() {
v.Add("names", arg)
}
if _, err := cli.stream("GET", "/images/get?"+v.Encode(), sopts); err != nil {
return err
}
}
return nil
}

View File

@@ -1,30 +1,36 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"sort"
"strings"
"text/tabwriter"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/stringutils"
"github.com/docker/docker/registry"
"github.com/docker/engine-api/types"
registrytypes "github.com/docker/engine-api/types/registry"
)
// ByStars sorts search results in ascending order by number of stars.
type ByStars []registry.SearchResult
func (r ByStars) Len() int { return len(r) }
func (r ByStars) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r ByStars) Less(i, j int) bool { return r[i].StarCount < r[j].StarCount }
// CmdSearch searches the Docker Hub for images.
//
// Usage: docker search [OPTIONS] TERM
func (cli *DockerCli) CmdSearch(args ...string) error {
cmd := Cli.Subcmd("search", []string{"TERM"}, Cli.DockerCommands["search"].Description, true)
noTrunc := cmd.Bool([]string{"-no-trunc"}, false, "Don't truncate output")
cmd := Cli.Subcmd("search", []string{"TERM"}, "Search the Docker Hub for images", true)
noTrunc := cmd.Bool([]string{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
trusted := cmd.Bool([]string{"#t", "#trusted", "#-trusted"}, false, "Only show trusted builds")
automated := cmd.Bool([]string{"-automated"}, false, "Only show automated builds")
stars := cmd.Uint([]string{"s", "-stars"}, 0, "Only displays with at least x stars")
stars := cmd.Uint([]string{"s", "#stars", "-stars"}, 0, "Only displays with at least x stars")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
@@ -33,36 +39,31 @@ func (cli *DockerCli) CmdSearch(args ...string) error {
v := url.Values{}
v.Set("term", name)
indexInfo, err := registry.ParseSearchIndexInfo(name)
// Resolve the Repository name from fqn to hostname + name
taglessRemote, _ := parsers.ParseRepositoryTag(name)
repoInfo, err := registry.ParseRepositoryInfo(taglessRemote)
if err != nil {
return err
}
authConfig := cli.resolveAuthConfig(indexInfo)
requestPrivilege := cli.registryAuthenticationPrivilegedFunc(indexInfo, "search")
encodedAuth, err := encodeAuthToBase64(authConfig)
rdr, _, err := cli.clientRequestAttemptLogin("GET", "/images/search?"+v.Encode(), nil, nil, repoInfo.Index, "search")
if err != nil {
return err
}
options := types.ImageSearchOptions{
Term: name,
RegistryAuth: encodedAuth,
}
defer rdr.Close()
unorderedResults, err := cli.client.ImageSearch(context.Background(), options, requestPrivilege)
if err != nil {
results := ByStars{}
if err := json.NewDecoder(rdr).Decode(&results); err != nil {
return err
}
results := searchResultsByStars(unorderedResults)
sort.Sort(results)
sort.Sort(sort.Reverse(results))
w := tabwriter.NewWriter(cli.out, 10, 1, 3, ' ', 0)
fmt.Fprintf(w, "NAME\tDESCRIPTION\tSTARS\tOFFICIAL\tAUTOMATED\n")
for _, res := range results {
if (*automated && !res.IsAutomated) || (int(*stars) > res.StarCount) {
if ((*automated || *trusted) && (!res.IsTrusted && !res.IsAutomated)) || (int(*stars) > res.StarCount) {
continue
}
desc := strings.Replace(res.Description, "\n", " ", -1)
@@ -84,10 +85,3 @@ func (cli *DockerCli) CmdSearch(args ...string) error {
w.Flush()
return nil
}
// SearchResultsByStars sorts search results in descending order by number of stars.
type searchResultsByStars []registrytypes.SearchResult
func (r searchResultsByStars) Len() int { return len(r) }
func (r searchResultsByStars) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r searchResultsByStars) Less(i, j int) bool { return r[j].StarCount < r[i].StarCount }

15
api/client/service.go Normal file
View File

@@ -0,0 +1,15 @@
// +build experimental
package client
import (
"os"
nwclient "github.com/docker/libnetwork/client"
)
func (cli *DockerCli) CmdService(args ...string) error {
nCli := nwclient.NewNetworkCli(cli.out, cli.err, nwclient.CallFunc(cli.callWrapper))
args = append([]string{"service"}, args...)
return nCli.Cmd(os.Args[0], args...)
}

View File

@@ -1,19 +1,18 @@
package client
import (
"encoding/json"
"fmt"
"io"
"net/url"
"os"
"strings"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/signal"
"github.com/docker/engine-api/types"
)
func (cli *DockerCli) forwardAllSignals(cid string) chan os.Signal {
@@ -21,7 +20,7 @@ func (cli *DockerCli) forwardAllSignals(cid string) chan os.Signal {
signal.CatchAll(sigc)
go func() {
for s := range sigc {
if s == signal.SIGCHLD || s == signal.SIGPIPE {
if s == signal.SIGCHLD {
continue
}
var sig string
@@ -33,10 +32,8 @@ func (cli *DockerCli) forwardAllSignals(cid string) chan os.Signal {
}
if sig == "" {
fmt.Fprintf(cli.err, "Unsupported signal: %v. Discarding.\n", s)
continue
}
if err := cli.client.ContainerKill(context.Background(), cid, sig); err != nil {
if _, _, err := readBody(cli.call("POST", fmt.Sprintf("/containers/%s/kill?signal=%s", cid, sig), nil, nil)); err != nil {
logrus.Debugf("Error sending signal: %s", err)
}
}
@@ -44,114 +41,130 @@ func (cli *DockerCli) forwardAllSignals(cid string) chan os.Signal {
return sigc
}
// CmdStart starts one or more containers.
// CmdStart starts one or more stopped containers.
//
// Usage: docker start [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdStart(args ...string) error {
cmd := Cli.Subcmd("start", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["start"].Description, true)
cmd := Cli.Subcmd("start", []string{"CONTAINER [CONTAINER...]"}, "Start one or more stopped containers", true)
attach := cmd.Bool([]string{"a", "-attach"}, false, "Attach STDOUT/STDERR and forward signals")
openStdin := cmd.Bool([]string{"i", "-interactive"}, false, "Attach container's STDIN")
detachKeys := cmd.String([]string{"-detach-keys"}, "", "Override the key sequence for detaching a container")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var (
cErr chan error
tty bool
)
if *attach || *openStdin {
// We're going to attach to a container.
// 1. Ensure we only have one container.
if cmd.NArg() > 1 {
return fmt.Errorf("You cannot start and attach multiple containers at once.")
}
// 2. Attach to the container.
containerID := cmd.Arg(0)
c, err := cli.client.ContainerInspect(context.Background(), containerID)
serverResp, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/json", nil, nil)
if err != nil {
return err
}
if !c.Config.Tty {
sigc := cli.forwardAllSignals(containerID)
defer serverResp.body.Close()
var c types.ContainerJSON
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return err
}
tty = c.Config.Tty
if !tty {
sigc := cli.forwardAllSignals(cmd.Arg(0))
defer signal.StopCatch(sigc)
}
if *detachKeys != "" {
cli.configFile.DetachKeys = *detachKeys
}
options := types.ContainerAttachOptions{
ContainerID: containerID,
Stream: true,
Stdin: *openStdin && c.Config.OpenStdin,
Stdout: true,
Stderr: true,
DetachKeys: cli.configFile.DetachKeys,
}
var in io.ReadCloser
if options.Stdin {
v := url.Values{}
v.Set("stream", "1")
if *openStdin && c.Config.OpenStdin {
v.Set("stdin", "1")
in = cli.in
}
resp, err := cli.client.ContainerAttach(context.Background(), options)
if err != nil {
return err
}
defer resp.Close()
if in != nil && c.Config.Tty {
if err := cli.setRawTerminal(); err != nil {
return err
}
defer cli.restoreTerminal(in)
}
v.Set("stdout", "1")
v.Set("stderr", "1")
cErr := promise.Go(func() error {
return cli.holdHijackedConnection(c.Config.Tty, in, cli.out, cli.err, resp)
hijacked := make(chan io.Closer)
// Block the return until the chan gets closed
defer func() {
logrus.Debugf("CmdStart() returned, defer waiting for hijack to finish.")
if _, ok := <-hijacked; ok {
fmt.Fprintln(cli.err, "Hijack did not finish (chan still open)")
}
cli.in.Close()
}()
cErr = promise.Go(func() error {
return cli.hijack("POST", "/containers/"+cmd.Arg(0)+"/attach?"+v.Encode(), tty, in, cli.out, cli.err, hijacked, nil)
})
// 3. Start the container.
if err := cli.client.ContainerStart(context.Background(), containerID); err != nil {
return err
// Acknowledge the hijack before starting
select {
case closer := <-hijacked:
// Make sure that the hijack gets closed when returning (results
// in closing the hijack chan and freeing server's goroutines)
if closer != nil {
defer closer.Close()
}
case err := <-cErr:
if err != nil {
return err
}
}
}
// 4. Wait for attachment to break.
if c.Config.Tty && cli.isTerminalOut {
if err := cli.monitorTtySize(containerID, false); err != nil {
var encounteredError error
var errNames []string
for _, name := range cmd.Args() {
_, _, err := readBody(cli.call("POST", "/containers/"+name+"/start", nil, nil))
if err != nil {
if !*attach && !*openStdin {
// attach and openStdin is false means it could be starting multiple containers
// when a container start failed, show the error message and start next
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
encounteredError = err
}
} else {
if !*attach && !*openStdin {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
}
if len(errNames) > 0 {
encounteredError = fmt.Errorf("Error: failed to start containers: %v", errNames)
}
if encounteredError != nil {
return encounteredError
}
if *openStdin || *attach {
if tty && cli.isTerminalOut {
if err := cli.monitorTtySize(cmd.Arg(0), false); err != nil {
fmt.Fprintf(cli.err, "Error monitoring TTY size: %s\n", err)
}
}
if attchErr := <-cErr; attchErr != nil {
return attchErr
}
_, status, err := getExitCode(cli, containerID)
_, status, err := getExitCode(cli, cmd.Arg(0))
if err != nil {
return err
}
if status != 0 {
return Cli.StatusError{StatusCode: status}
}
} else {
// We're not going to attach to anything.
// Start as many containers as we want.
return cli.startContainersWithoutAttachments(cmd.Args())
}
return nil
}
func (cli *DockerCli) startContainersWithoutAttachments(containerIDs []string) error {
var failedContainers []string
for _, containerID := range containerIDs {
if err := cli.client.ContainerStart(context.Background(), containerID); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
failedContainers = append(failedContainers, containerID)
} else {
fmt.Fprintf(cli.out, "%s\n", containerID)
}
}
if len(failedContainers) > 0 {
return fmt.Errorf("Error: failed to start containers: %v", strings.Join(failedContainers, ", "))
}
return nil
}

