Compare commits

..

37 Commits

Author SHA1 Message Date
unclejack 867b2a90c2 Merge pull request #4831 from unclejack/final_bump_v0.9.1
Bump to version 0.9.1
2014-03-26 00:00:37 +02:00
unclejack 3600720a36 Bump to version v0.9.1
Docker-DCO-1.1-Signed-off-by: Cristian Staretu <cristian.staretu@gmail.com> (github: unclejack)
2014-03-25 00:02:35 +02:00
Guillaume J. Charmes 8e681c34af Use BSD raw mode on darwin. Fixes nano, tmux and others
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)

Conflicts:
	pkg/term/termios_freebsd.go:
	  removed changes for this file because it's not present
2014-03-24 23:47:31 +02:00
Michael Crosby b60e316b67 Only unshare the mount namespace for execin
Fixes #4728
Docker-DCO-1.1-Signed-off-by: Michael Crosby <michael@crosbymichael.com> (github: crosbymichael)

Conflicts:
	pkg/libcontainer/nsinit/execin.go:
	  ns.logger.Println("unshare namespaces")' was removed already
2014-03-24 23:47:31 +02:00
unclejack a2fc45e627 retry to retrieve metadata on failure during pull
This makes Docker retry to retrieve the JSON metadata for the layers.
Docker will make 5 attempts to retrieve the metadata before failing and
it will increase the delay between attempts after each failed attempt.

This is a backport of PR #4387. The following changes had to be made:
-				img     *image.Image
+				img     *Image

-				img, err = image.NewImgJSON(imgJSON)
+				img, err = NewImgJSON(imgJSON)

Docker-DCO-1.1-Signed-off-by: Cristian Staretu <cristian.staretu@gmail.com> (github: unclejack)
2014-03-24 23:47:31 +02:00
Tianon Gravi 3f47f23fec Fix a lot of the sha256 and md5 stuff to be more DRY and extendible, and on more things (specifically, the tgz files too)
Docker-DCO-1.1-Signed-off-by: Andrew Page <admwiggin@gmail.com> (github: tianon)
2014-03-19 22:49:49 -06:00
Sven Dowideit e5974ae0ff missed a bug
Docker-DCO-1.1-Signed-off-by: Sven Dowideit <SvenDowideit@fosiki.com> (github: SvenDowideit)
2014-03-19 18:19:57 -06:00
Sven Dowideit ce002bf8da whitespace-blind
Docker-DCO-1.1-Signed-off-by: Sven Dowideit <SvenDowideit@fosiki.com> (github: SvenDowideit)

Docker-DCO-1.1-Signed-off-by: Andrew Page <admwiggin@gmail.com> (github: tianon)
2014-03-19 17:39:13 -06:00
Sven Dowideit 9cad4c2a1b Generate md5 and sha265 hashes when building, and upload them in hack/release.sh
Docker-DCO-1.1-Signed-off-by: Sven Dowideit <SvenDowideit@home.org.au> (github: SvenDowideit)

Docker-DCO-1.1-Signed-off-by: Andrew Page <admwiggin@gmail.com> (github: tianon)
2014-03-19 17:39:06 -06:00
unclejack 6056cce265 Merge pull request #4745 from unclejack/prepare_for_bump_v0.9.1
Prepare for bump v0.9.1
2014-03-19 20:00:55 +02:00
Guillaume J. Charmes cec512ccbb Make docker use the signal pkg with strings
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 20:29:37 +02:00
Guillaume J. Charmes d1e027291b Create portable signalMap
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 20:29:24 +02:00
Guillaume J. Charmes fda2934b6d Move signal to pkg
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 20:29:10 +02:00
Alexander Larsson 6fe06c6cc8 devmapper: Increase timeout in waitClose to 10sec
As reported in https://github.com/dotcloud/docker/issues/4389 we're
currently seeing timeouts in waitClose on some systems. We already
bumped the timeout in waitRemove() in
https://github.com/dotcloud/docker/issues/4504.

Docker-DCO-1.1-Signed-off-by: Alexander Larsson <alexl@redhat.com> (github: alexlarsson)
2014-03-18 19:15:44 +02:00
Alexander Larsson 63596f8c5d DeviceMapper: Succeed immediately when removing non-existant devices
We've seen situations where removal of "ID-init" failed during
container deletion (EBUSY), after removal of "ID" has succeeded. This
caused the container delete operation to fail, and on the next delete
attempt the removal of "ID" failed immediately with "does not exist".

Ideally we should not fail the ID-init removal, but its also non-ideal
to allow a state where the container is half-removed and we cannot
make progress deleting the container. So, we silently ignore not-exist
errors on device removal.

Docker-DCO-1.1-Signed-off-by: Alexander Larsson <alexl@redhat.com> (github: alexlarsson)
2014-03-18 18:57:31 +02:00
Vladimir Rutsky 3c9e23c297 Fix external link on security of containers
Docker-DCO-1.1-Signed-off-by: Vladimir Rutsky <altsysrq@gmail.com> (github: rutsky)
2014-03-18 18:08:59 +02:00
Victor Vieux 6908e0ba1e fix content-type detection in docker cp
Docker-DCO-1.1-Signed-off-by: Victor Vieux <victor.vieux@docker.com> (github: vieux)
2014-03-18 18:08:03 +02:00
Wes Morgan 816d55342d merge existing config when committing
Fixes #1141

Docker-DCO-1.1-Signed-off-by: Wes Morgan <cap10morgan@gmail.com> (github: cap10morgan)
2014-03-18 18:04:58 +02:00
Victor Vieux 2463be4dea fix panic in monitor
Docker-DCO-1.1-Signed-off-by: Victor Vieux <victor.vieux@docker.com> (github: vieux)
2014-03-18 17:04:56 +02:00
Isabel Jimenez 9e98ba72e1 adding configuration for timeout and disable it by default
Docker-DCO-1.1-Signed-off-by: Isabel Jimenez <contact@isabeljimenez.com> (github: jimenez)
2014-03-18 17:04:21 +02:00
Timothy Hobbs d1f7ebda16 Refactor out interface specific information from execdriver.Network
Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)
2014-03-18 16:52:57 +02:00
Timothy Hobbs 899f06bf79 Fix issue #4681 - No loopback interface within container when networking is disabled.
Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

Remove loopback code from veth strategy

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

Looback strategy: Get rid of uneeded code in Create
Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

Use append when building network strategy list

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

Swap loopback and veth strategies in Networks list

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

Revert "Swap loopback and veth strategies in Networks list"

This reverts commit 3b8b2c8454171d79bed5e9a80165172617e92fc7.

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)

When initializing networks, only return from the loop if there is an error

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)
2014-03-18 16:52:46 +02:00
Timothy Hobbs f020f1d31e Add failing test case for issue #4681
Add a failing test case for an issue where docker is not creating a loopback device if networking is dissabled.

Docker-DCO-1.1-Signed-off-by: Timothy Hobbs <timothyhobbs@seznam.cz> (github: https://github.com/timthelion)
2014-03-18 16:52:28 +02:00
Michael Crosby b9a0e886fc Send sigterm to child instead of sigkill
Docker-DCO-1.1-Signed-off-by: Michael Crosby <michael@crosbymichael.com> (github: crosbymichael)
2014-03-18 16:51:34 +02:00
Guillaume J. Charmes bcab23bbdc Have the exec driver and kernel version in non-debug mode in docker info
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 16:50:55 +02:00
Victor Vieux 7c35bfd78e as you could have multiple messages per line with streams, don't \r
Docker-DCO-1.1-Signed-off-by: Victor Vieux <victor.vieux@docker.com> (github: vieux)
2014-03-18 16:50:19 +02:00
Michael Crosby 78d66068a5 Always symlink /dev/ptmx for libcontainer
Docker-DCO-1.1-Signed-off-by: Michael Crosby <michael@crosbymichael.com> (github: crosbymichael)
2014-03-18 16:48:57 +02:00
Guillaume J. Charmes ab5b715964 Fix issue when /etc/apparmor.d does not exists
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 16:45:25 +02:00
unclejack d8448f27c6 don't leave empty cidFile behind
This makes `--cidfile` clean up empty container ID files. These are
left behind when creating the container fails.

Docker-DCO-1.1-Signed-off-by: Cristian Staretu <cristian.staretu@gmail.com> (github: unclejack)
2014-03-18 16:42:25 +02:00
Victor Vieux 463cf28788 improve deprecation message
Docker-DCO-1.1-Signed-off-by: Victor Vieux <victor.vieux@docker.com> (github: vieux)
2014-03-18 16:37:26 +02:00
Guillaume J. Charmes 6ce9daea0e Fix attach exit on darwin
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-18 16:35:48 +02:00
Alexander Larsson 39dbaa08dc devmapper: Increase sleep times and unlock while sleeping
We've seen some cases in the wild where waiting for unmount/deactivate
of devmapper devices taking a long time (several seconds). So, we increase
the sleeps to 10 seconds before we timeout. For instance:

https://github.com/dotcloud/docker/issues/4389

But, in order to not keep other processes blocked we unlock the global
dm lock while waiting to allow other devices to continue working.

Docker-DCO-1.1-Signed-off-by: Alexander Larsson <alexl@redhat.com> (github: alexlarsson)
2014-03-18 16:30:53 +02:00
Alexander Larsson d70e20cddd devmapper: Add per-device lock
We currently use a global lock to protect global data (like the
Devices map) as well as device data itself and access to
(non-threadsafe) libdevmapper.

This commit also adds a per-device lock, which will allow per-device
operations to temporarily release the global lock while e.g. waiting.
The per-device lock will make sure that nothing else accesses that
device while we're operating on it.