View File

@@ -1,208 +1,202 @@
package client
import (
"encoding/json"
"fmt"
"io"
"net/url"
"sort"
"strings"
"sync"
"text/tabwriter"
"time"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/events"
"github.com/docker/engine-api/types/filters"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/units"
)
type containerStats struct {
Name string
CPUPercentage float64
Memory float64
MemoryLimit float64
MemoryPercentage float64
NetworkRx float64
NetworkTx float64
mu sync.RWMutex
err error
}
func (s *containerStats) Collect(cli *DockerCli, streamStats bool) {
v := url.Values{}
if streamStats {
v.Set("stream", "1")
} else {
v.Set("stream", "0")
}
serverResp, err := cli.call("GET", "/containers/"+s.Name+"/stats?"+v.Encode(), nil, nil)
if err != nil {
s.mu.Lock()
s.err = err
s.mu.Unlock()
return
}
defer serverResp.body.Close()
var (
previousCPU uint64
previousSystem uint64
dec = json.NewDecoder(serverResp.body)
u = make(chan error, 1)
)
go func() {
for {
var v *types.Stats
if err := dec.Decode(&v); err != nil {
u <- err
return
}
var (
memPercent = float64(v.MemoryStats.Usage) / float64(v.MemoryStats.Limit) * 100.0
cpuPercent = 0.0
)
previousCPU = v.PreCpuStats.CpuUsage.TotalUsage
previousSystem = v.PreCpuStats.SystemUsage
cpuPercent = calculateCPUPercent(previousCPU, previousSystem, v)
s.mu.Lock()
s.CPUPercentage = cpuPercent
s.Memory = float64(v.MemoryStats.Usage)
s.MemoryLimit = float64(v.MemoryStats.Limit)
s.MemoryPercentage = memPercent
s.NetworkRx = float64(v.Network.RxBytes)
s.NetworkTx = float64(v.Network.TxBytes)
s.mu.Unlock()
u <- nil
if !streamStats {
return
}
}
}()
for {
select {
case <-time.After(2 * time.Second):
// zero out the values if we have not received an update within
// the specified duration.
s.mu.Lock()
s.CPUPercentage = 0
s.Memory = 0
s.MemoryPercentage = 0
s.mu.Unlock()
case err := <-u:
if err != nil {
s.mu.Lock()
s.err = err
s.mu.Unlock()
return
}
}
if !streamStats {
return
}
}
}
func (s *containerStats) Display(w io.Writer) error {
s.mu.RLock()
defer s.mu.RUnlock()
if s.err != nil {
return s.err
}
fmt.Fprintf(w, "%s\t%.2f%%\t%s/%s\t%.2f%%\t%s/%s\n",
s.Name,
s.CPUPercentage,
units.HumanSize(s.Memory), units.HumanSize(s.MemoryLimit),
s.MemoryPercentage,
units.HumanSize(s.NetworkRx), units.HumanSize(s.NetworkTx))
return nil
}
// CmdStats displays a live stream of resource usage statistics for one or more containers.
//
// This shows real-time information on CPU usage, memory usage, and network I/O.
//
// Usage: docker stats [OPTIONS] [CONTAINER...]
// Usage: docker stats CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdStats(args ...string) error {
cmd := Cli.Subcmd("stats", []string{"[CONTAINER...]"}, Cli.DockerCommands["stats"].Description, true)
all := cmd.Bool([]string{"a", "-all"}, false, "Show all containers (default shows just running)")
cmd := Cli.Subcmd("stats", []string{"CONTAINER [CONTAINER...]"}, "Display a live stream of one or more containers' resource usage statistics", true)
noStream := cmd.Bool([]string{"-no-stream"}, false, "Disable streaming stats and only pull the first result")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
names := cmd.Args()
showAll := len(names) == 0
closeChan := make(chan error)
// monitorContainerEvents watches for container creation and removal (only
// used when calling `docker stats` without arguments).
monitorContainerEvents := func(started chan<- struct{}, c chan events.Message) {
f := filters.NewArgs()
f.Add("type", "container")
options := types.EventsOptions{
Filters: f,
}
resBody, err := cli.client.Events(context.Background(), options)
// Whether we successfully subscribed to events or not, we can now
// unblock the main goroutine.
close(started)
if err != nil {
closeChan <- err
return
}
defer resBody.Close()
decodeEvents(resBody, func(event events.Message, err error) error {
if err != nil {
closeChan <- err
return nil
}
c <- event
return nil
})
}
// waitFirst is a WaitGroup to wait first stat data's reach for each container
waitFirst := &sync.WaitGroup{}
cStats := stats{}
// getContainerList simulates creation event for all previously existing
// containers (only used when calling `docker stats` without arguments).
getContainerList := func() {
options := types.ContainerListOptions{
All: *all,
}
cs, err := cli.client.ContainerList(context.Background(), options)
if err != nil {
closeChan <- err
}
for _, container := range cs {
s := &containerStats{Name: container.ID[:12]}
if cStats.add(s) {
waitFirst.Add(1)
go s.Collect(cli.client, !*noStream, waitFirst)
}
}
}
if showAll {
// If no names were specified, start a long running goroutine which
// monitors container events. We make sure we're subscribed before
// retrieving the list of running containers to avoid a race where we
// would "miss" a creation.
started := make(chan struct{})
eh := eventHandler{handlers: make(map[string]func(events.Message))}
eh.Handle("create", func(e events.Message) {
if *all {
s := &containerStats{Name: e.ID[:12]}
if cStats.add(s) {
waitFirst.Add(1)
go s.Collect(cli.client, !*noStream, waitFirst)
}
}
})
eh.Handle("start", func(e events.Message) {
s := &containerStats{Name: e.ID[:12]}
if cStats.add(s) {
waitFirst.Add(1)
go s.Collect(cli.client, !*noStream, waitFirst)
}
})
eh.Handle("die", func(e events.Message) {
if !*all {
cStats.remove(e.ID[:12])
}
})
eventChan := make(chan events.Message)
go eh.Watch(eventChan)
go monitorContainerEvents(started, eventChan)
defer close(eventChan)
<-started
// Start a short-lived goroutine to retrieve the initial list of
// containers.
getContainerList()
} else {
// Artificially send creation events for the containers we were asked to
// monitor (same code path than we use when monitoring all containers).
for _, name := range names {
s := &containerStats{Name: name}
if cStats.add(s) {
waitFirst.Add(1)
go s.Collect(cli.client, !*noStream, waitFirst)
}
}
// We don't expect any asynchronous errors: closeChan can be closed.
close(closeChan)
// Do a quick pause to detect any error with the provided list of
// container names.
time.Sleep(1500 * time.Millisecond)
var errs []string
cStats.mu.Lock()
for _, c := range cStats.cs {
c.mu.Lock()
if c.err != nil {
errs = append(errs, fmt.Sprintf("%s: %v", c.Name, c.err))
}
c.mu.Unlock()
}
cStats.mu.Unlock()
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, ", "))
}
}
// before print to screen, make sure each container get at least one valid stat data
waitFirst.Wait()
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
sort.Strings(names)
var (
cStats []*containerStats
w = tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
)
printHeader := func() {
if !*noStream {
fmt.Fprint(cli.out, "\033[2J")
fmt.Fprint(cli.out, "\033[H")
}
io.WriteString(w, "CONTAINER\tCPU %\tMEM USAGE / LIMIT\tMEM %\tNET I/O\tBLOCK I/O\tPIDS\n")
io.WriteString(w, "CONTAINER\tCPU %\tMEM USAGE/LIMIT\tMEM %\tNET I/O\n")
}
for _, n := range names {
s := &containerStats{Name: n}
cStats = append(cStats, s)
go s.Collect(cli, !*noStream)
}
// do a quick pause so that any failed connections for containers that do not exist are able to be
// evicted before we display the initial or default values.
time.Sleep(1500 * time.Millisecond)
var errs []string
for _, c := range cStats {
c.mu.Lock()
if c.err != nil {
errs = append(errs, fmt.Sprintf("%s: %v", c.Name, c.err))
}
c.mu.Unlock()
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, ", "))
}
for range time.Tick(500 * time.Millisecond) {
printHeader()
toRemove := []int{}
cStats.mu.Lock()
for i, s := range cStats.cs {
for i, s := range cStats {
if err := s.Display(w); err != nil && !*noStream {
toRemove = append(toRemove, i)
}
}
for j := len(toRemove) - 1; j >= 0; j-- {
i := toRemove[j]
cStats.cs = append(cStats.cs[:i], cStats.cs[i+1:]...)
cStats = append(cStats[:i], cStats[i+1:]...)
}
if len(cStats.cs) == 0 && !showAll {
if len(cStats) == 0 {
return nil
}
cStats.mu.Unlock()
w.Flush()
if *noStream {
break
}
select {
case err, ok := <-closeChan:
if ok {
if err != nil {
// this is suppressing "unexpected EOF" in the cli when the
// daemon restarts so it shutdowns cleanly
if err == io.ErrUnexpectedEOF {
return nil
}
return err
}
}
default:
// just skip
}
}
return nil
}
func calculateCPUPercent(previousCPU, previousSystem uint64, v *types.Stats) float64 {
var (
cpuPercent = 0.0
// calculate the change for the cpu usage of the container in between readings
cpuDelta = float64(v.CpuStats.CpuUsage.TotalUsage - previousCPU)
// calculate the change for the entire system between readings
systemDelta = float64(v.CpuStats.SystemUsage - previousSystem)
)
if systemDelta > 0.0 && cpuDelta > 0.0 {
cpuPercent = (cpuDelta / systemDelta) * float64(len(v.CpuStats.CpuUsage.PercpuUsage)) * 100.0
}
return cpuPercent
}

View File

@@ -1,219 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"io"
"strings"
"sync"
"time"
"github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
"github.com/docker/go-units"
"golang.org/x/net/context"
)
type containerStats struct {
Name string
CPUPercentage float64
Memory float64
MemoryLimit float64
MemoryPercentage float64
NetworkRx float64
NetworkTx float64
BlockRead float64
BlockWrite float64
PidsCurrent uint64
mu sync.RWMutex
err error
}
type stats struct {
mu sync.Mutex
cs []*containerStats
}
func (s *stats) add(cs *containerStats) bool {
s.mu.Lock()
defer s.mu.Unlock()
if _, exists := s.isKnownContainer(cs.Name); !exists {
s.cs = append(s.cs, cs)
return true
}
return false
}
func (s *stats) remove(id string) {
s.mu.Lock()
if i, exists := s.isKnownContainer(id); exists {
s.cs = append(s.cs[:i], s.cs[i+1:]...)
}
s.mu.Unlock()
}
func (s *stats) isKnownContainer(cid string) (int, bool) {
for i, c := range s.cs {
if c.Name == cid {
return i, true
}
}
return -1, false
}
func (s *containerStats) Collect(cli client.APIClient, streamStats bool, waitFirst *sync.WaitGroup) {
var (
getFirst bool
previousCPU uint64
previousSystem uint64
u = make(chan error, 1)
)
defer func() {
// if error happens and we get nothing of stats, release wait group whatever
if !getFirst {
getFirst = true
waitFirst.Done()
}
}()
responseBody, err := cli.ContainerStats(context.Background(), s.Name, streamStats)
if err != nil {
s.mu.Lock()
s.err = err
s.mu.Unlock()
return
}
defer responseBody.Close()
dec := json.NewDecoder(responseBody)
go func() {
for {
var v *types.StatsJSON
if err := dec.Decode(&v); err != nil {
u <- err
return
}
var memPercent = 0.0
var cpuPercent = 0.0
// MemoryStats.Limit will never be 0 unless the container is not running and we haven't
// got any data from cgroup
if v.MemoryStats.Limit != 0 {
memPercent = float64(v.MemoryStats.Usage) / float64(v.MemoryStats.Limit) * 100.0
}
previousCPU = v.PreCPUStats.CPUUsage.TotalUsage
previousSystem = v.PreCPUStats.SystemUsage
cpuPercent = calculateCPUPercent(previousCPU, previousSystem, v)
blkRead, blkWrite := calculateBlockIO(v.BlkioStats)
s.mu.Lock()
s.CPUPercentage = cpuPercent
s.Memory = float64(v.MemoryStats.Usage)
s.MemoryLimit = float64(v.MemoryStats.Limit)
s.MemoryPercentage = memPercent
s.NetworkRx, s.NetworkTx = calculateNetwork(v.Networks)
s.BlockRead = float64(blkRead)
s.BlockWrite = float64(blkWrite)
s.PidsCurrent = v.PidsStats.Current
s.mu.Unlock()
u <- nil
if !streamStats {
return
}
}
}()
for {
select {
case <-time.After(2 * time.Second):
// zero out the values if we have not received an update within
// the specified duration.
s.mu.Lock()
s.CPUPercentage = 0
s.Memory = 0
s.MemoryPercentage = 0
s.MemoryLimit = 0
s.NetworkRx = 0
s.NetworkTx = 0
s.BlockRead = 0
s.BlockWrite = 0
s.PidsCurrent = 0
s.mu.Unlock()
// if this is the first stat you get, release WaitGroup
if !getFirst {
getFirst = true
waitFirst.Done()
}
case err := <-u:
if err != nil {
s.mu.Lock()
s.err = err
s.mu.Unlock()
return
}
// if this is the first stat you get, release WaitGroup
if !getFirst {
getFirst = true
waitFirst.Done()
}
}
if !streamStats {
return
}
}
}
func (s *containerStats) Display(w io.Writer) error {
s.mu.RLock()
defer s.mu.RUnlock()
if s.err != nil {
return s.err
}
fmt.Fprintf(w, "%s\t%.2f%%\t%s / %s\t%.2f%%\t%s / %s\t%s / %s\t%d\n",
s.Name,
s.CPUPercentage,
units.HumanSize(s.Memory), units.HumanSize(s.MemoryLimit),
s.MemoryPercentage,
units.HumanSize(s.NetworkRx), units.HumanSize(s.NetworkTx),
units.HumanSize(s.BlockRead), units.HumanSize(s.BlockWrite),
s.PidsCurrent)
return nil
}
func calculateCPUPercent(previousCPU, previousSystem uint64, v *types.StatsJSON) float64 {
var (
cpuPercent = 0.0
// calculate the change for the cpu usage of the container in between readings
cpuDelta = float64(v.CPUStats.CPUUsage.TotalUsage) - float64(previousCPU)
// calculate the change for the entire system between readings
systemDelta = float64(v.CPUStats.SystemUsage) - float64(previousSystem)
)
if systemDelta > 0.0 && cpuDelta > 0.0 {
cpuPercent = (cpuDelta / systemDelta) * float64(len(v.CPUStats.CPUUsage.PercpuUsage)) * 100.0
}
return cpuPercent
}
func calculateBlockIO(blkio types.BlkioStats) (blkRead uint64, blkWrite uint64) {
for _, bioEntry := range blkio.IoServiceBytesRecursive {
switch strings.ToLower(bioEntry.Op) {
case "read":
blkRead = blkRead + bioEntry.Value
case "write":
blkWrite = blkWrite + bioEntry.Value
}
}
return
}
func calculateNetwork(network map[string]types.NetworkStats) (float64, float64) {
var rx, tx float64
for _, v := range network {
rx += float64(v.RxBytes)
tx += float64(v.TxBytes)
}
return rx, tx
}

View File

@@ -4,8 +4,6 @@ import (
"bytes"
"sync"
"testing"
"github.com/docker/engine-api/types"
)
func TestDisplay(t *testing.T) {
@@ -17,9 +15,6 @@ func TestDisplay(t *testing.T) {
MemoryPercentage: 100.0 / 2048.0 * 100.0,
NetworkRx: 100 * 1024 * 1024,
NetworkTx: 800 * 1024 * 1024,
BlockRead: 100 * 1024 * 1024,
BlockWrite: 800 * 1024 * 1024,
PidsCurrent: 1,
mu: sync.RWMutex{},
}
var b bytes.Buffer
@@ -27,21 +22,8 @@ func TestDisplay(t *testing.T) {
t.Fatalf("c.Display() gave error: %s", err)
}
got := b.String()
want := "app\t30.00%\t104.9 MB / 2.147 GB\t4.88%\t104.9 MB / 838.9 MB\t104.9 MB / 838.9 MB\t1\n"
want := "app\t30.00%\t104.9 MB/2.147 GB\t4.88%\t104.9 MB/838.9 MB\n"
if got != want {
t.Fatalf("c.Display() = %q, want %q", got, want)
}
}
func TestCalculBlockIO(t *testing.T) {
blkio := types.BlkioStats{
IoServiceBytesRecursive: []types.BlkioStatEntry{{8, 0, "read", 1234}, {8, 1, "read", 4567}, {8, 0, "write", 123}, {8, 1, "write", 456}},
}
blkRead, blkWrite := calculateBlockIO(blkio)
if blkRead != 5801 {
t.Fatalf("blkRead = %d, want 5801", blkRead)
}
if blkWrite != 579 {
t.Fatalf("blkWrite = %d, want 579", blkWrite)
}
}

View File

@@ -2,36 +2,40 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
"net/url"
"strconv"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdStop stops one or more containers.
// CmdStop stops one or more running containers.
//
// A running container is stopped by first sending SIGTERM and then SIGKILL if the container fails to stop within a grace period (the default is 10 seconds).
//
// Usage: docker stop [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdStop(args ...string) error {
cmd := Cli.Subcmd("stop", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["stop"].Description+".\nSending SIGTERM and then SIGKILL after a grace period", true)
cmd := Cli.Subcmd("stop", []string{"CONTAINER [CONTAINER...]"}, "Stop a running container by sending SIGTERM and then SIGKILL after a\ngrace period", true)
nSeconds := cmd.Int([]string{"t", "-time"}, 10, "Seconds to wait for stop before killing it")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
v := url.Values{}
v.Set("t", strconv.Itoa(*nSeconds))
var errNames []string
for _, name := range cmd.Args() {
if err := cli.client.ContainerStop(context.Background(), name, *nSeconds); err != nil {
errs = append(errs, err.Error())
_, _, err := readBody(cli.call("POST", "/containers/"+name+"/stop?"+v.Encode(), nil, nil))
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to stop containers: %v", errNames)
}
return nil
}

View File

@@ -1,46 +1,42 @@
package client
import (
"errors"
"golang.org/x/net/context"
"net/url"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/reference"
"github.com/docker/engine-api/types"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
)
// CmdTag tags an image into a repository.
//
// Usage: docker tag [OPTIONS] IMAGE[:TAG] [REGISTRYHOST/][USERNAME/]NAME[:TAG]
func (cli *DockerCli) CmdTag(args ...string) error {
cmd := Cli.Subcmd("tag", []string{"IMAGE[:TAG] [REGISTRYHOST/][USERNAME/]NAME[:TAG]"}, Cli.DockerCommands["tag"].Description, true)
force := cmd.Bool([]string{"#f", "#-force"}, false, "Force the tagging even if there's a conflict")
cmd := Cli.Subcmd("tag", []string{"IMAGE[:TAG] [REGISTRYHOST/][USERNAME/]NAME[:TAG]"}, "Tag an image into a repository", true)
force := cmd.Bool([]string{"f", "#force", "-force"}, false, "Force")
cmd.Require(flag.Exact, 2)
cmd.ParseFlags(args, true)
ref, err := reference.ParseNamed(cmd.Arg(1))
if err != nil {
var (
repository, tag = parsers.ParseRepositoryTag(cmd.Arg(1))
v = url.Values{}
)
//Check if the given image name can be resolved
if err := registry.ValidateRepositoryName(repository); err != nil {
return err
}
v.Set("repo", repository)
v.Set("tag", tag)
if _, isCanonical := ref.(reference.Canonical); isCanonical {
return errors.New("refusing to create a tag with a digest reference")
if *force {
v.Set("force", "1")
}
var tag string
if tagged, isTagged := ref.(reference.NamedTagged); isTagged {
tag = tagged.Tag()
if _, _, err := readBody(cli.call("POST", "/images/"+cmd.Arg(0)+"/tag?"+v.Encode(), nil, nil)); err != nil {
return err
}
options := types.ImageTagOptions{
ImageID: cmd.Arg(0),
RepositoryName: ref.Name(),
Tag: tag,
Force: *force,
}
return cli.client.ImageTag(context.Background(), options)
return nil
}

View File

@@ -1,12 +1,13 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"strings"
"text/tabwriter"
"golang.org/x/net/context"
"github.com/docker/docker/api/types"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
)
@@ -15,21 +16,28 @@ import (
//
// Usage: docker top CONTAINER
func (cli *DockerCli) CmdTop(args ...string) error {
cmd := Cli.Subcmd("top", []string{"CONTAINER [ps OPTIONS]"}, Cli.DockerCommands["top"].Description, true)
cmd := Cli.Subcmd("top", []string{"CONTAINER [ps OPTIONS]"}, "Display the running processes of a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var arguments []string
val := url.Values{}
if cmd.NArg() > 1 {
arguments = cmd.Args()[1:]
val.Set("ps_args", strings.Join(cmd.Args()[1:], " "))
}
procList, err := cli.client.ContainerTop(context.Background(), cmd.Arg(0), arguments)
serverResp, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/top?"+val.Encode(), nil, nil)
if err != nil {
return err
}
defer serverResp.body.Close()
procList := types.ContainerProcessList{}
if err := json.NewDecoder(serverResp.body).Decode(&procList); err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
fmt.Fprintln(w, strings.Join(procList.Titles, "\t"))

View File

@@ -1,48 +1,40 @@
package client
import (
"bufio"
"encoding/hex"
"encoding/json"
"errors"
"fmt"
"io"
"net"
"net/http"
"net/url"
"os"
"path"
"path/filepath"
"regexp"
"sort"
"strconv"
"strings"
"time"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
"github.com/docker/distribution/digest"
"github.com/docker/distribution/registry/client/auth"
"github.com/docker/distribution/registry/client/transport"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/distribution"
"github.com/docker/docker/pkg/jsonmessage"
"github.com/docker/docker/pkg/ansiescape"
"github.com/docker/docker/pkg/ioutils"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/reference"
"github.com/docker/docker/pkg/tlsconfig"
"github.com/docker/docker/registry"
apiclient "github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
registrytypes "github.com/docker/engine-api/types/registry"
"github.com/docker/go-connections/tlsconfig"
"github.com/docker/notary/client"
"github.com/docker/notary/passphrase"
"github.com/docker/notary/pkg/passphrase"
"github.com/docker/notary/trustmanager"
"github.com/docker/notary/tuf/data"
"github.com/docker/notary/tuf/signed"
"github.com/docker/notary/tuf/store"
"github.com/endophage/gotuf/data"
)
var (
releasesRole = path.Join(data.CanonicalTargetsRole, "releases")
untrusted bool
)
var untrusted bool
func addTrustedFlags(fs *flag.FlagSet, verify bool) {
var trusted bool
@@ -63,6 +55,8 @@ func isTrusted() bool {
return !untrusted
}
var targetRegexp = regexp.MustCompile(`([\S]+): digest: ([\S]+) size: ([\d]+)`)
type target struct {
reference registry.Reference
digest digest.Digest
@@ -85,43 +79,31 @@ func (cli *DockerCli) certificateDirectory(server string) (string, error) {
return filepath.Join(cliconfig.ConfigDir(), "tls", u.Host), nil
}
func trustServer(index *registrytypes.IndexInfo) (string, error) {
func trustServer(index *registry.IndexInfo) string {
if s := os.Getenv("DOCKER_CONTENT_TRUST_SERVER"); s != "" {
urlObj, err := url.Parse(s)
if err != nil || urlObj.Scheme != "https" {
return "", fmt.Errorf("valid https URL required for trust server, got %s", s)
if !strings.HasPrefix(s, "https://") {
return "https://" + s
}
return s, nil
return s
}
if index.Official {
return registry.NotaryServer, nil
return registry.NotaryServer
}
return "https://" + index.Name, nil
return "https://" + index.Name
}
type simpleCredentialStore struct {
auth types.AuthConfig
auth cliconfig.AuthConfig
}
func (scs simpleCredentialStore) Basic(u *url.URL) (string, string) {
return scs.auth.Username, scs.auth.Password
}
func (scs simpleCredentialStore) RefreshToken(u *url.URL, service string) string {
return scs.auth.IdentityToken
}
func (scs simpleCredentialStore) SetRefreshToken(*url.URL, string, string) {
}
// getNotaryRepository returns a NotaryRepository which stores all the
// information needed to operate on a notary repository.
// It creates a HTTP transport providing authentication support.
func (cli *DockerCli) getNotaryRepository(repoInfo *registry.RepositoryInfo, authConfig types.AuthConfig, actions ...string) (*client.NotaryRepository, error) {
server, err := trustServer(repoInfo.Index)
if err != nil {
return nil, err
func (cli *DockerCli) getNotaryRepository(repoInfo *registry.RepositoryInfo, authConfig cliconfig.AuthConfig) (*client.NotaryRepository, error) {
server := trustServer(repoInfo.Index)
if !strings.HasPrefix(server, "https://") {
return nil, errors.New("unsupported scheme: https required for trust server")
}
var cfg = tlsconfig.ClientDefault
@@ -151,7 +133,7 @@ func (cli *DockerCli) getNotaryRepository(repoInfo *registry.RepositoryInfo, aut
}
// Skip configuration headers since request is not going to Docker daemon
modifiers := registry.DockerHeaders(clientUserAgent(), http.Header{})
modifiers := registry.DockerHeaders(http.Header{})
authTransport := transport.NewTransport(base, modifiers...)
pingClient := &http.Client{
Transport: authTransport,
@@ -162,30 +144,24 @@ func (cli *DockerCli) getNotaryRepository(repoInfo *registry.RepositoryInfo, aut
if err != nil {
return nil, err
}
challengeManager := auth.NewSimpleChallengeManager()
resp, err := pingClient.Do(req)
if err != nil {
// Ignore error on ping to operate in offline mode
logrus.Debugf("Error pinging notary server %q: %s", endpointStr, err)
} else {
defer resp.Body.Close()
return nil, err
}
defer resp.Body.Close()
// Add response to the challenge manager to parse out
// authentication header and register authentication method
if err := challengeManager.AddResponse(resp); err != nil {
return nil, err
}
challengeManager := auth.NewSimpleChallengeManager()
if err := challengeManager.AddResponse(resp); err != nil {
return nil, err
}
creds := simpleCredentialStore{auth: authConfig}
tokenHandler := auth.NewTokenHandler(authTransport, creds, repoInfo.FullName(), actions...)
tokenHandler := auth.NewTokenHandler(authTransport, creds, repoInfo.CanonicalName, "push", "pull")
basicHandler := auth.NewBasicHandler(creds)
modifiers = append(modifiers, transport.RequestModifier(auth.NewAuthorizer(challengeManager, tokenHandler, basicHandler)))
tr := transport.NewTransport(base, modifiers...)
return client.NewNotaryRepository(cli.trustDirectory(), repoInfo.FullName(), server, tr, cli.getPassphraseRetriever())
return client.NewNotaryRepository(cli.trustDirectory(), repoInfo.CanonicalName, server, tr, cli.getPassphraseRetriever())
}
func convertTarget(t client.Target) (target, error) {
@@ -202,127 +178,88 @@ func convertTarget(t client.Target) (target, error) {
func (cli *DockerCli) getPassphraseRetriever() passphrase.Retriever {
aliasMap := map[string]string{
"root": "root",
"snapshot": "repository",
"targets": "repository",
"default": "repository",
"root": "offline",
"snapshot": "tagging",
"targets": "tagging",
}
baseRetriever := passphrase.PromptRetrieverWithInOut(cli.in, cli.out, aliasMap)
env := map[string]string{
"root": os.Getenv("DOCKER_CONTENT_TRUST_ROOT_PASSPHRASE"),
"snapshot": os.Getenv("DOCKER_CONTENT_TRUST_REPOSITORY_PASSPHRASE"),
"targets": os.Getenv("DOCKER_CONTENT_TRUST_REPOSITORY_PASSPHRASE"),
"default": os.Getenv("DOCKER_CONTENT_TRUST_REPOSITORY_PASSPHRASE"),
"root": os.Getenv("DOCKER_CONTENT_TRUST_OFFLINE_PASSPHRASE"),
"snapshot": os.Getenv("DOCKER_CONTENT_TRUST_TAGGING_PASSPHRASE"),
"targets": os.Getenv("DOCKER_CONTENT_TRUST_TAGGING_PASSPHRASE"),
}
// Backwards compatibility with old env names. We should remove this in 1.10
if env["root"] == "" {
if passphrase := os.Getenv("DOCKER_CONTENT_TRUST_OFFLINE_PASSPHRASE"); passphrase != "" {
env["root"] = passphrase
fmt.Fprintf(cli.err, "[DEPRECATED] The environment variable DOCKER_CONTENT_TRUST_OFFLINE_PASSPHRASE has been deprecated and will be removed in v1.10. Please use DOCKER_CONTENT_TRUST_ROOT_PASSPHRASE\n")
}
}
if env["snapshot"] == "" || env["targets"] == "" || env["default"] == "" {
if passphrase := os.Getenv("DOCKER_CONTENT_TRUST_TAGGING_PASSPHRASE"); passphrase != "" {
env["snapshot"] = passphrase
env["targets"] = passphrase
env["default"] = passphrase
fmt.Fprintf(cli.err, "[DEPRECATED] The environment variable DOCKER_CONTENT_TRUST_TAGGING_PASSPHRASE has been deprecated and will be removed in v1.10. Please use DOCKER_CONTENT_TRUST_REPOSITORY_PASSPHRASE\n")
}
}
return func(keyName string, alias string, createNew bool, numAttempts int) (string, bool, error) {
if v := env[alias]; v != "" {
return v, numAttempts > 1, nil
}
// For non-root roles, we can also try the "default" alias if it is specified
if v := env["default"]; v != "" && alias != data.CanonicalRootRole {
return v, numAttempts > 1, nil
}
return baseRetriever(keyName, alias, createNew, numAttempts)
}
}
func (cli *DockerCli) trustedReference(ref reference.NamedTagged) (reference.Canonical, error) {
repoInfo, err := registry.ParseRepositoryInfo(ref)
func (cli *DockerCli) trustedReference(repo string, ref registry.Reference) (registry.Reference, error) {
repoInfo, err := registry.ParseRepositoryInfo(repo)
if err != nil {
return nil, err
}
// Resolve the Auth config relevant for this server
authConfig := cli.resolveAuthConfig(repoInfo.Index)
authConfig := registry.ResolveAuthConfig(cli.configFile, repoInfo.Index)
notaryRepo, err := cli.getNotaryRepository(repoInfo, authConfig, "pull")
notaryRepo, err := cli.getNotaryRepository(repoInfo, authConfig)
if err != nil {
fmt.Fprintf(cli.out, "Error establishing connection to trust repository: %s\n", err)
return nil, err
}
t, err := notaryRepo.GetTargetByName(ref.Tag(), releasesRole, data.CanonicalTargetsRole)
t, err := notaryRepo.GetTargetByName(ref.String())
if err != nil {
return nil, err
}
// Only list tags in the top level targets role or the releases delegation role - ignore
// all other delegation roles
if t.Role != releasesRole && t.Role != data.CanonicalTargetsRole {
return nil, notaryError(repoInfo.FullName(), fmt.Errorf("No trust data for %s", ref.Tag()))
}
r, err := convertTarget(t.Target)
r, err := convertTarget(*t)
if err != nil {
return nil, err
}
return reference.WithDigest(ref, r.digest)
return registry.DigestReference(r.digest), nil
}
func (cli *DockerCli) tagTrusted(trustedRef reference.Canonical, ref reference.NamedTagged) error {
fmt.Fprintf(cli.out, "Tagging %s as %s\n", trustedRef.String(), ref.String())
func (cli *DockerCli) tagTrusted(repoInfo *registry.RepositoryInfo, trustedRef, ref registry.Reference) error {
fullName := trustedRef.ImageName(repoInfo.LocalName)
fmt.Fprintf(cli.out, "Tagging %s as %s\n", fullName, ref.ImageName(repoInfo.LocalName))
tv := url.Values{}
tv.Set("repo", repoInfo.LocalName)
tv.Set("tag", ref.String())
tv.Set("force", "1")
options := types.ImageTagOptions{
ImageID: trustedRef.String(),
RepositoryName: trustedRef.Name(),
Tag: ref.Tag(),
Force: true,
if _, _, err := readBody(cli.call("POST", "/images/"+fullName+"/tag?"+tv.Encode(), nil, nil)); err != nil {
return err
}
return cli.client.ImageTag(context.Background(), options)
return nil
}
func notaryError(repoName string, err error) error {
func notaryError(err error) error {
switch err.(type) {
case *json.SyntaxError:
logrus.Debugf("Notary syntax error: %s", err)
return fmt.Errorf("Error: no trust data available for remote repository %s. Try running notary server and setting DOCKER_CONTENT_TRUST_SERVER to its HTTPS address?", repoName)
case signed.ErrExpired:
return fmt.Errorf("Error: remote repository %s out-of-date: %v", repoName, err)
return errors.New("no trust data available for remote repository")
case client.ErrExpired:
return fmt.Errorf("remote repository out-of-date: %v", err)
case trustmanager.ErrKeyNotFound:
return fmt.Errorf("Error: signing keys for remote repository %s not found: %v", repoName, err)
case *net.OpError:
return fmt.Errorf("Error: error contacting notary server: %v", err)
case store.ErrMetaNotFound:
return fmt.Errorf("Error: trust data missing for remote repository %s or remote repository not found: %v", repoName, err)
case signed.ErrInvalidKeyType:
return fmt.Errorf("Warning: potential malicious behavior - trust data mismatch for remote repository %s: %v", repoName, err)
case signed.ErrNoKeys:
return fmt.Errorf("Error: could not find signing keys for remote repository %s, or could not decrypt signing key: %v", repoName, err)
case signed.ErrLowVersion:
return fmt.Errorf("Warning: potential malicious behavior - trust data version is lower than expected for remote repository %s: %v", repoName, err)
case signed.ErrRoleThreshold:
return fmt.Errorf("Warning: potential malicious behavior - trust data has insufficient signatures for remote repository %s: %v", repoName, err)
case client.ErrRepositoryNotExist:
return fmt.Errorf("Error: remote trust data does not exist for %s: %v", repoName, err)
case signed.ErrInsufficientSignatures:
return fmt.Errorf("Error: could not produce valid signature for %s. If Yubikey was used, was touch input provided?: %v", repoName, err)
return fmt.Errorf("signing keys not found: %v", err)
}
return err
}
func (cli *DockerCli) trustedPull(repoInfo *registry.RepositoryInfo, ref registry.Reference, authConfig types.AuthConfig, requestPrivilege apiclient.RequestPrivilegeFunc) error {