Docker-DCO-1.1-Signed-off-by: Alexander Larsson <alexl@redhat.com> (github: alexlarsson)
2014-03-18 16:30:28 +02:00
Tianon Gravi 2741b82cf3 Add variable for DOCKER_LOGFILE to sysvinit and use append instead of overwrite in opening the logfile
Docker-DCO-1.1-Signed-off-by: Andrew Page <admwiggin@gmail.com> (github: tianon)
2014-03-11 23:47:15 -06:00
Tianon Gravi 70ec7c6355 Fix init script cgroup mounting workarounds to be more similar to cgroupfs-mount and thus work properly
Docker-DCO-1.1-Signed-off-by: Andrew Page <admwiggin@gmail.com> (github: tianon)
2014-03-11 23:47:10 -06:00
Guillaume J. Charmes 0225a37951 Update parseLxcInfo to comply with new lxc1.0 format
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-11 23:46:43 -06:00
Guillaume J. Charmes 4bb1950089 Remove goroutine leak upon error
Docker-DCO-1.1-Signed-off-by: Guillaume J. Charmes <guillaume@charmes.net> (github: creack)
2014-03-11 23:46:36 -06:00
2037 changed files with 74390 additions and 225858 deletions
-2
View File
@@ -1,2 +0,0 @@
bundles
.gopath
+15 -24
View File
@@ -1,34 +1,25 @@
# Docker project generated files to ignore
# if you want to ignore files created by your editor/tools,
# please consider a global .gitignore https://help.github.com/articles/ignoring-files
*.exe
*.orig
*.rej
*.test
.*.swp
.DS_Store
.bashrc
.dotcloud
.flymake*
.git/
.gopath/
.hg/
.idea
.vagrant*
Vagrantfile
a.out
autogen/
bin
build_src
bundles/
docker/docker
docs/AWS_S3_BUCKET
docs/GITCOMMIT
docs/GIT_BRANCH
docs/VERSION
.*.swp
a.out
*.orig
build_src
.flymake*
.idea
.DS_Store
docs/_build
docs/_static
docs/_templates
docs/changed-files
pyenv
.gopath/
.dotcloud
*.test
bundles/
.hg/
.git/
vendor/pkg/
pyenv
Vagrantfile
+13 -97
View File
@@ -1,29 +1,4 @@
# Generate AUTHORS: hack/generate-authors.sh
# Tip for finding duplicates (besides scanning the output of AUTHORS for name
# duplicates that aren't also email duplicates): scan the output of:
# git log --format='%aE - %aN' | sort -uf
#
# For explanation on this file format: man git-shortlog
Patrick Stapleton <github@gdi2290.com>
Shishir Mahajan <shishir.mahajan@redhat.com> <smahajan@redhat.com>
Erwin van der Koogh <info@erronis.nl>
Ahmed Kamal <email.ahmedkamal@googlemail.com>
Tejesh Mehta <tejesh.mehta@gmail.com> <tj@init.me>
Cristian Staretu <cristian.staretu@gmail.com>
Cristian Staretu <cristian.staretu@gmail.com> <unclejacksons@gmail.com>
Cristian Staretu <cristian.staretu@gmail.com> <unclejack@users.noreply.github.com>
Marcus Linke <marcus.linke@gmx.de>
Aleksandrs Fadins <aleks@s-ko.net>
Christopher Latham <sudosurootdev@gmail.com>
Hu Keping <hukeping@huawei.com>
Wayne Chang <wayne@neverfear.org>
Chen Chao <cc272309126@gmail.com>
Daehyeok Mun <daehyeok@gmail.com>
<daehyeok@gmail.com> <daehyeok@daehyeokui-MacBook-Air.local>
<jt@yadutaf.fr> <admin@jtlebi.fr>
<jeff@docker.com> <jefferya@programmerq.net>
# Generate AUTHORS: git log --format='%aN <%aE>' | sort -uf
<charles.hooper@dotcloud.com> <chooper@plumata.com>
<daniel.mizyrycki@dotcloud.com> <daniel@dotcloud.com>
<daniel.mizyrycki@dotcloud.com> <mzdaniel@glidelink.net>
@@ -31,16 +6,14 @@ Guillaume J. Charmes <guillaume.charmes@docker.com> <charmes.guillaume@gmail.com
<guillaume.charmes@docker.com> <guillaume@dotcloud.com>
<guillaume.charmes@docker.com> <guillaume@docker.com>
<guillaume.charmes@docker.com> <guillaume.charmes@dotcloud.com>
<guillaume.charmes@docker.com> <guillaume@charmes.net>
<kencochrane@gmail.com> <KenCochrane@gmail.com>
Thatcher Peskens <thatcher@docker.com>
Thatcher Peskens <thatcher@docker.com> <thatcher@dotcloud.com>
Thatcher Peskens <thatcher@docker.com> dhrp <thatcher@gmx.net>
<sridharr@activestate.com> <github@srid.name>
Thatcher Peskens <thatcher@dotcloud.com> dhrp <thatcher@dotcloud.com>
Thatcher Peskens <thatcher@dotcloud.com> dhrp <thatcher@gmx.net>
Jérôme Petazzoni <jerome.petazzoni@dotcloud.com> jpetazzo <jerome.petazzoni@dotcloud.com>
Jérôme Petazzoni <jerome.petazzoni@dotcloud.com> <jp@enix.org>
Joffrey F <joffrey@docker.com>
Joffrey F <joffrey@docker.com> <joffrey@dotcloud.com>
Joffrey F <joffrey@docker.com> <f.joffrey@gmail.com>
Joffrey F <joffrey@dotcloud.com>
<joffrey@dotcloud.com> <f.joffrey@gmail.com>
Tim Terhorst <mynamewastaken+git@gmail.com>
Andy Smith <github@anarkystic.com>
<kalessin@kalessin.fr> <louis@dotcloud.com>
@@ -49,8 +22,8 @@ Andy Smith <github@anarkystic.com>
<victor.vieux@docker.com> <dev@vvieux.com>
<victor.vieux@docker.com> <victor@docker.com>
<victor.vieux@docker.com> <vieux@docker.com>
<victor.vieux@docker.com> <victorvieux@gmail.com>
<dominik@honnef.co> <dominikh@fork-bomb.org>
Thatcher Peskens <thatcher@dotcloud.com>
<ehanchrow@ine.com> <eric.hanchrow@gmail.com>
Walter Stanish <walter@pratyeka.org>
<daniel@gasienica.ch> <dgasienica@zynga.com>
@@ -73,72 +46,15 @@ Shih-Yuan Lee <fourdollars@gmail.com>
Daniel Mizyrycki <daniel.mizyrycki@dotcloud.com> root <root@vagrant-ubuntu-12.10.vagrantup.com>
Jean-Baptiste Dalido <jeanbaptiste@appgratis.com>
<proppy@google.com> <proppy@aminche.com>
<michael@docker.com> <michael@crosbymichael.com>
<michael@docker.com> <crosby.michael@gmail.com>
<michael@docker.com> <crosbymichael@gmail.com>
<github@developersupport.net> <github@metaliveblog.com>
<michael@crosbymichael.com> <crosby.michael@gmail.com>
<github@metaliveblog.com> <github@developersupport.net>
<brandon@ifup.org> <brandon@ifup.co>
<dano@spotify.com> <daniel.norberg@gmail.com>
<danny@codeaholics.org> <Danny.Yates@mailonline.co.uk>
<gurjeet@singh.im> <singh.gurjeet@gmail.com>
<shawn@churchofgit.com> <shawnlandden@gmail.com>
<sjoerd-github@linuxonly.nl> <sjoerd@byte.nl>
<solomon@docker.com> <solomon.hykes@dotcloud.com>
<solomon@docker.com> <solomon@dotcloud.com>
<solomon@docker.com> <s@docker.com>
Sven Dowideit <SvenDowideit@home.org.au>
Sven Dowideit <SvenDowideit@home.org.au> <SvenDowideit@fosiki.com>
Sven Dowideit <SvenDowideit@home.org.au> <SvenDowideit@docker.com>
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>
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>
Roberto G. Hashioka <roberto.hashioka@docker.com> <roberto_hashioka@hotmail.com>
<justin.p.simonelis@gmail.com> <justin.simonelis@PTS-JSIMON2.toronto.exclamation.com>
<taim@bosboot.org> <maztaim@users.noreply.github.com>
<viktor.vojnovski@amadeus.com> <vojnovski@gmail.com>
<vbatts@redhat.com> <vbatts@hashbangbash.com>
<altsysrq@gmail.com> <iamironbob@gmail.com>
Sridhar Ratnakumar <sridharr@activestate.com>
Sridhar Ratnakumar <sridharr@activestate.com> <github@srid.name>
Liang-Chi Hsieh <viirya@gmail.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>
Nathan LeClaire <nathan.leclaire@docker.com> <nathanleclaire@gmail.com>
<github@hollensbe.org> <erik+github@hollensbe.org>
<github@albersweb.de> <albers@users.noreply.github.com>
<lsm5@fedoraproject.org> <lsm5@redhat.com>
<marc@marc-abramowitz.com> <msabramo@gmail.com>
Matthew Heon <mheon@redhat.com> <mheon@mheonlaptop.redhat.com>
<bernat@luffy.cx> <vincent@bernat.im>
<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>
<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@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>
<github@gone.nl> <thaJeztah@users.noreply.github.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>
<solomon@dotcloud.com> <solomon.hykes@dotcloud.com>
<SvenDowideit@home.org.au> <SvenDowideit@fosiki.com>
Sven Dowideit <SvenDowideit@home.org.au> ¨Sven <¨SvenDowideit@home.org.au¨>
unclejack <unclejacksons@gmail.com> <unclejack@users.noreply.github.com>
+30
View File
@@ -0,0 +1,30 @@
# Note: right now we don't use go-specific features of travis.
# Later we might automate "go test" etc. (or do it inside a docker container...?)
language: go
go: 1.2
# Disable the normal go build.
install: true
before_script:
- env | sort
- sudo apt-get update -qq
- sudo apt-get install -qq python-yaml
- git remote add upstream git://github.com/dotcloud/docker.git
- upstream=master;
if [ "$TRAVIS_PULL_REQUEST" != false ]; then
upstream=$TRAVIS_BRANCH;
fi;
git fetch --append --no-tags upstream refs/heads/$upstream:refs/remotes/upstream/$upstream
# sometimes we have upstream master already as origin/master (PRs), but other times we don't, so let's just make sure we have a completely unambiguous way to specify "upstream master" from here out
# but if it's a PR against non-master, we need that upstream branch instead :)
- sudo pip install -r docs/requirements.txt
script:
- hack/travis/dco.py
- hack/travis/gofmt.py
- make -sC docs SPHINXOPTS=-q docs man
# vim:set sw=2 ts=2:
+31 -460
View File
@@ -1,235 +1,99 @@
# This file lists all individuals having contributed content to the repository.
# For how it is generated, see `hack/generate-authors.sh`.
# If you're submitting a patch, please add your name here in alphabetical order as part of the patch.
#
# For a list of active project maintainers, see the MAINTAINERS file.
#
Aanand Prasad <aanand.prasad@gmail.com>
Aaron Feng <aaron.feng@gmail.com>
Aaron Huslage <huslage@gmail.com>
Abel Muiño <amuino@gmail.com>
Abhinav Ajgaonkar <abhinav316@gmail.com>
Abin Shahab <ashahab@altiscale.com>
Adam Miller <admiller@redhat.com>
Adam Singer <financeCoding@gmail.com>
Aditya <aditya@netroy.in>
Adrian Mouat <adrian.mouat@gmail.com>
Adrien Folie <folie.adrien@gmail.com>
Ahmed Kamal <email.ahmedkamal@googlemail.com>
Ahmet Alp Balkan <ahmetb@microsoft.com>
Aidan Hobson Sayers <aidanhs@cantab.net>
AJ Bowen <aj@gandi.net>
Al Tobey <al@ooyala.com>
alambike <alambike@gmail.com>
Alan Thompson <cloojure@gmail.com>
Albert Callarisa <shark234@gmail.com>
Albert Zhang <zhgwenming@gmail.com>
Aleksa Sarai <cyphar@cyphar.com>
Aleksandrs Fadins <aleks@s-ko.net>
Alex Gaynor <alex.gaynor@gmail.com>
Alex Warhawk <ax.warhawk@gmail.com>
Alexander Boyd <alex@opengroove.org>
Alexander Larsson <alexl@redhat.com>
Alexander Morozov <lk4d4@docker.com>
Alexander Shopov <ash@kambanaria.org>
Alexandr Morozov <lk4d4@docker.com>
Alexey Kotlyarov <alexey@infoxchange.net.au>
Alexey Shamrin <shamrin@gmail.com>
Alex Gaynor <alex.gaynor@gmail.com>
Alexis THOMAS <fr.alexisthomas@gmail.com>
almoehi <almoehi@users.noreply.github.com>
amangoel <amangoel@gmail.com>
Amit Bakshi <ambakshi@gmail.com>
Anand Patil <anand.prabhakar.patil@gmail.com>
AnandkumarPatel <anandkumarpatel@gmail.com>
Andre Dublin <81dublin@gmail.com>
Al Tobey <al@ooyala.com>
Andrea Luzzardi <aluzzardi@gmail.com>
Andrea Turli <andrea.turli@gmail.com>
Andreas Köhler <andi5.py@gmx.net>
Andreas Savvides <andreas@editd.com>
Andreas Tiefenthaler <at@an-ti.eu>
Andrew C. Bodine <acbodine@us.ibm.com>
Andrew Duckworth <grillopress@gmail.com>
Andrew France <andrew@avito.co.uk>
Andrew Macgregor <andrew.macgregor@agworld.com.au>
Andrew Munsell <andrew@wizardapps.net>
Andrew Weiss <andrew.weiss@outlook.com>
Andrew Williams <williams.andrew@gmail.com>
Andrews Medina <andrewsmedina@gmail.com>
Andrey Petrov <andrey.petrov@shazow.net>
Andrey Stolbovsky <andrey.stolbovsky@gmail.com>
Andy Chambers <anchambers@paypal.com>
andy diller <dillera@gmail.com>
Andy Goldstein <agoldste@redhat.com>
Andy Kipp <andy@rstudio.com>
Andy Rothfusz <github@developersupport.net>
Andy Rothfusz <github@metaliveblog.com>
Andy Smith <github@anarkystic.com>
Andy Wilson <wilson.andrew.j+github@gmail.com>
Ankush Agarwal <ankushagarwal11@gmail.com>
Anthony Baire <Anthony.Baire@irisa.fr>
Anthony Bishopric <git@anthonybishopric.com>
Anton Löfgren <anton.lofgren@gmail.com>
Anton Nikitin <anton.k.nikitin@gmail.com>
Antony Messerli <amesserl@rackspace.com>
apocas <petermdias@gmail.com>
ArikaChen <eaglesora@gmail.com>
Arnaud Porterie <arnaud.porterie@docker.com>
Arthur Gautier <baloo@gandi.net>
Asbjørn Enge <asbjorn@hanafjedle.net>
averagehuman <averagehuman@users.noreply.github.com>
Avi Miller <avi.miller@oracle.com>
Barnaby Gray <barnaby@pickle.me.uk>
Barry Allard <barry.allard@gmail.com>
Bartłomiej Piotrowski <b@bpiotrowski.pl>
bdevloed <boris.de.vloed@gmail.com>
Ben Firshman <ben@firshman.co.uk>
Benoit Chesneau <bchesneau@gmail.com>
Ben Sargent <ben@brokendigits.com>
Ben Toews <mastahyeti@gmail.com>
Ben Wiklund <ben@daisyowl.com>
Benjamin Atkin <ben@benatkin.com>
Benoit Chesneau <bchesneau@gmail.com>
Bernerd Schaefer <bj.schaefer@gmail.com>
Bert Goethals <bert@bertg.be>
Bhiraj Butala <abhiraj.butala@gmail.com>
bin liu <liubin0329@users.noreply.github.com>
Blake Geno <blakegeno@gmail.com>
Bouke Haarsma <bouke@webatoom.nl>
Boyd Hemphill <boyd@feedmagnet.com>
Brandon Liu <bdon@bdon.org>
Brandon Philips <brandon@ifup.org>
Brandon Rhodes <brandon@rhodesmill.org>
Brett Kochendorfer <brett.kochendorfer@gmail.com>
Brian (bex) Exelbierd <bexelbie@redhat.com>
Brian Dorsey <brian@dorseys.org>
Brian Flad <bflad417@gmail.com>
Brian Goff <cpuguy83@gmail.com>
Brian McCallister <brianm@skife.org>
Brian Olsen <brian@maven-group.org>
Brian Shumate <brian@couchbase.com>
Brice Jaglin <bjaglin@teads.tv>
Briehan Lombaard <briehan.lombaard@gmail.com>
Bruno Bigras <bigras.bruno@gmail.com>
Bruno Binet <bruno.binet@gmail.com>
Bruno Renié <brutasse@gmail.com>
Bryan Bess <squarejaw@bsbess.com>
Bryan Matsuo <bryan.matsuo@gmail.com>
Bryan Murphy <bmurphy1976@gmail.com>
Burke Libbey <burke@libbey.me>
Byung Kang <byung.kang.ctr@amrdec.army.mil>
Caleb Spare <cespare@gmail.com>
Calen Pennington <cale@edx.org>
Cameron Boehmer <cameron.boehmer@gmail.com>
Carl X. Su <bcbcarl@gmail.com>
Charles Hooper <charles.hooper@dotcloud.com>
Charles Lindsay <chaz@chazomatic.us>
Charles Merriam <charles.merriam@gmail.com>
Charlie Lewis <charliel@lab41.org>
Chen Chao <cc272309126@gmail.com>
Chewey <prosto-chewey@users.noreply.github.com>
Chia-liang Kao <clkao@clkao.org>
Chris Alfonso <calfonso@redhat.com>
Chris Armstrong <chris@opdemand.com>
Chris Snow <chsnow123@gmail.com>
Chris St. Pierre <chris.a.st.pierre@gmail.com>
chrismckinnel <chris.mckinnel@tangentlabs.co.uk>
Christian Berendt <berendt@b1-systems.de>
Christian Stefanescu <st.chris@gmail.com>
ChristoperBiscardi <biscarch@sketcht.com>
Christophe Troestler <christophe.Troestler@umons.ac.be>
Christopher Currie <codemonkey+github@gmail.com>
Christopher Latham <sudosurootdev@gmail.com>
Christopher Rigor <crigor@gmail.com>
Chun Chen <chenchun.feed@gmail.com>
Ciro S. Costa <ciro.costa@usp.br>
Christophe Troestler <christophe.Troestler@umons.ac.be>
Clayton Coleman <ccoleman@redhat.com>
Colin Dunklau <colin.dunklau@gmail.com>
Colin Rice <colin@daedrum.net>
Colin Walters <walters@verbum.org>
Cory Forsyth <cory.forsyth@gmail.com>
cressie176 <github@stephen-cresswell.net>
Cristian Staretu <cristian.staretu@gmail.com>
Cruceru Calin-Cristian <crucerucalincristian@gmail.com>
Daan van Berkel <daan.v.berkel.1980@gmail.com>
Daehyeok Mun <daehyeok@gmail.com>
Dafydd Crosby <dtcrsby@gmail.com>
Dan Buch <d.buch@modcloth.com>
Dan Cotora <dan@bluevision.ro>
Dan Griffin <dgriffin@peer1.com>
Dan Hirsch <thequux@upstandinghackers.com>
Dan Keder <dan.keder@gmail.com>
Dan McPherson <dmcphers@redhat.com>
Dan Stine <sw@stinemail.com>
Dan Walsh <dwalsh@redhat.com>
Dan Williams <me@deedubs.com>
Daniel Exner <dex@dragonslave.de>
Daniel Farrell <dfarrell@redhat.com>
Daniel Garcia <daniel@danielgarcia.info>
Daniel Gasienica <daniel@gasienica.ch>
Daniel Menet <membership@sontags.ch>
Daniel Mizyrycki <daniel.mizyrycki@dotcloud.com>
Daniel Norberg <dano@spotify.com>
Daniel Nordberg <dnordberg@gmail.com>
Daniel Robinson <gottagetmac@gmail.com>
Daniel Von Fange <daniel@leancoder.com>
Daniel YC Lin <dlin.tw@gmail.com>
Daniel, Dao Quang Minh <dqminh89@gmail.com>
Danny Berger <dpb587@gmail.com>
Danny Yates <danny@codeaholics.org>
Darren Coxall <darren@darrencoxall.com>
Darren Shepherd <darren.s.shepherd@gmail.com>
David Anderson <dave@natulte.net>
David Calavera <david.calavera@gmail.com>
David Corking <dmc-source@dcorking.com>
David Gageot <david@gageot.net>
David Gebler <davidgebler@gmail.com>
David Mat <david@davidmat.com>
David Mcanulty <github@hellspark.com>
David Pelaez <pelaez89@gmail.com>
David Röthlisberger <david@rothlis.net>
David Sissitka <me@dsissitka.com>
Davide Ceretti <davide.ceretti@hogarthww.com>
Dawn Chen <dawnchen@google.com>
decadent <decadent@users.noreply.github.com>
Deni Bertovic <deni@kset.org>
Derek <crq@kernel.org>
Derek <crquan@gmail.com>
Derek McGowan <derek@mcgstyle.net>
Deric Crago <deric.crago@gmail.com>
Deshi Xiao <dxiao@redhat.com>
Dinesh Subhraveti <dineshs@altiscale.com>
Djibril Koné <kone.djibril@gmail.com>
dkumor <daniel@dkumor.com>
Dmitry Demeshchuk <demeshchuk@gmail.com>
Dmitry V. Krivenok <krivenok.dmitry@gmail.com>
Dolph Mathews <dolph.mathews@gmail.com>
Dominik Honnef <dominik@honnef.co>
Don Kjer <don.kjer@gmail.com>
Don Spaulding <donspauldingii@gmail.com>
Doug Davis <dug@us.ibm.com>
doug tangren <d.tangren@gmail.com>
Dr Nic Williams <drnicwilliams@gmail.com>
dragon788 <dragon788@users.noreply.github.com>
Dražen Lučanin <kermit666@gmail.com>
Dr Nic Williams <drnicwilliams@gmail.com>
Dustin Sallings <dustin@spy.net>
Edmund Wagner <edmund-wagner@web.de>
Eiichi Tsukata <devel@etsukata.com>
Eike Herzbach <eike@herzbach.net>
Eivind Uggedal <eivind@uggedal.com>
Elias Probst <mail@eliasprobst.eu>
Emil Hernvall <emil@quench.at>
Emily Maier <emily@emilymaier.net>
Emily Rose <emily@contactvibe.com>
Eric Hanchrow <ehanchrow@ine.com>
Eric Lee <thenorthsecedes@gmail.com>
Eric Myhre <hash@exultant.us>
Eric Paris <eparis@redhat.com>
Eric Windisch <ewindisch@docker.com>
Erik Dubbelboer <erik@dubbelboer.com>
Erik Hollensbe <github@hollensbe.org>
Erik Inge Bolsø <knan@redpill-linpro.com>
Erik Kristensen <erik@erikkristensen.com>
Erno Hopearuoho <erno.hopearuoho@gmail.com>
Erwin van der Koogh <info@erronis.nl>
Eugene Yakubovich <eugene.yakubovich@coreos.com>
eugenkrizo <eugen.krizo@gmail.com>
Evan Carmi <carmi@users.noreply.github.com>
Evan Hazlett <ejhazlett@gmail.com>
Evan Krall <krall@yelp.com>
Evan Phoenix <evan@fallingsnow.net>
Evan Wies <evan@neomantra.net>
@@ -240,534 +104,241 @@ Fabio Rehm <fgrehm@gmail.com>
Fabrizio Regini <freegenie@gmail.com>
Faiz Khan <faizkhan00@gmail.com>
Fareed Dudhia <fareeddudhia@googlemail.com>
Felix Rabe <felix@rabe.io>
Fernando <fermayo@gmail.com>
Filipe Brandenburger <filbranden@google.com>
Flavio Castelli <fcastelli@suse.com>
FLGMwt <ryan.stelly@live.com>
Francisco Carriedo <fcarriedo@gmail.com>
Francisco Souza <f@souza.cc>
Frank Macreery <frank@macreery.com>
Frank Rosquin <frank.rosquin+github@gmail.com>
Fred Lifton <fred.lifton@docker.com>
Frederick F. Kautz IV <fkautz@alumni.cmu.edu>
Frederik Loeffert <frederik@zitrusmedia.de>
Freek Kalter <freek@kalteronline.org>
Gabe Rosenhouse <gabe@missionst.com>
Gabor Nagy <mail@aigeruth.hu>
Gabriel Monroy <gabriel@opdemand.com>
Galen Sampson <galen.sampson@gmail.com>
Gareth Rushgrove <gareth@morethanseven.net>
gautam, prasanna <prasannagautam@gmail.com>
Geoffrey Bachelet <grosfrais@gmail.com>
George Xie <georgexsh@gmail.com>
Gereon Frey <gereon.frey@dynport.de>
German DZ <germ@ndz.com.ar>
Gert van Valkenhoef <g.h.m.van.valkenhoef@rug.nl>
Giuseppe Mazzotta <gdm85@users.noreply.github.com>
Gleb Fotengauer-Malinovskiy <glebfm@altlinux.org>
Gleb M Borisov <borisov.gleb@gmail.com>
Glyn Normington <gnormington@gopivotal.com>
Goffert van Gool <goffert@phusion.nl>
golubbe <ben.golub@dotcloud.com>
Graydon Hoare <graydon@pobox.com>
Greg Thornton <xdissent@me.com>
grunny <mwgrunny@gmail.com>
Guilherme Salgado <gsalgado@gmail.com>
Guillaume Dufour <gdufour.prestataire@voyages-sncf.com>
Guillaume J. Charmes <guillaume.charmes@docker.com>
Gurjeet Singh <gurjeet@singh.im>
Guruprasad <lgp171188@gmail.com>
Hans Rødtang <hansrodtang@gmail.com>
Harald Albers <github@albersweb.de>
Harley Laue <losinggeneration@gmail.com>
Hector Castro <hectcastro@gmail.com>
Henning Sprang <henning.sprang@gmail.com>
Hobofan <goisser94@gmail.com>
Hollie Teal <hollie@docker.com>
Hu Keping <hukeping@huawei.com>
Hu Tao <hutao@cn.fujitsu.com>
Huayi Zhang <irachex@gmail.com>
Hugo Duncan <hugo@hugoduncan.org>
Hunter Blanks <hunter@twilio.com>
Huu Nguyen <huu@prismskylabs.com>
hyeongkyu.lee <hyeongkyu.lee@navercorp.com>
Ian Babrou <ibobrik@gmail.com>
Ian Bishop <ianbishop@pace7.com>
Ian Bull <irbull@gmail.com>
Ian Main <imain@redhat.com>
Ian Truslove <ian.truslove@gmail.com>
Igor Dolzhikov <bluesriverz@gmail.com>
ILYA Khlopotov <ilya.khlopotov@gmail.com>
inglesp <peter.inglesby@gmail.com>
Isaac Dupree <antispam@idupree.com>
Isabel Jimenez <contact.isabeljimenez@gmail.com>
Isao Jonas <isao.jonas@gmail.com>
Ivan Fraixedes <ifcdev@gmail.com>
Jack Danger Canty <jackdanger@squareup.com>
Jacob Atzen <jacob@jacobatzen.dk>
Jacob Edelman <edelman.jd@gmail.com>
Jake Moshenko <jake@devtable.com>
jakedt <jake@devtable.com>
James Allen <jamesallen0108@gmail.com>
James Carr <james.r.carr@gmail.com>
James DeFelice <james.defelice@ishisystems.com>
James Harrison Fisher <jameshfisher@gmail.com>
James Kyle <james@jameskyle.org>
James Mills <prologic@shortcircuit.net.au>
James Turnbull <james@lovedthanlost.net>
Jan Keromnes <janx@linux.com>
Jan Pazdziora <jpazdziora@redhat.com>
Jan Toebes <jan@toebes.info>
Jaroslaw Zabiello <hipertracker@gmail.com>
jaseg <jaseg@jaseg.net>
Jason Giedymin <jasong@apache.org>
Jason Hall <imjasonh@gmail.com>
Jason Livesay <ithkuil@gmail.com>
Jason McVetta <jason.mcvetta@gmail.com>
Jason Plum <jplum@devonit.com>
Jean-Baptiste Barth <jeanbaptiste.barth@gmail.com>
Jean-Baptiste Dalido <jeanbaptiste@appgratis.com>
Jean-Paul Calderone <exarkun@twistedmatrix.com>
Jean-Tiare Le Bigot <jt@yadutaf.fr>
Jeff Anderson <jeff@docker.com>
Jeff Lindsay <progrium@gmail.com>
Jeff Welch <whatthejeff@gmail.com>
Jeffrey Bolle <jeffreybolle@gmail.com>
Jeremy Grosser <jeremy@synack.me>
Jérôme Petazzoni <jerome.petazzoni@dotcloud.com>
Jesse Dubay <jesse@thefortytwo.net>
Jessica Frazelle <jess@docker.com>
Jezeniel Zapanta <jpzapanta22@gmail.com>
Jilles Oldenbeuving <ojilles@gmail.com>
Jim Alateras <jima@comware.com.au>
Jim Perrin <jperrin@centos.org>
Jimmy Cuadra <jimmy@jimmycuadra.com>
Jiří Župka <jzupka@redhat.com>
Joe Beda <joe.github@bedafamily.com>
Joe Ferguson <joe@infosiftr.com>
Joe Shaw <joe@joeshaw.org>
Joe Van Dyk <joe@tanga.com>
Joel Friedly <joelfriedly@gmail.com>
Joel Handwell <joelhandwell@gmail.com>
Joffrey F <joffrey@docker.com>
Joffrey F <joffrey@dotcloud.com>
Johan Euphrosine <proppy@google.com>
Johan Rydberg <johan.rydberg@gmail.com>
Johannes 'fish' Ziemke <github@freigeist.org>
Johan Rydberg <johan.rydberg@gmail.com>
John Costa <john.costa@gmail.com>
John Feminella <jxf@jxf.me>
John Gardiner Myers <jgmyers@proofpoint.com>
John Gossman <johngos@microsoft.com>
John OBrien III <jobrieniii@yahoo.com>
John Warwick <jwarwick@gmail.com>
Jon Wedaman <jweede@gmail.com>
Jonas Pfenniger <jonas@pfenniger.name>
Jonathan A. Sternberg <jonathansternberg@gmail.com>
Jonathan Boulle <jonathanboulle@gmail.com>
Jonathan Camp <jonathan@irondojo.com>
Jonathan McCrohan <jmccrohan@gmail.com>
Jonathan Mueller <j.mueller@apoveda.ch>
Jonathan Pares <jonathanpa@users.noreply.github.com>
Jonathan Rudenberg <jonathan@titanous.com>
Jon Wedaman <jweede@gmail.com>
Joost Cassee <joost@cassee.net>
Jordan Arentsen <blissdev@gmail.com>
Jordan Sissel <jls@semicomplete.com>
Joseph Anthony Pasquale Holsten <joseph@josephholsten.com>
Joseph Hager <ajhager@gmail.com>
Josh <jokajak@gmail.com>
Josh Hawn <josh.hawn@docker.com>
Josh Poimboeuf <jpoimboe@redhat.com>
Josiah Kiehl <jkiehl@riotgames.com>
JP <jpellerin@leapfrogonline.com>
Julian Taylor <jtaylor.debian@googlemail.com>
Julien Barbier <write0@gmail.com>
Julien Bordellier <julienbordellier@gmail.com>
Julien Dubois <julien.dubois@gmail.com>
Justin Force <justin.force@gmail.com>
Justin Plock <jplock@users.noreply.github.com>
Justin Simonelis <justin.p.simonelis@gmail.com>
Jyrki Puttonen <jyrkiput@gmail.com>
Jérôme Petazzoni <jerome.petazzoni@dotcloud.com>
Jörg Thalheim <joerg@higgsboson.tk>
Kamil Domanski <kamil@domanski.co>
Karan Lyons <karan@karanlyons.com>
Karl Grzeszczak <karlgrz@gmail.com>
Kato Kazuyoshi <kato.kazuyoshi@gmail.com>
Kawsar Saiyeed <kawsar.saiyeed@projiris.com>
Keli Hu <dev@keli.hu>
Ken Cochrane <kencochrane@gmail.com>
Ken ICHIKAWA <ichikawa.ken@jp.fujitsu.com>
Kevin "qwazerty" Houdebert <kevin.houdebert@gmail.com>
Kevin Clark <kevin.clark@gmail.com>
Kevin J. Lynagh <kevin@keminglabs.com>
Kevin Menard <kevin@nirvdrum.com>
Kevin Wallace <kevin@pentabarf.net>
Keyvan Fatehi <keyvanfatehi@gmail.com>
kies <lleelm@gmail.com>
kim0 <email.ahmedkamal@googlemail.com>
Kim BKC Carlbacker <kim.carlbacker@gmail.com>
Kimbro Staken <kstaken@kstaken.com>
Kiran Gangadharan <kiran.daredevil@gmail.com>
knappe <tyler.knappe@gmail.com>
Kohei Tsuruta <coheyxyz@gmail.com>
Konrad Kleine <konrad.wilhelm.kleine@gmail.com>
Konstantin Pelykh <kpelykh@zettaset.com>
Krasimir Georgiev <support@vip-consult.co.uk>
krrg <krrgithub@gmail.com>
Kyle Conroy <kyle.j.conroy@gmail.com>
kyu <leehk1227@gmail.com>
Lachlan Coote <lcoote@vmware.com>
Lajos Papp <lajos.papp@sequenceiq.com>
Lakshan Perera <lakshan@laktek.com>
lalyos <lalyos@yahoo.com>
Lance Chen <cyen0312@gmail.com>
Lars R. Damerow <lars@pixar.com>
Laurie Voss <github@seldo.com>
leeplay <hyeongkyu.lee@navercorp.com>
Lei Jitang <leijitang@huawei.com>
Len Weincier <len@cloudafrica.net>
Leszek Kowalski <github@leszekkowalski.pl>
Levi Gross <levi@levigross.com>
Lewis Marshall <lewis@lmars.net>
Lewis Peckover <lew+github@lew.io>
Liang-Chi Hsieh <viirya@gmail.com>
limsy <seongyeol37@gmail.com>
Lokesh Mandvekar <lsm5@fedoraproject.org>
Lorenz Leutgeb <lorenz.leutgeb@gmail.com>
Lokesh Mandvekar <lsm5@redhat.com>
Louis Opter <kalessin@kalessin.fr>
lukaspustina <lukas.pustina@centerdevice.com>
lukemarsden <luke@digital-crocus.com>
Lénaïc Huard <lhuard@amadeus.com>
Madhu Venugopal <madhu@socketplane.io>
Mahesh Tiyyagura <tmahesh@gmail.com>
Malte Janduda <mail@janduda.net>
Manfred Zabarauskas <manfredas@zabarauskas.com>
Manuel Meurer <manuel@krautcomputing.com>
Manuel Woelker <github@manuel.woelker.org>
Marc Abramowitz <marc@marc-abramowitz.com>
Marc Kuo <kuomarc2@gmail.com>
Marc Tamsky <mtamsky@gmail.com>
Marco Hennings <marco.hennings@freiheit.com>
Marcus Farkas <toothlessgear@finitebox.com>
Marcus Linke <marcus.linke@gmx.de>
Marcus Ramberg <marcus@nordaaker.com>
Marek Goldmann <marek.goldmann@gmail.com>
Marianna <mtesselh@gmail.com>
Marius Voila <marius.voila@gmail.com>
Mark Allen <mrallen1@yahoo.com>
Mark McGranaghan <mmcgrana@gmail.com>
Marko Mikulicic <mmikulicic@gmail.com>
Marko Tibold <marko@tibold.nl>
Markus Fix <lispmeister@gmail.com>
Martijn van Oosterhout <kleptog@svana.org>
Martin Honermeyer <maze@strahlungsfrei.de>
Martin Redmond <martin@tinychat.com>
Mary Anthony <moxieandmore@gmail.com>
Mason Malone <mason.malone@gmail.com>
Mateusz Sulima <sulima.mateusz@gmail.com>
Mathias Monnerville <mathias@monnerville.com>
Mathieu Le Marec - Pasquet <kiorky@cryptelium.net>
Matt Apperson <me@mattapperson.com>
Matt Bachmann <bachmann.matt@gmail.com>
Matt Haggard <haggardii@gmail.com>
Matthew Heon <mheon@redhat.com>
Matthew Mueller <mattmuelle@gmail.com>
Matthew Riley <mattdr@google.com>
Matthias Klumpp <matthias@tenstral.net>
Matthias Kühnle <git.nivoc@neverbox.com>
mattymo <raytrac3r@gmail.com>
mattyw <mattyw@me.com>
Max Shytikov <mshytikov@gmail.com>
Maxim Treskin <zerthurd@gmail.com>
Maxime Petazzoni <max@signalfuse.com>
Maxim Treskin <zerthurd@gmail.com>
meejah <meejah@meejah.ca>
Mengdi Gao <usrgdd@gmail.com>
Mert Yazıcıoğlu <merty@users.noreply.github.com>
Michael Brown <michael@netdirect.ca>
Michael Crosby <michael@docker.com>
Michael Crosby <michael@crosbymichael.com>
Michael Gorsuch <gorsuch@github.com>
Michael Hudson-Doyle <michael.hudson@linaro.org>
Michael Neale <michael.neale@gmail.com>
Michael Prokop <github@michael-prokop.at>
Michael Scharf <github@scharf.gr>
Michael Stapelberg <michael+gh@stapelberg.de>
Michael Steinert <mike.steinert@gmail.com>
Michael Thies <michaelthies78@gmail.com>
Michal Jemala <michal.jemala@gmail.com>
Michal Minar <miminar@redhat.com>
Michaël Pailloncy <mpapo.dev@gmail.com>
Michiel@unhosted <michiel@unhosted.org>
Miguel Angel Fernández <elmendalerenda@gmail.com>
Mike Chelen <michael.chelen@gmail.com>
Mike Gaffney <mike@uberu.com>
Mike MacCana <mike.maccana@gmail.com>
Mike Naberezny <mike@naberezny.com>
Mike Snitzer <snitzer@redhat.com>
Mikhail Sobolev <mss@mawhrin.net>
Mohit Soni <mosoni@ebay.com>
Morgante Pell <morgante.pell@morgante.net>
Morten Siebuhr <sbhr@sbhr.dk>
Mrunal Patel <mrunalp@gmail.com>
mschurenko <matt.schurenko@gmail.com>
Mustafa Akın <mustafa91@gmail.com>
Médi-Rémi Hashim <medimatrix@users.noreply.github.com>
Nan Monnand Deng <monnand@gmail.com>
Naoki Orii <norii@cs.cmu.edu>
Nate Eagleson <nate@nateeag.com>
Nate Jones <nate@endot.org>
Nathan Hsieh <hsieh.nathan@gmail.com>
Nathan Kleyn <nathan@nathankleyn.com>
Nathan LeClaire <nathan.leclaire@docker.com>
Neal McBurnett <neal@mcburnett.org>
Nelson Chen <crazysim@gmail.com>
Niall O'Higgins <niallo@unworkable.org>
Nicholas E. Rabenau <nerab@gmx.at>
Nick Payne <nick@kurai.co.uk>
Nick Stenning <nick.stenning@digital.cabinet-office.gov.uk>
Nick Stinemates <nick@stinemates.org>
Nicolas De loof <nicolas.deloof@gmail.com>
Nicolas Dudebout <nicolas.dudebout@gatech.edu>
Nicolas Goy <kuon@goyman.com>
Nicolas Kaiser <nikai@nikai.net>
NikolaMandic <mn080202@gmail.com>
noducks <onemannoducks@gmail.com>
Nolan Darilek <nolan@thewordnerd.info>
nzwsch <hi@nzwsch.com>
O.S. Tezer <ostezer@gmail.com>
OddBloke <daniel@daniel-watkins.co.uk>
odk- <github@odkurzacz.org>
Oguz Bilgic <fisyonet@gmail.com>
Oh Jinkyun <tintypemolly@gmail.com>
Ole Reifschneider <mail@ole-reifschneider.de>
Olivier Gambier <dmp42@users.noreply.github.com>
O.S.Tezer <ostezer@gmail.com>
pandrew <letters@paulnotcom.se>
panticz <mail@konczalski.de>
Pascal Borreli <pascal@borreli.com>
Pascal Hartig <phartig@rdrei.net>
Patrick Hemmer <patrick.hemmer@gmail.com>
Patrick Stapleton <github@gdi2290.com>
pattichen <craftsbear@gmail.com>
Paul <paul9869@gmail.com>
Paul Annesley <paul@annesley.cc>
Paul Bowsher <pbowsher@globalpersonals.co.uk>
Paul Hammond <paul@paulhammond.org>
Paul Jimenez <pj@place.org>
Paul Lietar <paul@lietar.net>
Paul Morie <pmorie@gmail.com>
Paul Nasrat <pnasrat@gmail.com>
Paul Weaver <pauweave@cisco.com>
Pavlos Ratis <dastergon@gentoo.org>
Peter Bourgon <peter@bourgon.org>
Paul <paul9869@gmail.com>
Peter Braden <peterbraden@peterbraden.co.uk>
Peter Ericson <pdericson@gmail.com>
Peter Salvatore <peter@psftw.com>
Peter Waller <p@pwaller.net>