var refs []target
func (cli *DockerCli) trustedPull(repoInfo *registry.RepositoryInfo, ref registry.Reference, authConfig cliconfig.AuthConfig) error {
var (
v = url.Values{}
refs = []target{}
)
notaryRepo, err := cli.getNotaryRepository(repoInfo, authConfig, "pull")
notaryRepo, err := cli.getNotaryRepository(repoInfo, authConfig)
if err != nil {
fmt.Fprintf(cli.out, "Error establishing connection to trust repository: %s\n", err)
return err
@@ -330,39 +267,24 @@ func (cli *DockerCli) trustedPull(repoInfo *registry.RepositoryInfo, ref registr
if ref.String() == "" {
// List all targets
targets, err := notaryRepo.ListTargets(releasesRole, data.CanonicalTargetsRole)
targets, err := notaryRepo.ListTargets()
if err != nil {
return notaryError(repoInfo.FullName(), err)
return notaryError(err)
}
for _, tgt := range targets {
t, err := convertTarget(tgt.Target)
t, err := convertTarget(*tgt)
if err != nil {
fmt.Fprintf(cli.out, "Skipping target for %q\n", repoInfo.Name())
continue
}
// Only list tags in the top level targets role or the releases delegation role - ignore
// all other delegation roles
if tgt.Role != releasesRole && tgt.Role != data.CanonicalTargetsRole {
fmt.Fprintf(cli.out, "Skipping target for %q\n", repoInfo.LocalName)
continue
}
refs = append(refs, t)
}
if len(refs) == 0 {
return notaryError(repoInfo.FullName(), fmt.Errorf("No trusted tags for %s", repoInfo.FullName()))
}
} else {
t, err := notaryRepo.GetTargetByName(ref.String(), releasesRole, data.CanonicalTargetsRole)
t, err := notaryRepo.GetTargetByName(ref.String())
if err != nil {
return notaryError(repoInfo.FullName(), err)
return notaryError(err)
}
// Only get the tag if it's in the top level targets role or the releases delegation role
// ignore it if it's in any other delegation roles
if t.Role != releasesRole && t.Role != data.CanonicalTargetsRole {
return notaryError(repoInfo.FullName(), fmt.Errorf("No trust data for %s", ref.String()))
}
logrus.Debugf("retrieving target for %s role\n", t.Role)
r, err := convertTarget(t.Target)
r, err := convertTarget(*t)
if err != nil {
return err
@@ -370,190 +292,163 @@ func (cli *DockerCli) trustedPull(repoInfo *registry.RepositoryInfo, ref registr
refs = append(refs, r)
}
v.Set("fromImage", repoInfo.LocalName)
for i, r := range refs {
displayTag := r.reference.String()
if displayTag != "" {
displayTag = ":" + displayTag
}
fmt.Fprintf(cli.out, "Pull (%d of %d): %s%s@%s\n", i+1, len(refs), repoInfo.Name(), displayTag, r.digest)
fmt.Fprintf(cli.out, "Pull (%d of %d): %s%s@%s\n", i+1, len(refs), repoInfo.LocalName, displayTag, r.digest)
v.Set("tag", r.digest.String())
if err := cli.imagePullPrivileged(authConfig, repoInfo.Name(), r.digest.String(), requestPrivilege); err != nil {
_, _, err = cli.clientRequestAttemptLogin("POST", "/images/create?"+v.Encode(), nil, cli.out, repoInfo.Index, "pull")
if err != nil {
return err
}
// If reference is not trusted, tag by trusted reference
if !r.reference.HasDigest() {
tagged, err := reference.WithTag(repoInfo, r.reference.String())
if err != nil {
return err
}
trustedRef, err := reference.WithDigest(repoInfo, r.digest)
if err != nil {
return err
}
if err := cli.tagTrusted(trustedRef, tagged); err != nil {
if err := cli.tagTrusted(repoInfo, registry.DigestReference(r.digest), r.reference); err != nil {
return err
}
}
}
return nil
}
func (cli *DockerCli) trustedPush(repoInfo *registry.RepositoryInfo, tag string, authConfig types.AuthConfig, requestPrivilege apiclient.RequestPrivilegeFunc) error {
responseBody, err := cli.imagePushPrivileged(authConfig, repoInfo.Name(), tag, requestPrivilege)
func selectKey(keys map[string]string) string {
if len(keys) == 0 {
return ""
}
keyIDs := []string{}
for k := range keys {
keyIDs = append(keyIDs, k)
}
// TODO(dmcgowan): let user choose if multiple keys, now pick consistently
sort.Strings(keyIDs)
return keyIDs[0]
}
func targetStream(in io.Writer) (io.WriteCloser, <-chan []target) {
r, w := io.Pipe()
out := io.MultiWriter(in, w)
targetChan := make(chan []target)
go func() {
targets := []target{}
scanner := bufio.NewScanner(r)
scanner.Split(ansiescape.ScanANSILines)
for scanner.Scan() {
line := scanner.Bytes()
if matches := targetRegexp.FindSubmatch(line); len(matches) == 4 {
dgst, err := digest.ParseDigest(string(matches[2]))
if err != nil {
// Line does match what is expected, continue looking for valid lines
logrus.Debugf("Bad digest value %q in matched line, ignoring\n", string(matches[2]))
continue
}
s, err := strconv.ParseInt(string(matches[3]), 10, 64)
if err != nil {
// Line does match what is expected, continue looking for valid lines
logrus.Debugf("Bad size value %q in matched line, ignoring\n", string(matches[3]))
continue
}
targets = append(targets, target{
reference: registry.ParseReference(string(matches[1])),
digest: dgst,
size: s,
})
}
}
targetChan <- targets
}()
return ioutils.NewWriteCloserWrapper(out, w.Close), targetChan
}
func (cli *DockerCli) trustedPush(repoInfo *registry.RepositoryInfo, tag string, authConfig cliconfig.AuthConfig) error {
streamOut, targetChan := targetStream(cli.out)
v := url.Values{}
v.Set("tag", tag)
_, _, err := cli.clientRequestAttemptLogin("POST", "/images/"+repoInfo.LocalName+"/push?"+v.Encode(), nil, streamOut, repoInfo.Index, "push")
// Close stream channel to finish target parsing
if err := streamOut.Close(); err != nil {
return err
}
// Check error from request
if err != nil {
return err
}
defer responseBody.Close()
// Get target results
targets := <-targetChan
// If it is a trusted push we would like to find the target entry which match the
// tag provided in the function and then do an AddTarget later.
target := &client.Target{}
// Count the times of calling for handleTarget,
// if it is called more that once, that should be considered an error in a trusted push.
cnt := 0
handleTarget := func(aux *json.RawMessage) {
cnt++
if cnt > 1 {
// handleTarget should only be called one. This will be treated as an error.
return
}
var pushResult distribution.PushResult
err := json.Unmarshal(*aux, &pushResult)
if err == nil && pushResult.Tag != "" && pushResult.Digest.Validate() == nil {
h, err := hex.DecodeString(pushResult.Digest.Hex())
if err != nil {
target = nil
return
}
target.Name = registry.ParseReference(pushResult.Tag).String()
target.Hashes = data.Hashes{string(pushResult.Digest.Algorithm()): h}
target.Length = int64(pushResult.Size)
}
}
// We want trust signatures to always take an explicit tag,
// otherwise it will act as an untrusted push.
if tag == "" {
if err = jsonmessage.DisplayJSONMessagesStream(responseBody, cli.out, cli.outFd, cli.isTerminalOut, nil); err != nil {
return err
}
fmt.Fprintln(cli.out, "No tag specified, skipping trust metadata push")
fmt.Fprintf(cli.out, "No tag specified, skipping trust metadata push\n")
return nil
}
if len(targets) == 0 {
fmt.Fprintf(cli.out, "No targets found, skipping trust metadata push\n")
return nil
}
if err = jsonmessage.DisplayJSONMessagesStream(responseBody, cli.out, cli.outFd, cli.isTerminalOut, handleTarget); err != nil {
return err
}
fmt.Fprintf(cli.out, "Signing and pushing trust metadata\n")
if cnt > 1 {
return fmt.Errorf("internal error: only one call to handleTarget expected")
}
if target == nil {
fmt.Fprintln(cli.out, "No targets found, please provide a specific tag in order to sign it")
return nil
}
fmt.Fprintln(cli.out, "Signing and pushing trust metadata")
repo, err := cli.getNotaryRepository(repoInfo, authConfig, "push", "pull")
repo, err := cli.getNotaryRepository(repoInfo, authConfig)
if err != nil {
fmt.Fprintf(cli.out, "Error establishing connection to notary repository: %s\n", err)
return err
}
// get the latest repository metadata so we can figure out which roles to sign
_, err = repo.Update(false)
switch err.(type) {
case client.ErrRepoNotInitialized, client.ErrRepositoryNotExist:
keys := repo.CryptoService.ListKeys(data.CanonicalRootRole)
var rootKeyID string
// always select the first root key
if len(keys) > 0 {
sort.Strings(keys)
rootKeyID = keys[0]
} else {
rootPublicKey, err := repo.CryptoService.Create(data.CanonicalRootRole, "", data.ECDSAKey)
if err != nil {
return err
}
rootKeyID = rootPublicKey.ID()
for _, target := range targets {
h, err := hex.DecodeString(target.digest.Hex())
if err != nil {
return err
}
// Initialize the notary repository with a remotely managed snapshot key
if err := repo.Initialize(rootKeyID, data.CanonicalSnapshotRole); err != nil {
return notaryError(repoInfo.FullName(), err)
t := &client.Target{
Name: target.reference.String(),
Hashes: data.Hashes{
string(target.digest.Algorithm()): h,
},
Length: int64(target.size),
}
if err := repo.AddTarget(t); err != nil {
return err
}
fmt.Fprintf(cli.out, "Finished initializing %q\n", repoInfo.FullName())
err = repo.AddTarget(target, data.CanonicalTargetsRole)
case nil:
// already initialized and we have successfully downloaded the latest metadata
err = cli.addTargetToAllSignableRoles(repo, target)
default:
return notaryError(repoInfo.FullName(), err)
}
if err == nil {
err = repo.Publish()
err = repo.Publish()
if _, ok := err.(*client.ErrRepoNotInitialized); !ok {
return notaryError(err)
}
if err != nil {
fmt.Fprintf(cli.out, "Failed to sign %q:%s - %s\n", repoInfo.FullName(), tag, err.Error())
return notaryError(repoInfo.FullName(), err)
ks := repo.KeyStoreManager
keys := ks.RootKeyStore().ListKeys()
rootKey := selectKey(keys)
if rootKey == "" {
rootKey, err = ks.GenRootKey("ecdsa")
if err != nil {
return err
}
}
fmt.Fprintf(cli.out, "Successfully signed %q:%s\n", repoInfo.FullName(), tag)
return nil
}
// Attempt to add the image target to all the top level delegation roles we can
// (based on whether we have the signing key and whether the role's path allows
// us to).
// If there are no delegation roles, we add to the targets role.
func (cli *DockerCli) addTargetToAllSignableRoles(repo *client.NotaryRepository, target *client.Target) error {
var signableRoles []string
// translate the full key names, which includes the GUN, into just the key IDs
allCanonicalKeyIDs := make(map[string]struct{})
for fullKeyID := range repo.CryptoService.ListAllKeys() {
allCanonicalKeyIDs[path.Base(fullKeyID)] = struct{}{}
}
allDelegationRoles, err := repo.GetDelegationRoles()
cryptoService, err := ks.GetRootCryptoService(rootKey)
if err != nil {
return err
}
// if there are no delegation roles, then just try to sign it into the targets role
if len(allDelegationRoles) == 0 {
return repo.AddTarget(target, data.CanonicalTargetsRole)
if err := repo.Initialize(cryptoService); err != nil {
return notaryError(err)
}
fmt.Fprintf(cli.out, "Finished initializing %q\n", repoInfo.CanonicalName)
// there are delegation roles, find every delegation role we have a key for, and
// attempt to sign into into all those roles.
for _, delegationRole := range allDelegationRoles {
// We do not support signing any delegation role that isn't a direct child of the targets role.
// Also don't bother checking the keys if we can't add the target
// to this role due to path restrictions
if path.Dir(delegationRole.Name) != data.CanonicalTargetsRole || !delegationRole.CheckPaths(target.Name) {
continue
}
for _, canonicalKeyID := range delegationRole.KeyIDs {
if _, ok := allCanonicalKeyIDs[canonicalKeyID]; ok {
signableRoles = append(signableRoles, delegationRole.Name)
break
}
}
}
if len(signableRoles) == 0 {
return fmt.Errorf("no valid signing keys for delegation roles")
}
return repo.AddTarget(target, signableRoles...)
return notaryError(repo.Publish())
}

View File

@@ -1,56 +0,0 @@
package client
import (
"os"
"testing"
"github.com/docker/docker/registry"
registrytypes "github.com/docker/engine-api/types/registry"
)
func unsetENV() {
os.Unsetenv("DOCKER_CONTENT_TRUST")
os.Unsetenv("DOCKER_CONTENT_TRUST_SERVER")
}
func TestENVTrustServer(t *testing.T) {
defer unsetENV()
indexInfo := &registrytypes.IndexInfo{Name: "testserver"}
if err := os.Setenv("DOCKER_CONTENT_TRUST_SERVER", "https://notary-test.com:5000"); err != nil {
t.Fatal("Failed to set ENV variable")
}
output, err := trustServer(indexInfo)
expectedStr := "https://notary-test.com:5000"
if err != nil || output != expectedStr {
t.Fatalf("Expected server to be %s, got %s", expectedStr, output)
}
}
func TestHTTPENVTrustServer(t *testing.T) {
defer unsetENV()
indexInfo := &registrytypes.IndexInfo{Name: "testserver"}
if err := os.Setenv("DOCKER_CONTENT_TRUST_SERVER", "http://notary-test.com:5000"); err != nil {
t.Fatal("Failed to set ENV variable")
}
_, err := trustServer(indexInfo)
if err == nil {
t.Fatal("Expected error with invalid scheme")
}
}
func TestOfficialTrustServer(t *testing.T) {
indexInfo := &registrytypes.IndexInfo{Name: "testserver", Official: true}
output, err := trustServer(indexInfo)
if err != nil || output != registry.NotaryServer {
t.Fatalf("Expected server to be %s, got %s", registry.NotaryServer, output)
}
}
func TestNonOfficialTrustServer(t *testing.T) {
indexInfo := &registrytypes.IndexInfo{Name: "testserver", Official: false}
output, err := trustServer(indexInfo)
expectedStr := "https://" + indexInfo.Name
if err != nil || output != expectedStr {
t.Fatalf("Expected server to be %s, got %s", expectedStr, output)
}
}