Phil <underscorephil@gmail.com>
Phil Estes <estesp@linux.vnet.ibm.com>
Peter Waller <peter@scraperwiki.com>
Phil Spitler <pspitler@gmail.com>
Philipp Weissensteiner <mail@philippweissensteiner.com>
Phillip Alexander <git@phillipalexander.io>
Piergiuliano Bossi <pgbossi@gmail.com>
Pierre <py@poujade.org>
Pierre Wacrenier <pierre.wacrenier@gmail.com>
Pierre-Alain RIVIERE <pariviere@ippon.fr>
Piotr Bogdan <ppbogdan@gmail.com>
pixelistik <pixelistik@users.noreply.github.com>
Porjo <porjo38@yahoo.com.au>
Prasanna Gautam <prasannagautam@gmail.com>
Przemek Hejman <przemyslaw.hejman@gmail.com>
pysqz <randomq@126.com>
Qiang Huang <h.huangqiang@huawei.com>
Quentin Brossard <qbrossard@gmail.com>
r0n22 <cameron.regan@gmail.com>
Rafal Jeczalik <rjeczalik@gmail.com>
Rafe Colton <rafael.colton@gmail.com>
Rajat Pandit <rp@rajatpandit.com>
Rajdeep Dua <dua_rajdeep@yahoo.com>
Ralph Bean <rbean@redhat.com>
Ramkumar Ramachandra <artagnon@gmail.com>
Ramon van Alteren <ramon@vanalteren.nl>
Recursive Madman <recursive.madman@gmx.de>
Remi Rampin <remirampin@gmail.com>
Renato Riccieri Santos Zannon <renato.riccieri@gmail.com>
rgstephens <greg@udon.org>
Rhys Hiltner <rhys@twitch.tv>
Richard Harvey <richard@squarecows.com>
Richard Metzler <richard@paadee.com>
Richo Healey <richo@psych0tik.net>
Rick Bradley <rick@users.noreply.github.com>
Rick van de Loo <rickvandeloo@gmail.com>
Robert Bachmann <rb@robertbachmann.at>
Robert Bittle <guywithnose@gmail.com>
Robert Obryk <robryk@gmail.com>
Roberto G. Hashioka <roberto.hashioka@docker.com>
Robin Speekenbrink <robin@kingsquare.nl>
robpc <rpcann@gmail.com>
Roberto Hashioka <roberto_hashioka@hotmail.com>
Rodrigo Vaz <rodrigo.vaz@gmail.com>
Roel Van Nyen <roel.vannyen@gmail.com>
Roger Peppe <rogpeppe@gmail.com>
Rohit Jnagal <jnagal@google.com>
Roland Huß <roland@jolokia.org>
Roland Moriz <rmoriz@users.noreply.github.com>
Ron Smits <ron.smits@gmail.com>
Rovanion Luckey <rovanion.luckey@gmail.com>
Rudolph Gottesheim <r.gottesheim@loot.at>
Ryan Anderson <anderson.ryanc@gmail.com>
Ryan Aslett <github@mixologic.com>
Ryan Detzel <ryan.detzel@gmail.com>
Ryan Fowler <rwfowler@gmail.com>
Ryan O'Donnell <odonnellryanc@gmail.com>
Ryan Seto <ryanseto@yak.net>
Ryan Thomas <rthomas@atlassian.com>
Rémy Greinhofer <remy.greinhofer@livelovely.com>
Sam Alba <sam.alba@gmail.com>
Sam Bailey <cyprix@cyprix.com.au>
Sam J Sharpe <sam.sharpe@digital.cabinet-office.gov.uk>
Sam Reis <sreis@atlassian.com>
Sam Rijs <srijs@airpost.net>
Sami Wagiaalla <swagiaal@redhat.com>
Samuel Andaya <samuel@andaya.net>
Samuel PHAN <samuel-phan@users.noreply.github.com>
Satnam Singh <satnam@raintown.org>
satoru <satorulogic@gmail.com>
Satoshi Amemiya <satoshi_amemiya@voyagegroup.com>
Scott Bessler <scottbessler@gmail.com>
Scott Collier <emailscottcollier@gmail.com>
Scott Johnston <scott@docker.com>
Scott Stamp <scottstamp851@gmail.com>
Scott Walls <sawalls@umich.edu>
Sean Cronin <seancron@gmail.com>
Sean P. Kane <skane@newrelic.com>
Sebastiaan van Stijn <github@gone.nl>
Senthil Kumar Selvaraj <senthil.thecoder@gmail.com>
SeongJae Park <sj38.park@gmail.com>
Shane Canon <scanon@lbl.gov>
shaunol <shaunol@gmail.com>
Shawn Landden <shawn@churchofgit.com>
Shawn Siefkas <shawn.siefkas@meredith.com>
Shih-Yuan Lee <fourdollars@gmail.com>
Shishir Mahajan <shishir.mahajan@redhat.com>
shuai-z <zs.broccoli@gmail.com>
shin- <joffrey@docker.com>
Silas Sewell <silas@sewell.org>
Simon Taranto <simon.taranto@gmail.com>
Sindhu S <sindhus@live.in>
Sjoerd Langkemper <sjoerd-github@linuxonly.nl>
Solomon Hykes <solomon@docker.com>
Solomon Hykes <solomon@dotcloud.com>
Song Gao <song@gao.io>
Soulou <leo@unbekandt.eu>
soulshake <amy@gandi.net>
Sridatta Thatipamala <sthatipamala@gmail.com>
Sridhar Ratnakumar <sridharr@activestate.com>
Srini Brahmaroutu <sbrahma@us.ibm.com>
Srini Brahmaroutu <srbrahma@us.ibm.com>
Steeve Morin <steeve.morin@gmail.com>
Stefan Praszalowicz <stefan@greplin.com>
Stephen Crosby <stevecrozz@gmail.com>
Steven Burgess <steven.a.burgess@hotmail.com>
Steven Merrill <steven.merrill@gmail.com>
Sven Dowideit <SvenDowideit@home.org.au>
sudosurootdev <sudosurootdev@gmail.com>
Sven Dowideit <svendowideit@home.org.au>
Sylvain Bellemare <sylvain.bellemare@ezeep.com>
Sébastien <sebastien@yoozio.com>
Sébastien Luttringer <seblu@seblu.net>
Sébastien Stormacq <sebsto@users.noreply.github.com>
tang0th <tang0th@gmx.com>
Tatsuki Sugiura <sugi@nemui.org>
Tatsushi Inagaki <e29253@jp.ibm.com>
Ted M. Young <tedyoung@gmail.com>
Tehmasp Chaudhri <tehmasp@gmail.com>
Tejesh Mehta <tejesh.mehta@gmail.com>
Thatcher Peskens <thatcher@docker.com>
Thatcher Peskens <thatcher@dotcloud.com>
Thermionix <bond711@gmail.com>
Thijs Terlouw <thijsterlouw@gmail.com>
Thomas Bikeev <thomas.bikeev@mac.com>
Thomas Frössman <thomasf@jossystem.se>
Thomas Hansen <thomas.hansen@gmail.com>
Thomas LEVEIL <thomasleveil@gmail.com>
Thomas Orozco <thomas@orozco.fr>
Thomas Schroeter <thomas@cliqz.com>
Tianon Gravi <admwiggin@gmail.com>
Tibor Vass <teabee89@gmail.com>
Tim Bosse <taim@bosboot.org>
Tim Hockin <thockin@google.com>
Tim Ruffles <oi@truffles.me.uk>
Tim Smith <timbot@google.com>
Tim Bosse <maztaim@users.noreply.github.com>
Tim Terhorst <mynamewastaken+git@gmail.com>
Timothy Hobbs <timothyhobbs@seznam.cz>
tjwebb123 <tjwebb123@users.noreply.github.com>
tobe <tobegit3hub@gmail.com>
Tobias Bieniek <Tobias.Bieniek@gmx.de>
Tobias Gesellchen <tobias@gesellix.de>
Tobias Schmidt <ts@soundcloud.com>
Tobias Schwab <tobias.schwab@dynport.de>
Todd Lunter <tlunter@gmail.com>
Tom Fotherby <tom+github@peopleperhour.com>
Tom Hulihan <hulihan.tom159@gmail.com>
Tom Maaswinkel <tom.maaswinkel@12wiki.eu>
Tomas Tomecek <ttomecek@redhat.com>
Tomasz Lipinski <tlipinski@users.noreply.github.com>
Tomasz Nurkiewicz <nurkiewicz@gmail.com>
Tommaso Visconti <tommaso.visconti@gmail.com>
Tonis Tiigi <tonistiigi@gmail.com>
Tony Daws <tony@daws.ca>
Torstein Husebø <torstein@huseboe.net>
tpng <benny.tpng@gmail.com>
Travis Cline <travis.cline@gmail.com>
Trent Ogren <tedwardo2@gmail.com>
Tyler Brock <tyler.brock@gmail.com>
Tzu-Jung Lee <roylee17@gmail.com>
Ulysse Carion <ulyssecarion@gmail.com>
unknown <sebastiaan@ws-key-sebas3.dpi1.dpi>
Vaidas Jablonskis <jablonskis@gmail.com>
unclejack <unclejacksons@gmail.com>
vgeta <gopikannan.venugopalsamy@gmail.com>
Victor Coisne <victor.coisne@dotcloud.com>
Victor Lyuboslavsky <victor@victoreda.com>
Victor Marmol <vmarmol@google.com>
Victor Vieux <victor.vieux@docker.com>
Viktor Vojnovski <viktor.vojnovski@amadeus.com>
Vincent Batts <vbatts@redhat.com>
Vincent Bernat <bernat@luffy.cx>
Vincent Bernat <Vincent.Bernat@exoscale.ch>
Vincent Giersch <vincent.giersch@ovh.net>
Vincent Mayers <vincent.mayers@inbloom.org>
Vincent Woo <me@vincentwoo.com>
Vinod Kulkarni <vinod.kulkarni@gmail.com>
Vishal Doshi <vishal.doshi@gmail.com>
Vishnu Kannan <vishnuk@google.com>
Vitor Monteiro <vmrmonteiro@gmail.com>
Vivek Agarwal <me@vivek.im>
Vivek Dasgupta <vdasgupt@redhat.com>
Vivek Goyal <vgoyal@redhat.com>
Vladimir Bulyga <xx@ccxx.cc>
Vladimir Kirillov <proger@wilab.org.ua>
Vladimir Rutsky <altsysrq@gmail.com>
Vojtech Vitek (V-Teq) <vvitek@redhat.com>
waitingkuo <waitingkuo0527@gmail.com>
Walter Leibbrandt <github@wrl.co.za>
Vladimir Rutsky <iamironbob@gmail.com>
Walter Stanish <walter@pratyeka.org>
Ward Vandewege <ward@jhvc.com>
WarheadsSE <max@warheads.net>
Wayne Chang <wayne@neverfear.org>
Wes Morgan <cap10morgan@gmail.com>
Will Dietz <w@wdtz.org>
William Delanoue <william.delanoue@gmail.com>
Will Rouesnel <w.rouesnel@gmail.com>
Will Weaver <monkey@buildingbananas.com>
William Delanoue <william.delanoue@gmail.com>
William Henry <whenry@redhat.com>
William Riancho <wr.wllm@gmail.com>
William Thurston <thurstw@amazon.com>
Xiuming Chen <cc@cxm.cc>
xuzhaokui <cynicholas@gmail.com>
Yang Bai <hamo.by@gmail.com>
Yasunori Mahata <nori@mahata.net>
Yohei Ueda <yohei@jp.ibm.com>
Yurii Rashkovskii <yrashk@gmail.com>
Zac Dover <zdover@redhat.com>
Zach Borboa <zachborboa@gmail.com>
Zain Memon <zain@inzain.net>
Zaiste! <oh@zaiste.net>
Zane DeGraffenried <zane.deg@gmail.com>
Zilin Du <zilin.du@gmail.com>
zimbatm <zimbatm@zimbatm.com>
Zoltan Tombol <zoltan.tombol@gmail.com>
zqh <zqhxuyuan@gmail.com>
Álex González <agonzalezro@gmail.com>
Álvaro Lázaro <alvaro.lazaro.g@gmail.com>
尹吉峰 <jifeng.yin@gmail.com>
+8 -451
View File
@@ -1,448 +1,5 @@
# Changelog
## 1.7.0 (2015-06-16)
#### Runtime
+ Experimental feature: support for out-of-process volume plugins
+ Experimental feature: support for out-of-process network plugins
* Logging: syslog logging driver is available
* The userland proxy can be disabled in favor of hairpin NAT using the daemons `--userland-proxy=false` flag
* The `exec` command supports the `-u|--user` flag to specify the new process owner
+ Default gateway for containers can be specified daemon-wide using the `--default-gateway` and `--default-gateway-v6` flags
+ The CPU CFS (Completely Fair Scheduler) quota can be set in `docker run` using `--cpu-quota`
+ Container block IO can be controlled in `docker run` using`--blkio-weight`
+ ZFS support
+ The `docker logs` command supports a `--since` argument
+ UTS namespace can be shared with the host with `docker run --uts=host`
#### Quality
* Networking stack was entirely rewritten as part of the libnetwork effort
* Engine internals refactoring (shout out to the contributors?)
* Volumes code was entirely rewritten to support the plugins effort
+ Sending SIGUSR1 to a daemon will dump all goroutines stacks without exiting
#### Build
+ Support ${variable:-value} and ${variable:+value} syntax for environment variables
+ Support resource management flags `--cgroup-parent`, `--cpu-period`, `--cpu-quota`, `--cpuset-cpus`, `--cpuset-mems`
+ git context changes with branches and directories
* The .dockerignore file support exclusion rules
#### Distribution
+ Client support for v2 mirroring support for the official registry
#### Bugfixes
* Firewalld is now supported and will automatically be used when available
* mounting --device recursively
## 1.6.2 (2015-05-13)
#### Runtime
- Revert change prohibiting mounting into /sys
## 1.6.1 (2015-05-07)
#### Security
- Fix read/write /proc paths (CVE-2015-3630)
- Prohibit VOLUME /proc and VOLUME / (CVE-2015-3631)
- Fix opening of file-descriptor 1 (CVE-2015-3627)
- Fix symlink traversal on container respawn allowing local privilege escalation (CVE-2015-3629)
- Prohibit mount of /sys
#### Runtime
- Update Apparmor policy to not allow mounts
## 1.6.0 (2015-04-07)
#### Builder
+ Building images from an image ID
+ Build containers with resource constraints, ie `docker build --cpu-shares=100 --memory=1024m...`
+ `commit --change` to apply specified Dockerfile instructions while committing the image
+ `import --change` to apply specified Dockerfile instructions while importing the image
+ Builds no longer continue in the background when canceled with CTRL-C
#### Client
+ Windows Support
#### Runtime
+ Container and image Labels
+ `--cgroup-parent` for specifying a parent cgroup to place container cgroup within
+ Logging drivers, `json-file`, `syslog`, or `none`
+ Pulling images by ID
+ `--ulimit` to set the ulimit on a container
+ `--default-ulimit` option on the daemon which applies to all created containers (and overwritten by `--ulimit` on run)
## 1.5.0 (2015-02-10)
#### Builder
+ Dockerfile to use for a given `docker build` can be specified with the `-f` flag
* Dockerfile and .dockerignore files can be themselves excluded as part of the .dockerignore file, thus preventing modifications to these files invalidating ADD or COPY instructions cache
* ADD and COPY instructions accept relative paths
* Dockerfile `FROM scratch` instruction is now interpreted as a no-base specifier
* Improve performance when exposing a large number of ports
#### Hack
+ Allow client-side only integration tests for Windows
* Include docker-py integration tests against Docker daemon as part of our test suites
#### Packaging
+ Support for the new version of the registry HTTP API
* Speed up `docker push` for images with a majority of already existing layers
- Fixed contacting a private registry through a proxy
#### Remote API
+ A new endpoint will stream live container resource metrics and can be accessed with the `docker stats` command
+ Containers can be renamed using the new `rename` endpoint and the associated `docker rename` command
* Container `inspect` endpoint show the ID of `exec` commands running in this container
* Container `inspect` endpoint show the number of times Docker auto-restarted the container
* New types of event can be streamed by the `events` endpoint: OOM (container died with out of memory), exec_create, and exec_start'
- Fixed returned string fields which hold numeric characters incorrectly omitting surrounding double quotes
#### Runtime
+ 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
* 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
- Fixed container exiting on out of memory to return an invalid exit code
#### Other
* The HTTP_PROXY, HTTPS_PROXY, and NO_PROXY environment variables are properly taken into account by the client when connecting to the Docker daemon
## 1.4.1 (2014-12-15)
#### Runtime
- Fix issue with volumes-from and bind mounts not being honored after create
## 1.4.0 (2014-12-11)
#### 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:
`ENV name=value name2=value2...`
+ New Overlayfs Storage Driver
+ `docker info` now returns an `ID` and `Name` field
+ Filter events by event name, container, or image
+ `docker cp` now supports copying from container volumes
- Fixed `docker tag`, so it honors `--force` when overriding a tag for existing
image.
## 1.3.3 (2014-12-11)
#### Security
- Fix path traversal vulnerability in processing of absolute symbolic links (CVE-2014-9356)
- Fix decompression of xz image archives, preventing privilege escalation (CVE-2014-9357)
- Validate image IDs (CVE-2014-9358)
#### Runtime
- Fix an issue when image archives are being read slowly
#### Client
- Fix a regression related to stdin redirection
- Fix a regression with `docker cp` when destination is the current directory
## 1.3.2 (2014-11-20)
#### Security
- Fix tar breakout vulnerability
* Extractions are now sandboxed chroot
- Security options are no longer committed to images
#### Runtime
- Fix deadlock in `docker ps -f exited=1`
- Fix a bug when `--volumes-from` references a container that failed to start
#### Registry
+ `--insecure-registry` now accepts CIDR notation such as 10.1.0.0/16
* Private registries whose IPs fall in the 127.0.0.0/8 range do no need the `--insecure-registry` flag
- Skip the experimental registry v2 API when mirroring is enabled
## 1.3.1 (2014-10-28)
#### Security
* Prevent fallback to SSL protocols < TLS 1.0 for client, daemon and registry
+ Secure HTTPS connection to registries with certificate verification and without HTTP fallback unless `--insecure-registry` is specified
#### Runtime
- Fix issue where volumes would not be shared
#### Client
- Fix issue with `--iptables=false` not automatically setting `--ip-masq=false`
- Fix docker run output to non-TTY stdout
#### Builder
- Fix escaping `$` for environment variables
- Fix issue with lowercase `onbuild` Dockerfile instruction
- Restrict environment variable expansion to `ENV`, `ADD`, `COPY`, `WORKDIR`, `EXPOSE`, `VOLUME` and `USER`
## 1.3.0 (2014-10-14)
#### Notable features since 1.2.0
+ Docker `exec` allows you to run additional processes inside existing containers
+ Docker `create` gives you the ability to create a container via the CLI without executing a process
+ `--security-opts` options to allow user to customize container labels and apparmor profiles
+ Docker `ps` filters
- Wildcard support to COPY/ADD
+ Move production URLs to get.docker.com from get.docker.io
+ Allocate IP address on the bridge inside a valid CIDR
+ Use drone.io for PR and CI testing
+ Ability to setup an official registry mirror
+ Ability to save multiple images with docker `save`
## 1.2.0 (2014-08-20)
#### Runtime
+ Make /etc/hosts /etc/resolv.conf and /etc/hostname editable at runtime
+ Auto-restart containers using policies
+ Use /var/lib/docker/tmp for large temporary files
+ `--cap-add` and `--cap-drop` to tweak what linux capability you want
+ `--device` to use devices in containers
#### Client
+ `docker search` on private registries
+ Add `exited` filter to `docker ps --filter`
* `docker rm -f` now kills instead of stop
+ Support for IPv6 addresses in `--dns` flag
#### Proxy
+ Proxy instances in separate processes
* Small bug fix on UDP proxy
## 1.1.2 (2014-07-23)
#### Runtime
+ Fix port allocation for existing containers
+ Fix containers restart on daemon restart
#### Packaging
+ Fix /etc/init.d/docker issue on Debian
## 1.1.1 (2014-07-09)
#### Builder
* Fix issue with ADD
## 1.1.0 (2014-07-03)
#### Notable features since 1.0.1
+ Add `.dockerignore` support
+ Pause containers during `docker commit`
+ Add `--tail` to `docker logs`
#### Builder
+ Allow a tar file as context for `docker build`
* Fix issue with white-spaces and multi-lines in `Dockerfiles`
#### Runtime
* Overall performance improvements
* Allow `/` as source of `docker run -v`
* Fix port allocation
* Fix bug in `docker save`
* Add links information to `docker inspect`
#### Client
* Improve command line parsing for `docker commit`
#### Remote API
* Improve status code for the `start` and `stop` endpoints
## 1.0.1 (2014-06-19)
#### Notable features since 1.0.0
* Enhance security for the LXC driver
#### Builder
* Fix `ONBUILD` instruction passed to grandchildren
#### Runtime
* Fix events subscription
* Fix /etc/hostname file with host networking
* Allow `-h` and `--net=none`
* Fix issue with hotplug devices in `--privileged`
#### Client
* Fix artifacts with events
* Fix a panic with empty flags
* Fix `docker cp` on Mac OS X
#### Miscellaneous
* Fix compilation on Mac OS X
* Fix several races
## 1.0.0 (2014-06-09)
#### Notable features since 0.12.0
* Production support
## 0.12.0 (2014-06-05)
#### Notable features since 0.11.0
* 40+ various improvements to stability, performance and usability
* New `COPY` Dockerfile instruction to allow copying a local file from the context into the container without ever extracting if the file is a tar file
* Inherit file permissions from the host on `ADD`
* New `pause` and `unpause` commands to allow pausing and unpausing of containers using cgroup freezer
* The `images` command has a `-f`/`--filter` option to filter the list of images
* Add `--force-rm` to clean up after a failed build
* Standardize JSON keys in Remote API to CamelCase
* Pull from a docker run now assumes `latest` tag if not specified
* Enhance security on Linux capabilities and device nodes
## 0.11.1 (2014-05-07)
#### Registry
- Fix push and pull to private registry
## 0.11.0 (2014-05-07)
#### Notable features since 0.10.0
* SELinux support for mount and process labels
* Linked containers can be accessed by hostname
* Use the net `--net` flag to allow advanced network configuration such as host networking so that containers can use the host's network interfaces
* Add a ping endpoint to the Remote API to do healthchecks of your docker daemon
* Logs can now be returned with an optional timestamp
* Docker now works with registries that support SHA-512
* Multiple registry endpoints are supported to allow registry mirrors
## 0.10.0 (2014-04-08)
#### Builder
- Fix printing multiple messages on a single line. Fixes broken output during builds.
- Follow symlinks inside container's root for ADD build instructions.
- Fix EXPOSE caching.
#### Documentation
- Add the new options of `docker ps` to the documentation.
- Add the options of `docker restart` to the documentation.
- Update daemon docs and help messages for --iptables and --ip-forward.
- Updated apt-cacher-ng docs example.
- Remove duplicate description of --mtu from docs.
- Add missing -t and -v for `docker images` to the docs.
- Add fixes to the cli docs.
- Update libcontainer docs.
- Update images in docs to remove references to AUFS and LXC.
- Update the nodejs_web_app in the docs to use the new epel RPM address.
- Fix external link on security of containers.
- Update remote API docs.
- Add image size to history docs.
- Be explicit about binding to all interfaces in redis example.
- Document DisableNetwork flag in the 1.10 remote api.
- Document that `--lxc-conf` is lxc only.
- Add chef usage documentation.
- Add example for an image with multiple for `docker load`.
- Explain what `docker run -a` does in the docs.
#### Contrib
- Add variable for DOCKER_LOGFILE to sysvinit and use append instead of overwrite in opening the logfile.
- Fix init script cgroup mounting workarounds to be more similar to cgroupfs-mount and thus work properly.
- Remove inotifywait hack from the upstart host-integration example because it's not necessary any more.
- Add check-config script to contrib.
- Fix fish shell completion.
#### Hack
* Clean up "go test" output from "make test" to be much more readable/scannable.
* Excluse more "definitely not unit tested Go source code" directories from hack/make/test.
+ Generate md5 and sha256 hashes when building, and upload them via hack/release.sh.
- Include contributed completions in Ubuntu PPA.
+ Add cli integration tests.
* Add tweaks to the hack scripts to make them simpler.
#### Remote API
+ Add TLS auth support for API.
* Move git clone from daemon to client.
- Fix content-type detection in docker cp.
* Split API into 2 go packages.
#### Runtime
* Support hairpin NAT without going through Docker server.
- devicemapper: succeed immediately when removing non-existing devices.
- devicemapper: improve handling of devicemapper devices (add per device lock, increase sleep time and unlock while sleeping).
- devicemapper: increase timeout in waitClose to 10 seconds.
- devicemapper: ensure we shut down thin pool cleanly.
- devicemapper: pass info, rather than hash to activateDeviceIfNeeded, deactivateDevice, setInitialized, deleteDevice.
- devicemapper: avoid AB-BA deadlock.
- devicemapper: make shutdown better/faster.
- improve alpha sorting in mflag.
- Remove manual http cookie management because the cookiejar is being used.
- Use BSD raw mode on Darwin. Fixes nano, tmux and others.
- Add FreeBSD support for the client.
- Merge auth package into registry.
- Add deprecation warning for -t on `docker pull`.
- Remove goroutine leak on error.
- Update parseLxcInfo to comply with new lxc1.0 format.
- Fix attach exit on darwin.
- Improve deprecation message.
- Retry to retrieve the layer metadata up to 5 times for `docker pull`.
- Only unshare the mount namespace for execin.
- Merge existing config when committing.
- Disable daemon startup timeout.
- Fix issue #4681: add loopback interface when networking is disabled.
- Add failing test case for issue #4681.
- Send SIGTERM to child, instead of SIGKILL.
- Show the driver and the kernel version in `docker info` even when not in debug mode.
- Always symlink /dev/ptmx for libcontainer. This fixes console related problems.
- Fix issue caused by the absence of /etc/apparmor.d.
- Don't leave empty cidFile behind when failing to create the container.
- Mount cgroups automatically if they're not mounted already.
- Use mock for search tests.
- Update to double-dash everywhere.
- Move .dockerenv parsing to lxc driver.
- Move all bind-mounts in the container inside the namespace.
- Don't use separate bind mount for container.
- Always symlink /dev/ptmx for libcontainer.
- Don't kill by pid for other drivers.
- Add initial logging to libcontainer.
* Sort by port in `docker ps`.
- Move networking drivers into runtime top level package.
+ Add --no-prune to `docker rmi`.
+ Add time since exit in `docker ps`.
- graphdriver: add build tags.
- Prevent allocation of previously allocated ports & prevent improve port allocation.
* Add support for --since/--before in `docker ps`.
- Clean up container stop.
+ Add support for configurable dns search domains.
- Add support for relative WORKDIR instructions.
- Add --output flag for docker save.
- Remove duplication of DNS entries in config merging.
- Add cpuset.cpus to cgroups and native driver options.
- Remove docker-ci.
- Promote btrfs. btrfs is no longer considered experimental.
- Add --input flag to `docker load`.
- Return error when existing bridge doesn't match IP address.
- Strip comments before parsing line continuations to avoid interpreting instructions as comments.
- Fix TestOnlyLoopbackExistsWhenUsingDisableNetworkOption to ignore "DOWN" interfaces.
- Add systemd implementation of cgroups and make containers show up as systemd units.
- Fix commit and import when no repository is specified.
- Remount /var/lib/docker as --private to fix scaling issue.
- Use the environment's proxy when pinging the remote registry.
- Reduce error level from harmless errors.
* Allow --volumes-from to be individual files.
- Fix expanding buffer in StdCopy.
- Set error regardless of attach or stdin. This fixes #3364.
- Add support for --env-file to load environment variables from files.
- Symlink /etc/mtab and /proc/mounts.
- Allow pushing a single tag.
- Shut down containers cleanly at shutdown and wait forever for the containers to shut down. This makes container shutdown on daemon shutdown work properly via SIGTERM.
- Don't throw error when starting an already running container.
- Fix dynamic port allocation limit.
- remove setupDev from libcontainer.
- Add API version to `docker version`.
- Return correct exit code when receiving signal and make SIGQUIT quit without cleanup.
- Fix --volumes-from mount failure.
- Allow non-privileged containers to create device nodes.
- Skip login tests because of external dependency on a hosted service.
- Deprecate `docker images --tree` and `docker images --viz`.
- Deprecate `docker insert`.
- Include base abstraction for apparmor. This fixes some apparmor related problems on Ubuntu 14.04.
- Add specific error message when hitting 401 over HTTP on push.
- Fix absolute volume check.
- Remove volumes-from from the config.
- Move DNS options to hostconfig.
- Update the apparmor profile for libcontainer.
- Add deprecation notice for `docker commit -run`.
## 0.9.1 (2014-03-24)
#### Builder
@@ -523,7 +80,7 @@
- Add newlines to the JSON stream functions.
#### Runtime
* Do not ping the registry from the CLI. All requests to registries flow through the daemon.
* Do not ping the registry from the CLI. All requests to registres flow through the daemon.
- Check for nil information return in the lxc driver. This fixes panics with older lxc versions.
- Devicemapper: cleanups and fix for unmount. Fixes two problems which were causing unmount to fail intermittently.
- Devicemapper: remove directory when removing device. Directories don't get left behind when removing the device.
@@ -1115,7 +672,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
+ Add domainname support
+ Implement image filtering with path.Match
* Remove unnecessary warnings
* Remove unnecesasry warnings
* Remove os/user dependency
* Only mount the hostname file when the config exists
* Handle signals within the `docker login` command
@@ -1138,7 +695,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
+ Hack: Vendor all dependencies
* Remote API: Bump to v1.5
* Packaging: Break down hack/make.sh into small scripts, one per 'bundle': test, binary, ubuntu etc.
* Documentation: General improvements
* Documentation: General improvments
## 0.6.1 (2013-08-23)
@@ -1408,7 +965,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
* Prevent rm of running containers
* Use go1.1 cookiejar
- Fix issue detaching from running TTY container
- Forbid parallel push/pull for a single image/repo. Fixes #311
- Forbid parralel push/pull for a single image/repo. Fixes #311
- Fix race condition within Run command when attaching.
#### Client
@@ -1524,7 +1081,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
+ Add caching to docker builder
+ Add support for docker builder with native API as top level command
+ Implement ENV within docker builder
- Check the command existence prior create and add Unit tests for the case
- Check the command existance prior create and add Unit tests for the case
* use any whitespaces instead of tabs
#### Runtime
@@ -1563,13 +1120,13 @@ With the ongoing changes to the networking and execution subsystems of docker te
#### Runtime
- Fix the command existence check
- Fix the command existance check
- strings.Split may return an empty string on no match
- Fix an index out of range crash if cgroup memory is not
#### Documentation
* Various improvements
* Various improvments
+ New example: sharing data between 2 couchdb databases
#### Other
@@ -1599,7 +1156,7 @@ With the ongoing changes to the networking and execution subsystems of docker te
## 0.2.0 (2013-04-23)
- Runtime: ghost containers can be killed and waited for
* Documentation: update install instructions
* Documentation: update install intructions
- Packaging: fix Vagrantfile
- Development: automate releasing binaries and ubuntu packages
+ Add a changelog
+132 -328
View File
@@ -1,387 +1,191 @@
# Contributing to Docker
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/project/who-written-for/).
Want to hack on Docker? Awesome! Here are instructions to get you
started. They are probably not perfect, please let us know if anything
feels wrong or incomplete.
![Contributors guide](docs/sources/static_files/contributors.png)
## Reporting Issues
This page contains information about reporting issues as well as some tips and
guidelines useful to experienced open source contributors. Finally, make sure
you read our [community guidelines](#docker-community-guidelines) before you
start participating.
When reporting [issues](https://github.com/dotcloud/docker/issues)
on GitHub please include your host OS (Ubuntu 12.04, Fedora 19, etc),
the output of `uname -a` and the output of `docker version` along with
the output of `docker info`. Please include the steps required to reproduce
the problem if possible and applicable.
This information will help us review and fix your issue faster.
## Topics
## Build Environment
* [Reporting Security Issues](#reporting-security-issues)
* [Design and Cleanup Proposals](#design-and-cleanup-proposals)
* [Reporting Issues](#reporting-other-issues)
* [Quick Contribution Tips and Guidelines](#quick-contribution-tips-and-guidelines)
* [Community Guidelines](#docker-community-guidelines)
For instructions on setting up your development environment, please
see our dedicated [dev environment setup
docs](http://docs.docker.io/en/latest/contributing/devenvironment/).
## Reporting security issues
## Contribution guidelines
The Docker maintainers take security seriously. If you discover a security
issue, please bring it to their attention right away!
### Pull requests are always welcome
Please **DO NOT** file a public issue, instead send your report privately to
[security@docker.com](mailto:security@docker.com),
We are always thrilled to receive pull requests, and do our best to
process them as fast as possible. Not sure if that typo is worth a pull
request? Do it! We will appreciate it.
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
us know We currently do not offer a paid security bounty program, but are not
ruling it out in the future.
If your pull request is not accepted on the first try, don't be
discouraged! If there's a problem with the implementation, hopefully you
received feedback on what to improve.
We're trying very hard to keep Docker lean and focused. We don't want it
to do everything for everybody. This means that we might decide against
incorporating a new feature. However, there might be a way to implement
that feature *on top of* docker.
## Reporting other issues
### Discuss your design on the mailing list
A great way to contribute to the project is to send a detailed report when you
encounter an issue. We always appreciate a well-written, thorough bug report,
and will thank you for it!
We recommend discussing your plans [on the mailing
list](https://groups.google.com/forum/?fromgroups#!forum/docker-dev)
before starting to code - especially for more ambitious contributions.
This gives other contributors a chance to point you in the right
direction, give feedback on your design, and maybe point out if someone
else is working on the same thing.
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, add a quick "+1" or "I have this problem too." Doing this
helps prioritize the most common problems and requests.
### Create issues...
When reporting issues, please include your host OS (Ubuntu 12.04, Fedora 19,
etc). Please include:
Any significant improvement should be documented as [a GitHub
issue](https://github.com/dotcloud/docker/issues) before anybody
starts working on it.
* The output of `uname -a`.
* The output of `docker version`.
* The output of `docker -D info`.
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.
**Issue Report Template**:
```
Description of problem:
`docker version`:
`docker info`:
`uname -a`:
Environment details (AWS, VirtualBox, physical, etc.):
How reproducible:
Steps to Reproduce:
1.
2.
3.
Actual Results:
Expected Results:
Additional info:
```
##Quick contribution tips and guidelines
This section gives the experienced contributor some tips and guidelines.
###Pull requests are always welcome
Not sure if that typo is worth a pull request? Found a bug and know how to fix
it? Do it! We will appreciate it. Any significant improvement should be
documented as [a GitHub issue](https://github.com/docker/docker/issues) before
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/project/make-a-contribution/).
### Design and cleanup proposals
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/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.
However, there might be a way to implement that feature *on top of* Docker.
### Talking to other Docker users and contributors
<table class="tg">
<col width="45%">
<col width="65%">
<tr>
<td>Internet&nbsp;Relay&nbsp;Chat&nbsp;(IRC)</td>
<td>
<p>
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
<a href="https://botbot.me/freenode/docker/#" target="_blank">our chat archives</a>.
</p>
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>
<td>Google Groups</td>
<td>
There are two groups.
<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.
</td>
</tr>
<tr>
<td>Twitter</td>
<td>
You can follow <a href="https://twitter.com/docker/" target="_blank">Docker's Twitter feed</a>
to get updates on our products. You can also tweet us questions or just
share blogs or stories.
</td>
</tr>
<tr>
<td>Stack Overflow</td>
<td>
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>
</tr>
</table>
### ...but check for existing issues first!
Please take a moment to check that an issue doesn't already exist
documenting your bug report or improvement proposal. If it does, it
never hurts to add a quick "+1" or "I have this problem too". This will
help prioritize the most common problems and requests.
### Conventions
Fork the repository and make changes on your fork in a feature branch:
Fork the repo and make changes on your fork in a feature branch:
- If it's a bug fix branch, name it XXXX-something where XXXX is the number of
the issue.
- If it's a feature branch, create an enhancement issue to announce
your intentions, and name it XXXX-something where XXXX is the number of the
issue.
- If it's a bugfix branch, name it XXX-something where XXX is the number of the
issue
- If it's a feature branch, create an enhancement issue to announce your
intentions, and name it XXX-something where XXX is the number of the issue.
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/project/test-and-docs/) on your branch before
submitting a pull request.
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 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/project/doc-style) and instructions on [building
the documentation](https://docs.docker.com/project/test-and-docs/#build-and-test-the-documentation).
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 ``docs/README.md`` for more
information on building the docs and how docs get released.
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
committing your changes. Most editors have plug-ins that do this automatically.
and maintenance. Always run `go fmt` before committing your changes. Most
editors have plugins that do this automatically, and there's also a git
pre-commit hook:
Pull request descriptions should be as clear as possible and include a reference
to all the issues that they address.
```
curl -o .git/hooks/pre-commit https://raw.github.com/edsrzf/gofmt-git-hook/master/fmt-check && chmod +x .git/hooks/pre-commit
```
Commit messages must start with a capitalized and short summary (max. 50 chars)
written in the imperative, followed by an optional, more detailed explanatory
text which is separated from the summary by an empty line.
Pull requests descriptions should be as clear as possible and include a
reference to all the issues that they address.
Pull requests must not contain commits from other users or branches.
Code review comments may be added to your pull request. Discuss, then make the
suggested modifications and push additional commits to your feature branch. Post
a comment after pushing. New commits show up in the pull request automatically,
but the reviewers are notified only when you comment.
suggested modifications and push additional commits to your feature branch. Be
sure to post a comment after pushing. The new commits will show up in the pull
request automatically, but the reviewers will not be notified unless you
comment.
Pull requests must be cleanly rebased on top of master without multiple branches
mixed into the PR.
Before the pull request is merged, make sure that you squash your commits into
logical units of work using `git rebase -i` and `git push -f`. After every
commit the test suite should be passing. Include documentation changes in the
same commit so that a revert would remove all traces of the feature or fix.
**Git tip**: If your PR no longer merges cleanly, use `rebase master` in your
feature branch to update your pull request rather than `merge master`.
Commits that fix or close an issue should include a reference like `Closes #XXX`
or `Fixes #XXX`, which will automatically close the issue when merged.
Before you make a pull request, squash your commits into logical units of work
using `git rebase -i` and `git push -f`. A logical unit of work is a consistent
set of patches that should be reviewed together: for example, upgrading the
version of a vendored dependency and taking advantage of its now available new
feature constitute two separate units of work. Implementing a new function and
calling it in another file constitute a single logical unit of work. The very
high majory 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/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.
Include an issue reference like `Closes #XXXX` or `Fixes #XXXX` in commits that
close an issue. Including references automatically closes the issue on a merge.
Please do not add yourself to the `AUTHORS` file, as it is regenerated regularly
from the Git history.
Add your name to the AUTHORS file, but make sure the list is sorted and your
name and email address match your git configuration. The AUTHORS file is
regenerated occasionally from the git commit history, so a mismatch may result
in your changes being overwritten.
### Merge approval
Docker maintainers use LGTM (Looks Good To Me) in comments on the code review to
indicate acceptance.
Docker maintainers use LGTM (looks good to me) in comments on the code review
to indicate acceptance.
A change requires LGTMs from an absolute majority of the maintainers of each
component affected. For example, if a change affects `docs/` and `registry/`, it
needs an absolute majority from the maintainers of `docs/` AND, separately, an
absolute majority of the maintainers of `registry/`.
component affected. For example, if a change affects docs/ and registry/, it
needs an absolute majority from the maintainers of docs/ AND, separately, an
absolute majority of the maintainers of registry.
For more details, see the [MAINTAINERS](MAINTAINERS) page.
For more details see [MAINTAINERS.md](hack/MAINTAINERS.md)
### Sign your work
The sign-off is a simple line at the end of the explanation for the patch. Your
signature certifies that you wrote the patch or otherwise have the right to pass
it on as an open-source patch. The rules are pretty simple: if you can certify
the below (from [developercertificate.org](http://developercertificate.org/)):
The sign-off is a simple line at the end of the explanation for the
patch, which certifies that you wrote it or otherwise have the right to
pass it on as an open-source patch. The rules are pretty simple: if you
can certify the below:
```
Developer Certificate of Origin
Version 1.1
Docker Developer Certificate of Origin 1.1
Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
660 York Street, Suite 102,
San Francisco, CA 94110 USA
By making a contribution to the Docker Project ("Project"), I represent and
warrant that:
Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.
a. The contribution was created in whole or in part by me and I have the right
to submit the contribution on my own behalf or on behalf of a third party who
has authorized me to submit this contribution to the Project; or
Developer's Certificate of Origin 1.1
b. The contribution is based upon previous work that, to the best of my
knowledge, is covered under an appropriate open source license and I have the
right and authorization to submit that work with modifications, whether
created in whole or in part by me, under the same open source license (unless
I am permitted to submit under a different license) that I have identified in
the contribution; or
By making a contribution to this project, I certify that:
c. The contribution was provided directly to me by some other person who
represented and warranted (a) or (b) and I have not modified it.
(a) The contribution was created in whole or in part by me and I
have the right to submit it under the open source license
indicated in the file; or
(b) The contribution is based upon previous work that, to the best
of my knowledge, is covered under an appropriate open source
license and I have the right under that license to submit that
work with modifications, whether created in whole or in part
by me, under the same open source license (unless I am
permitted to submit under a different license), as indicated
in the file; or
(c) The contribution was provided directly to me by some other
person who certified (a), (b) or (c) and I have not modified
it.
(d) I understand and agree that this project and the contribution
are public and that a record of the contribution (including all
personal information I submit with it, including my sign-off) is
maintained indefinitely and may be redistributed consistent with
this project or the open source license(s) involved.
d. I understand and agree that this Project and the contribution are publicly
known and that a record of the contribution (including all personal
information I submit with it, including my sign-off record) is maintained
indefinitely and may be redistributed consistent with this Project or the open
source license(s) involved.
```
Then you just add a line to every git commit message:
then you just add a line to every git commit message:
Signed-off-by: Joe Smith <joe.smith@email.com>
Docker-DCO-1.1-Signed-off-by: Joe Smith <joe.smith@email.com> (github: github_handle)
Use your real name (sorry, no pseudonyms or anonymous contributions.)
using your real name (sorry, no pseudonyms or anonymous contributions.)
If you set your `user.name` and `user.email` git configs, you can sign your
commit automatically with `git commit -s`.
One way to automate this, is customise your get ``commit.template`` by adding
a ``prepare-commit-msg`` hook to your docker checkout:
Note that the old-style `Docker-DCO-1.1-Signed-off-by: ...` format is still
accepted, so there is no need to update outstanding pull requests to the new
format right away, but please do adjust your processes for future contributions.
```
curl -o .git/hooks/prepare-commit-msg https://raw.github.com/dotcloud/docker/master/contrib/prepare-commit-msg.hook && chmod +x .git/hooks/prepare-commit-msg
```
* Note: the above script expects to find your GitHub user name in ``git config --get github.user``
#### Small patch exception
There are several exceptions to the signing requirement. Currently these are:
* Your patch fixes spelling or grammar errors.
* Your patch is a single line change to documentation.
If you have any questions, please refer to the FAQ in the [docs](http://docs.docker.io)
### How can I become a maintainer?
* 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
* Step 1: learn the component inside out
* Step 2: make yourself useful by contributing code, bugfixes, support etc.
* Step 3: volunteer on the irc channel (#docker@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 accomodate 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
your help to keep it that way. To help with this we've come up with some general
guidelines for the community as a whole:
* Be nice: Be courteous, respectful and polite to fellow community members:
no regional, racial, gender, or other abuse will be tolerated. We like
nice people way better than mean ones!
* Encourage diversity and participation: Make everyone in our community feel
welcome, regardless of their background and the extent of their
contributions, and do everything possible to encourage participation in
our community.
* Keep it legal: Basically, don't get us in trouble. Share only content that
you own, do not share private or sensitive information, and don't break
the law.
* Stay on topic: Make sure that you are posting to the correct channel and
avoid off-topic discussions. Remember when you update an issue or respond
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.
### Guideline violations — 3 strikes method
The point of this section is not to find opportunities to punish people, but we
do need a fair way to deal with people who are making our community suck.
1. First occurrence: We'll give you a friendly, but public reminder that the
behavior is inappropriate according to our guidelines.
2. Second occurrence: We will send you a private message with a warning that
any additional violations will result in removal from the community.
3. Third occurrence: Depending on the violation, we may need to delete or ban
your account.
**Notes:**
* Obvious spammers are banned on first occurrence. If we don't do this, we'll
have spam all over the place.
* Violations are forgiven after 6 months of good behavior, and we won't hold a
grudge.
* People who commit minor infractions will get some education, rather than
hammering them in the 3 strikes process.
* The rules apply equally to everyone in the community, no matter how much
you've contributed.
* Extreme violations of a threatening, abusive, destructive or illegal nature
will be addressed immediately and are not subject to 3 strikes or forgiveness.
* Contact abuse@docker.com to report abuse or appeal violations. In the case of
appeals, we know that mistakes happen, and we'll work with you to come up with a
fair solution if there has been a misunderstanding.
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!
+31 -125
View File
@@ -6,13 +6,13 @@
# docker build -t docker .
#
# # 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
# docker run -v `pwd`:/go/src/github.com/dotcloud/docker -privileged -i -t docker bash
#
# # Run the test suite:
# docker run --privileged docker hack/make.sh test
# docker run -privileged docker hack/make.sh test
#
# # Publish a release:
# docker run --privileged \
# docker run -privileged \
# -e AWS_S3_BUCKET=baz \
# -e AWS_ACCESS_KEY=foo \
# -e AWS_SECRET_KEY=bar \
@@ -23,18 +23,15 @@
# the case. Therefore, you don't have to disable it anymore.
#
FROM ubuntu:14.04
MAINTAINER Tianon Gravi <admwiggin@gmail.com> (@tianon)
RUN apt-key adv --keyserver hkp://p80.pool.sks-keyservers.net --recv-keys E871F18B51E0147C77796AC81196BA81F6B0FC61
RUN echo deb http://ppa.launchpad.net/zfs-native/stable/ubuntu trusty main > /etc/apt/sources.list.d/zfs.list
docker-version 0.6.1
FROM ubuntu:13.10
MAINTAINER Tianon Gravi <admwiggin@gmail.com> (@tianon)
# Packaged dependencies
RUN apt-get update && apt-get install -y \
apparmor \
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -yq \
apt-utils \
aufs-tools \
automake \
bash-completion \
btrfs-tools \
build-essential \
curl \
@@ -45,146 +42,55 @@ RUN apt-get update && apt-get install -y \
libcap-dev \
libsqlite3-dev \
mercurial \
parallel \
python-mock \
python-pip \
python-websocket \
reprepro \
ruby1.9.1 \
ruby1.9.1-dev \
s3cmd=1.1.0* \
ubuntu-zfs \
libzfs-dev \
--no-install-recommends
# Get and compile LXC 0.8 (since it is the most stable)
RUN git clone --no-checkout https://github.com/lxc/lxc.git /usr/local/lxc && cd /usr/local/lxc && git checkout -q lxc-0.8.0
RUN cd /usr/local/lxc && ./autogen.sh && ./configure --disable-docs && make && make install
# Get lvm2 source for compiling statically
RUN git clone -b v2_02_103 https://git.fedorahosted.org/git/lvm2.git /usr/local/lvm2
RUN git clone --no-checkout https://git.fedorahosted.org/git/lvm2.git /usr/local/lvm2 && cd /usr/local/lvm2 && git checkout -q v2_02_103
# see https://git.fedorahosted.org/cgit/lvm2.git/refs/tags for release tags
# note: we don't use "git clone -b" above because it then spews big nasty warnings about 'detached HEAD' state that we can't silence as easily as we can silence them using "git checkout" directly
# Compile and install lvm2
RUN cd /usr/local/lvm2 \
&& ./configure --enable-static_link \
&& make device-mapper \
&& make install_device-mapper
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 lxc
ENV LXC_VERSION 1.0.7
RUN mkdir -p /usr/src/lxc \
&& curl -sSL https://linuxcontainers.org/downloads/lxc/lxc-${LXC_VERSION}.tar.gz | tar -v -C /usr/src/lxc/ -xz --strip-components=1
RUN cd /usr/src/lxc \
&& ./configure \
&& make \
&& make install \
&& ldconfig
# Install Go
ENV GO_VERSION 1.4.2
RUN curl -sSL https://golang.org/dl/go${GO_VERSION}.src.tar.gz | tar -v -C /usr/local -xz \
&& mkdir -p /go/bin
ENV PATH /go/bin:/usr/local/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/docker/docker/vendor
RUN cd /usr/local/go/src && ./make.bash --no-clean 2>&1
RUN curl -s https://go.googlecode.com/files/go1.2.1.src.tar.gz | tar -v -C /usr/local -xz
ENV PATH /usr/local/go/bin:$PATH
ENV GOPATH /go:/go/src/github.com/dotcloud/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/386 \
freebsd/amd64 freebsd/386 freebsd/arm \
windows/amd64 windows/386
ENV DOCKER_CROSSPLATFORMS linux/386 linux/arm darwin/amd64 darwin/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
ENV GOARM 5
RUN cd /usr/local/go/src && bash -xc '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
# 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
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
RUN go get code.google.com/p/go.tools/cmd/cover
# 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 d957768537c5af40e4f4cd96871f7b2bde9e2923
RUN set -x \
&& git clone https://github.com/docker/distribution.git /go/src/github.com/docker/distribution \
&& (cd /go/src/github.com/docker/distribution && git checkout -q $REGISTRY_COMMIT) \
&& GOPATH=/go/src/github.com/docker/distribution/Godeps/_workspace:/go \
go build -o /go/bin/registry-v2 github.com/docker/distribution/cmd/registry \
&& rm -rf /go/src/github.com/docker/distribution/
# Get the "docker-py" source so we can run their integration tests
ENV DOCKER_PY_COMMIT 91985b239764fe54714fa0a93d52aa362357d251
RUN git clone https://github.com/docker/docker-py.git /docker-py \
&& cd /docker-py \
&& git checkout -q $DOCKER_PY_COMMIT
RUN gem install --no-rdoc --no-ri fpm --version 1.0.2
# Setup s3cmd config
RUN { \
echo '[default]'; \
echo 'access_key=$AWS_ACCESS_KEY'; \
echo 'secret_key=$AWS_SECRET_KEY'; \
} > ~/.s3cfg
RUN /bin/echo -e '[default]\naccess_key=$AWS_ACCESS_KEY\nsecret_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'
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 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.sh /go/src/github.com/docker/docker/contrib/
RUN ./contrib/download-frozen-image.sh /docker-frozen-images \
busybox:latest@4986bf8c15363d1c5d15512d5266f8777bfba4974ac56e3270e7760f6f0a8125 \
hello-world:frozen@e45a5af57b00862e5ef5782a9925979a02ba2b12dff832fd0991335f4a11e5c5 \
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 \
&& git clone -b v1.0.1 https://github.com/cpuguy83/go-md2man.git /go/src/github.com/cpuguy83/go-md2man \
&& git clone -b v1.2 https://github.com/russross/blackfriday.git /go/src/github.com/russross/blackfriday
# Download toml validator
ENV TOMLV_COMMIT 9baf8a8a9f2ed20a8e54160840c492f937eeaf9a
RUN set -x \
&& git clone https://github.com/BurntSushi/toml.git /go/src/github.com/BurntSushi/toml \
&& (cd /go/src/github.com/BurntSushi/toml && git checkout -q $TOMLV_COMMIT)
# copy vendor/ because go-md2man needs golang.org/x/net
COPY vendor /go/src/github.com/docker/docker/vendor
RUN go install -v github.com/cpuguy83/go-md2man \
github.com/BurntSushi/toml/cmd/tomlv
VOLUME /var/lib/docker
WORKDIR /go/src/github.com/dotcloud/docker
ENV DOCKER_BUILDTAGS apparmor
# Wrap all commands in the "docker-in-docker" script to allow nested containers
ENTRYPOINT ["hack/dind"]
ENTRYPOINT ["hack/dind"]
# Upload docker source
COPY . /go/src/github.com/docker/docker
ADD . /go/src/github.com/dotcloud/docker
-34
View File
@@ -1,34 +0,0 @@
# docker build -t docker:simple -f Dockerfile.simple .
# docker run --rm docker:simple hack/make.sh dynbinary
# docker run --rm --privileged docker:simple hack/dind hack/make.sh test-unit
# docker run --rm --privileged -v /var/lib/docker docker:simple hack/dind hack/make.sh dynbinary test-integration-cli
# This represents the bare minimum required to build and test Docker.
FROM debian:jessie
# compile and runtime deps
# https://github.com/docker/docker/blob/master/project/PACKAGERS.md#build-dependencies
# https://github.com/docker/docker/blob/master/project/PACKAGERS.md#runtime-dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
btrfs-tools \
curl \
gcc \
git \
golang \
libdevmapper-dev \
libsqlite3-dev \
\
ca-certificates \
e2fsprogs \
iptables \
procps \
xz-utils \
\
aufs-tools \
lxc \
&& rm -rf /var/lib/apt/lists/*
ENV AUTO_GOPATH 1
WORKDIR /usr/src/docker
COPY . /usr/src/docker
+24
View File
@@ -0,0 +1,24 @@
## FIXME
This file is a loose collection of things to improve in the codebase, for the internal
use of the maintainers.
They are not big enough to be in the roadmap, not user-facing enough to be github issues,
and not important enough to be discussed in the mailing list.
They are just like FIXME comments in the source code, except we're not sure where in the source
to put them - so we put them here :)
* Run linter on codebase
* Unify build commands and regular commands
* Move source code into src/ subdir for clarity
* docker build: on non-existent local path for ADD, don't show full absolute path on the host
* use size header for progress bar in pull
* Clean up context upload in build!!!
* Parallel pull
* Upgrade dockerd without stopping containers
* Simple command to remove all untagged images (`docker rmi $(docker images | awk '/^<none>/ { print $3 }')`)
* Simple command to clean up containers for disk space
* Clean up the ProgressReader api, it's a PITA to use
+14 -3
View File
@@ -1,7 +1,7 @@
Apache License
Version 2.0, January 2004
https://www.apache.org/licenses/
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
@@ -176,13 +176,24 @@
END OF TERMS AND CONDITIONS
Copyright 2013-2015 Docker, Inc.
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
https://www.apache.org/licenses/LICENSE-2.0
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
+7 -702
View File
@@ -1,702 +1,7 @@
# Docker maintainers file
#
# 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.
[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.
* A pull request can be in 1 of 5 distinct states, for each of which there is a corresponding label
that needs to be applied. `Rules.review.states` contains the list of states with possible targets
for each.
"""
# Triage
[Rules.review.states.0-needs-triage]
# Maintainers are expected to triage new incoming pull requests by removing
# the `0-triage` label and adding the correct labels (e.g. `1-design-review`)
# potentially skipping some steps depending on the kind of pull request.
# Use common sense for judging.
#
# Checking for DCO should be done at this stage.
#
# If an owner, responsible for closing or merging, can be assigned to the PR,
# the better.
close = "e.g. unresponsive contributor without DCO"
3-docs-review = "non-proposal documentation-only change"
2-code-review = "e.g. trivial bugfix"
1-design-review = "general case"
# Design review
[Rules.review.states.1-needs-design-review]
# Maintainers are expected to comment on the design of the pull request.
# Review of documentation is expected only in the context of design validation,
# not for stylistic changes.
#
# Ideally, documentation should reflect the expected behavior of the code.
# No code review should take place in this step.
#
# Once design is approved, a maintainer should make sure to remove this label
# and add the next one.
close = "design rejected"
3-docs-review = "proposals with only documentation changes"
2-code-review = "general case"
# Code review
[Rules.review.states.2-needs-code-review]
# Maintainers are expected to review the code and ensure that it is good
# quality and in accordance with the documentation in the PR.
#
# If documentation is absent but expected, maintainers should ask for documentation.
#
# All tests should pass.
#
# Once code is approved according to the rules of the subsystem, a maintainer
# should make sure to remove this label and add the next one.
close = ""
1-design-review = "raises design concerns"
4-merge = "trivial change not impacting documentation"
3-docs-review = "general case"
# Docs review
[Rules.review.states.3-needs-docs-review]
# Maintainers are expected to review the documentation in its bigger context,
# ensuring consistency, completeness, validity, and breadth of coverage across
# all extent and new documentation.
#
# They should ask for any editorial change that makes the documentation more
# consistent and easier to understand.
#
# Changes and additions to docs must be reviewed and approved (LGTM'd) by a minimum of
# two docs sub-project maintainers. If the docs change originates with a docs
# maintainer, only one additional LGTM is required (since we assume a docs maintainer
# approves of their own PR).
#
# Once documentation is approved (see below), a maintainer should make sure to remove this
# label and add the next one.
close = ""
2-code-review = "requires more code changes"
1-design-review = "raises design concerns"
4-merge = "general case"
# Merge
[Rules.review.states.4-needs-merge]
# Maintainers are expected to merge this pull request as soon as possible.
# They can ask for a rebase, or carry the pull request themselves.
# These should be the easy PRs to merge.
close = "carry PR"
merge = ""
[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
# can't solve, they show up and fix it with bizarre devices and weaponry.
# They have final say on technical implementation and coding style.
# They are ultimately responsible for quality in all its forms: usability polish,
# bugfixes, performance, stability, etc. When ownership can cleanly be passed to
# a subsystem, they are responsible for doing so and holding the
# subsystem maintainers accountable. If ownership is unclear, they are the de facto owners.
# 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 = [
"crosbymichael",
"erikh",
"estesp",
"icecrime",
"jfrazelle",
"lk4d4",
"runcom",
"tibor",
"unclejack",
"vbatts",
"vieux",
"vishh"
]
[Org.Subsystems]
# 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]
# The curators help ensure that incoming issues and pull requests are properly triaged and
# that our various contribution and reviewing processes are respected. With their knowledge of
# the repository activity, they can also guide contributors to relevant material or
# discussions.
#
# They are neither code nor docs reviewers, so they are never expected to merge. They can
# however:
# - 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 = [
"thajeztah"