View File

@@ -2,9 +2,6 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -14,21 +11,22 @@ import (
//
// Usage: docker unpause CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdUnpause(args ...string) error {
cmd := Cli.Subcmd("unpause", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["unpause"].Description, true)
cmd := Cli.Subcmd("unpause", []string{"CONTAINER [CONTAINER...]"}, "Unpause all processes within a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
var errNames []string
for _, name := range cmd.Args() {
if err := cli.client.ContainerUnpause(context.Background(), name); err != nil {
errs = append(errs, err.Error())
if _, _, err := readBody(cli.call("POST", fmt.Sprintf("/containers/%s/unpause", name), nil, nil)); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to unpause containers: %v", errNames)
}
return nil
}

View File

@@ -1,117 +0,0 @@
package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types/container"
"github.com/docker/go-units"
)
// CmdUpdate updates resources of one or more containers.
//
// Usage: docker update [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdUpdate(args ...string) error {
cmd := Cli.Subcmd("update", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["update"].Description, true)
flBlkioWeight := cmd.Uint16([]string{"-blkio-weight"}, 0, "Block IO (relative weight), between 10 and 1000")
flCPUPeriod := cmd.Int64([]string{"-cpu-period"}, 0, "Limit CPU CFS (Completely Fair Scheduler) period")
flCPUQuota := cmd.Int64([]string{"-cpu-quota"}, 0, "Limit CPU CFS (Completely Fair Scheduler) quota")
flCpusetCpus := cmd.String([]string{"-cpuset-cpus"}, "", "CPUs in which to allow execution (0-3, 0,1)")
flCpusetMems := cmd.String([]string{"-cpuset-mems"}, "", "MEMs in which to allow execution (0-3, 0,1)")
flCPUShares := cmd.Int64([]string{"#c", "-cpu-shares"}, 0, "CPU shares (relative weight)")
flMemoryString := cmd.String([]string{"m", "-memory"}, "", "Memory limit")
flMemoryReservation := cmd.String([]string{"-memory-reservation"}, "", "Memory soft limit")
flMemorySwap := cmd.String([]string{"-memory-swap"}, "", "Swap limit equal to memory plus swap: '-1' to enable unlimited swap")
flKernelMemory := cmd.String([]string{"-kernel-memory"}, "", "Kernel memory limit")
flRestartPolicy := cmd.String([]string{"-restart"}, "", "Restart policy to apply when a container exits")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
if cmd.NFlag() == 0 {
return fmt.Errorf("You must provide one or more flags when using this command.")
}
var err error
var flMemory int64
if *flMemoryString != "" {
flMemory, err = units.RAMInBytes(*flMemoryString)
if err != nil {
return err
}
}
var memoryReservation int64
if *flMemoryReservation != "" {
memoryReservation, err = units.RAMInBytes(*flMemoryReservation)
if err != nil {
return err
}
}
var memorySwap int64
if *flMemorySwap != "" {
if *flMemorySwap == "-1" {
memorySwap = -1
} else {
memorySwap, err = units.RAMInBytes(*flMemorySwap)
if err != nil {
return err
}
}
}
var kernelMemory int64
if *flKernelMemory != "" {
kernelMemory, err = units.RAMInBytes(*flKernelMemory)
if err != nil {
return err
}
}
var restartPolicy container.RestartPolicy
if *flRestartPolicy != "" {
restartPolicy, err = opts.ParseRestartPolicy(*flRestartPolicy)
if err != nil {
return err
}
}
resources := container.Resources{
BlkioWeight: *flBlkioWeight,
CpusetCpus: *flCpusetCpus,
CpusetMems: *flCpusetMems,
CPUShares: *flCPUShares,
Memory: flMemory,
MemoryReservation: memoryReservation,
MemorySwap: memorySwap,
KernelMemory: kernelMemory,
CPUPeriod: *flCPUPeriod,
CPUQuota: *flCPUQuota,
}
updateConfig := container.UpdateConfig{
Resources: resources,
RestartPolicy: restartPolicy,
}
names := cmd.Args()
var errs []string
for _, name := range names {
if err := cli.client.ContainerUpdate(context.Background(), name, updateConfig); err != nil {
errs = append(errs, err.Error())
} else {
fmt.Fprintf(cli.out, "%s\n", name)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
}
return nil
}

View File

@@ -1,119 +1,324 @@
package client
import (
"bytes"
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"io"
"io/ioutil"
"net/http"
"net/url"
"os"
gosignal "os/signal"
"path/filepath"
"runtime"
"strconv"
"strings"
"time"
"golang.org/x/net/context"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api"
"github.com/docker/docker/api/types"
"github.com/docker/docker/autogen/dockerversion"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/pkg/jsonmessage"
"github.com/docker/docker/pkg/signal"
"github.com/docker/docker/pkg/stdcopy"
"github.com/docker/docker/pkg/term"
"github.com/docker/docker/registry"
"github.com/docker/engine-api/client"
"github.com/docker/engine-api/types"
registrytypes "github.com/docker/engine-api/types/registry"
)
func (cli *DockerCli) electAuthServer() string {
// The daemon `/info` endpoint informs us of the default registry being
// used. This is essential in cross-platforms environment, where for
// example a Linux client might be interacting with a Windows daemon, hence
// the default registry URL might be Windows specific.
serverAddress := registry.IndexServer
if info, err := cli.client.Info(context.Background()); err != nil {
fmt.Fprintf(cli.out, "Warning: failed to get default registry endpoint from daemon (%v). Using system default: %s\n", err, serverAddress)
} else {
serverAddress = info.IndexServerAddress
}
return serverAddress
var (
errConnectionRefused = errors.New("Cannot connect to the Docker daemon. Is 'docker -d' running on this host?")
)
type serverResponse struct {
body io.ReadCloser
header http.Header
statusCode int
}
// encodeAuthToBase64 serializes the auth configuration as JSON base64 payload
func encodeAuthToBase64(authConfig types.AuthConfig) (string, error) {
buf, err := json.Marshal(authConfig)
if err != nil {
return "", err
}
return base64.URLEncoding.EncodeToString(buf), nil
// HTTPClient creates a new HTTP client with the cli's client transport instance.
func (cli *DockerCli) HTTPClient() *http.Client {
return &http.Client{Transport: cli.transport}
}
func (cli *DockerCli) registryAuthenticationPrivilegedFunc(index *registrytypes.IndexInfo, cmdName string) client.RequestPrivilegeFunc {
return func() (string, error) {
fmt.Fprintf(cli.out, "\nPlease login prior to %s:\n", cmdName)
indexServer := registry.GetAuthConfigKey(index)
authConfig, err := cli.configureAuth("", "", indexServer, false)
if err != nil {
return "", err
func (cli *DockerCli) encodeData(data interface{}) (*bytes.Buffer, error) {
params := bytes.NewBuffer(nil)
if data != nil {
if err := json.NewEncoder(params).Encode(data); err != nil {
return nil, err
}
return encodeAuthToBase64(authConfig)
}
return params, nil
}
func (cli *DockerCli) clientRequest(method, path string, in io.Reader, headers map[string][]string) (*serverResponse, error) {
serverResp := &serverResponse{
body: nil,
statusCode: -1,
}
expectedPayload := (method == "POST" || method == "PUT")
if expectedPayload && in == nil {
in = bytes.NewReader([]byte{})
}
req, err := http.NewRequest(method, fmt.Sprintf("%s/v%s%s", cli.basePath, api.Version, path), in)
if err != nil {
return serverResp, err
}
// Add CLI Config's HTTP Headers BEFORE we set the Docker headers
// then the user can't change OUR headers
for k, v := range cli.configFile.HTTPHeaders {
req.Header.Set(k, v)
}
req.Header.Set("User-Agent", "Docker-Client/"+dockerversion.VERSION+" ("+runtime.GOOS+")")
req.URL.Host = cli.addr
req.URL.Scheme = cli.scheme
if headers != nil {
for k, v := range headers {
req.Header[k] = v
}
}
if expectedPayload && req.Header.Get("Content-Type") == "" {
req.Header.Set("Content-Type", "text/plain")
}
resp, err := cli.HTTPClient().Do(req)
if resp != nil {
serverResp.statusCode = resp.StatusCode
}
if err != nil {
if strings.Contains(err.Error(), "connection refused") {
return serverResp, errConnectionRefused
}
if cli.tlsConfig == nil {
return serverResp, fmt.Errorf("%v.\n* Are you trying to connect to a TLS-enabled daemon without TLS?\n* Is your docker daemon up and running?", err)
}
if cli.tlsConfig != nil && strings.Contains(err.Error(), "remote error: bad certificate") {
return serverResp, fmt.Errorf("The server probably has client authentication (--tlsverify) enabled. Please check your TLS client certification settings: %v", err)
}
return serverResp, fmt.Errorf("An error occurred trying to connect: %v", err)
}
if serverResp.statusCode < 200 || serverResp.statusCode >= 400 {
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return serverResp, err
}
if len(body) == 0 {
return serverResp, fmt.Errorf("Error: request returned %s for API route and version %s, check if the server supports the requested API version", http.StatusText(serverResp.statusCode), req.URL)
}
return serverResp, fmt.Errorf("Error response from daemon: %s", bytes.TrimSpace(body))
}
serverResp.body = resp.Body
serverResp.header = resp.Header
return serverResp, nil
}
func (cli *DockerCli) clientRequestAttemptLogin(method, path string, in io.Reader, out io.Writer, index *registry.IndexInfo, cmdName string) (io.ReadCloser, int, error) {
cmdAttempt := func(authConfig cliconfig.AuthConfig) (io.ReadCloser, int, error) {
buf, err := json.Marshal(authConfig)
if err != nil {
return nil, -1, err
}
registryAuthHeader := []string{
base64.URLEncoding.EncodeToString(buf),
}
// begin the request
serverResp, err := cli.clientRequest(method, path, in, map[string][]string{
"X-Registry-Auth": registryAuthHeader,
})
if err == nil && out != nil {
// If we are streaming output, complete the stream since
// errors may not appear until later.
err = cli.streamBody(serverResp.body, serverResp.header.Get("Content-Type"), true, out, nil)
}
if err != nil {
// Since errors in a stream appear after status 200 has been written,
// we may need to change the status code.
if strings.Contains(err.Error(), "Authentication is required") ||
strings.Contains(err.Error(), "Status 401") ||
strings.Contains(err.Error(), "401 Unauthorized") ||
strings.Contains(err.Error(), "status code 401") {
serverResp.statusCode = http.StatusUnauthorized
}
}
return serverResp.body, serverResp.statusCode, err
}
// Resolve the Auth config relevant for this server
authConfig := registry.ResolveAuthConfig(cli.configFile, index)
body, statusCode, err := cmdAttempt(authConfig)
if statusCode == http.StatusUnauthorized {
fmt.Fprintf(cli.out, "\nPlease login prior to %s:\n", cmdName)
if err = cli.CmdLogin(index.GetAuthConfigKey()); err != nil {
return nil, -1, err
}
authConfig = registry.ResolveAuthConfig(cli.configFile, index)
return cmdAttempt(authConfig)
}
return body, statusCode, err
}
func (cli *DockerCli) callWrapper(method, path string, data interface{}, headers map[string][]string) (io.ReadCloser, http.Header, int, error) {
sr, err := cli.call(method, path, data, headers)
return sr.body, sr.header, sr.statusCode, err
}
func (cli *DockerCli) call(method, path string, data interface{}, headers map[string][]string) (*serverResponse, error) {
params, err := cli.encodeData(data)
if err != nil {
sr := &serverResponse{
body: nil,
header: nil,
statusCode: -1,
}
return sr, nil
}
if data != nil {
if headers == nil {
headers = make(map[string][]string)
}
headers["Content-Type"] = []string{"application/json"}
}
serverResp, err := cli.clientRequest(method, path, params, headers)
return serverResp, err
}
type streamOpts struct {
rawTerminal bool
in io.Reader
out io.Writer
err io.Writer
headers map[string][]string
}
func (cli *DockerCli) stream(method, path string, opts *streamOpts) (*serverResponse, error) {
serverResp, err := cli.clientRequest(method, path, opts.in, opts.headers)
if err != nil {
return serverResp, err
}
return serverResp, cli.streamBody(serverResp.body, serverResp.header.Get("Content-Type"), opts.rawTerminal, opts.out, opts.err)
}
func (cli *DockerCli) streamBody(body io.ReadCloser, contentType string, rawTerminal bool, stdout, stderr io.Writer) error {
defer body.Close()
if api.MatchesContentType(contentType, "application/json") {
return jsonmessage.DisplayJSONMessagesStream(body, stdout, cli.outFd, cli.isTerminalOut)
}
if stdout != nil || stderr != nil {
// When TTY is ON, use regular copy
var err error
if rawTerminal {
_, err = io.Copy(stdout, body)
} else {
_, err = stdcopy.StdCopy(stdout, stderr, body)
}
logrus.Debugf("[stream] End of stdout")
return err
}
return nil
}
func (cli *DockerCli) resizeTty(id string, isExec bool) {
height, width := cli.getTtySize()
cli.resizeTtyTo(id, height, width, isExec)
}
func (cli *DockerCli) resizeTtyTo(id string, height, width int, isExec bool) {
if height == 0 && width == 0 {
return
}
v := url.Values{}
v.Set("h", strconv.Itoa(height))
v.Set("w", strconv.Itoa(width))
options := types.ResizeOptions{
ID: id,
Height: height,
Width: width,
}
var err error
if isExec {
err = cli.client.ContainerExecResize(context.Background(), options)
path := ""
if !isExec {
path = "/containers/" + id + "/resize?"
} else {
err = cli.client.ContainerResize(context.Background(), options)
path = "/exec/" + id + "/resize?"
}
if err != nil {
if _, _, err := readBody(cli.call("POST", path+v.Encode(), nil, nil)); err != nil {
logrus.Debugf("Error resize: %s", err)
}
}
func waitForExit(cli *DockerCli, containerID string) (int, error) {
serverResp, err := cli.call("POST", "/containers/"+containerID+"/wait", nil, nil)
if err != nil {
return -1, err
}
defer serverResp.body.Close()
var res types.ContainerWaitResponse
if err := json.NewDecoder(serverResp.body).Decode(&res); err != nil {
return -1, err
}
return res.StatusCode, nil
}
// getExitCode perform an inspect on the container. It returns
// the running state and the exit code.
func getExitCode(cli *DockerCli, containerID string) (bool, int, error) {
c, err := cli.client.ContainerInspect(context.Background(), containerID)
serverResp, err := cli.call("GET", "/containers/"+containerID+"/json", nil, nil)
if err != nil {
// If we can't connect, then the daemon probably died.
if err != client.ErrConnectionFailed {
if err != errConnectionRefused {
return false, -1, err
}
return false, -1, nil
}
defer serverResp.body.Close()
var c types.ContainerJSON
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return false, -1, err
}
return c.State.Running, c.State.ExitCode, nil
}
// getExecExitCode perform an inspect on the exec command. It returns
// the running state and the exit code.
func getExecExitCode(cli *DockerCli, execID string) (bool, int, error) {
resp, err := cli.client.ContainerExecInspect(context.Background(), execID)
serverResp, err := cli.call("GET", "/exec/"+execID+"/json", nil, nil)
if err != nil {
// If we can't connect, then the daemon probably died.
if err != client.ErrConnectionFailed {
if err != errConnectionRefused {
return false, -1, err
}
return false, -1, nil
}
return resp.Running, resp.ExitCode, nil
defer serverResp.body.Close()
//TODO: Should we reconsider having a type in api/types?
//this is a response to exex/id/json not container
var c struct {
Running bool
ExitCode int
}
if err := json.NewDecoder(serverResp.body).Decode(&c); err != nil {
return false, -1, err
}
return c.Running, c.ExitCode, nil
}
func (cli *DockerCli) monitorTtySize(id string, isExec bool) error {
@@ -159,44 +364,16 @@ func (cli *DockerCli) getTtySize() (int, int) {
return int(ws.Height), int(ws.Width)
}
func copyToFile(outfile string, r io.Reader) error {
tmpFile, err := ioutil.TempFile(filepath.Dir(outfile), ".docker_temp_")
func readBody(serverResp *serverResponse, err error) ([]byte, int, error) {
if serverResp.body != nil {
defer serverResp.body.Close()
}
if err != nil {
return err
return nil, serverResp.statusCode, err
}
tmpPath := tmpFile.Name()
_, err = io.Copy(tmpFile, r)
tmpFile.Close()
body, err := ioutil.ReadAll(serverResp.body)
if err != nil {
os.Remove(tmpPath)
return err
return nil, -1, err
}
if err = os.Rename(tmpPath, outfile); err != nil {
os.Remove(tmpPath)
return err
}
return nil
}
// resolveAuthConfig is like registry.ResolveAuthConfig, but if using the
// default index, it uses the default index name for the daemon's platform,
// not the client's platform.
func (cli *DockerCli) resolveAuthConfig(index *registrytypes.IndexInfo) types.AuthConfig {
configKey := index.Name
if index.Official {
configKey = cli.electAuthServer()
}
a, _ := getCredentials(cli.configFile, configKey)
return a
}
func (cli *DockerCli) retrieveAuthConfigs() map[string]types.AuthConfig {
acs, _ := getAllCredentials(cli.configFile)
return acs
return body, serverResp.statusCode, nil
}