]
[people]
# A reference list of all people associated with the project.
# All other sections should refer to people by their canonical key
# in the people section.
# ADD YOURSELF HERE IN ALPHABETICAL ORDER
[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.cpuguy83]
Name = "Brian Goff"
Email = "cpuguy83@gmail.com"
Github = "cpuguy83"
[people.crosbymichael]
Name = "Michael Crosby"
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.jfrazelle]
Name = "Jessie Frazelle"
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.moxiegirl]
Name = "Mary Anthony"
Email = "mary.anthony@docker.com"
GitHub = "moxiegirl"
[people.nathanmccauley]
Name = "Nathan McCauley"
Email = "nathan.mccauley@docker.com"
GitHub = "nathanmccauley"
[people.runcom]
Name = "Antonio Murdaca"
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"
GitHub = "SvenDowideit"
[people.thajeztah]
Name = "Sebastiaan van Stijn"
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"
GitHub = "tianon"
[people.tibor]
Name = "Tibor Vass"
Email = "tibor@docker.com"
GitHub = "tiborvass"
[people.vbatts]
Name = "Vincent Batts"
Email = "vbatts@redhat.com"
GitHub = "vbatts"
[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"
GitHub = "vishh"
Solomon Hykes <solomon@dotcloud.com> (@shykes)
Guillaume Charmes <guillaume@dotcloud.com> (@creack)
Victor Vieux <vieux@docker.com> (@vieux)
Michael Crosby <michael@crosbymichael.com> (@crosbymichael)
.travis.yml: Tianon Gravi <admwiggin@gmail.com> (@tianon)
Dockerfile: Tianon Gravi <admwiggin@gmail.com> (@tianon)
Makefile: Tianon Gravi <admwiggin@gmail.com> (@tianon)
+10 -66
View File
@@ -1,43 +1,9 @@
.PHONY: all binary build cross default docs docs-build docs-shell shell test test-unit test-integration-cli test-docker-py validate
.PHONY: all binary build cross default docs docs-build docs-shell shell test test-integration
# 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 DOCKER_CLIENTONLY \
-e DOCKER_EXECDRIVER \
-e DOCKER_EXPERIMENTAL \
-e DOCKER_GRAPHDRIVER \
-e DOCKER_STORAGE_OPTS \
-e DOCKER_USERLANDPROXY \
-e TESTDIRS \
-e TESTFLAGS \
-e TIMEOUT
# note: we _cannot_ add "-e DOCKER_BUILDTAGS" here because even if it's unset in the shell, that would shadow the "ENV DOCKER_BUILDTAGS" set in our Dockerfile, which is very important for our official builds
# to allow `make BIND_DIR=. shell` or `make BIND_DIR= test`
# (default to no bind mount if DOCKER_HOST is set)
# note: BINDDIR is supported for backwards-compatibility here
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)")
# to allow `make DOCSDIR=docs docs-shell` (to create a bind mount in docs)
DOCS_MOUNT := $(if $(DOCSDIR),-v $(CURDIR)/$(DOCSDIR):/$(DOCSDIR))
# to allow `make DOCSPORT=9000 docs`
DOCSPORT := 8000
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_RUN_DOCKER := docker run --rm -it --privileged $(DOCKER_ENVS) $(DOCKER_MOUNT) "$(DOCKER_IMAGE)"
DOCKER_RUN_DOCS := docker run --rm -it $(DOCS_MOUNT) -e AWS_S3_BUCKET -e NOCACHE
# for some docs workarounds (see below in "docs-build" target)
GITCOMMIT := $(shell git rev-parse --short HEAD 2>/dev/null)
GIT_BRANCH := $(shell git rev-parse --abbrev-ref HEAD)
DOCKER_IMAGE := docker:$(GIT_BRANCH)
DOCKER_DOCS_IMAGE := docker-docs:$(GIT_BRANCH)
DOCKER_RUN_DOCKER := docker run --rm -i -t --privileged -e TESTFLAGS -v "$(CURDIR)/bundles:/go/src/github.com/dotcloud/docker/bundles" "$(DOCKER_IMAGE)"
default: binary
@@ -51,33 +17,16 @@ cross: build
$(DOCKER_RUN_DOCKER) hack/make.sh binary cross
docs: docs-build
$(DOCKER_RUN_DOCS) -p $(if $(DOCSPORT),$(DOCSPORT):)8000 "$(DOCKER_DOCS_IMAGE)" mkdocs serve
docker run --rm -i -t -p 8000:8000 "$(DOCKER_DOCS_IMAGE)"
docs-shell: docs-build
$(DOCKER_RUN_DOCS) -p $(if $(DOCSPORT),$(DOCSPORT):)8000 "$(DOCKER_DOCS_IMAGE)" bash
docs-release: docs-build
$(DOCKER_RUN_DOCS) -e OPTIONS -e BUILD_ROOT -e DISTRIBUTION_ID \
-v $(CURDIR)/docs/awsconfig:/docs/awsconfig \
"$(DOCKER_DOCS_IMAGE)" ./release.sh
docs-test: docs-build
$(DOCKER_RUN_DOCS) "$(DOCKER_DOCS_IMAGE)" ./test.sh
docker run --rm -i -t -p 8000:8000 "$(DOCKER_DOCS_IMAGE)" bash
test: build
$(DOCKER_RUN_DOCKER) hack/make.sh binary cross test-unit test-integration-cli test-docker-py
$(DOCKER_RUN_DOCKER) hack/make.sh test test-integration
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-gofmt validate-test validate-toml validate-vet
test-integration: build
$(DOCKER_RUN_DOCKER) hack/make.sh test-integration
shell: build
$(DOCKER_RUN_DOCKER) bash
@@ -86,11 +35,6 @@ build: bundles
docker build -t "$(DOCKER_IMAGE)" .
docs-build:
cp ./VERSION docs/VERSION
echo "$(GIT_BRANCH)" > docs/GIT_BRANCH
# echo "$(AWS_S3_BUCKET)" > docs/AWS_S3_BUCKET
echo "$(GITCOMMIT)" > docs/GITCOMMIT
docker pull docs/base
docker build -t "$(DOCKER_DOCS_IMAGE)" docs
bundles:
+6 -6
View File
@@ -1,7 +1,7 @@
Docker
Copyright 2012-2015 Docker, Inc.
Copyright 2012-2014 Docker, Inc.
This product includes software developed at Docker, Inc. (https://www.docker.com).
This product includes software developed at Docker, Inc. (http://www.docker.com).
This product contains software (https://github.com/kr/pty) developed
by Keith Rarick, licensed under the MIT License.
@@ -10,10 +10,10 @@ The following is courtesy of our legal counsel:
Use and transfer of Docker may be subject to certain restrictions by the
United States and other governments.
United States and other governments.
It is your responsibility to ensure that your use and/or transfer does not
violate applicable laws.
violate applicable laws.
For more information, please see https://www.bis.doc.gov
For more information, please see http://www.bis.doc.gov
See also https://www.apache.org/dev/crypto.html and/or seek legal counsel.
See also http://www.apache.org/dev/crypto.html and/or seek legal counsel.
+41 -144
View File
@@ -2,29 +2,23 @@ Docker: the Linux container engine
==================================
Docker is an open source project to pack, ship and run any application
as a lightweight container.
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
This means that they can run anywhere, from your laptop to the largest
EC2 compute instance and everything in between - and they don't require
you to use a particular language, framework or packaging system. That
that you 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
databases and backend services without depending on a particular stack
or provider.
Docker began as an open-source implementation of the deployment engine which
powers [dotCloud](https://dotcloud.com), a popular Platform-as-a-Service.
Docker is an open-source implementation of the deployment engine which
powers [dotCloud](http://dotcloud.com), a popular Platform-as-a-Service.
It benefits directly from the experience accumulated over several years
of large-scale operation and support of hundreds of thousands of
applications and databases.
![Docker L](docs/sources/static_files/docker-logo-compressed.png "Docker")
## Security Disclosure
Security is very important to us. If you have any issue regarding security,
please disclose the information responsibly by sending an email to
security@docker.com and not by creating a github issue.
![Docker L](docs/theme/docker/static/img/dockerlogo-h.png "Docker")
## Better than VMs
@@ -56,27 +50,27 @@ By contrast, Docker relies on a different sandboxing method known as
*containerization*. Unlike traditional virtualization, containerization
takes place at the kernel level. Most modern operating system kernels
now support the primitives necessary for containerization, including
Linux with [openvz](https://openvz.org),
Linux with [openvz](http://openvz.org),
[vserver](http://linux-vserver.org) and more recently
[lxc](http://lxc.sourceforge.net), Solaris with
[zones](https://docs.oracle.com/cd/E26502_01/html/E29024/preface-1.html#scrolltoc),
[zones](http://docs.oracle.com/cd/E26502_01/html/E29024/preface-1.html#scrolltoc)
and FreeBSD with
[Jails](https://www.freebsd.org/doc/handbook/jails.html).
[Jails](http://www.freebsd.org/doc/handbook/jails.html).
Docker builds on top of these low-level primitives to offer developers a
portable format and runtime environment that solves all four problems.
portable format and runtime environment that solves all 4 problems.
Docker containers are small (and their transfer can be optimized with
layers), they have basically zero memory and cpu overhead, they are
completely portable, and are designed from the ground up with an
completely portable and are designed from the ground up with an
application-centric design.
Perhaps best of all, because Docker operates at the OS level, it can still be
The best part: because Docker operates at the OS level, it can still be
run inside a VM!
## Plays well with others
Docker does not require you to buy into a particular programming
language, framework, packaging system, or configuration language.
Docker does not require that you buy into a particular programming
language, framework, packaging system or configuration language.
Is your application a Unix process? Does it use files, tcp connections,
environment variables, standard Unix streams and command-line arguments
@@ -100,22 +94,22 @@ This is usually difficult for several reasons:
typically don't work well with each other, requiring awkward
custom integrations.
* *Conflicting dependencies*. Different applications may depend on
* Conflicting dependencies. Different applications may depend on
different versions of the same dependency. Packaging tools handle
these situations with various degrees of ease - but they all
handle them in different and incompatible ways, which again forces
the developer to do extra work.
* *Custom dependencies*. A developer may need to prepare a custom
* Custom dependencies. A developer may need to prepare a custom
version of their application's dependency. Some packaging systems
can handle custom versions of a dependency, others can't - and all
of them handle it differently.
Docker solves the problem of dependency hell by giving the developer a simple
way to express *all* their application's dependencies in one place, while
streamlining the process of assembling them. If this makes you think of
[XKCD 927](https://xkcd.com/927/), don't worry. Docker doesn't
Docker solves dependency hell by giving the developer a simple way to
express *all* their application's dependencies in one place, and
streamline the process of assembling them. If this makes you think of
[XKCD 927](http://xkcd.com/927/), don't worry. Docker doesn't
*replace* your favorite packaging systems. It simply orchestrates
their use in a simple and repeatable way. How does it do that? With
layers.
@@ -131,8 +125,9 @@ Here's a typical Docker build process:
```bash
FROM ubuntu:12.04
RUN apt-get update && apt-get install -y python python-pip curl
RUN curl -sSL https://github.com/shykes/helloflask/archive/master.tar.gz | tar -xzv
RUN apt-get update
RUN apt-get install -q -y python python-pip curl
RUN curl -L https://github.com/shykes/helloflask/archive/master.tar.gz | tar -xzv
RUN cd helloflask-master && pip install -r requirements.txt
```
@@ -147,19 +142,20 @@ 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.
We also offer an interactive tutorial for quickly learning the basics of
using Docker.
For up-to-date install instructions, see the [Docs](https://docs.docker.com).
For up-to-date install instructions and online tutorials, see the
[Getting Started page](http://www.docker.io/gettingstarted/).
Usage examples
==============
Docker can be used to run short-lived commands, long-running daemons
(app servers, databases, etc.), interactive shell sessions, etc.
(app servers, databases etc.), interactive shell sessions, etc.
You can find a [list of real-world
examples](https://docs.docker.com/examples/) in the
examples](http://docs.docker.io/en/latest/examples/) in the
documentation.
Under the hood
@@ -171,125 +167,26 @@ Under the hood, Docker is built on the following components:
[cgroup](http://blog.dotcloud.com/kernel-secrets-from-the-paas-garage-part-24-c)
and
[namespacing](http://blog.dotcloud.com/under-the-hood-linux-kernels-on-dotcloud-part)
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/docker/libcontainer/blob/master/SPEC.md)
capabilities of the Linux kernel;
* The [Go](http://golang.org) programming language.
Contributing to Docker
======================
[![GoDoc](https://godoc.org/github.com/docker/docker?status.svg)](https://godoc.org/github.com/docker/docker)
[![Jenkins Build Status](https://jenkins.dockerproject.com/job/Docker%20Master/badge/icon)](https://jenkins.dockerproject.com/job/Docker%20Master/)
Want to hack on Docker? Awesome! There are instructions to get you
started [here](CONTRIBUTING.md).
Want to hack on Docker? Awesome! We have [instructions to help you get
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.
Getting the development builds
==============================
Want to run Docker from a master build? You can download
master builds at [master.dockerproject.com](https://master.dockerproject.com).
They are updated with each commit merged into the master branch.
Don't know how to use that super cool new feature in the master build? Check
out the master docs at
[docs.master.dockerproject.com](http://docs.master.dockerproject.com).
How the project is run
======================
Docker is a very, very active project. If you want to learn more about how it is run,
or want to get more involved, the best place to start is [the project directory](https://github.com/docker/docker/tree/master/project).
We are always open to suggestions on process improvements, and are always looking for more maintainers.
### Talking to other Docker users and contributors
<table class="tg">
<col width="45%">
<col width="65%">
<tr>
<td>Internet&nbsp;Relay&nbsp;Chat&nbsp;(IRC)</td>
<td>
<p>
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
<a href="https://botbot.me/freenode/docker/#" target="_blank">our chat archives</a>.
</p>
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>
<td>Google Groups</td>
<td>
There are two groups.
<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.
</td>
</tr>
<tr>
<td>Twitter</td>
<td>
You can follow <a href="https://twitter.com/docker/" target="_blank">Docker's Twitter feed</a>
to get updates on our products. You can also tweet us questions or just
share blogs or stories.
</td>
</tr>
<tr>
<td>Stack Overflow</td>
<td>
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>
</tr>
</table>
They are probably not perfect, please let us know if anything feels
wrong or incomplete.
### Legal
*Brought to you courtesy of our legal counsel. For more context,
please see the [NOTICE](https://github.com/docker/docker/blob/master/NOTICE) document in this repo.*
please see the Notice document.*
Use and transfer of Docker may be subject to certain restrictions by the
United States and other governments.
United States and other governments.
It is your responsibility to ensure that your use and/or transfer does not
violate applicable laws.
violate applicable laws.
For more information, please see https://www.bis.doc.gov
Licensing
=========
Docker is licensed under the Apache License, Version 2.0. See
[LICENSE](https://github.com/docker/docker/blob/master/LICENSE) for the full
license text.
Other Docker Related Projects
=============================
There are a number of projects under development that are based on Docker's
core technology. These projects expand the tooling built around the
Docker platform to broaden its application and utility.
* [Docker Registry](https://github.com/docker/distribution): Registry
server for Docker (hosting/delivery of repositories and images)
* [Docker Machine](https://github.com/docker/machine): Machine management
for a container-centric world
* [Docker Swarm](https://github.com/docker/swarm): A Docker-native clustering
system
* [Docker Compose](https://github.com/docker/compose) (formerly Fig):
Define and run multi-container apps
* [Kitematic](https://github.com/kitematic/kitematic): The easiest way to use
Docker on a Mac
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.
For more information, please see http://www.bis.doc.gov
+1 -1
View File
@@ -1 +1 @@
1.7.0-rc1
0.9.1
+1
View File
@@ -0,0 +1 @@
Victor Vieux <vieux@docker.com> (@vieux)
-5
View File
@@ -1,5 +0,0 @@
This directory contains code pertaining to the Docker API:
- Used by the docker client when communicating with the docker daemon
- Used by third party tools wishing to interface with the docker daemon
+46
View File
@@ -1,6 +1,9 @@
package api
import (
"fmt"
"net/http"
"net/http/httptest"
"testing"
)
@@ -17,3 +20,46 @@ func TestJsonContentType(t *testing.T) {
t.Fail()
}
}
func TestGetBoolParam(t *testing.T) {
if ret, err := getBoolParam("true"); err != nil || !ret {
t.Fatalf("true -> true, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam("True"); err != nil || !ret {
t.Fatalf("True -> true, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam("1"); err != nil || !ret {
t.Fatalf("1 -> true, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam(""); err != nil || ret {
t.Fatalf("\"\" -> false, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam("false"); err != nil || ret {
t.Fatalf("false -> false, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam("0"); err != nil || ret {
t.Fatalf("0 -> false, nil | got %t %s", ret, err)
}
if ret, err := getBoolParam("faux"); err == nil || ret {
t.Fatalf("faux -> false, err | got %t %s", ret, err)
}
}
func TesthttpError(t *testing.T) {
r := httptest.NewRecorder()
httpError(r, fmt.Errorf("No such method"))
if r.Code != http.StatusNotFound {
t.Fatalf("Expected %d, got %d", http.StatusNotFound, r.Code)
}
httpError(r, fmt.Errorf("This accound hasn't been activated"))
if r.Code != http.StatusForbidden {
t.Fatalf("Expected %d, got %d", http.StatusForbidden, r.Code)
}
httpError(r, fmt.Errorf("Some error"))
if r.Code != http.StatusInternalServerError {
t.Fatalf("Expected %d, got %d", http.StatusInternalServerError, r.Code)
}
}
+2462
View File
File diff suppressed because it is too large Load Diff
-83
View File
@@ -1,83 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"io"
"net/url"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/signal"
)
// CmdAttach attaches to a running container.
//
// Usage: docker attach [OPTIONS] CONTAINER
func (cli *DockerCli) CmdAttach(args ...string) error {
var (
cmd = cli.Subcmd("attach", "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)
name := cmd.Arg(0)
stream, _, err := cli.call("GET", "/containers/"+name+"/json", nil, nil)
if err != nil {
return err
}
var c types.ContainerJSON
if err := json.NewDecoder(stream).Decode(&c); err != nil {
return err
}
if !c.State.Running {
return fmt.Errorf("You cannot attach to a stopped container, start it first")
}
if err := cli.CheckTtyInput(!*noStdin, c.Config.Tty); err != nil {
return err
}
if c.Config.Tty && cli.isTerminalOut {
if err := cli.monitorTtySize(cmd.Arg(0), false); err != nil {
logrus.Debugf("Error monitoring TTY size: %s", err)
}
}
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, cmd.Arg(0))
if err != nil {
return err
}
if status != 0 {
return StatusError{StatusCode: status}
}
return nil
}
-310
View File
@@ -1,310 +0,0 @@
package client
import (
"bufio"
"encoding/base64"
"encoding/json"
"fmt"
"io"
"io/ioutil"
"net/http"
"net/url"
"os"
"os/exec"
"path"
"path/filepath"
"runtime"
"strconv"
"strings"
"github.com/docker/docker/api"
"github.com/docker/docker/graph/tags"
"github.com/docker/docker/pkg/archive"
"github.com/docker/docker/pkg/fileutils"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
"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/units"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/registry"
"github.com/docker/docker/utils"
)
const (
tarHeaderSize = 512
)
// CmdBuild builds a new image from the source code at a given path.
//
// If '-' is provided instead of a path or URL, Docker will build an image from either a Dockerfile or tar archive read from STDIN.
//
// Usage: docker build [OPTIONS] PATH | URL | -
func (cli *DockerCli) CmdBuild(args ...string) error {
cmd := cli.Subcmd("build", "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"}, "", "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")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
var (
context archive.Archive
isRemote bool
err error
)
_, err = exec.LookPath("git")
hasGit := err == nil
if cmd.Arg(0) == "-" {
// As a special case, 'docker build -' will build from either an empty context with the
// contents of stdin as a Dockerfile, or a tar-ed context from stdin.
buf := bufio.NewReader(cli.in)
magic, err := buf.Peek(tarHeaderSize)
if err != nil && err != io.EOF {
return fmt.Errorf("failed to peek context header from STDIN: %v", err)
}
if !archive.IsArchive(magic) {
dockerfile, err := ioutil.ReadAll(buf)
if err != nil {
return fmt.Errorf("failed to read Dockerfile from STDIN: %v", err)
}
// -f option has no meaning when we're reading it from stdin,
// so just use our default Dockerfile name
*dockerfileName = api.DefaultDockerfileName
context, err = archive.Generate(*dockerfileName, string(dockerfile))
} else {
context = ioutil.NopCloser(buf)
}
} else if urlutil.IsURL(cmd.Arg(0)) && (!urlutil.IsGitURL(cmd.Arg(0)) || !hasGit) {
isRemote = true
} else {
root := cmd.Arg(0)
if urlutil.IsGitURL(root) {
root, err = utils.GitClone(root)
if err != nil {
return err
}
defer os.RemoveAll(root)
}
if _, err := os.Stat(root); err != nil {
return err
}
absRoot, err := filepath.Abs(root)
if err != nil {
return err
}
filename := *dockerfileName // path to Dockerfile
if *dockerfileName == "" {
// No -f/--file was specified so use the default
*dockerfileName = api.DefaultDockerfileName
filename = filepath.Join(absRoot, *dockerfileName)
// Just to be nice ;-) look for 'dockerfile' too but only
// use it if we found it, otherwise ignore this check
if _, err = os.Lstat(filename); os.IsNotExist(err) {
tmpFN := path.Join(absRoot, strings.ToLower(*dockerfileName))
if _, err = os.Lstat(tmpFN); err == nil {
*dockerfileName = strings.ToLower(*dockerfileName)
filename = tmpFN
}
}
}
origDockerfile := *dockerfileName // used for error msg
if filename, err = filepath.Abs(filename); err != nil {
return err
}
// Verify that 'filename' is within the build context
filename, err = symlink.FollowSymlinkInScope(filename, absRoot)
if err != nil {
return fmt.Errorf("The Dockerfile (%s) must be within the build context (%s)", origDockerfile, root)
}
// Now reset the dockerfileName to be relative to the build context
*dockerfileName, err = filepath.Rel(absRoot, filename)
if err != nil {
return err
}
// And canonicalize dockerfile name to a platform-independent one
*dockerfileName, err = archive.CanonicalTarNameForPath(*dockerfileName)
if err != nil {
return fmt.Errorf("Cannot canonicalize dockerfile path %s: %v", *dockerfileName, err)
}
if _, err = os.Lstat(filename); os.IsNotExist(err) {
return fmt.Errorf("Cannot locate Dockerfile: %s", origDockerfile)
}
var includes = []string{"."}
excludes, err := utils.ReadDockerIgnore(path.Join(root, ".dockerignore"))
if err != nil {
return 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.
keepThem1, _ := fileutils.Matches(".dockerignore", excludes)
keepThem2, _ := fileutils.Matches(*dockerfileName, excludes)
if keepThem1 || keepThem2 {
includes = append(includes, ".dockerignore", *dockerfileName)
}
if err := utils.ValidateContextDirectory(root, excludes); err != nil {
return fmt.Errorf("Error checking context is accessible: '%s'. Please check permissions and try again.", err)
}
options := &archive.TarOptions{
Compression: archive.Uncompressed,
ExcludePatterns: excludes,
IncludeFiles: includes,
}
context, err = archive.TarWithOptions(root, options)
if err != nil {
return err
}
}
// windows: show error message about modified file permissions
// FIXME: this is not a valid warning when the daemon is running windows. should be removed once docker engine for windows can build.
if runtime.GOOS == "windows" {
fmt.Fprintln(cli.err, `SECURITY WARNING: You are building a Docker image from Windows against a Linux 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.`)
}
var body io.Reader
// Setup an upload progress bar
// FIXME: ProgressReader shouldn't be this annoying to use
if context != nil {
sf := streamformatter.NewStreamFormatter()
body = 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 != "" {
parsedMemory, err := units.RAMInBytes(*flMemoryString)
if err != nil {
return err
}
memory = parsedMemory
}
var memorySwap int64
if *flMemorySwap != "" {
if *flMemorySwap == "-1" {
memorySwap = -1
} else {
parsedMemorySwap, err := units.RAMInBytes(*flMemorySwap)
if err != nil {
return err
}
memorySwap = parsedMemorySwap
}
}
// Send the build context
v := &url.Values{}
//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
}
}
}
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")
}
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", *dockerfileName)
headers := http.Header(make(map[string][]string))
buf, err := json.Marshal(cli.configFile.AuthConfigs)
if err != nil {
return err
}
headers.Add("X-Registry-Config", base64.URLEncoding.EncodeToString(buf))
if context != nil {
headers.Set("Content-Type", "application/tar")
}
sopts := &streamOpts{
rawTerminal: true,
in: body,
out: cli.out,
headers: headers,
}
err = cli.stream("POST", fmt.Sprintf("/build?%s", v.Encode()), sopts)
if jerr, ok := err.(*jsonmessage.JSONError); ok {
// If no error code is set, default to 1
if jerr.Code == 0 {
jerr.Code = 1
}
return StatusError{Status: jerr.Message, StatusCode: jerr.Code}
}
return err
}
-203
View File
@@ -1,203 +0,0 @@
package client
import (
"crypto/tls"
"encoding/json"
"errors"
"fmt"
"io"
"net/http"
"os"
"path/filepath"
"reflect"
"strings"
"text/template"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/pkg/homedir"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/term"
"github.com/docker/docker/utils"
)
// DockerCli represents the docker command line client.
// Instances of the client can be returned from NewDockerCli.
type DockerCli struct {
// proto holds the client protocol i.e. unix.
proto string
// addr holds the client address.
addr string
// configFile has the client configuration file
configFile *cliconfig.ConfigFile
// in holds the input stream and closer (io.ReadCloser) for the client.
in io.ReadCloser
// out holds the output stream (io.Writer) for the client.
out io.Writer
// err holds the error stream (io.Writer) for the client.
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 dindicates whether the client's STDOUT is a TTY
isTerminalOut bool
// transport holds the client transport instance.
transport *http.Transport
}
var funcMap = template.FuncMap{
"json": func(v interface{}) string {
a, _ := json.Marshal(v)
return string(a)
},
}
func (cli *DockerCli) Out() io.Writer {
return cli.out
}
func (cli *DockerCli) Err() io.Writer {
return cli.err
}
func (cli *DockerCli) getMethod(args ...string) (func(...string) error, bool) {
camelArgs := make([]string, len(args))
for i, s := range args {
if len(s) == 0 {
return nil, false
}
camelArgs[i] = strings.ToUpper(s[:1]) + strings.ToLower(s[1:])
}
methodName := "Cmd" + strings.Join(camelArgs, "")
method := reflect.ValueOf(cli).MethodByName(methodName)
if !method.IsValid() {
return nil, false
}
return method.Interface().(func(...string) error), true
}
// Cmd executes the specified command.
func (cli *DockerCli) Cmd(args ...string) error {
if len(args) > 1 {
method, exists := cli.getMethod(args[:2]...)
if exists {
return method(args[2:]...)
}
}
if len(args) > 0 {
method, exists := cli.getMethod(args[0])
if !exists {
return fmt.Errorf("docker: '%s' is not a docker command. See 'docker --help'.", args[0])
}
return method(args[1:]...)
}
return cli.CmdHelp()
}
// Subcmd is a subcommand of the main "docker" command.
// A subcommand represents an action that can be performed
// from the Docker command line client.
//
// To see all available subcommands, run "docker --help".
func (cli *DockerCli) Subcmd(name, signature, description string, exitOnError bool) *flag.FlagSet {
var errorHandling flag.ErrorHandling
if exitOnError {
errorHandling = flag.ExitOnError
} else {
errorHandling = flag.ContinueOnError
}
flags := flag.NewFlagSet(name, errorHandling)
flags.Usage = func() {
options := ""
if signature != "" {
signature = " " + signature
}
if flags.FlagCountUndeprecated() > 0 {
options = " [OPTIONS]"
}
fmt.Fprintf(cli.out, "\nUsage: docker %s%s%s\n\n%s\n\n", name, options, signature, description)
flags.SetOutput(cli.out)
flags.PrintDefaults()
os.Exit(0)
}
return flags
}
// CheckTtyInput checks if we are trying to attach to a container tty
// from a non-tty client input stream, and if so, returns an error.
func (cli *DockerCli) CheckTtyInput(attachStdin, ttyMode bool) error {
// In order to attach to a container tty, input stream for the client must
// be a tty itself: redirecting or piping the client standard input is
// incompatible with `docker run -t`, `docker exec -t` or `docker attach`.
if ttyMode && attachStdin && !cli.isTerminalIn {
return errors.New("cannot enable tty mode on non tty input")
}
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.
// The client will be given a 32-second timeout (see https://github.com/docker/docker/pull/8035).
func NewDockerCli(in io.ReadCloser, out, err io.Writer, keyFile string, proto, addr string, tlsConfig *tls.Config) *DockerCli {
var (
inFd uintptr
outFd uintptr
isTerminalIn = false
isTerminalOut = false
scheme = "http"
)
if tlsConfig != nil {
scheme = "https"
}
if in != nil {
inFd, isTerminalIn = term.GetFdInfo(in)
}
if out != nil {
outFd, isTerminalOut = term.GetFdInfo(out)
}
if err == nil {
err = out
}
// The transport is created here for reuse during the client session.
tr := &http.Transport{
TLSClientConfig: tlsConfig,
}
utils.ConfigureTCPTransport(tr, proto, addr)
configFile, e := cliconfig.Load(filepath.Join(homedir.Get(), ".docker"))
if e != nil {
fmt.Fprintf(err, "WARNING: Error loading config file:%v\n", e)
}
return &DockerCli{
proto: proto,
addr: addr,
configFile: configFile,
in: in,
out: out,
err: err,
keyFile: keyFile,
inFd: inFd,
outFd: outFd,
isTerminalIn: isTerminalIn,
isTerminalOut: isTerminalOut,
tlsConfig: tlsConfig,
scheme: scheme,
transport: tr,
}
}
-17
View File
@@ -1,17 +0,0 @@
// Package client provides a command-line interface for Docker.
//
// Run "docker help SUBCOMMAND" or "docker SUBCOMMAND --help" to see more information on any Docker subcommand, including the full list of options supported for the subcommand.
// See https://docs.docker.com/installation/ for instructions on installing Docker.
package client
import "fmt"
// An StatusError reports an unsuccessful exit by a command.
type StatusError struct {
Status string
StatusCode int
}
func (e StatusError) Error() string {
return fmt.Sprintf("Status: %s, Code: %d", e.Status, e.StatusCode)
}
-80
View File
@@ -1,80 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"github.com/docker/docker/api/types"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"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", "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"}, "", "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", "#-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)
repository, tag = parsers.ParseRepositoryTag(cmd.Arg(1))
)
//Check if the given image name can be resolved
if repository != "" {
if err := registry.ValidateRepositoryName(repository); err != nil {
return err
}
}
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 = &runconfig.Config{}
if err := json.Unmarshal([]byte(*flConfig), config); err != nil {
return err
}
}
stream, _, err := cli.call("POST", "/commit?"+v.Encode(), config, nil)
if err != nil {
return err
}
if err := json.NewDecoder(stream).Decode(&response); err != nil {
return err
}
fmt.Fprintln(cli.out, response.ID)
return nil
}
-57
View File
@@ -1,57 +0,0 @@
package client
import (
"fmt"
"io"
"strings"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/archive"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdCp copies files/folders from a path on the container to a directory on the host running the command.
//
// If HOSTDIR is '-', the data is written as a tar file to STDOUT.
//
// Usage: docker cp CONTAINER:PATH HOSTDIR
func (cli *DockerCli) CmdCp(args ...string) error {
cmd := cli.Subcmd("cp", "CONTAINER:PATH HOSTDIR|-", "Copy files/folders from a PATH on the container to a HOSTDIR on the host\nrunning the command. Use '-' to write the data as a tar file to STDOUT.", true)
cmd.Require(flag.Exact, 2)
cmd.ParseFlags(args, true)
// deal with path name with `:`
info := strings.SplitN(cmd.Arg(0), ":", 2)
if len(info) != 2 {
return fmt.Errorf("Error: Path not specified")
}
cfg := &types.CopyConfig{
Resource: info[1],
}
stream, statusCode, err := cli.call("POST", "/containers/"+info[0]+"/copy", cfg, nil)
if stream != nil {
defer stream.Close()
}
if statusCode == 404 {
return fmt.Errorf("No such container: %v", info[0])
}
if err != nil {
return err
}
hostPath := cmd.Arg(1)
if statusCode == 200 {
if hostPath == "-" {
_, err = io.Copy(cli.out, stream)
} else {
err = archive.Untar(stream, hostPath, &archive.TarOptions{NoLchown: true})
}
if err != nil {
return err
}
}
return nil
}
-159
View File
@@ -1,159 +0,0 @@
package client
import (
"encoding/base64"
"encoding/json"
"fmt"
"io"
"net/url"
"os"
"strings"
"github.com/docker/docker/api/types"
"github.com/docker/docker/graph/tags"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
"github.com/docker/docker/runconfig"
"github.com/docker/docker/utils"
)
func (cli *DockerCli) pullImage(image string) error {
return cli.pullImageCustomOut(image, cli.out)
}
func (cli *DockerCli) pullImageCustomOut(image string, out io.Writer) error {
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(repos)
if err != nil {
return err
}
// 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
}
registryAuthHeader := []string{
base64.URLEncoding.EncodeToString(buf),
}
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
}
return nil
}
type cidFile struct {
path string
file *os.File
written bool
}
func newCIDFile(path string) (*cidFile, error) {
if _, err := os.Stat(path); err == nil {
return nil, fmt.Errorf("Container ID file found, make sure the other container isn't running or delete %s", path)
}
f, err := os.Create(path)
if err != nil {
return nil, fmt.Errorf("Failed to create the container ID file: %s", err)
}
return &cidFile{path: path, file: f}, nil
}
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
if containerIDFile, err = newCIDFile(cidfile); err != nil {
return nil, err
}
defer containerIDFile.Close()
}
//create the container
stream, statusCode, err := cli.call("POST", "/containers/create?"+containerValues.Encode(), mergedConfig, nil)
//if image not found try to pull it
if statusCode == 404 && strings.Contains(err.Error(), config.Image) {
repo, tag := parsers.ParseRepositoryTag(config.Image)
if tag == "" {
tag = tags.DEFAULTTAG
}
fmt.Fprintf(cli.err, "Unable to find image '%s' locally\n", utils.ImageReference(repo, tag))
// 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
}
// Retry
if stream, _, err = cli.call("POST", "/containers/create?"+containerValues.Encode(), mergedConfig, nil); err != nil {
return nil, err
}
} else if err != nil {
return nil, err
}
var response types.ContainerCreateResponse
if err := json.NewDecoder(stream).Decode(&response); err != nil {
return nil, err
}
for _, warning := range response.Warnings {
fmt.Fprintf(cli.err, "WARNING: %s\n", warning)
}
if containerIDFile != nil {
if err = containerIDFile.Write(response.ID); err != nil {
return nil, err
}
}
return &response, nil
}
// CmdCreate creates a new container from a given image.
//
// Usage: docker create [OPTIONS] IMAGE [COMMAND] [ARG...]
func (cli *DockerCli) CmdCreate(args ...string) error {
cmd := cli.Subcmd("create", "IMAGE [COMMAND] [ARG...]", "Create a new container", true)
// These are flags not stored in Config/HostConfig
var (
flName = cmd.String([]string{"-name"}, "", "Assign a name to the container")
)
config, hostConfig, cmd, err := runconfig.Parse(cmd, args)
if err != nil {
cmd.ReportError(err.Error(), true)
}
if config.Image == "" {
cmd.Usage()
return nil
}
response, err := cli.createContainer(config, hostConfig, hostConfig.ContainerIDFile, *flName)
if err != nil {
return err
}
fmt.Fprintf(cli.out, "%s\n", response.ID)
return nil
}
-52
View File
@@ -1,52 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/archive"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdDiff shows changes on a container's filesystem.
//
// Each changed file is printed on a separate line, prefixed with a single
// character that indicates the status of the file: C (modified), A (added),
// or D (deleted).
//
// Usage: docker diff CONTAINER
func (cli *DockerCli) CmdDiff(args ...string) error {
cmd := cli.Subcmd("diff", "CONTAINER", "Inspect changes on a container's filesystem", true)
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
if cmd.Arg(0) == "" {
return fmt.Errorf("Container name cannot be empty")
}
rdr, _, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/changes", nil, nil)
if err != nil {
return err
}
changes := []types.ContainerChange{}
if err := json.NewDecoder(rdr).Decode(&changes); err != nil {
return err
}
for _, change := range changes {
var kind string
switch change.Kind {
case archive.ChangeModify:
kind = "C"
case archive.ChangeAdd:
kind = "A"
case archive.ChangeDelete:
kind = "D"
}
fmt.Fprintf(cli.out, "%s %s\n", kind, change.Path)
}
return nil
}
-60
View File
@@ -1,60 +0,0 @@
package client
import (
"net/url"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"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", "", "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")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
var (
v = url.Values{}
eventFilterArgs = filters.Args{}
)
// Consolidate all filter flags, and sanity check them early.
// They'll get process in the daemon/server.
for _, f := range flFilter.GetAll() {
var err error
eventFilterArgs, err = filters.ParseFlag(f, eventFilterArgs)
if err != nil {
return err
}
}
if *since != "" {
v.Set("since", timeutils.GetTimestamp(*since))
}
if *until != "" {
v.Set("until", timeutils.GetTimestamp(*until))
}
if len(eventFilterArgs) > 0 {
filterJSON, err := filters.ToParam(eventFilterArgs)
if err != nil {
return err
}
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
}
-131
View File
@@ -1,131 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"io"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
"github.com/docker/docker/pkg/promise"
"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", "CONTAINER COMMAND [ARG...]", "Run a command in a running container", true)
execConfig, err := runconfig.ParseExec(cmd, args)
// just in case the ParseExec does not exit
if execConfig.Container == "" || err != nil {
return StatusError{StatusCode: 1}
}
stream, _, err := cli.call("POST", "/containers/"+execConfig.Container+"/exec", execConfig, nil)
if err != nil {
return err
}
var response types.ContainerExecCreateResponse
if err := json.NewDecoder(stream).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 {
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()
// fmt.Fprintf(cli.out, "%s\n", execID)
return nil
}
// Interactive exec requested.
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
}
if execConfig.AttachStdout {
out = cli.out
}
if execConfig.AttachStderr {
if execConfig.Tty {
stderr = cli.out
} else {
stderr = cli.err
}
}
errCh = promise.Go(func() error {
return cli.hijack("POST", "/exec/"+execID+"/start", execConfig.Tty, in, out, stderr, hijacked, execConfig)
})
// 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
}
}
if execConfig.Tty && cli.isTerminalIn {
if err := cli.monitorTtySize(execID, true); err != nil {
fmt.Fprintf(cli.err, "Error monitoring TTY size: %s\n", err)
}
}
if err := <-errCh; err != nil {
logrus.Debugf("Error hijack: %s", err)
return err
}
var status int
if _, status, err = getExecExitCode(cli, execID); err != nil {
return err
}
if status != 0 {
return StatusError{StatusCode: status}
}
return nil
}
-46
View File
@@ -1,46 +0,0 @@
package client
import (
"errors"
"io"
"os"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdExport exports a filesystem as a tar archive.
//
// The tar archive is streamed to STDOUT by default or written to a file.
//
// Usage: docker export [OPTIONS] CONTAINER
func (cli *DockerCli) CmdExport(args ...string) error {
cmd := cli.Subcmd("export", "CONTAINER", "Export a filesystem as a tar archive (streamed to STDOUT by default)", true)
outfile := cmd.String([]string{"o", "-output"}, "", "Write to a file, instead of STDOUT")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
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.")
}
image := cmd.Arg(0)
sopts := &streamOpts{
rawTerminal: true,
out: output,
}
if err := cli.stream("GET", "/containers/"+image+"/export", sopts); err != nil {
return err
}
return nil
}
-34
View File
@@ -1,34 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdHelp displays information on a Docker command.
//
// If more than one command is specified, information is only shown for the first command.
//
// Usage: docker help COMMAND or docker COMMAND --help
func (cli *DockerCli) CmdHelp(args ...string) error {
if len(args) > 1 {
method, exists := cli.getMethod(args[:2]...)
if exists {
method("--help")
return nil
}
}
if len(args) > 0 {
method, exists := cli.getMethod(args[0])
if !exists {
return fmt.Errorf("docker: '%s' is not a docker command. See 'docker --help'.", args[0])
}
method("--help")
return nil
}
flag.Usage()
return nil
}
-257
View File
@@ -1,257 +0,0 @@
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/docker/pkg/term"
)
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
}
}
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("/v%s%s", api.APIVERSION, 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)
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 conn, ok := rwc.(interface {
CloseWrite() error
}); ok {
if err := conn.CloseWrite(); err != nil {
logrus.Debugf("Couldn't send EOF: %s", err)
}
}
// Discard errors due to pipe interruption
return nil
})
if stdout != nil || stderr != 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
}
-73
View File
@@ -1,73 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"text/tabwriter"
"time"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/stringutils"
"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", "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{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
rdr, _, err := cli.call("GET", "/images/"+cmd.Arg(0)+"/history", nil, nil)
if err != nil {
return err
}
history := []types.ImageHistory{}
if err := json.NewDecoder(rdr).Decode(&history); err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
if !*quiet {
fmt.Fprintln(w, "IMAGE\tCREATED\tCREATED BY\tSIZE\tCOMMENT")
}
for _, entry := range history {
if *noTrunc {
fmt.Fprintf(w, entry.ID)
} else {
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))
}
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
}
-126
View File
@@ -1,126 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"text/tabwriter"
"time"
"github.com/docker/docker/api/types"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"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", "[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{"#notrunc", "-no-trunc"}, false, "Don't truncate output")
showDigests := cmd.Bool([]string{"-digests"}, false, "Show digests")
flFilter := opts.NewListOpts(nil)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Filter output based on conditions provided")
cmd.Require(flag.Max, 1)
cmd.ParseFlags(args, true)
// Consolidate all filter flags, and sanity check them early.
// They'll get process in the daemon/server.
imageFilterArgs := filters.Args{}
for _, f := range flFilter.GetAll() {
var err error
imageFilterArgs, err = filters.ParseFlag(f, imageFilterArgs)
if err != nil {
return err
}
}
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 {
// FIXME rename this parameter, to not be confused with the filters flag
v.Set("filter", matchName)
}
if *all {
v.Set("all", "1")
}
rdr, _, err := cli.call("GET", "/images/json?"+v.Encode(), nil, nil)
if err != nil {
return err
}
images := []types.Image{}
if err := json.NewDecoder(rdr).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 {
fmt.Fprintln(w, "REPOSITORY\tTAG\tIMAGE ID\tCREATED\tVIRTUAL SIZE")
}
}
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)
}
}
}
if !*quiet {
w.Flush()
}
return nil
}
-64
View File
@@ -1,64 +0,0 @@
package client
import (
"fmt"
"io"
"net/url"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"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.
//
// The URL argument is the address of a tarball (.tar, .tar.gz, .tgz, .bzip, .tar.xz, .txz) file. If the URL is '-', then the tar file is read from STDIN.
//
// Usage: docker import [OPTIONS] URL [REPOSITORY[:TAG]]
func (cli *DockerCli) CmdImport(args ...string) error {
cmd := cli.Subcmd("import", "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")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var (
v = url.Values{}
src = cmd.Arg(0)
repository = cmd.Arg(1)
)
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 'URL|- [REPOSITORY [TAG]]' has been deprecated. Please use URL|- [REPOSITORY[:TAG]]\n")
v.Set("tag", cmd.Arg(2))
}
if repository != "" {
//Check if the given image name can be resolved
repo, _ := parsers.ParseRepositoryTag(repository)
if err := registry.ValidateRepositoryName(repo); err != nil {
return err
}
}
var in io.Reader
if src == "-" {
in = cli.in
}
sopts := &streamOpts{
rawTerminal: true,
in: in,
out: cli.out,
}
return cli.stream("POST", "/images/create?"+v.Encode(), sopts)
}
-96
View File
@@ -1,96 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
"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", "", "Display system-wide information", true)
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, false)
rdr, _, err := cli.call("GET", "/info", nil, nil)
if err != nil {
return err
}
info := &types.Info{}
if err := json.NewDecoder(rdr).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, "Images: %d\n", info.Images)
fmt.Fprintf(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])
}
}
fmt.Fprintf(cli.out, "Execution Driver: %s\n", info.ExecutionDriver)
fmt.Fprintf(cli.out, "Logging Driver: %s\n", info.LoggingDriver)
fmt.Fprintf(cli.out, "Kernel Version: %s\n", info.KernelVersion)
fmt.Fprintf(cli.out, "Operating System: %s\n", info.OperatingSystem)
fmt.Fprintf(cli.out, "CPUs: %d\n", info.NCPU)
fmt.Fprintf(cli.out, "Total Memory: %s\n", units.BytesSize(float64(info.MemTotal)))
fmt.Fprintf(cli.out, "Name: %s\n", info.Name)
fmt.Fprintf(cli.out, "ID: %s\n", info.ID)
if info.Debug {
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)
}
if info.HttpProxy != "" {
fmt.Fprintf(cli.out, "Http Proxy: %s\n", info.HttpProxy)
}
if info.HttpsProxy != "" {
fmt.Fprintf(cli.out, "Https Proxy: %s\n", info.HttpsProxy)
}
if info.NoProxy != "" {
fmt.Fprintf(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)
}
}
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.Labels != nil {
fmt.Fprintln(cli.out, "Labels:")
for _, attribute := range info.Labels {
fmt.Fprintf(cli.out, " %s\n", attribute)
}
}
if info.ExperimentalBuild {
fmt.Fprintf(cli.out, "Experimental: true\n")
}
return nil
}
-121
View File
@@ -1,121 +0,0 @@
package client
import (
"bytes"
"encoding/json"
"fmt"
"io"
"strings"
"text/template"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
)
// 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", "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")
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var tmpl *template.Template
if *tmplStr != "" {
var err error
if tmpl, err = template.New("").Funcs(funcMap).Parse(*tmplStr); err != nil {
return StatusError{StatusCode: 64,
Status: "Template parsing error: " + err.Error()}
}
}
indented := new(bytes.Buffer)
indented.WriteString("[\n")
status := 0
isImage := false
for _, name := range cmd.Args() {
obj, _, err := readBody(cli.call("GET", "/containers/"+name+"/json", nil, nil))
if err != nil {
obj, _, err = readBody(cli.call("GET", "/images/"+name+"/json", nil, nil))
isImage = true
if err != nil {
if strings.Contains(err.Error(), "No such") {
fmt.Fprintf(cli.err, "Error: No such image or container: %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 {
if _, err := io.Copy(cli.out, indented); err != nil {
return err
}
}
if status != 0 {
return StatusError{StatusCode: status}
}
return nil
}
-32
View File
@@ -1,32 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdKill kills one or more running container using SIGKILL or a specified signal.
//
// Usage: docker kill [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdKill(args ...string) error {
cmd := cli.Subcmd("kill", "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 errNames []string
for _, name := range cmd.Args() {
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(errNames) > 0 {
return fmt.Errorf("Error: failed to kill containers: %v", errNames)
}
return nil
}
-41
View File
@@ -1,41 +0,0 @@
package client
import (
"io"
"os"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdLoad loads an image from a tar archive.
//
// The tar archive is read from STDIN by default, or from a tar archive file.
//
// Usage: docker load [OPTIONS]
func (cli *DockerCli) CmdLoad(args ...string) error {
cmd := cli.Subcmd("load", "", "Load an image from a tar archive on STDIN", true)
infile := cmd.String([]string{"i", "-input"}, "", "Read from a tar archive file, instead of STDIN")
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, true)
var (
input io.Reader = cli.in
err error
)
if *infile != "" {
input, err = os.Open(*infile)
if err != nil {
return err
}
}
sopts := &streamOpts{
rawTerminal: true,
in: input,
out: cli.out,
}
if err := cli.stream("POST", "/images/load", sopts); err != nil {
return err
}
return nil
}
-144
View File
@@ -1,144 +0,0 @@
package client
import (
"bufio"
"encoding/json"
"fmt"
"io"
"os"
"strings"
"github.com/docker/docker/api/types"
"github.com/docker/docker/cliconfig"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/term"
"github.com/docker/docker/registry"
)
// 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", "[SERVER]", "Register or log in to a Docker registry server, if no server is\nspecified \""+registry.IndexServerAddress()+"\" is the default.", true)
cmd.Require(flag.Max, 1)
var username, password, email string
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)
serverAddress := registry.IndexServerAddress()
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 {
// 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
stream, statusCode, err := cli.call("POST", "/auth", cli.configFile.AuthConfigs[serverAddress], nil)
if 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
}
var response types.AuthResponse
if err := json.NewDecoder(stream).Decode(&response); err != nil {
// Upon error, remove entry
delete(cli.configFile.AuthConfigs, serverAddress)
return 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.Fprintf(cli.out, "%s\n", response.Status)
}
return nil
}
-36
View File
@@ -1,36 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/registry"
)
// CmdLogout logs a user out from a Docker registry.
//
// If no server is specified, the user will be logged out from the registry's index server.
//
// Usage: docker logout [SERVER]
func (cli *DockerCli) CmdLogout(args ...string) error {
cmd := cli.Subcmd("logout", "[SERVER]", "Log out from a Docker registry, if no server is\nspecified \""+registry.IndexServerAddress()+"\" is the default.", true)
cmd.Require(flag.Max, 1)
cmd.ParseFlags(args, false)
serverAddress := registry.IndexServerAddress()
if len(cmd.Args()) > 0 {
serverAddress = cmd.Arg(0)
}
if _, ok := cli.configFile.AuthConfigs[serverAddress]; !ok {
fmt.Fprintf(cli.out, "Not logged in to %s\n", serverAddress)
} else {
fmt.Fprintf(cli.out, "Remove login credentials for %s\n", serverAddress)
delete(cli.configFile.AuthConfigs, serverAddress)
if err := cli.configFile.Save(); err != nil {
return fmt.Errorf("Failed to save docker config: %v", err)
}
}
return nil
}
-68
View File
@@ -1,68 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/timeutils"
)
// CmdLogs fetches the logs of a given container.
//
// docker logs [OPTIONS] CONTAINER
func (cli *DockerCli) CmdLogs(args ...string) error {
var (
cmd = cli.Subcmd("logs", "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")
tail = cmd.String([]string{"-tail"}, "all", "Number of lines to show from the end of the logs")
)
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
name := cmd.Arg(0)
stream, _, err := cli.call("GET", "/containers/"+name+"/json", nil, nil)
if err != nil {
return err
}
var c types.ContainerJSON
if err := json.NewDecoder(stream).Decode(&c); err != nil {
return err
}
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))
}
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,
}
return cli.stream("GET", "/containers/"+name+"/logs?"+v.Encode(), sopts)
}
-30
View File
@@ -1,30 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdPause pauses all processes within one or more containers.
//
// Usage: docker pause CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdPause(args ...string) error {
cmd := cli.Subcmd("pause", "CONTAINER [CONTAINER...]", "Pause all processes within a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, false)
var errNames []string
for _, name := range cmd.Args() {
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(errNames) > 0 {
return fmt.Errorf("Error: failed to pause containers: %v", errNames)
}
return nil
}
-64
View File
@@ -1,64 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"strings"
"github.com/docker/docker/nat"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdPort lists port mappings for a container.
// If a private port is specified, it also shows the public-facing port that is NATed to the private port.
//
// Usage: docker port CONTAINER [PRIVATE_PORT[/PROTO]]
func (cli *DockerCli) CmdPort(args ...string) error {
cmd := cli.Subcmd("port", "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)
stream, _, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/json", nil, nil)
if err != nil {
return err
}
var c struct {
NetworkSettings struct {
Ports nat.PortMap
}
}
if err := json.NewDecoder(stream).Decode(&c); err != nil {
return err
}
if cmd.NArg() == 2 {
var (
port = cmd.Arg(1)
proto = "tcp"
parts = strings.SplitN(port, "/", 2)
)
if len(parts) == 2 && len(parts[1]) != 0 {
port = parts[0]
proto = parts[1]
}
natPort := port + "/" + proto
if frontends, exists := c.NetworkSettings.Ports[nat.Port(port+"/"+proto)]; exists && frontends != nil {
for _, frontend := range frontends {
fmt.Fprintf(cli.out, "%s:%s\n", frontend.HostIp, frontend.HostPort)
}
return nil
}
return fmt.Errorf("Error: No public port '%s' published for %s", natPort, cmd.Arg(0))
}
for from, frontends := range c.NetworkSettings.Ports {
for _, frontend := range frontends {
fmt.Fprintf(cli.out, "%s -> %s:%s\n", from, frontend.HostIp, frontend.HostPort)
}
}
return nil
}
-175
View File
@@ -1,175 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"strconv"
"strings"
"text/tabwriter"
"time"
"github.com/docker/docker/api"
"github.com/docker/docker/api/types"
"github.com/docker/docker/opts"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers/filters"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/stringutils"
"github.com/docker/docker/pkg/units"
)
// CmdPs outputs a list of Docker containers.
//
// Usage: docker ps [OPTIONS]
func (cli *DockerCli) CmdPs(args ...string) error {
var (
err error
psFilterArgs = filters.Args{}
v = url.Values{}
cmd = cli.Subcmd("ps", "", "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{"#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")
flFilter = opts.NewListOpts(nil)
)
cmd.Require(flag.Exact, 0)
cmd.Var(&flFilter, []string{"f", "-filter"}, "Filter output based on conditions provided")
cmd.ParseFlags(args, true)
if *last == -1 && *nLatest {
*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() {
if psFilterArgs, err = filters.ParseFlag(f, psFilterArgs); err != nil {
return err
}
}
if len(psFilterArgs) > 0 {
filterJSON, err := filters.ToParam(psFilterArgs)
if err != nil {
return err
}
v.Set("filters", filterJSON)
}
rdr, _, err := cli.call("GET", "/containers/json?"+v.Encode(), nil, nil)
if err != nil {
return err
}
containers := []types.Container{}
if err := json.NewDecoder(rdr).Decode(&containers); err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
if !*quiet {
fmt.Fprint(w, "CONTAINER ID\tIMAGE\tCOMMAND\tCREATED\tSTATUS\tPORTS\tNAMES")
if *size {
fmt.Fprintln(w, "\tSIZE")
} else {
fmt.Fprint(w, "\n")
}
}
stripNamePrefix := func(ss []string) []string {
for i, s := range ss {
ss[i] = s[1:]
}
return ss
}
for _, container := range containers {
ID := container.ID
if !*noTrunc {
ID = stringid.TruncateID(ID)
}
if *quiet {
fmt.Fprintln(w, ID)
continue
}
var (
names = stripNamePrefix(container.Names)
command = strconv.Quote(container.Command)
)
if !*noTrunc {
command = stringutils.Truncate(command, 20)
// only display the default name for the container with notrunc is passed
for _, name := range names {
if len(strings.Split(name, "/")) == 1 {
names = []string{name}
break
}
}
}
image := container.Image
if image == "" {
image = "<no image>"
}
fmt.Fprintf(w, "%s\t%s\t%s\t%s ago\t%s\t%s\t%s\t", ID, image, command,
units.HumanDuration(time.Now().UTC().Sub(time.Unix(int64(container.Created), 0))),
container.Status, api.DisplayablePorts(container.Ports), strings.Join(names, ","))
if *size {
if container.SizeRootFs > 0 {
fmt.Fprintf(w, "%s (virtual %s)\n", units.HumanSize(float64(container.SizeRw)), units.HumanSize(float64(container.SizeRootFs)))
} else {
fmt.Fprintf(w, "%s\n", units.HumanSize(float64(container.SizeRw)))
}
continue
}
fmt.Fprint(w, "\n")
}
if !*quiet {
w.Flush()
}
return nil
}
-47
View File
@@ -1,47 +0,0 @@
package client
import (
"fmt"
"net/url"
"github.com/docker/docker/graph/tags"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
"github.com/docker/docker/utils"
)
// 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", "NAME[:TAG|@DIGEST]", "Pull an image or a repository from the registry", true)
allTags := cmd.Bool([]string{"a", "-all-tags"}, false, "Download all tagged images in the repository")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
var (
v = url.Values{}
remote = cmd.Arg(0)
newRemote = remote
)
taglessRemote, tag := parsers.ParseRepositoryTag(remote)
if tag == "" && !*allTags {
newRemote = utils.ImageReference(taglessRemote, tags.DEFAULTTAG)
}
if tag != "" && *allTags {
return fmt.Errorf("tag can't be used with --all-tags/-a")
}
v.Set("fromImage", newRemote)
// Resolve the Repository name from fqn to RepositoryInfo
repoInfo, err := registry.ParseRepositoryInfo(taglessRemote)
if err != nil {
return err
}
_, _, err = cli.clientRequestAttemptLogin("POST", "/images/create?"+v.Encode(), nil, cli.out, repoInfo.Index, "pull")
return err
}
-49
View File
@@ -1,49 +0,0 @@
package client
import (
"fmt"
"net/url"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/registry"
)
// 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", "NAME[:TAG]", "Push an image or a repository to the registry", true)
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
name := cmd.Arg(0)
remote, tag := parsers.ParseRepositoryTag(name)
// Resolve the Repository name from fqn to RepositoryInfo
repoInfo, err := registry.ParseRepositoryInfo(remote)
if err != nil {
return err
}
// Resolve the Auth config relevant for this server
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)
}
v := url.Values{}
v.Set("tag", tag)
_, _, err = cli.clientRequestAttemptLogin("POST", "/images/"+remote+"/push?"+v.Encode(), nil, cli.out, repoInfo.Index, "push")
return err
}
-25
View File
@@ -1,25 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdRename renames a container.
//
// Usage: docker rename OLD_NAME NEW_NAME
func (cli *DockerCli) CmdRename(args ...string) error {