View File

@@ -1,23 +1,21 @@
package client
import (
"encoding/json"
"runtime"
"text/template"
"time"
"golang.org/x/net/context"
"github.com/docker/docker/api"
"github.com/docker/docker/api/types"
"github.com/docker/docker/autogen/dockerversion"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/dockerversion"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/utils"
"github.com/docker/docker/utils/templates"
"github.com/docker/engine-api/types"
)
var versionTemplate = `Client:
var VersionTemplate = `Client:
Version: {{.Client.Version}}
API version: {{.Client.APIVersion}}
API version: {{.Client.ApiVersion}}
Go version: {{.Client.GoVersion}}
Git commit: {{.Client.GitCommit}}
Built: {{.Client.BuildTime}}
@@ -26,70 +24,73 @@ var versionTemplate = `Client:
Server:
Version: {{.Server.Version}}
API version: {{.Server.APIVersion}}
API version: {{.Server.ApiVersion}}
Go version: {{.Server.GoVersion}}
Git commit: {{.Server.GitCommit}}
Built: {{.Server.BuildTime}}
OS/Arch: {{.Server.Os}}/{{.Server.Arch}}{{if .Server.Experimental}}
Experimental: {{.Server.Experimental}}{{end}}{{end}}`
type VersionData struct {
Client types.Version
ServerOK bool
Server types.Version
}
// CmdVersion shows Docker version information.
//
// Available version information is shown for: client Docker version, client API version, client Go version, client Git commit, client OS/Arch, server Docker version, server API version, server Go version, server Git commit, and server OS/Arch.
//
// Usage: docker version
func (cli *DockerCli) CmdVersion(args ...string) (err error) {
cmd := Cli.Subcmd("version", nil, Cli.DockerCommands["version"].Description, true)
cmd := Cli.Subcmd("version", nil, "Show the Docker version information.", true)
tmplStr := cmd.String([]string{"f", "#format", "-format"}, "", "Format the output using the given go template")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
templateFormat := versionTemplate
if *tmplStr != "" {
templateFormat = *tmplStr
if *tmplStr == "" {
*tmplStr = VersionTemplate
}
var tmpl *template.Template
if tmpl, err = templates.Parse(templateFormat); err != nil {
if tmpl, err = template.New("").Funcs(funcMap).Parse(*tmplStr); err != nil {
return Cli.StatusError{StatusCode: 64,
Status: "Template parsing error: " + err.Error()}
}
vd := types.VersionResponse{
Client: &types.Version{
Version: dockerversion.Version,
APIVersion: cli.client.ClientVersion(),
vd := VersionData{
Client: types.Version{
Version: dockerversion.VERSION,
ApiVersion: api.Version,
GoVersion: runtime.Version(),
GitCommit: dockerversion.GitCommit,
BuildTime: dockerversion.BuildTime,
GitCommit: dockerversion.GITCOMMIT,
BuildTime: dockerversion.BUILDTIME,
Os: runtime.GOOS,
Arch: runtime.GOARCH,
Experimental: utils.ExperimentalBuild(),
},
}
serverVersion, err := cli.client.ServerVersion(context.Background())
if err == nil {
vd.Server = &serverVersion
}
// first we need to make BuildTime more human friendly
t, errTime := time.Parse(time.RFC3339Nano, vd.Client.BuildTime)
if errTime == nil {
vd.Client.BuildTime = t.Format(time.ANSIC)
}
if vd.ServerOK() {
t, errTime = time.Parse(time.RFC3339Nano, vd.Server.BuildTime)
if errTime == nil {
vd.Server.BuildTime = t.Format(time.ANSIC)
defer func() {
if err2 := tmpl.Execute(cli.out, vd); err2 != nil && err == nil {
err = err2
}
cli.out.Write([]byte{'\n'})
}()
serverResp, err := cli.call("GET", "/version", nil, nil)
if err != nil {
return err
}
if err2 := tmpl.Execute(cli.out, vd); err2 != nil && err == nil {
err = err2
defer serverResp.body.Close()
if err = json.NewDecoder(serverResp.body).Decode(&vd.Server); err != nil {
return Cli.StatusError{StatusCode: 1,
Status: "Error reading remote version: " + err.Error()}
}
cli.out.Write([]byte{'\n'})
return err
vd.ServerOK = true
return
}

View File

@@ -1,177 +0,0 @@
package client
import (
"fmt"
"sort"
"text/tabwriter"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/filters"
)
// CmdVolume is the parent subcommand for all volume commands
//
// Usage: docker volume <COMMAND> <OPTS>
func (cli *DockerCli) CmdVolume(args ...string) error {
description := Cli.DockerCommands["volume"].Description + "\n\nCommands:\n"
commands := [][]string{
{"create", "Create a volume"},
{"inspect", "Return low-level information on a volume"},
{"ls", "List volumes"},
{"rm", "Remove a volume"},
}
for _, cmd := range commands {
description += fmt.Sprintf(" %-25.25s%s\n", cmd[0], cmd[1])
}
description += "\nRun 'docker volume COMMAND --help' for more information on a command"
cmd := Cli.Subcmd("volume", []string{"[COMMAND]"}, description, false)
cmd.Require(flag.Exact, 0)
err := cmd.ParseFlags(args, true)
cmd.Usage()
return err
}
// CmdVolumeLs outputs a list of Docker volumes.
//
// Usage: docker volume ls [OPTIONS]
func (cli *DockerCli) CmdVolumeLs(args ...string) error {
cmd := Cli.Subcmd("volume ls", nil, "List volumes", true)
quiet := cmd.Bool([]string{"q", "-quiet"}, false, "Only display volume names")
flFilter := opts.NewListOpts(nil)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Provide filter values (i.e. 'dangling=true')")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
volFilterArgs := filters.NewArgs()
for _, f := range flFilter.GetAll() {
var err error
volFilterArgs, err = filters.ParseFlag(f, volFilterArgs)
if err != nil {
return err
}
}
volumes, err := cli.client.VolumeList(context.Background(), volFilterArgs)
if err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
if !*quiet {
for _, warn := range volumes.Warnings {
fmt.Fprintln(cli.err, warn)
}
fmt.Fprintf(w, "DRIVER \tVOLUME NAME")
fmt.Fprintf(w, "\n")
}
sort.Sort(byVolumeName(volumes.Volumes))
for _, vol := range volumes.Volumes {
if *quiet {
fmt.Fprintln(w, vol.Name)
continue
}
fmt.Fprintf(w, "%s\t%s\n", vol.Driver, vol.Name)
}
w.Flush()
return nil
}
type byVolumeName []*types.Volume
func (r byVolumeName) Len() int { return len(r) }
func (r byVolumeName) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r byVolumeName) Less(i, j int) bool {
return r[i].Name < r[j].Name
}
// CmdVolumeInspect displays low-level information on one or more volumes.
//
// Usage: docker volume inspect [OPTIONS] VOLUME [VOLUME...]
func (cli *DockerCli) CmdVolumeInspect(args ...string) error {
cmd := Cli.Subcmd("volume inspect", []string{"VOLUME [VOLUME...]"}, "Return low-level information on a volume", true)
tmplStr := cmd.String([]string{"f", "-format"}, "", "Format the output using the given go template")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
if err := cmd.Parse(args); err != nil {
return nil
}
inspectSearcher := func(name string) (interface{}, []byte, error) {
i, err := cli.client.VolumeInspect(context.Background(), name)
return i, nil, err
}
return cli.inspectElements(*tmplStr, cmd.Args(), inspectSearcher)
}
// CmdVolumeCreate creates a new volume.
//
// Usage: docker volume create [OPTIONS]
func (cli *DockerCli) CmdVolumeCreate(args ...string) error {
cmd := Cli.Subcmd("volume create", nil, "Create a volume", true)
flDriver := cmd.String([]string{"d", "-driver"}, "local", "Specify volume driver name")
flName := cmd.String([]string{"-name"}, "", "Specify volume name")
flDriverOpts := opts.NewMapOpts(nil, nil)
cmd.Var(flDriverOpts, []string{"o", "-opt"}, "Set driver specific options")
flLabels := opts.NewListOpts(nil)
cmd.Var(&flLabels, []string{"-label"}, "Set metadata for a volume")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
volReq := types.VolumeCreateRequest{
Driver: *flDriver,
DriverOpts: flDriverOpts.GetAll(),
Name: *flName,
Labels: runconfigopts.ConvertKVStringsToMap(flLabels.GetAll()),
}
vol, err := cli.client.VolumeCreate(context.Background(), volReq)
if err != nil {
return err
}
fmt.Fprintf(cli.out, "%s\n", vol.Name)
return nil
}
// CmdVolumeRm removes one or more volumes.
//
// Usage: docker volume rm VOLUME [VOLUME...]
func (cli *DockerCli) CmdVolumeRm(args ...string) error {
cmd := Cli.Subcmd("volume rm", []string{"VOLUME [VOLUME...]"}, "Remove a volume", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var status = 0
for _, name := range cmd.Args() {
if err := cli.client.VolumeRemove(context.Background(), name); err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
status = 1
continue
}
fmt.Fprintf(cli.out, "%s\n", name)
}
if status != 0 {
return Cli.StatusError{StatusCode: status}
}
return nil
}

View File

@@ -2,9 +2,6 @@ package client
import (
"fmt"
"strings"
"golang.org/x/net/context"
Cli "github.com/docker/docker/cli"
flag "github.com/docker/docker/pkg/mflag"
@@ -16,22 +13,23 @@ import (
//
// Usage: docker wait CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdWait(args ...string) error {
cmd := Cli.Subcmd("wait", []string{"CONTAINER [CONTAINER...]"}, Cli.DockerCommands["wait"].Description, true)
cmd := Cli.Subcmd("wait", []string{"CONTAINER [CONTAINER...]"}, "Block until a container stops, then print its exit code.", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errs []string
var errNames []string
for _, name := range cmd.Args() {
status, err := cli.client.ContainerWait(context.Background(), name)
status, err := waitForExit(cli, name)
if err != nil {
errs = append(errs, err.Error())
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%d\n", status)
}
}
if len(errs) > 0 {
return fmt.Errorf("%s", strings.Join(errs, "\n"))
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to wait containers: %v", errNames)
}
return nil
}

View File

@@ -5,66 +5,50 @@ import (
"mime"
"path/filepath"
"sort"
"strconv"
"strings"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/system"
"github.com/docker/docker/pkg/version"
"github.com/docker/engine-api/types"
"github.com/docker/libtrust"
)
// Common constants for daemon and client.
const (
// Version of Current REST API
DefaultVersion version.Version = "1.23"
// Current REST API version
Version version.Version = "1.20"
// MinVersion represents Minimum REST API version supported
// Minimun REST API version supported
MinVersion version.Version = "1.12"
// NoBaseImageSpecifier is the symbol used by the FROM
// command to specify that no base image is to be used.
NoBaseImageSpecifier string = "scratch"
// Default filename with Docker commands, read by docker build
DefaultDockerfileName string = "Dockerfile"
)
// byPortInfo is a temporary type used to sort types.Port by its fields
type byPortInfo []types.Port
type ByPrivatePort []types.Port
func (r byPortInfo) Len() int { return len(r) }
func (r byPortInfo) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r byPortInfo) Less(i, j int) bool {
if r[i].PrivatePort != r[j].PrivatePort {
return r[i].PrivatePort < r[j].PrivatePort
}
func (r ByPrivatePort) Len() int { return len(r) }
func (r ByPrivatePort) Swap(i, j int) { r[i], r[j] = r[j], r[i] }
func (r ByPrivatePort) Less(i, j int) bool { return r[i].PrivatePort < r[j].PrivatePort }
if r[i].IP != r[j].IP {
return r[i].IP < r[j].IP
}
if r[i].PublicPort != r[j].PublicPort {
return r[i].PublicPort < r[j].PublicPort
}
return r[i].Type < r[j].Type
}
// DisplayablePorts returns formatted string representing open ports of container
// e.g. "0.0.0.0:80->9090/tcp, 9988/tcp"
// it's used by command 'docker ps'
func DisplayablePorts(ports []types.Port) string {
type portGroup struct {
first int
last int
}
groupMap := make(map[string]*portGroup)
var result []string
var hostMappings []string
var groupMapKeys []string
sort.Sort(byPortInfo(ports))
var (
result = []string{}
hostMappings = []string{}
firstInGroupMap map[string]int
lastInGroupMap map[string]int
)
firstInGroupMap = make(map[string]int)
lastInGroupMap = make(map[string]int)
sort.Sort(ByPrivatePort(ports))
for _, port := range ports {
current := port.PrivatePort
portKey := port.Type
var (
current = port.PrivatePort
portKey = port.Type
firstInGroup int
lastInGroup int
)
if port.IP != "" {
if port.PublicPort != current {
hostMappings = append(hostMappings, fmt.Sprintf("%s:%d->%d/%s", port.IP, port.PublicPort, port.PrivatePort, port.Type))
@@ -72,41 +56,45 @@ func DisplayablePorts(ports []types.Port) string {
}
portKey = fmt.Sprintf("%s/%s", port.IP, port.Type)
}
group := groupMap[portKey]
firstInGroup = firstInGroupMap[portKey]
lastInGroup = lastInGroupMap[portKey]
if group == nil {
groupMap[portKey] = &portGroup{first: current, last: current}
// record order that groupMap keys are created
groupMapKeys = append(groupMapKeys, portKey)
continue
}
if current == (group.last + 1) {
group.last = current
if firstInGroup == 0 {
firstInGroupMap[portKey] = current
lastInGroupMap[portKey] = current
continue
}
result = append(result, formGroup(portKey, group.first, group.last))
groupMap[portKey] = &portGroup{first: current, last: current}
if current == (lastInGroup + 1) {
lastInGroupMap[portKey] = current
continue
}
result = append(result, FormGroup(portKey, firstInGroup, lastInGroup))
firstInGroupMap[portKey] = current
lastInGroupMap[portKey] = current
}
for _, portKey := range groupMapKeys {
g := groupMap[portKey]
result = append(result, formGroup(portKey, g.first, g.last))
for portKey, firstInGroup := range firstInGroupMap {
result = append(result, FormGroup(portKey, firstInGroup, lastInGroupMap[portKey]))
}
result = append(result, hostMappings...)
return strings.Join(result, ", ")
}
func formGroup(key string, start, last int) string {
parts := strings.Split(key, "/")
groupType := parts[0]
var ip string
func FormGroup(key string, start, last int) string {
var (
group string
parts = strings.Split(key, "/")
groupType = parts[0]
ip = ""
)
if len(parts) > 1 {
ip = parts[0]
groupType = parts[1]
}
group := strconv.Itoa(start)
if start != last {
group = fmt.Sprintf("%s-%d", group, last)
if start == last {
group = fmt.Sprintf("%d", start)
} else {
group = fmt.Sprintf("%d-%d", start, last)
}
if ip != "" {
group = fmt.Sprintf("%s:%s->%s", ip, group, group)
@@ -114,7 +102,6 @@ func formGroup(key string, start, last int) string {
return fmt.Sprintf("%s/%s", group, groupType)
}
// MatchesContentType validates the content type against the expected one
func MatchesContentType(contentType, expectedType string) bool {
mimetype, _, err := mime.ParseMediaType(contentType)
if err != nil {

View File

@@ -1,341 +0,0 @@
package api
import (
"io/ioutil"
"path/filepath"
"testing"
"os"
"github.com/docker/engine-api/types"
)
type ports struct {
ports []types.Port
expected string
}
// DisplayablePorts
func TestDisplayablePorts(t *testing.T) {
cases := []ports{
{
[]types.Port{
{
PrivatePort: 9988,
Type: "tcp",
},
},
"9988/tcp"},
{
[]types.Port{
{
PrivatePort: 9988,
Type: "udp",
},
},
"9988/udp",
},
{
[]types.Port{
{
IP: "0.0.0.0",
PrivatePort: 9988,
Type: "tcp",
},
},
"0.0.0.0:0->9988/tcp",
},
{
[]types.Port{
{
PrivatePort: 9988,
PublicPort: 8899,
Type: "tcp",
},
},
"9988/tcp",
},
{
[]types.Port{
{
IP: "4.3.2.1",
PrivatePort: 9988,
PublicPort: 8899,
Type: "tcp",
},
},
"4.3.2.1:8899->9988/tcp",
},
{
[]types.Port{
{
IP: "4.3.2.1",
PrivatePort: 9988,
PublicPort: 9988,
Type: "tcp",
},
},
"4.3.2.1:9988->9988/tcp",
},
{
[]types.Port{
{
PrivatePort: 9988,
Type: "udp",
}, {
PrivatePort: 9988,
Type: "udp",
},
},
"9988/udp, 9988/udp",
},
{
[]types.Port{
{
IP: "1.2.3.4",
PublicPort: 9998,
PrivatePort: 9998,
Type: "udp",
}, {
IP: "1.2.3.4",
PublicPort: 9999,
PrivatePort: 9999,
Type: "udp",
},
},
"1.2.3.4:9998-9999->9998-9999/udp",
},
{
[]types.Port{
{
IP: "1.2.3.4",
PublicPort: 8887,
PrivatePort: 9998,
Type: "udp",
}, {
IP: "1.2.3.4",
PublicPort: 8888,
PrivatePort: 9999,
Type: "udp",
},
},
"1.2.3.4:8887->9998/udp, 1.2.3.4:8888->9999/udp",
},
{
[]types.Port{
{
PrivatePort: 9998,
Type: "udp",
}, {
PrivatePort: 9999,
Type: "udp",
},
},
"9998-9999/udp",
},
{
[]types.Port{
{
IP: "1.2.3.4",
PrivatePort: 6677,
PublicPort: 7766,
Type: "tcp",
}, {
PrivatePort: 9988,
PublicPort: 8899,
Type: "udp",
},
},
"9988/udp, 1.2.3.4:7766->6677/tcp",
},
{
[]types.Port{
{
IP: "1.2.3.4",
PrivatePort: 9988,
PublicPort: 8899,
Type: "udp",
}, {
IP: "1.2.3.4",
PrivatePort: 9988,
PublicPort: 8899,
Type: "tcp",
}, {
IP: "4.3.2.1",
PrivatePort: 2233,
PublicPort: 3322,
Type: "tcp",
},
},
"4.3.2.1:3322->2233/tcp, 1.2.3.4:8899->9988/tcp, 1.2.3.4:8899->9988/udp",
},
{
[]types.Port{
{
PrivatePort: 9988,
PublicPort: 8899,
Type: "udp",
}, {
IP: "1.2.3.4",
PrivatePort: 6677,
PublicPort: 7766,
Type: "tcp",
}, {
IP: "4.3.2.1",
PrivatePort: 2233,
PublicPort: 3322,
Type: "tcp",
},
},
"9988/udp, 4.3.2.1:3322->2233/tcp, 1.2.3.4:7766->6677/tcp",
},
{
[]types.Port{
{
PrivatePort: 80,
Type: "tcp",
}, {
PrivatePort: 1024,
Type: "tcp",
}, {
PrivatePort: 80,
Type: "udp",
}, {
PrivatePort: 1024,
Type: "udp",
}, {
IP: "1.1.1.1",
PublicPort: 80,
PrivatePort: 1024,
Type: "tcp",
}, {
IP: "1.1.1.1",
PublicPort: 80,
PrivatePort: 1024,
Type: "udp",
}, {
IP: "1.1.1.1",
PublicPort: 1024,
PrivatePort: 80,
Type: "tcp",
}, {
IP: "1.1.1.1",
PublicPort: 1024,
PrivatePort: 80,
Type: "udp",
}, {
IP: "2.1.1.1",
PublicPort: 80,
PrivatePort: 1024,
Type: "tcp",
}, {
IP: "2.1.1.1",
PublicPort: 80,
PrivatePort: 1024,
Type: "udp",
}, {
IP: "2.1.1.1",
PublicPort: 1024,
PrivatePort: 80,
Type: "tcp",
}, {
IP: "2.1.1.1",
PublicPort: 1024,
PrivatePort: 80,
Type: "udp",
},
},
"80/tcp, 80/udp, 1024/tcp, 1024/udp, 1.1.1.1:1024->80/tcp, 1.1.1.1:1024->80/udp, 2.1.1.1:1024->80/tcp, 2.1.1.1:1024->80/udp, 1.1.1.1:80->1024/tcp, 1.1.1.1:80->1024/udp, 2.1.1.1:80->1024/tcp, 2.1.1.1:80->1024/udp",
},
}
for _, port := range cases {
actual := DisplayablePorts(port.ports)
if port.expected != actual {
t.Fatalf("Expected %s, got %s.", port.expected, actual)
}
}
}
// MatchesContentType
func TestJsonContentType(t *testing.T) {
if !MatchesContentType("application/json", "application/json") {
t.Fail()
}
if !MatchesContentType("application/json; charset=utf-8", "application/json") {
t.Fail()
}
if MatchesContentType("dockerapplication/json", "application/json") {
t.Fail()
}
}
// LoadOrCreateTrustKey
func TestLoadOrCreateTrustKeyInvalidKeyFile(t *testing.T) {
tmpKeyFolderPath, err := ioutil.TempDir("", "api-trustkey-test")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(tmpKeyFolderPath)
tmpKeyFile, err := ioutil.TempFile(tmpKeyFolderPath, "keyfile")
if err != nil {
t.Fatal(err)
}
if _, err := LoadOrCreateTrustKey(tmpKeyFile.Name()); err == nil {
t.Fatalf("expected an error, got nothing.")
}
}
func TestLoadOrCreateTrustKeyCreateKey(t *testing.T) {
tmpKeyFolderPath, err := ioutil.TempDir("", "api-trustkey-test")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(tmpKeyFolderPath)
// Without the need to create the folder hierarchy
tmpKeyFile := filepath.Join(tmpKeyFolderPath, "keyfile")
if key, err := LoadOrCreateTrustKey(tmpKeyFile); err != nil || key == nil {
t.Fatalf("expected a new key file, got : %v and %v", err, key)
}
if _, err := os.Stat(tmpKeyFile); err != nil {
t.Fatalf("Expected to find a file %s, got %v", tmpKeyFile, err)
}
// With the need to create the folder hierarchy as tmpKeyFie is in a path
// where some folders do not exist.
tmpKeyFile = filepath.Join(tmpKeyFolderPath, "folder/hierarchy/keyfile")
if key, err := LoadOrCreateTrustKey(tmpKeyFile); err != nil || key == nil {
t.Fatalf("expected a new key file, got : %v and %v", err, key)
}
if _, err := os.Stat(tmpKeyFile); err != nil {
t.Fatalf("Expected to find a file %s, got %v", tmpKeyFile, err)
}
// With no path at all
defer os.Remove("keyfile")
if key, err := LoadOrCreateTrustKey("keyfile"); err != nil || key == nil {
t.Fatalf("expected a new key file, got : %v and %v", err, key)
}
if _, err := os.Stat("keyfile"); err != nil {
t.Fatalf("Expected to find a file keyfile, got %v", err)
}
}
func TestLoadOrCreateTrustKeyLoadValidKey(t *testing.T) {
tmpKeyFile := filepath.Join("fixtures", "keyfile")
if key, err := LoadOrCreateTrustKey(tmpKeyFile); err != nil || key == nil {
t.Fatalf("expected a key file, got : %v and %v", err, key)
}
}

View File

@@ -1,7 +0,0 @@
-----BEGIN EC PRIVATE KEY-----
keyID: AWX2:I27X:WQFX:IOMK:CNAK:O7PW:VYNB:ZLKC:CVAE:YJP2:SI4A:XXAY
MHcCAQEEILHTRWdcpKWsnORxSFyBnndJ4ROU41hMtr/GCiLVvwBQoAoGCCqGSM49
AwEHoUQDQgAElpVFbQ2V2UQKajqdE3fVxJ+/pE/YuEFOxWbOxF2be19BY209/iky
NzeFFK7SLpQ4CBJ7zDVXOHsMzrkY/GquGA==
-----END EC PRIVATE KEY-----

56
api/server/form.go Normal file
View File

@@ -0,0 +1,56 @@
package server
import (
"fmt"
"net/http"
"strconv"
"strings"
)
func boolValue(r *http.Request, k string) bool {
s := strings.ToLower(strings.TrimSpace(r.FormValue(k)))
return !(s == "" || s == "0" || s == "no" || s == "false" || s == "none")
}
// boolValueOrDefault returns the default bool passed if the query param is
// missing, otherwise it's just a proxy to boolValue above
func boolValueOrDefault(r *http.Request, k string, d bool) bool {
if _, ok := r.Form[k]; !ok {
return d
}
return boolValue(r, k)
}
func int64ValueOrZero(r *http.Request, k string) int64 {
val, err := strconv.ParseInt(r.FormValue(k), 10, 64)
if err != nil {
return 0
}
return val
}
type archiveOptions struct {
name string
path string
}
func archiveFormValues(r *http.Request, vars map[string]string) (archiveOptions, error) {
if vars == nil {
return archiveOptions{}, fmt.Errorf("Missing parameter")
}
if err := parseForm(r); err != nil {
return archiveOptions{}, err
}
name := vars["name"]
path := r.Form.Get("path")
switch {
case name == "":
return archiveOptions{}, fmt.Errorf("bad parameter: 'name' cannot be empty")
case path == "":
return archiveOptions{}, fmt.Errorf("bad parameter: 'path' cannot be empty")
}
return archiveOptions{name, path}, nil
}

View File

@@ -1,4 +1,4 @@
package httputils
package server
import (
"net/http"
@@ -26,7 +26,7 @@ func TestBoolValue(t *testing.T) {
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
a := BoolValue(r, "test")
a := boolValue(r, "test")
if a != e {
t.Fatalf("Value: %s, expected: %v, actual: %v", c, e, a)
}
@@ -35,7 +35,7 @@ func TestBoolValue(t *testing.T) {
func TestBoolValueOrDefault(t *testing.T) {
r, _ := http.NewRequest("GET", "", nil)
if !BoolValueOrDefault(r, "queryparam", true) {
if !boolValueOrDefault(r, "queryparam", true) {
t.Fatal("Expected to get true default value, got false")
}
@@ -43,7 +43,7 @@ func TestBoolValueOrDefault(t *testing.T) {
v.Set("param", "")
r, _ = http.NewRequest("GET", "", nil)
r.Form = v
if BoolValueOrDefault(r, "param", true) {
if boolValueOrDefault(r, "param", true) {
t.Fatal("Expected not to get true")
}
}
@@ -62,44 +62,9 @@ func TestInt64ValueOrZero(t *testing.T) {
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
a := Int64ValueOrZero(r, "test")
a := int64ValueOrZero(r, "test")
if a != e {
t.Fatalf("Value: %s, expected: %v, actual: %v", c, e, a)
}
}
}
func TestInt64ValueOrDefault(t *testing.T) {
cases := map[string]int64{
"": -1,
"-1": -1,
"42": 42,
}
for c, e := range cases {
v := url.Values{}
v.Set("test", c)
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
a, err := Int64ValueOrDefault(r, "test", -1)
if a != e {
t.Fatalf("Value: %s, expected: %v, actual: %v", c, e, a)
}
if err != nil {
t.Fatalf("Error should be nil, but received: %s", err)
}
}
}
func TestInt64ValueOrDefaultWithError(t *testing.T) {
v := url.Values{}
v.Set("test", "invalid")
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
_, err := Int64ValueOrDefault(r, "test", -1)
if err == nil {
t.Fatalf("Expected an error.")
}
}