cmd := cli.Subcmd("rename", "OLD_NAME NEW_NAME", "Rename a container", true)
cmd.Require(flag.Exact, 2)
cmd.ParseFlags(args, true)
oldName := cmd.Arg(0)
newName := cmd.Arg(1)
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)
}
return nil
}
-38
View File
@@ -1,38 +0,0 @@
package client
import (
"fmt"
"net/url"
"strconv"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdRestart restarts one or more running containers.
//
// Usage: docker stop [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdRestart(args ...string) error {
cmd := cli.Subcmd("restart", "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)
v := url.Values{}
v.Set("t", strconv.Itoa(*nSeconds))
var errNames []string
for _, name := range cmd.Args() {
_, _, 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(errNames) > 0 {
return fmt.Errorf("Error: failed to restart containers: %v", errNames)
}
return nil
}
-54
View File
@@ -1,54 +0,0 @@
package client
import (
"fmt"
"net/url"
"strings"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdRm removes one or more containers.
//
// Usage: docker rm [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdRm(args ...string) error {
cmd := cli.Subcmd("rm", "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", "-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)
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, "/")
_, _, 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(errNames) > 0 {
return fmt.Errorf("Error: failed to remove containers: %v", errNames)
}
return nil
}
-59
View File
@@ -1,59 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdRmi removes all images with the specified name(s).
//
// Usage: docker rmi [OPTIONS] IMAGE [IMAGE...]
func (cli *DockerCli) CmdRmi(args ...string) error {
var (
cmd = cli.Subcmd("rmi", "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)
cmd.ParseFlags(args, true)
v := url.Values{}
if *force {
v.Set("force", "1")
}
if *noprune {
v.Set("noprune", "1")
}
var errNames []string
for _, name := range cmd.Args() {
rdr, _, err := cli.call("DELETE", "/images/"+name+"?"+v.Encode(), nil, nil)
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
dels := []types.ImageDelete{}
if err := json.NewDecoder(rdr).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)
} else {
fmt.Fprintf(cli.out, "Untagged: %s\n", del.Untagged)
}
}
}
}
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to remove images: %v", errNames)
}
return nil
}
-246
View File
@@ -1,246 +0,0 @@
package client
import (
"fmt"
"io"
"net/url"
"os"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/signal"
"github.com/docker/docker/runconfig"
"github.com/docker/libnetwork/resolvconf/dns"
)
func (cid *cidFile) Close() error {
cid.file.Close()
if !cid.written {
if err := os.Remove(cid.path); err != nil {
return fmt.Errorf("failed to remove the CID file '%s': %s \n", cid.path, err)
}
}
return nil
}
func (cid *cidFile) Write(id string) error {
if _, err := cid.file.Write([]byte(id)); err != nil {
return fmt.Errorf("Failed to write the container ID to the file: %s", err)
}
cid.written = true
return nil
}
// 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", "IMAGE [COMMAND] [ARG...]", "Run a command in a new container", true)
// These are flags not stored in Config/HostConfig
var (
flAutoRemove = cmd.Bool([]string{"-rm"}, false, "Automatically remove the container when it exits")
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")
flAttach *opts.ListOpts
ErrConflictAttachDetach = fmt.Errorf("Conflicting options: -a and -d")
ErrConflictRestartPolicyAndAutoRemove = fmt.Errorf("Conflicting options: --restart and --rm")
ErrConflictDetachAutoRemove = fmt.Errorf("Conflicting options: --rm and -d")
)
config, hostConfig, cmd, err := runconfig.Parse(cmd, args)
// just in case the Parse does not exit
if err != nil {
cmd.ReportError(err.Error(), true)
}
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 {
if dns.IsLocalhost(dnsIP) {
fmt.Fprintf(cli.err, "WARNING: Localhost DNS setting (--dns=%s) may fail in containers.\n", dnsIP)
break
}
}
}
if config.Image == "" {
cmd.Usage()
return nil
}
if !*flDetach {
if err := cli.CheckTtyInput(config.AttachStdin, config.Tty); err != nil {
return err
}
} else {
if fl := cmd.Lookup("-attach"); fl != nil {
flAttach = fl.Value.(*opts.ListOpts)
if flAttach.Len() != 0 {
return ErrConflictAttachDetach
}
}
if *flAutoRemove {
return ErrConflictDetachAutoRemove
}
config.AttachStdin = false
config.AttachStdout = false
config.AttachStderr = false
config.StdinOnce = false
}
// Disable flSigProxy when in TTY mode
sigProxy := *flSigProxy
if config.Tty {
sigProxy = false
}
createResponse, err := cli.createContainer(config, hostConfig, hostConfig.ContainerIDFile, *flName)
if err != nil {
return err
}
if sigProxy {
sigc := cli.forwardAllSignals(createResponse.ID)
defer signal.StopCatch(sigc)
}
var (
waitDisplayID chan struct{}
errCh chan error
)
if !config.AttachStdout && !config.AttachStderr {
// Make this asynchronous to allow the client to write to stdin before having to read the ID
waitDisplayID = make(chan struct{})
go func() {
defer close(waitDisplayID)
fmt.Fprintf(cli.out, "%s\n", createResponse.ID)
}()
}
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
}
}
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()
}
case err := <-errCh:
if err != nil {
logrus.Debugf("Error hijack: %s", err)
return 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 = 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 {
if err := cli.monitorTtySize(createResponse.ID, false); err != nil {
fmt.Fprintf(cli.err, "Error monitoring TTY size: %s\n", err)
}
}
if errCh != nil {
if err := <-errCh; err != nil {
logrus.Debugf("Error hijack: %s", err)
return err
}
}
// Detached mode: wait for the id to be displayed and return.
if !config.AttachStdout && !config.AttachStderr {
// Detached mode
<-waitDisplayID
return nil
}
var status int
// Attached mode
if *flAutoRemove {
// Autoremove: wait for the container to finish, retrieve
// the exit code and remove the container
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
}
} else {
// 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 = waitForExit(cli, createResponse.ID); err != nil {
return err
}
} else {
// In TTY mode, there is a race: if the process dies too slowly, the state could
// be updated after the getExitCode call and result in the wrong exit code being reported
if _, status, err = getExitCode(cli, createResponse.ID); err != nil {
return err
}
}
}
if status != 0 {
return StatusError{StatusCode: status}
}
return nil
}
-57
View File
@@ -1,57 +0,0 @@
package client
import (
"errors"
"io"
"net/url"
"os"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdSave saves one or more images to a tar archive.
//
// The tar archive is written to STDOUT by default, or written to a file.
//
// Usage: docker save [OPTIONS] IMAGE [IMAGE...]
func (cli *DockerCli) CmdSave(args ...string) error {
cmd := cli.Subcmd("save", "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)
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.")
}
sopts := &streamOpts{
rawTerminal: true,
out: output,
}
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
}
-84
View File
@@ -1,84 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"sort"
"strings"
"text/tabwriter"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/stringutils"
"github.com/docker/docker/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", "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", "-stars"}, 0, "Only displays with at least x stars")
cmd.Require(flag.Exact, 1)
cmd.ParseFlags(args, true)
name := cmd.Arg(0)
v := url.Values{}
v.Set("term", name)
// Resolve the Repository name from fqn to hostname + name
taglessRemote, _ := parsers.ParseRepositoryTag(name)
repoInfo, err := registry.ParseRepositoryInfo(taglessRemote)
if err != nil {
return err
}
rdr, _, err := cli.clientRequestAttemptLogin("GET", "/images/search?"+v.Encode(), nil, nil, repoInfo.Index, "search")
if err != nil {
return err
}
results := ByStars{}
if err := json.NewDecoder(rdr).Decode(&results); err != nil {
return err
}
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 || *trusted) && (!res.IsTrusted && !res.IsAutomated)) || (int(*stars) > res.StarCount) {
continue
}
desc := strings.Replace(res.Description, "\n", " ", -1)
desc = strings.Replace(desc, "\r", " ", -1)
if !*noTrunc && len(desc) > 45 {
desc = stringutils.Truncate(desc, 42) + "..."
}
fmt.Fprintf(w, "%s\t%s\t%d\t", res.Name, desc, res.StarCount)
if res.IsOfficial {
fmt.Fprint(w, "[OK]")
}
fmt.Fprint(w, "\t")
if res.IsAutomated || res.IsTrusted {
fmt.Fprint(w, "[OK]")
}
fmt.Fprint(w, "\n")
}
w.Flush()
return nil
}
-167
View File
@@ -1,167 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"io"
"net/url"
"os"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/signal"
)
func (cli *DockerCli) forwardAllSignals(cid string) chan os.Signal {
sigc := make(chan os.Signal, 128)
signal.CatchAll(sigc)
go func() {
for s := range sigc {
if s == signal.SIGCHLD {
continue
}
var sig string
for sigStr, sigN := range signal.SignalMap {
if sigN == s {
sig = sigStr
break
}
}
if sig == "" {
fmt.Fprintf(cli.err, "Unsupported signal: %v. Discarding.\n", s)
}
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)
}
}
}()
return sigc
}
// CmdStart starts one or more stopped containers.
//
// Usage: docker start [OPTIONS] CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdStart(args ...string) error {
var (
cErr chan error
tty bool
cmd = cli.Subcmd("start", "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")
)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
if *attach || *openStdin {
if cmd.NArg() > 1 {
return fmt.Errorf("You cannot start and attach multiple containers at once.")
}
stream, _, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/json", nil, nil)
if err != nil {
return err
}
var c types.ContainerJSON
if err := json.NewDecoder(stream).Decode(&c); err != nil {
return err
}
tty = c.Config.Tty
if !tty {
sigc := cli.forwardAllSignals(cmd.Arg(0))
defer signal.StopCatch(sigc)
}
var in io.ReadCloser
v := url.Values{}
v.Set("stream", "1")
if *openStdin && c.Config.OpenStdin {
v.Set("stdin", "1")
in = cli.in
}
v.Set("stdout", "1")
v.Set("stderr", "1")
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)
})
// 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
}
}
}
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, cmd.Arg(0))
if err != nil {
return err
}
if status != 0 {
return StatusError{StatusCode: status}
}
}
return nil
}
-202
View File
@@ -1,202 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"io"
"net/url"
"sort"
"strings"
"sync"
"text/tabwriter"
"time"
"github.com/docker/docker/api/types"
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")
}
stream, _, 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 stream.Close()
var (
previousCPU uint64
previousSystem uint64
start = true
dec = json.NewDecoder(stream)
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
)
if !start {
cpuPercent = calculateCPUPercent(previousCPU, previousSystem, v)
}
start = false
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()
previousCPU = v.CpuStats.CpuUsage.TotalUsage
previousSystem = v.CpuStats.SystemUsage
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 CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdStats(args ...string) error {
cmd := cli.Subcmd("stats", "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()
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\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(500 * 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{}
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 = append(cStats[:i], cStats[i+1:]...)
}
if len(cStats) == 0 {
return nil
}
w.Flush()
if *noStream {
break
}
}
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
}
-29
View File
@@ -1,29 +0,0 @@
package client
import (
"bytes"
"sync"
"testing"
)
func TestDisplay(t *testing.T) {
c := &containerStats{
Name: "app",
CPUPercentage: 30.0,
Memory: 100 * 1024 * 1024.0,
MemoryLimit: 2048 * 1024 * 1024.0,
MemoryPercentage: 100.0 / 2048.0 * 100.0,
NetworkRx: 100 * 1024 * 1024,
NetworkTx: 800 * 1024 * 1024,
mu: sync.RWMutex{},
}
var b bytes.Buffer
if err := c.Display(&b); err != nil {
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\n"
if got != want {
t.Fatalf("c.Display() = %q, want %q", got, want)
}
}
-40
View File
@@ -1,40 +0,0 @@
package client
import (
"fmt"
"net/url"
"strconv"
flag "github.com/docker/docker/pkg/mflag"
)
// 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", "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)
v := url.Values{}
v.Set("t", strconv.Itoa(*nSeconds))
var errNames []string
for _, name := range cmd.Args() {
_, _, 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(errNames) > 0 {
return fmt.Errorf("Error: failed to stop containers: %v", errNames)
}
return nil
}
-41
View File
@@ -1,41 +0,0 @@
package client
import (
"net/url"
flag "github.com/docker/docker/pkg/mflag"
"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", "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)
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 *force {
v.Set("force", "1")
}
if _, _, err := readBody(cli.call("POST", "/images/"+cmd.Arg(0)+"/tag?"+v.Encode(), nil, nil)); err != nil {
return err
}
return nil
}
-46
View File
@@ -1,46 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"net/url"
"strings"
"text/tabwriter"
"github.com/docker/docker/api/types"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdTop displays the running processes of a container.
//
// Usage: docker top CONTAINER
func (cli *DockerCli) CmdTop(args ...string) error {
cmd := cli.Subcmd("top", "CONTAINER [ps OPTIONS]", "Display the running processes of a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
val := url.Values{}
if cmd.NArg() > 1 {
val.Set("ps_args", strings.Join(cmd.Args()[1:], " "))
}
stream, _, err := cli.call("GET", "/containers/"+cmd.Arg(0)+"/top?"+val.Encode(), nil, nil)
if err != nil {
return err
}
procList := types.ContainerProcessList{}
if err := json.NewDecoder(stream).Decode(&procList); err != nil {
return err
}
w := tabwriter.NewWriter(cli.out, 20, 1, 3, ' ', 0)
fmt.Fprintln(w, strings.Join(procList.Titles, "\t"))
for _, proc := range procList.Processes {
fmt.Fprintln(w, strings.Join(proc, "\t"))
}
w.Flush()
return nil
}
-30
View File
@@ -1,30 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdUnpause unpauses all processes within a container, for one or more containers.
//
// Usage: docker unpause CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdUnpause(args ...string) error {
cmd := cli.Subcmd("unpause", "CONTAINER [CONTAINER...]", "Unpause all processes within a container", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, false)
var errNames []string
for _, name := range cmd.Args() {
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(errNames) > 0 {
return fmt.Errorf("Error: failed to unpause containers: %v", errNames)
}
return nil
}
-345
View File
@@ -1,345 +0,0 @@
package client
import (
"bytes"
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"io"
"io/ioutil"
"net/http"
"net/url"
"os"
gosignal "os/signal"
"runtime"
"strconv"
"strings"
"time"
"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"
)
var (
errConnectionRefused = errors.New("Cannot connect to the Docker daemon. Is 'docker -d' running on this host?")
)
// HTTPClient creates a new HTP client with the cli's client transport instance.
func (cli *DockerCli) HTTPClient() *http.Client {
return &http.Client{Transport: cli.transport}
}
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 params, nil
}
func (cli *DockerCli) clientRequest(method, path string, in io.Reader, headers map[string][]string) (io.ReadCloser, string, int, error) {
expectedPayload := (method == "POST" || method == "PUT")
if expectedPayload && in == nil {
in = bytes.NewReader([]byte{})
}
req, err := http.NewRequest(method, fmt.Sprintf("/v%s%s", api.APIVERSION, path), in)
if err != nil {
return nil, "", -1, 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)
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)
statusCode := -1
if resp != nil {
statusCode = resp.StatusCode
}
if err != nil {
if strings.Contains(err.Error(), "connection refused") {
return nil, "", statusCode, errConnectionRefused
}
if cli.tlsConfig == nil {
return nil, "", statusCode, fmt.Errorf("%v. Are you trying to connect to a TLS-enabled daemon without TLS?", err)
}
return nil, "", statusCode, fmt.Errorf("An error occurred trying to connect: %v", err)
}
if statusCode < 200 || statusCode >= 400 {
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return nil, "", statusCode, err
}
if len(body) == 0 {
return nil, "", statusCode, fmt.Errorf("Error: request returned %s for API route and version %s, check if the server supports the requested API version", http.StatusText(statusCode), req.URL)
}
return nil, "", statusCode, fmt.Errorf("Error response from daemon: %s", bytes.TrimSpace(body))
}
return resp.Body, resp.Header.Get("Content-Type"), statusCode, 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
body, contentType, statusCode, 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(body, contentType, 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(), "status code 401") {
statusCode = http.StatusUnauthorized
}
}
return body, 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) call(method, path string, data interface{}, headers map[string][]string) (io.ReadCloser, int, error) {
params, err := cli.encodeData(data)
if err != nil {
return nil, -1, err
}
if data != nil {
if headers == nil {
headers = make(map[string][]string)
}
headers["Content-Type"] = []string{"application/json"}
}
body, _, statusCode, err := cli.clientRequest(method, path, params, headers)
return body, statusCode, 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) error {
body, contentType, _, err := cli.clientRequest(method, path, opts.in, opts.headers)
if err != nil {
return err
}
return cli.streamBody(body, contentType, 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()
if height == 0 && width == 0 {
return
}
v := url.Values{}
v.Set("h", strconv.Itoa(height))
v.Set("w", strconv.Itoa(width))
path := ""
if !isExec {
path = "/containers/" + id + "/resize?"
} else {
path = "/exec/" + id + "/resize?"
}
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) {
stream, _, err := cli.call("POST", "/containers/"+containerID+"/wait", nil, nil)
if err != nil {
return -1, err
}
var res types.ContainerWaitResponse
if err := json.NewDecoder(stream).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) {
stream, _, err := cli.call("GET", "/containers/"+containerID+"/json", nil, nil)
if err != nil {
// If we can't connect, then the daemon probably died.
if err != errConnectionRefused {
return false, -1, err
}
return false, -1, nil
}
var c types.ContainerJSON
if err := json.NewDecoder(stream).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) {
stream, _, err := cli.call("GET", "/exec/"+execID+"/json", nil, nil)
if err != nil {
// If we can't connect, then the daemon probably died.
if err != errConnectionRefused {
return false, -1, err
}
return false, -1, nil
}
//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(stream).Decode(&c); err != nil {
return false, -1, err
}
return c.Running, c.ExitCode, nil
}
func (cli *DockerCli) monitorTtySize(id string, isExec bool) error {
cli.resizeTty(id, isExec)
if runtime.GOOS == "windows" {
go func() {
prevH, prevW := cli.getTtySize()
for {
time.Sleep(time.Millisecond * 250)
h, w := cli.getTtySize()
if prevW != w || prevH != h {
cli.resizeTty(id, isExec)
}
prevH = h
prevW = w
}
}()
} else {
sigchan := make(chan os.Signal, 1)
gosignal.Notify(sigchan, signal.SIGWINCH)
go func() {
for range sigchan {
cli.resizeTty(id, isExec)
}
}()
}
return nil
}
func (cli *DockerCli) getTtySize() (int, int) {
if !cli.isTerminalOut {
return 0, 0
}
ws, err := term.GetWinsize(cli.outFd)
if err != nil {
logrus.Debugf("Error getting size: %s", err)
if ws == nil {
return 0, 0
}
}
return int(ws.Height), int(ws.Width)
}
func readBody(stream io.ReadCloser, statusCode int, err error) ([]byte, int, error) {
if stream != nil {
defer stream.Close()
}
if err != nil {
return nil, statusCode, err
}
body, err := ioutil.ReadAll(stream)
if err != nil {
return nil, -1, err
}
return body, statusCode, nil
}
-55
View File
@@ -1,55 +0,0 @@
package client
import (
"encoding/json"
"fmt"
"runtime"
"github.com/docker/docker/api"
"github.com/docker/docker/api/types"
"github.com/docker/docker/autogen/dockerversion"
flag "github.com/docker/docker/pkg/mflag"
)
// 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) error {
cmd := cli.Subcmd("version", "", "Show the Docker version information.", true)
cmd.Require(flag.Exact, 0)
cmd.ParseFlags(args, false)
if dockerversion.VERSION != "" {
fmt.Fprintf(cli.out, "Client version: %s\n", dockerversion.VERSION)
}
fmt.Fprintf(cli.out, "Client API version: %s\n", api.APIVERSION)
fmt.Fprintf(cli.out, "Go version (client): %s\n", runtime.Version())
if dockerversion.GITCOMMIT != "" {
fmt.Fprintf(cli.out, "Git commit (client): %s\n", dockerversion.GITCOMMIT)
}
fmt.Fprintf(cli.out, "OS/Arch (client): %s/%s\n", runtime.GOOS, runtime.GOARCH)
stream, _, err := cli.call("GET", "/version", nil, nil)
if err != nil {
return err
}
var v types.Version
if err := json.NewDecoder(stream).Decode(&v); err != nil {
fmt.Fprintf(cli.err, "Error reading remote version: %s\n", err)
return err
}
fmt.Fprintf(cli.out, "Server version: %s\n", v.Version)
if v.ApiVersion != "" {
fmt.Fprintf(cli.out, "Server API version: %s\n", v.ApiVersion)
}
fmt.Fprintf(cli.out, "Go version (server): %s\n", v.GoVersion)
fmt.Fprintf(cli.out, "Git commit (server): %s\n", v.GitCommit)
fmt.Fprintf(cli.out, "OS/Arch (server): %s/%s\n", v.Os, v.Arch)
return nil
}
-34
View File
@@ -1,34 +0,0 @@
package client
import (
"fmt"
flag "github.com/docker/docker/pkg/mflag"
)
// CmdWait blocks until a container stops, then prints its exit code.
//
// If more than one container is specified, this will wait synchronously on each container.
//
// Usage: docker wait CONTAINER [CONTAINER...]
func (cli *DockerCli) CmdWait(args ...string) error {
cmd := cli.Subcmd("wait", "CONTAINER [CONTAINER...]", "Block until a container stops, then print its exit code.", true)
cmd.Require(flag.Min, 1)
cmd.ParseFlags(args, true)
var errNames []string
for _, name := range cmd.Args() {
status, err := waitForExit(cli, name)
if err != nil {
fmt.Fprintf(cli.err, "%s\n", err)
errNames = append(errNames, name)
} else {
fmt.Fprintf(cli.out, "%d\n", status)
}
}
if len(errNames) > 0 {
return fmt.Errorf("Error: failed to wait containers: %v", errNames)
}
return nil
}
+21 -104
View File
@@ -2,126 +2,43 @@ package api
import (
"fmt"
"github.com/dotcloud/docker/engine"
"github.com/dotcloud/docker/utils"
"mime"
"path/filepath"
"sort"
"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/libtrust"
)
// Common constants for daemon and client.
const (
APIVERSION version.Version = "1.19" // Current REST API version
DefaultDockerfileName string = "Dockerfile" // Default filename with Docker commands, read by docker build
APIVERSION = "1.10"
DEFAULTHTTPHOST = "127.0.0.1"
DEFAULTUNIXSOCKET = "/var/run/docker.sock"
)
type ByPrivatePort []types.Port
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 }
func DisplayablePorts(ports []types.Port) string {
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 {
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))
continue
}
portKey = fmt.Sprintf("%s/%s", port.IP, port.Type)
}
firstInGroup = firstInGroupMap[portKey]
lastInGroup = lastInGroupMap[portKey]
if firstInGroup == 0 {
firstInGroupMap[portKey] = current
lastInGroupMap[portKey] = current
continue
}
if current == (lastInGroup + 1) {
lastInGroupMap[portKey] = current
continue
}
result = append(result, FormGroup(portKey, firstInGroup, lastInGroup))
firstInGroupMap[portKey] = current
lastInGroupMap[portKey] = current
func ValidateHost(val string) (string, error) {
host, err := utils.ParseHost(DEFAULTHTTPHOST, DEFAULTUNIXSOCKET, val)
if err != nil {
return val, err
}
for portKey, firstInGroup := range firstInGroupMap {
result = append(result, FormGroup(portKey, firstInGroup, lastInGroupMap[portKey]))
}
result = append(result, hostMappings...)
return strings.Join(result, ", ")
return host, nil
}
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]
//TODO remove, used on < 1.5 in getContainersJSON
func displayablePorts(ports *engine.Table) string {
result := []string{}
for _, port := range ports.Data {
if port.Get("IP") == "" {
result = append(result, fmt.Sprintf("%d/%s", port.GetInt("PublicPort"), port.Get("Type")))
} else {
result = append(result, fmt.Sprintf("%s:%d->%d/%s", port.Get("IP"), port.GetInt("PublicPort"), port.GetInt("PrivatePort"), port.Get("Type")))
}
}
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)
}
return fmt.Sprintf("%s/%s", group, groupType)
return strings.Join(result, ", ")
}
func MatchesContentType(contentType, expectedType string) bool {
mimetype, _, err := mime.ParseMediaType(contentType)
if err != nil {
logrus.Errorf("Error parsing media type: %s error: %v", contentType, err)
utils.Errorf("Error parsing media type: %s error: %s", contentType, err.Error())
}
return err == nil && mimetype == expectedType
}
// LoadOrCreateTrustKey attempts to load the libtrust key at the given path,
// otherwise generates a new one
func LoadOrCreateTrustKey(trustKeyPath string) (libtrust.PrivateKey, error) {
err := system.MkdirAll(filepath.Dir(trustKeyPath), 0700)
if err != nil {
return nil, err
}
trustKey, err := libtrust.LoadKeyFile(trustKeyPath)
if err == libtrust.ErrKeyFileDoesNotExist {
trustKey, err = libtrust.GenerateECP256PrivateKey()
if err != nil {
return nil, fmt.Errorf("Error generating key: %s", err)
}
if err := libtrust.SaveKey(trustKeyPath, trustKey); err != nil {
return nil, fmt.Errorf("Error saving key file: %s", err)
}
} else if err != nil {
return nil, fmt.Errorf("Error loading key file %s: %s", trustKeyPath, err)
}
return trustKey, nil
}
+18
View File
@@ -0,0 +1,18 @@
package api
import (
"github.com/dotcloud/docker/nat"
"github.com/dotcloud/docker/runconfig"
)
type Container struct {
Config runconfig.Config
HostConfig runconfig.HostConfig
State struct {
Running bool
ExitCode int
}
NetworkSettings struct {
Ports nat.PortMap
}
}
+1222
View File
File diff suppressed because it is too large Load Diff
-20
View File
@@ -1,20 +0,0 @@
package server
import (
"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")
}
func int64ValueOrZero(r *http.Request, k string) int64 {
val, err := strconv.ParseInt(r.FormValue(k), 10, 64)
if err != nil {
return 0
}
return val
}
-55
View File
@@ -1,55 +0,0 @@
package server
import (
"net/http"
"net/url"
"testing"
)
func TestBoolValue(t *testing.T) {
cases := map[string]bool{
"": false,
"0": false,
"no": false,
"false": false,
"none": false,
"1": true,
"yes": true,
"true": true,
"one": true,
"100": true,
}
for c, e := range cases {
v := url.Values{}
v.Set("test", c)
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
a := boolValue(r, "test")
if a != e {
t.Fatalf("Value: %s, expected: %v, actual: %v", c, e, a)
}
}
}
func TestInt64ValueOrZero(t *testing.T) {
cases := map[string]int64{
"": 0,
"asdf": 0,
"0": 0,
"1": 1,
}
for c, e := range cases {
v := url.Values{}
v.Set("test", c)
r, _ := http.NewRequest("POST", "", nil)
r.Form = v
a := int64ValueOrZero(r, "test")
if a != e {
t.Fatalf("Value: %s, expected: %v, actual: %v", c, e, a)
}
}
}
-38
View File
@@ -1,38 +0,0 @@
package server
import (
"expvar"
"fmt"
"net/http"
"net/http/pprof"
"github.com/gorilla/mux"
)
func ProfilerSetup(mainRouter *mux.Router, path string) {
var r = mainRouter.PathPrefix(path).Subrouter()
r.HandleFunc("/vars", expVars)
r.HandleFunc("/pprof/", pprof.Index)
r.HandleFunc("/pprof/cmdline", pprof.Cmdline)
r.HandleFunc("/pprof/profile", pprof.Profile)
r.HandleFunc("/pprof/symbol", pprof.Symbol)
r.HandleFunc("/pprof/block", pprof.Handler("block").ServeHTTP)
r.HandleFunc("/pprof/heap", pprof.Handler("heap").ServeHTTP)
r.HandleFunc("/pprof/goroutine", pprof.Handler("goroutine").ServeHTTP)
r.HandleFunc("/pprof/threadcreate", pprof.Handler("threadcreate").ServeHTTP)
}
// Replicated from expvar.go as not public.
func expVars(w http.ResponseWriter, r *http.Request) {
first := true
w.Header().Set("Content-Type", "application/json; charset=utf-8")
fmt.Fprintf(w, "{\n")
expvar.Do(func(kv expvar.KeyValue) {
if !first {
fmt.Fprintf(w, ",\n")
}
first = false
fmt.Fprintf(w, "%q: %s", kv.Key, kv.Value)
})
fmt.Fprintf(w, "\n}\n")
}
-1575
View File
File diff suppressed because it is too large Load Diff
-78
View File
@@ -1,78 +0,0 @@
// +build linux
package server
import (
"fmt"
"net"
"net/http"
"github.com/docker/docker/daemon"
"github.com/docker/docker/pkg/sockets"
"github.com/docker/docker/pkg/systemd"
)
// newServer sets up the required serverCloser and does protocol specific checking.
func (s *Server) newServer(proto, addr string) (serverCloser, error) {
var (
err error
l net.Listener
)
switch proto {
case "fd":
ls, err := systemd.ListenFD(addr)
if err != nil {
return nil, err
}
chErrors := make(chan error, len(ls))
// We don't want to start serving on these sockets until the
// daemon is initialized and installed. Otherwise required handlers
// won't be ready.
<-s.start
// Since ListenFD will return one or more sockets we have
// to create a go func to spawn off multiple serves
for i := range ls {
listener := ls[i]
go func() {
httpSrv := http.Server{Handler: s.router}
chErrors <- httpSrv.Serve(listener)
}()
}
for i := 0; i < len(ls); i++ {
if err := <-chErrors; err != nil {
return nil, err
}
}
return nil, nil
case "tcp":
l, err = s.initTcpSocket(addr)
if err != nil {
return nil, err
}
case "unix":
if l, err = sockets.NewUnixSocket(addr, s.cfg.SocketGroup, s.start); err != nil {
return nil, err
}
default:
return nil, fmt.Errorf("Invalid protocol format: %q", proto)
}
return &HttpServer{
&http.Server{
Addr: addr,
Handler: s.router,
},
l,
}, nil
}
func (s *Server) AcceptConnections(d *daemon.Daemon) {
// Tell the init daemon we are accepting requests
s.daemon = d
go systemd.SdNotify("READY=1")
// close the lock so the listeners start accepting connections