View File

@@ -1,69 +0,0 @@
package httputils
import (
"net/http"
"strings"
"github.com/Sirupsen/logrus"
)
// httpStatusError is an interface
// that errors with custom status codes
// implement to tell the api layer
// which response status to set.
type httpStatusError interface {
HTTPErrorStatusCode() int
}
// inputValidationError is an interface
// that errors generated by invalid
// inputs can implement to tell the
// api layer to set a 400 status code
// in the response.
type inputValidationError interface {
IsValidationError() bool
}
// WriteError decodes a specific docker error and sends it in the response.
func WriteError(w http.ResponseWriter, err error) {
if err == nil || w == nil {
logrus.WithFields(logrus.Fields{"error": err, "writer": w}).Error("unexpected HTTP error handling")
return
}
var statusCode int
errMsg := err.Error()
switch e := err.(type) {
case httpStatusError:
statusCode = e.HTTPErrorStatusCode()
case inputValidationError:
statusCode = http.StatusBadRequest
default:
// FIXME: this is brittle and should not be necessary, but we still need to identify if
// there are errors falling back into this logic.
// If we need to differentiate between different possible error types,
// we should create appropriate error types that implement the httpStatusError interface.
errStr := strings.ToLower(errMsg)
for keyword, status := range map[string]int{
"not found": http.StatusNotFound,
"no such": http.StatusNotFound,
"bad parameter": http.StatusBadRequest,
"conflict": http.StatusConflict,
"impossible": http.StatusNotAcceptable,
"wrong login/password": http.StatusUnauthorized,
"hasn't been activated": http.StatusForbidden,
} {
if strings.Contains(errStr, keyword) {
statusCode = status
break
}
}
}
if statusCode == 0 {
statusCode = http.StatusInternalServerError
}
http.Error(w, errMsg, statusCode)
}

View File

@@ -1,73 +0,0 @@
package httputils
import (
"fmt"
"net/http"
"path/filepath"
"strconv"
"strings"
)
// BoolValue transforms a form value in different formats into a boolean type.
func BoolValue(r *http.Request, k string) bool {
s := strings.ToLower(strings.TrimSpace(r.FormValue(k)))
return !(s == "" || s == "0" || s == "no" || s == "false" || s == "none")
}
// BoolValueOrDefault returns the default bool passed if the query param is
// missing, otherwise it's just a proxy to boolValue above
func BoolValueOrDefault(r *http.Request, k string, d bool) bool {
if _, ok := r.Form[k]; !ok {
return d
}
return BoolValue(r, k)
}
// Int64ValueOrZero parses a form value into an int64 type.
// It returns 0 if the parsing fails.
func Int64ValueOrZero(r *http.Request, k string) int64 {
val, err := Int64ValueOrDefault(r, k, 0)
if err != nil {
return 0
}
return val
}
// Int64ValueOrDefault parses a form value into an int64 type. If there is an
// error, returns the error. If there is no value returns the default value.
func Int64ValueOrDefault(r *http.Request, field string, def int64) (int64, error) {
if r.Form.Get(field) != "" {
value, err := strconv.ParseInt(r.Form.Get(field), 10, 64)
if err != nil {
return value, err
}
return value, nil
}
return def, nil
}
// ArchiveOptions stores archive information for different operations.
type ArchiveOptions struct {
Name string
Path string
}
// ArchiveFormValues parses form values and turns them into ArchiveOptions.
// It fails if the archive name and path are not in the request.
func ArchiveFormValues(r *http.Request, vars map[string]string) (ArchiveOptions, error) {
if err := ParseForm(r); err != nil {
return ArchiveOptions{}, err
}
name := vars["name"]
path := filepath.FromSlash(r.Form.Get("path"))
switch {
case name == "":
return ArchiveOptions{}, fmt.Errorf("bad parameter: 'name' cannot be empty")
case path == "":
return ArchiveOptions{}, fmt.Errorf("bad parameter: 'path' cannot be empty")
}
return ArchiveOptions{name, path}, nil
}

View File

@@ -1,107 +0,0 @@
package httputils
import (
"encoding/json"
"fmt"
"io"
"net/http"
"strings"
"golang.org/x/net/context"
"github.com/docker/docker/api"
"github.com/docker/docker/pkg/version"
)
// APIVersionKey is the client's requested API version.
const APIVersionKey = "api-version"
// UAStringKey is used as key type for user-agent string in net/context struct
const UAStringKey = "upstream-user-agent"
// APIFunc is an adapter to allow the use of ordinary functions as Docker API endpoints.
// Any function that has the appropriate signature can be registered as a API endpoint (e.g. getVersion).
type APIFunc func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error
// HijackConnection interrupts the http response writer to get the
// underlying connection and operate with it.
func HijackConnection(w http.ResponseWriter) (io.ReadCloser, io.Writer, error) {
conn, _, err := w.(http.Hijacker).Hijack()
if err != nil {
return nil, nil, err
}
// Flush the options to make sure the client sets the raw mode
conn.Write([]byte{})
return conn, conn, nil
}
// CloseStreams ensures that a list for http streams are properly closed.
func CloseStreams(streams ...interface{}) {
for _, stream := range streams {
if tcpc, ok := stream.(interface {
CloseWrite() error
}); ok {
tcpc.CloseWrite()
} else if closer, ok := stream.(io.Closer); ok {
closer.Close()
}
}
}
// CheckForJSON makes sure that the request's Content-Type is application/json.
func CheckForJSON(r *http.Request) error {
ct := r.Header.Get("Content-Type")
// No Content-Type header is ok as long as there's no Body
if ct == "" {
if r.Body == nil || r.ContentLength == 0 {
return nil
}
}
// Otherwise it better be json
if api.MatchesContentType(ct, "application/json") {
return nil
}
return fmt.Errorf("Content-Type specified (%s) must be 'application/json'", ct)
}
// ParseForm ensures the request form is parsed even with invalid content types.
// If we don't do this, POST method without Content-type (even with empty body) will fail.
func ParseForm(r *http.Request) error {
if r == nil {
return nil
}
if err := r.ParseForm(); err != nil && !strings.HasPrefix(err.Error(), "mime:") {
return err
}
return nil
}
// ParseMultipartForm ensures the request form is parsed, even with invalid content types.
func ParseMultipartForm(r *http.Request) error {
if err := r.ParseMultipartForm(4096); err != nil && !strings.HasPrefix(err.Error(), "mime:") {
return err
}
return nil
}
// WriteJSON writes the value v to the http response stream as json with standard json encoding.
func WriteJSON(w http.ResponseWriter, code int, v interface{}) error {
w.Header().Set("Content-Type", "application/json")
w.WriteHeader(code)
return json.NewEncoder(w).Encode(v)
}
// VersionFromContext returns an API version from the context using APIVersionKey.
// It panics if the context value does not have version.Version type.
func VersionFromContext(ctx context.Context) (ver version.Version) {
if ctx == nil {
return
}
val := ctx.Value(APIVersionKey)
if val == nil {
return
}
return val.(version.Version)
}

View File

@@ -1,41 +0,0 @@
package server
import (
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api"
"github.com/docker/docker/api/server/httputils"
"github.com/docker/docker/api/server/middleware"
"github.com/docker/docker/dockerversion"
"github.com/docker/docker/pkg/authorization"
)
// handleWithGlobalMiddlwares wraps the handler function for a request with
// the server's global middlewares. The order of the middlewares is backwards,
// meaning that the first in the list will be evaluated last.
func (s *Server) handleWithGlobalMiddlewares(handler httputils.APIFunc) httputils.APIFunc {
next := handler
handleVersion := middleware.NewVersionMiddleware(dockerversion.Version, api.DefaultVersion, api.MinVersion)
next = handleVersion(next)
if s.cfg.EnableCors {
handleCORS := middleware.NewCORSMiddleware(s.cfg.CorsHeaders)
next = handleCORS(next)
}
handleUserAgent := middleware.NewUserAgentMiddleware(s.cfg.Version)
next = handleUserAgent(next)
// Only want this on debug level
if s.cfg.Logging && logrus.GetLevel() == logrus.DebugLevel {
next = middleware.DebugRequestMiddleware(next)
}
if len(s.cfg.AuthorizationPluginNames) > 0 {
s.authZPlugins = authorization.NewPlugins(s.cfg.AuthorizationPluginNames)
handleAuthorization := middleware.NewAuthorizationMiddleware(s.authZPlugins)
next = handleAuthorization(next)
}
return next
}

View File

@@ -1,42 +0,0 @@
package middleware
import (
"net/http"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/server/httputils"
"github.com/docker/docker/pkg/authorization"
"golang.org/x/net/context"
)
// NewAuthorizationMiddleware creates a new Authorization middleware.
func NewAuthorizationMiddleware(plugins []authorization.Plugin) Middleware {
return func(handler httputils.APIFunc) httputils.APIFunc {
return func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
// FIXME: fill when authN gets in
// User and UserAuthNMethod are taken from AuthN plugins
// Currently tracked in https://github.com/docker/docker/pull/13994
user := ""
userAuthNMethod := ""
authCtx := authorization.NewCtx(plugins, user, userAuthNMethod, r.Method, r.RequestURI)
if err := authCtx.AuthZRequest(w, r); err != nil {
logrus.Errorf("AuthZRequest for %s %s returned error: %s", r.Method, r.RequestURI, err)
return err
}
rw := authorization.NewResponseModifier(w)
if err := handler(ctx, rw, r, vars); err != nil {
logrus.Errorf("Handler for %s %s returned error: %s", r.Method, r.RequestURI, err)
return err
}
if err := authCtx.AuthZResponse(rw, r); err != nil {
logrus.Errorf("AuthZResponse for %s %s returned error: %s", r.Method, r.RequestURI, err)
return err
}
return nil
}
}
}

View File

@@ -1,33 +0,0 @@
package middleware
import (
"net/http"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/server/httputils"
"golang.org/x/net/context"
)
// NewCORSMiddleware creates a new CORS middleware.
func NewCORSMiddleware(defaultHeaders string) Middleware {
return func(handler httputils.APIFunc) httputils.APIFunc {
return func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
// If "api-cors-header" is not given, but "api-enable-cors" is true, we set cors to "*"
// otherwise, all head values will be passed to HTTP handler
corsHeaders := defaultHeaders
if corsHeaders == "" {
corsHeaders = "*"
}
writeCorsHeaders(w, r, corsHeaders)
return handler(ctx, w, r, vars)
}
}
}
func writeCorsHeaders(w http.ResponseWriter, r *http.Request, corsHeaders string) {
logrus.Debugf("CORS header is enabled and set to: %s", corsHeaders)
w.Header().Add("Access-Control-Allow-Origin", corsHeaders)
w.Header().Add("Access-Control-Allow-Headers", "Origin, X-Requested-With, Content-Type, Accept, X-Registry-Auth")
w.Header().Add("Access-Control-Allow-Methods", "HEAD, GET, POST, DELETE, PUT, OPTIONS")
}

View File

@@ -1,56 +0,0 @@
package middleware
import (
"bufio"
"encoding/json"
"io"
"net/http"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/server/httputils"
"github.com/docker/docker/pkg/ioutils"
"golang.org/x/net/context"
)
// DebugRequestMiddleware dumps the request to logger
func DebugRequestMiddleware(handler httputils.APIFunc) httputils.APIFunc {
return func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
logrus.Debugf("Calling %s %s", r.Method, r.RequestURI)
if r.Method != "POST" {
return handler(ctx, w, r, vars)
}
if err := httputils.CheckForJSON(r); err != nil {
return handler(ctx, w, r, vars)
}
maxBodySize := 4096 // 4KB
if r.ContentLength > int64(maxBodySize) {
return handler(ctx, w, r, vars)
}
body := r.Body
bufReader := bufio.NewReaderSize(body, maxBodySize)
r.Body = ioutils.NewReadCloserWrapper(bufReader, func() error { return body.Close() })
b, err := bufReader.Peek(maxBodySize)
if err != io.EOF {
// either there was an error reading, or the buffer is full (in which case the request is too large)
return handler(ctx, w, r, vars)
}
var postForm map[string]interface{}
if err := json.Unmarshal(b, &postForm); err == nil {
if _, exists := postForm["password"]; exists {
postForm["password"] = "*****"
}
formStr, errMarshal := json.Marshal(postForm)
if errMarshal == nil {
logrus.Debugf("form data: %s", string(formStr))
} else {
logrus.Debugf("form data: %q", postForm)
}
}
return handler(ctx, w, r, vars)
}
}

View File

@@ -1,7 +0,0 @@
package middleware
import "github.com/docker/docker/api/server/httputils"
// Middleware is an adapter to allow the use of ordinary functions as Docker API filters.
// Any function that has the appropriate signature can be registered as a middleware.
type Middleware func(handler httputils.APIFunc) httputils.APIFunc

View File

@@ -1,37 +0,0 @@
package middleware
import (
"net/http"
"strings"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/server/httputils"
"github.com/docker/docker/pkg/version"
"golang.org/x/net/context"
)
// NewUserAgentMiddleware creates a new UserAgent middleware.
func NewUserAgentMiddleware(versionCheck string) Middleware {
serverVersion := version.Version(versionCheck)
return func(handler httputils.APIFunc) httputils.APIFunc {
return func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
ctx = context.WithValue(ctx, httputils.UAStringKey, r.Header.Get("User-Agent"))
if strings.Contains(r.Header.Get("User-Agent"), "Docker-Client/") {
userAgent := strings.Split(r.Header.Get("User-Agent"), "/")
// v1.20 onwards includes the GOOS of the client after the version
// such as Docker/1.7.0 (linux)
if len(userAgent) == 2 && strings.Contains(userAgent[1], " ") {
userAgent[1] = strings.Split(userAgent[1], " ")[0]
}
if len(userAgent) == 2 && !serverVersion.Equal(version.Version(userAgent[1])) {
logrus.Debugf("Client and server don't have the same version (client: %s, server: %s)", userAgent[1], serverVersion)
}
}
return handler(ctx, w, r, vars)
}
}
}

View File

@@ -1,45 +0,0 @@
package middleware
import (
"fmt"
"net/http"
"runtime"
"github.com/docker/docker/api/server/httputils"
"github.com/docker/docker/pkg/version"
"golang.org/x/net/context"
)
type badRequestError struct {
error
}
func (badRequestError) HTTPErrorStatusCode() int {
return http.StatusBadRequest
}
// NewVersionMiddleware creates a new Version middleware.
func NewVersionMiddleware(versionCheck string, defaultVersion, minVersion version.Version) Middleware {
serverVersion := version.Version(versionCheck)
return func(handler httputils.APIFunc) httputils.APIFunc {
return func(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
apiVersion := version.Version(vars["version"])
if apiVersion == "" {
apiVersion = defaultVersion
}
if apiVersion.GreaterThan(defaultVersion) {
return badRequestError{fmt.Errorf("client is newer than server (client API version: %s, server API version: %s)", apiVersion, defaultVersion)}
}
if apiVersion.LessThan(minVersion) {
return badRequestError{fmt.Errorf("client version %s is too old. Minimum supported API version is %s, please upgrade your client to a newer version", apiVersion, minVersion)}
}
header := fmt.Sprintf("Docker/%s (%s)", serverVersion, runtime.GOOS)
w.Header().Set("Server", header)
ctx = context.WithValue(ctx, httputils.APIVersionKey, apiVersion)
return handler(ctx, w, r, vars)
}
}
}

Some files were not shown because too many files have changed in this diff Show More