select {
case <-s.start:
default:
close(s.start)
}
}
-45
View File
@@ -1,45 +0,0 @@
// +build windows
package server
import (
"errors"
"net"
"net/http"
"github.com/docker/docker/daemon"
)
// NewServer sets up the required Server and does protocol specific checking.
func (s *Server) newServer(proto, addr string) (Server, error) {
var (
err error
l net.Listener
)
switch proto {
case "tcp":
l, err = s.initTcpSocket(addr)
if err != nil {
return nil, err
}
default:
return nil, errors.New("Invalid protocol format. Windows only supports tcp.")
}
return &HttpServer{
&http.Server{
Addr: addr,
Handler: s.router,
},
l,
}, nil
}
func (s *Server) AcceptConnections(d *daemon.Daemon) {
s.daemon = d
// close the lock so the listeners start accepting connections
select {
case <-s.start:
default:
close(s.start)
}
}
-87
View File
@@ -1,87 +0,0 @@
// This package is used for API stability in the types and response to the
// consumers of the API stats endpoint.
package types
import "time"
type ThrottlingData struct {
// Number of periods with throttling active
Periods uint64 `json:"periods"`
// Number of periods when the container hit its throttling limit.
ThrottledPeriods uint64 `json:"throttled_periods"`
// Aggregate time the container was throttled for in nanoseconds.
ThrottledTime uint64 `json:"throttled_time"`
}
// All CPU stats are aggregated since container inception.
type CpuUsage struct {
// Total CPU time consumed.
// Units: nanoseconds.
TotalUsage uint64 `json:"total_usage"`
// Total CPU time consumed per core.
// Units: nanoseconds.
PercpuUsage []uint64 `json:"percpu_usage"`
// Time spent by tasks of the cgroup in kernel mode.
// Units: nanoseconds.
UsageInKernelmode uint64 `json:"usage_in_kernelmode"`
// Time spent by tasks of the cgroup in user mode.
// Units: nanoseconds.
UsageInUsermode uint64 `json:"usage_in_usermode"`
}
type CpuStats struct {
CpuUsage CpuUsage `json:"cpu_usage"`
SystemUsage uint64 `json:"system_cpu_usage"`
ThrottlingData ThrottlingData `json:"throttling_data,omitempty"`
}
type MemoryStats struct {
// current res_counter usage for memory
Usage uint64 `json:"usage"`
// maximum usage ever recorded.
MaxUsage uint64 `json:"max_usage"`
// TODO(vishh): Export these as stronger types.
// all the stats exported via memory.stat.
Stats map[string]uint64 `json:"stats"`
// number of times memory usage hits limits.
Failcnt uint64 `json:"failcnt"`
Limit uint64 `json:"limit"`
}
type BlkioStatEntry struct {
Major uint64 `json:"major"`
Minor uint64 `json:"minor"`
Op string `json:"op"`
Value uint64 `json:"value"`
}
type BlkioStats struct {
// number of bytes tranferred to and from the block device
IoServiceBytesRecursive []BlkioStatEntry `json:"io_service_bytes_recursive"`
IoServicedRecursive []BlkioStatEntry `json:"io_serviced_recursive"`
IoQueuedRecursive []BlkioStatEntry `json:"io_queue_recursive"`
IoServiceTimeRecursive []BlkioStatEntry `json:"io_service_time_recursive"`
IoWaitTimeRecursive []BlkioStatEntry `json:"io_wait_time_recursive"`
IoMergedRecursive []BlkioStatEntry `json:"io_merged_recursive"`
IoTimeRecursive []BlkioStatEntry `json:"io_time_recursive"`
SectorsRecursive []BlkioStatEntry `json:"sectors_recursive"`
}
type Network struct {
RxBytes uint64 `json:"rx_bytes"`
RxPackets uint64 `json:"rx_packets"`
RxErrors uint64 `json:"rx_errors"`
RxDropped uint64 `json:"rx_dropped"`
TxBytes uint64 `json:"tx_bytes"`
TxPackets uint64 `json:"tx_packets"`
TxErrors uint64 `json:"tx_errors"`
TxDropped uint64 `json:"tx_dropped"`
}
type Stats struct {
Read time.Time `json:"read"`
Network Network `json:"network,omitempty"`
CpuStats CpuStats `json:"cpu_stats,omitempty"`
MemoryStats MemoryStats `json:"memory_stats,omitempty"`
BlkioStats BlkioStats `json:"blkio_stats,omitempty"`
}
-221
View File
@@ -1,221 +0,0 @@
package types
import (
"time"
"github.com/docker/docker/daemon/network"
"github.com/docker/docker/pkg/version"
"github.com/docker/docker/runconfig"
)
// ContainerCreateResponse contains the information returned to a client on the
// creation of a new container.
type ContainerCreateResponse struct {
// ID is the ID of the created container.
ID string `json:"Id"`
// Warnings are any warnings encountered during the creation of the container.
Warnings []string `json:"Warnings"`
}
// POST /containers/{name:.*}/exec
type ContainerExecCreateResponse struct {
// ID is the exec ID.
ID string `json:"Id"`
}
// POST /auth
type AuthResponse struct {
// Status is the authentication status
Status string `json:"Status"`
}
// POST "/containers/"+containerID+"/wait"
type ContainerWaitResponse struct {
// StatusCode is the status code of the wait job
StatusCode int `json:"StatusCode"`
}
// POST "/commit?container="+containerID
type ContainerCommitResponse struct {
ID string `json:"Id"`
}
// GET "/containers/{name:.*}/changes"
type ContainerChange struct {
Kind int
Path string
}
// GET "/images/{name:.*}/history"
type ImageHistory struct {
ID string `json:"Id"`
Created int64
CreatedBy string
Tags []string
Size int64
Comment string
}
// DELETE "/images/{name:.*}"
type ImageDelete struct {
Untagged string `json:",omitempty"`
Deleted string `json:",omitempty"`
}
// GET "/images/json"
type Image struct {
ID string `json:"Id"`
ParentId string
RepoTags []string
RepoDigests []string
Created int
Size int
VirtualSize int
Labels map[string]string
}
// GET "/images/{name:.*}/json"
type ImageInspect struct {
Id string
Parent string
Comment string
Created time.Time
Container string
ContainerConfig *runconfig.Config
DockerVersion string
Author string
Config *runconfig.Config
Architecture string
Os string
Size int64
VirtualSize int64
}
// GET "/containers/json"
type Port struct {
IP string
PrivatePort int
PublicPort int
Type string
}
type Container struct {
ID string `json:"Id"`
Names []string `json:",omitempty"`
Image string `json:",omitempty"`
Command string `json:",omitempty"`
Created int `json:",omitempty"`
Ports []Port `json:",omitempty"`
SizeRw int `json:",omitempty"`
SizeRootFs int `json:",omitempty"`
Labels map[string]string `json:",omitempty"`
Status string `json:",omitempty"`
}
// POST "/containers/"+containerID+"/copy"
type CopyConfig struct {
Resource string
}
// GET "/containers/{name:.*}/top"
type ContainerProcessList struct {
Processes [][]string
Titles []string
}
type Version struct {
Version string
ApiVersion version.Version
GitCommit string
GoVersion string
Os string
Arch string
KernelVersion string `json:",omitempty"`
}
// GET "/info"
type Info struct {
ID string
Containers int
Images int
Driver string
DriverStatus [][2]string
MemoryLimit bool
SwapLimit bool
CpuCfsPeriod bool
CpuCfsQuota bool
IPv4Forwarding bool
Debug bool
NFd int
OomKillDisable bool
NGoroutines int
SystemTime string
ExecutionDriver string
LoggingDriver string
NEventsListener int
KernelVersion string
OperatingSystem string
IndexServerAddress string
RegistryConfig interface{}
InitSha1 string
InitPath string
NCPU int
MemTotal int64
DockerRootDir string
HttpProxy string
HttpsProxy string
NoProxy string
Name string
Labels []string
ExperimentalBuild bool
}
// This struct is a temp struct used by execStart
// Config fields is part of ExecConfig in runconfig package
type ExecStartCheck struct {
// ExecStart will first check if it's detached
Detach bool
// Check if there's a tty
Tty bool
}
type ContainerState struct {
Running bool
Paused bool
Restarting bool
OOMKilled bool
Dead bool
Pid int
ExitCode int
Error string
StartedAt time.Time
FinishedAt time.Time
}
// GET "/containers/{name:.*}/json"
type ContainerJSON struct {
Id string
Created time.Time
Path string
Args []string
Config *runconfig.Config
State *ContainerState
Image string
NetworkSettings *network.Settings
ResolvConfPath string
HostnamePath string
HostsPath string
LogPath string
Name string
RestartCount int
Driver string
ExecDriver string
MountLabel string
ProcessLabel string
Volumes map[string]string
VolumesRW map[string]bool
AppArmorProfile string
ExecIDs []string
HostConfig *runconfig.HostConfig
}
@@ -1,2 +1 @@
Michael Crosby <michael@crosbymichael.com> (@crosbymichael)
Guillaume J. Charmes <guillaume@docker.com> (@creack)
+205 -383
View File
@@ -1,27 +1,22 @@
package archive
import (
"archive/tar"
"bufio"
"bytes"
"compress/bzip2"
"compress/gzip"
"errors"
"fmt"
"github.com/dotcloud/docker/pkg/system"
"github.com/dotcloud/docker/utils"
"github.com/dotcloud/docker/vendor/src/code.google.com/p/go/src/pkg/archive/tar"
"io"
"io/ioutil"
"os"
"os/exec"
"path"
"path/filepath"
"runtime"
"strings"
"syscall"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/pkg/fileutils"
"github.com/docker/docker/pkg/pools"
"github.com/docker/docker/pkg/promise"
"github.com/docker/docker/pkg/system"
)
type (
@@ -29,28 +24,13 @@ type (
ArchiveReader io.Reader
Compression int
TarOptions struct {
IncludeFiles []string
ExcludePatterns []string
Compression Compression
NoLchown bool
Name string
Includes []string
Compression Compression
}
// Archiver allows the reuse of most utility functions of this package
// with a pluggable Untar function.
Archiver struct {
Untar func(io.Reader, string, *TarOptions) error
}
// breakoutError is used to differentiate errors related to breaking out
// When testing archive breakout in the unit tests, this error is expected
// in order for the test to pass.
breakoutError error
)
var (
ErrNotImplemented = errors.New("Function not implemented")
defaultArchiver = &Archiver{Untar}
)
const (
@@ -60,27 +40,27 @@ const (
Xz
)
func IsArchive(header []byte) bool {
compression := DetectCompression(header)
if compression != Uncompressed {
return true
}
r := tar.NewReader(bytes.NewBuffer(header))
_, err := r.Next()
return err == nil
}
func DetectCompression(source []byte) Compression {
sourceLen := len(source)
for compression, m := range map[Compression][]byte{
Bzip2: {0x42, 0x5A, 0x68},
Gzip: {0x1F, 0x8B, 0x08},
Xz: {0xFD, 0x37, 0x7A, 0x58, 0x5A, 0x00},
} {
if len(source) < len(m) {
logrus.Debugf("Len too short")
fail := false
if len(m) > sourceLen {
utils.Debugf("Len too short")
continue
}
if bytes.Compare(m, source[:len(m)]) == 0 {
i := 0
for _, b := range m {
if b != source[i] {
fail = true
break
}
i++
}
if !fail {
return compression
}
}
@@ -94,52 +74,43 @@ func xzDecompress(archive io.Reader) (io.ReadCloser, error) {
}
func DecompressStream(archive io.Reader) (io.ReadCloser, error) {
p := pools.BufioReader32KPool
buf := p.Get(archive)
bs, err := buf.Peek(10)
if err != nil {
return nil, err
buf := make([]byte, 10)
totalN := 0
for totalN < 10 {
n, err := archive.Read(buf[totalN:])
if err != nil {
if err == io.EOF {
return nil, fmt.Errorf("Tarball too short")
}
return nil, err
}
totalN += n
utils.Debugf("[tar autodetect] n: %d", n)
}
compression := DetectCompression(buf)
wrap := io.MultiReader(bytes.NewReader(buf), archive)
compression := DetectCompression(bs)
switch compression {
case Uncompressed:
readBufWrapper := p.NewReadCloserWrapper(buf, buf)
return readBufWrapper, nil
return ioutil.NopCloser(wrap), nil
case Gzip:
gzReader, err := gzip.NewReader(buf)
if err != nil {
return nil, err
}
readBufWrapper := p.NewReadCloserWrapper(buf, gzReader)
return readBufWrapper, nil
return gzip.NewReader(wrap)
case Bzip2:
bz2Reader := bzip2.NewReader(buf)
readBufWrapper := p.NewReadCloserWrapper(buf, bz2Reader)
return readBufWrapper, nil
return ioutil.NopCloser(bzip2.NewReader(wrap)), nil
case Xz:
xzReader, err := xzDecompress(buf)
if err != nil {
return nil, err
}
readBufWrapper := p.NewReadCloserWrapper(buf, xzReader)
return readBufWrapper, nil
return xzDecompress(wrap)
default:
return nil, fmt.Errorf("Unsupported compression format %s", (&compression).Extension())
}
}
func CompressStream(dest io.WriteCloser, compression Compression) (io.WriteCloser, error) {
p := pools.BufioWriter32KPool
buf := p.Get(dest)
switch compression {
case Uncompressed:
writeBufWrapper := p.NewWriteCloserWrapper(buf, buf)
return writeBufWrapper, nil
return utils.NopWriteCloser(dest), nil
case Gzip:
gzWriter := gzip.NewWriter(dest)
writeBufWrapper := p.NewWriteCloserWrapper(buf, gzWriter)
return writeBufWrapper, nil
return gzip.NewWriter(dest), nil
case Bzip2, Xz:
// archive/bzip2 does not support writing, and there is no xz support at all
// However, this is not a problem as docker only currently generates gzipped tars
@@ -163,30 +134,7 @@ func (compression *Compression) Extension() string {
return ""
}
type tarAppender struct {
TarWriter *tar.Writer
Buffer *bufio.Writer
// for hardlink mapping
SeenFiles map[uint64]string
}
// canonicalTarName provides a platform-independent and consistent posix-style
//path for files and directories to be archived regardless of the platform.
func canonicalTarName(name string, isDir bool) (string, error) {
name, err := CanonicalTarNameForPath(name)
if err != nil {
return "", err
}
// suffix with '/' for directories
if isDir && !strings.HasSuffix(name, "/") {
name += "/"
}
return name, nil
}
func (ta *tarAppender) addTarFile(path, name string) error {
func addTarFile(path, name string, tw *tar.Writer) error {
fi, err := os.Lstat(path)
if err != nil {
return err
@@ -203,31 +151,22 @@ func (ta *tarAppender) addTarFile(path, name string) error {
if err != nil {
return err
}
hdr.Mode = int64(chmodTarEntry(os.FileMode(hdr.Mode)))
name, err = canonicalTarName(name, fi.IsDir())
if err != nil {
return fmt.Errorf("tar: cannot canonicalize path: %v", err)
if fi.IsDir() && !strings.HasSuffix(name, "/") {
name = name + "/"
}
hdr.Name = name
nlink, inode, err := setHeaderForSpecialDevice(hdr, ta, name, fi.Sys())
if err != nil {
return err
}
// if it's a regular file and has more than 1 link,
// it's hardlinked, so set the type flag accordingly
if fi.Mode().IsRegular() && nlink > 1 {
// a link should have a name that it links too
// and that linked name should be first in the tar archive
if oldpath, ok := ta.SeenFiles[inode]; ok {
hdr.Typeflag = tar.TypeLink
hdr.Linkname = oldpath
hdr.Size = 0 // This Must be here for the writer math to add up!
} else {
ta.SeenFiles[inode] = name
stat, ok := fi.Sys().(*syscall.Stat_t)
if ok {
// Currently go does not fill in the major/minors
if stat.Mode&syscall.S_IFBLK == syscall.S_IFBLK ||
stat.Mode&syscall.S_IFCHR == syscall.S_IFCHR {
hdr.Devmajor = int64(major(uint64(stat.Rdev)))
hdr.Devminor = int64(minor(uint64(stat.Rdev)))
}
}
capability, _ := system.Lgetxattr(path, "security.capability")
@@ -236,33 +175,26 @@ func (ta *tarAppender) addTarFile(path, name string) error {
hdr.Xattrs["security.capability"] = string(capability)
}
if err := ta.TarWriter.WriteHeader(hdr); err != nil {
if err := tw.WriteHeader(hdr); err != nil {
return err
}
if hdr.Typeflag == tar.TypeReg {
file, err := os.Open(path)
if err != nil {
return err
}
ta.Buffer.Reset(ta.TarWriter)
defer ta.Buffer.Reset(nil)
_, err = io.Copy(ta.Buffer, file)
file.Close()
if err != nil {
return err
}
err = ta.Buffer.Flush()
if err != nil {
if file, err := os.Open(path); err != nil {
return err
} else {
_, err := io.Copy(tw, file)
if err != nil {
return err
}
file.Close()
}
}
return nil
}
func createTarFile(path, extractDir string, hdr *tar.Header, reader io.Reader, Lchown bool) error {
func createTarFile(path, extractDir string, hdr *tar.Header, reader io.Reader) error {
// hdr.Mode is in linux format, which we can use for sycalls,
// but for os.Foo() calls we need the mode converted to os.FileMode,
// so use hdrInfo.Mode() (they differ for e.g. setuid bits)
@@ -291,48 +223,40 @@ func createTarFile(path, extractDir string, hdr *tar.Header, reader io.Reader, L
file.Close()
case tar.TypeBlock, tar.TypeChar, tar.TypeFifo:
// Handle this is an OS-specific way
if err := handleTarTypeBlockCharFifo(hdr, path); err != nil {
mode := uint32(hdr.Mode & 07777)
switch hdr.Typeflag {
case tar.TypeBlock:
mode |= syscall.S_IFBLK
case tar.TypeChar:
mode |= syscall.S_IFCHR
case tar.TypeFifo:
mode |= syscall.S_IFIFO
}
if err := syscall.Mknod(path, mode, int(mkdev(hdr.Devmajor, hdr.Devminor))); err != nil {
return err
}
case tar.TypeLink:
targetPath := filepath.Join(extractDir, hdr.Linkname)
// check for hardlink breakout
if !strings.HasPrefix(targetPath, extractDir) {
return breakoutError(fmt.Errorf("invalid hardlink %q -> %q", targetPath, hdr.Linkname))
}
if err := os.Link(targetPath, path); err != nil {
if err := os.Link(filepath.Join(extractDir, hdr.Linkname), path); err != nil {
return err
}
case tar.TypeSymlink:
// path -> hdr.Linkname = targetPath
// e.g. /extractDir/path/to/symlink -> ../2/file = /extractDir/path/2/file
targetPath := filepath.Join(filepath.Dir(path), hdr.Linkname)
// the reason we don't need to check symlinks in the path (with FollowSymlinkInScope) is because
// that symlink would first have to be created, which would be caught earlier, at this very check:
if !strings.HasPrefix(targetPath, extractDir) {
return breakoutError(fmt.Errorf("invalid symlink %q -> %q", path, hdr.Linkname))
}
if err := os.Symlink(hdr.Linkname, path); err != nil {
return err
}
case tar.TypeXGlobalHeader:
logrus.Debugf("PAX Global Extended Headers found and ignored")
utils.Debugf("PAX Global Extended Headers found and ignored")
return nil
default:
return fmt.Errorf("Unhandled tar header type %d\n", hdr.Typeflag)
}
// Lchown is not supported on Windows
if runtime.GOOS != "windows" {
if err := os.Lchown(path, hdr.Uid, hdr.Gid); err != nil && Lchown {
return err
}
if err := os.Lchown(path, hdr.Uid, hdr.Gid); err != nil {
return err
}
for key, value := range hdr.Xattrs {
@@ -343,24 +267,20 @@ func createTarFile(path, extractDir string, hdr *tar.Header, reader io.Reader, L
// There is no LChmod, so ignore mode for symlink. Also, this
// must happen after chown, as that can modify the file mode
if err := handleLChmod(hdr, path, hdrInfo); err != nil {
return err
if hdr.Typeflag != tar.TypeSymlink {
if err := os.Chmod(path, hdrInfo.Mode()); err != nil {
return err
}
}
ts := []syscall.Timespec{timeToTimespec(hdr.AccessTime), timeToTimespec(hdr.ModTime)}
// syscall.UtimesNano doesn't support a NOFOLLOW flag atm
if hdr.Typeflag == tar.TypeLink {
if fi, err := os.Lstat(hdr.Linkname); err == nil && (fi.Mode()&os.ModeSymlink == 0) {
if err := system.UtimesNano(path, ts); err != nil && err != system.ErrNotSupportedPlatform {
return err
}
}
} else if hdr.Typeflag != tar.TypeSymlink {
if err := system.UtimesNano(path, ts); err != nil && err != system.ErrNotSupportedPlatform {
// syscall.UtimesNano doesn't support a NOFOLLOW flag atm, and
if hdr.Typeflag != tar.TypeSymlink {
if err := system.UtimesNano(path, ts); err != nil {
return err
}
} else {
if err := system.LUtimesNano(path, ts); err != nil && err != system.ErrNotSupportedPlatform {
if err := system.LUtimesNano(path, ts); err != nil {
return err
}
}
@@ -370,19 +290,28 @@ func createTarFile(path, extractDir string, hdr *tar.Header, reader io.Reader, L
// Tar creates an archive from the directory at `path`, and returns it as a
// stream of bytes.
func Tar(path string, compression Compression) (io.ReadCloser, error) {
return TarWithOptions(path, &TarOptions{Compression: compression})
return TarFilter(path, &TarOptions{Compression: compression})
}
// TarWithOptions creates an archive from the directory at `path`, only including files whose relative
// paths are included in `options.IncludeFiles` (if non-nil) or not in `options.ExcludePatterns`.
func TarWithOptions(srcPath string, options *TarOptions) (io.ReadCloser, error) {
patterns, patDirs, exceptions, err := fileutils.CleanPatterns(options.ExcludePatterns)
if err != nil {
return nil, err
func escapeName(name string) string {
escaped := make([]byte, 0)
for i, c := range []byte(name) {
if i == 0 && c == '/' {
continue
}
// all printable chars except "-" which is 0x2d
if (0x20 <= c && c <= 0x7E) && c != 0x2d {
escaped = append(escaped, c)
} else {
escaped = append(escaped, fmt.Sprintf("\\%03o", c)...)
}
}
return string(escaped)
}
// Tar creates an archive from the directory at `path`, only including files whose relative
// paths are included in `filter`. If `filter` is nil, then all files are included.
func TarFilter(srcPath string, options *TarOptions) (io.ReadCloser, error) {
pipeReader, pipeWriter := io.Pipe()
compressWriter, err := CompressStream(pipeWriter, options.Compression)
@@ -390,108 +319,73 @@ func TarWithOptions(srcPath string, options *TarOptions) (io.ReadCloser, error)
return nil, err
}
go func() {
ta := &tarAppender{
TarWriter: tar.NewWriter(compressWriter),
Buffer: pools.BufioWriter32KPool.Get(nil),
SeenFiles: make(map[uint64]string),
}
// this buffer is needed for the duration of this piped stream
defer pools.BufioWriter32KPool.Put(ta.Buffer)
tw := tar.NewWriter(compressWriter)
go func() {
// In general we log errors here but ignore them because
// during e.g. a diff operation the container can continue
// mutating the filesystem and we can see transient errors
// from this
if options.IncludeFiles == nil {
options.IncludeFiles = []string{"."}
if options.Includes == nil {
options.Includes = []string{"."}
}
seen := make(map[string]bool)
var renamedRelFilePath string // For when tar.Options.Name is set
for _, include := range options.IncludeFiles {
for _, include := range options.Includes {
filepath.Walk(filepath.Join(srcPath, include), func(filePath string, f os.FileInfo, err error) error {
if err != nil {
logrus.Debugf("Tar: Can't stat file %s to tar: %s", srcPath, err)
utils.Debugf("Tar: Can't stat file %s to tar: %s\n", srcPath, err)
return nil
}
relFilePath, err := filepath.Rel(srcPath, filePath)
if err != nil || (relFilePath == "." && f.IsDir()) {
// Error getting relative path OR we are looking
// at the root path. Skip in both situations.
if err != nil {
return nil
}
skip := false
// If "include" is an exact match for the current file
// then even if there's an "excludePatterns" pattern that
// matches it, don't skip it. IOW, assume an explicit 'include'
// is asking for that file no matter what - which is true
// for some files, like .dockerignore and Dockerfile (sometimes)
if include != relFilePath {
skip, err = fileutils.OptimizedMatches(relFilePath, patterns, patDirs)
if err != nil {
logrus.Debugf("Error matching %s", relFilePath, err)
return err
}
}
if skip {
if !exceptions && f.IsDir() {
return filepath.SkipDir
}
return nil
}
if seen[relFilePath] {
return nil
}
seen[relFilePath] = true
// Rename the base resource
if options.Name != "" && filePath == srcPath+"/"+filepath.Base(relFilePath) {
renamedRelFilePath = relFilePath
}
// Set this to make sure the items underneath also get renamed
if options.Name != "" {
relFilePath = strings.Replace(relFilePath, renamedRelFilePath, options.Name, 1)
}
if err := ta.addTarFile(filePath, relFilePath); err != nil {
logrus.Debugf("Can't add file %s to tar: %s", filePath, err)
if err := addTarFile(filePath, relFilePath, tw); err != nil {
utils.Debugf("Can't add file %s to tar: %s\n", srcPath, err)
}
return nil
})
}
// Make sure to check the error on Close.
if err := ta.TarWriter.Close(); err != nil {
logrus.Debugf("Can't close tar writer: %s", err)
if err := tw.Close(); err != nil {
utils.Debugf("Can't close tar writer: %s\n", err)
}
if err := compressWriter.Close(); err != nil {
logrus.Debugf("Can't close compress writer: %s", err)
utils.Debugf("Can't close compress writer: %s\n", err)
}
if err := pipeWriter.Close(); err != nil {
logrus.Debugf("Can't close pipe writer: %s", err)
utils.Debugf("Can't close pipe writer: %s\n", err)
}
}()
return pipeReader, nil
}
func Unpack(decompressedArchive io.Reader, dest string, options *TarOptions) error {
// Untar reads a stream of bytes from `archive`, parses it as a tar archive,
// and unpacks it into the directory at `path`.
// The archive may be compressed with one of the following algorithms:
// identity (uncompressed), gzip, bzip2, xz.
// FIXME: specify behavior when target path exists vs. doesn't exist.
func Untar(archive io.Reader, dest string, options *TarOptions) error {
if archive == nil {
return fmt.Errorf("Empty archive")
}
decompressedArchive, err := DecompressStream(archive)
if err != nil {
return err
}
defer decompressedArchive.Close()
tr := tar.NewReader(decompressedArchive)
trBuf := pools.BufioReader32KPool.Get(nil)
defer pools.BufioReader32KPool.Put(trBuf)
var dirs []*tar.Header
// Iterate through the files in the archive.
loop:
for {
hdr, err := tr.Next()
if err == io.EOF {
@@ -503,21 +397,14 @@ loop:
}
// Normalize name, for safety and for a simple is-root check
// This keeps "../" as-is, but normalizes "/../" to "/"
hdr.Name = filepath.Clean(hdr.Name)
for _, exclude := range options.ExcludePatterns {
if strings.HasPrefix(hdr.Name, exclude) {
continue loop
}
}
if !strings.HasSuffix(hdr.Name, "/") {
// Not the root directory, ensure that the parent directory exists
parent := filepath.Dir(hdr.Name)
parentPath := filepath.Join(dest, parent)
if _, err := os.Lstat(parentPath); err != nil && os.IsNotExist(err) {
err = system.MkdirAll(parentPath, 0777)
err = os.MkdirAll(parentPath, 600)
if err != nil {
return err
}
@@ -525,30 +412,20 @@ loop:
}
path := filepath.Join(dest, hdr.Name)
rel, err := filepath.Rel(dest, path)
if err != nil {
return err
}
if strings.HasPrefix(rel, "../") {
return breakoutError(fmt.Errorf("%q is outside of %q", hdr.Name, dest))
}
// If path exits we almost always just want to remove and replace it
// The only exception is when it is a directory *and* the file from
// the layer is also a directory. Then we want to merge them (i.e.
// just apply the metadata from the layer).
if fi, err := os.Lstat(path); err == nil {
if fi.IsDir() && hdr.Name == "." {
continue
}
if !(fi.IsDir() && hdr.Typeflag == tar.TypeDir) {
if err := os.RemoveAll(path); err != nil {
return err
}
}
}
trBuf.Reset(tr)
if err := createTarFile(path, dest, hdr, trBuf, !options.NoLchown); err != nil {
if err := createTarFile(path, dest, hdr, tr); err != nil {
return err
}
@@ -566,143 +443,57 @@ loop:
return err
}
}
return nil
}
// Untar reads a stream of bytes from `archive`, parses it as a tar archive,
// and unpacks it into the directory at `dest`.
// The archive may be compressed with one of the following algorithms:
// identity (uncompressed), gzip, bzip2, xz.
// FIXME: specify behavior when target path exists vs. doesn't exist.
func Untar(archive io.Reader, dest string, options *TarOptions) error {
if archive == nil {
return fmt.Errorf("Empty archive")
}
dest = filepath.Clean(dest)
if options == nil {
options = &TarOptions{}
}
if options.ExcludePatterns == nil {
options.ExcludePatterns = []string{}
}
decompressedArchive, err := DecompressStream(archive)
if err != nil {
return err
}
defer decompressedArchive.Close()
return Unpack(decompressedArchive, dest, options)
}
func (archiver *Archiver) TarUntar(src, dst string) error {
logrus.Debugf("TarUntar(%s %s)", src, dst)
archive, err := TarWithOptions(src, &TarOptions{Compression: Uncompressed})
// TarUntar is a convenience function which calls Tar and Untar, with
// the output of one piped into the other. If either Tar or Untar fails,
// TarUntar aborts and returns the error.
func TarUntar(src string, dst string) error {
utils.Debugf("TarUntar(%s %s)", src, dst)
archive, err := TarFilter(src, &TarOptions{Compression: Uncompressed})
if err != nil {
return err
}
defer archive.Close()
return archiver.Untar(archive, dst, nil)
}
// TarUntar is a convenience function which calls Tar and Untar, with the output of one piped into the other.
// If either Tar or Untar fails, TarUntar aborts and returns the error.
func TarUntar(src, dst string) error {
return defaultArchiver.TarUntar(src, dst)
}
func (archiver *Archiver) UntarPath(src, dst string) error {
archive, err := os.Open(src)
if err != nil {
return err
}
defer archive.Close()
if err := archiver.Untar(archive, dst, nil); err != nil {
return err
}
return nil
return Untar(archive, dst, nil)
}
// UntarPath is a convenience function which looks for an archive
// at filesystem path `src`, and unpacks it at `dst`.
func UntarPath(src, dst string) error {
return defaultArchiver.UntarPath(src, dst)
}
func (archiver *Archiver) CopyWithTar(src, dst string) error {
srcSt, err := os.Stat(src)
archive, err := os.Open(src)
if err != nil {
return err
}
if !srcSt.IsDir() {
return archiver.CopyFileWithTar(src, dst)
}
// Create dst, copy src's content into it
logrus.Debugf("Creating dest directory: %s", dst)
if err := system.MkdirAll(dst, 0755); err != nil && !os.IsExist(err) {
defer archive.Close()
if err := Untar(archive, dst, nil); err != nil {
return err
}
logrus.Debugf("Calling TarUntar(%s, %s)", src, dst)
return archiver.TarUntar(src, dst)
return nil
}
// CopyWithTar creates a tar archive of filesystem path `src`, and
// unpacks it at filesystem path `dst`.
// The archive is streamed directly with fixed buffering and no
// intermediary disk IO.
//
func CopyWithTar(src, dst string) error {
return defaultArchiver.CopyWithTar(src, dst)
}
func (archiver *Archiver) CopyFileWithTar(src, dst string) (err error) {
logrus.Debugf("CopyFileWithTar(%s, %s)", src, dst)
srcSt, err := os.Stat(src)
if err != nil {
return err
}
if srcSt.IsDir() {
return fmt.Errorf("Can't copy a directory")
if !srcSt.IsDir() {
return CopyFileWithTar(src, dst)
}
// Clean up the trailing slash
if dst[len(dst)-1] == os.PathSeparator {
dst = filepath.Join(dst, filepath.Base(src))
}
// Create the holding directory if necessary
if err := system.MkdirAll(filepath.Dir(dst), 0700); err != nil && !os.IsExist(err) {
// Create dst, copy src's content into it
utils.Debugf("Creating dest directory: %s", dst)
if err := os.MkdirAll(dst, 0755); err != nil && !os.IsExist(err) {
return err
}
r, w := io.Pipe()
errC := promise.Go(func() error {
defer w.Close()
srcF, err := os.Open(src)
if err != nil {
return err
}
defer srcF.Close()
hdr, err := tar.FileInfoHeader(srcSt, "")
if err != nil {
return err
}
hdr.Name = filepath.Base(dst)
hdr.Mode = int64(chmodTarEntry(os.FileMode(hdr.Mode)))
tw := tar.NewWriter(w)
defer tw.Close()
if err := tw.WriteHeader(hdr); err != nil {
return err
}
if _, err := io.Copy(tw, srcF); err != nil {
return err
}
return nil
})
defer func() {
if er := <-errC; err != nil {
err = er
}
}()
return archiver.Untar(r, filepath.Dir(dst), nil)
utils.Debugf("Calling TarUntar(%s, %s)", src, dst)
return TarUntar(src, dst)
}
// CopyFileWithTar emulates the behavior of the 'cp' command-line
@@ -712,7 +503,54 @@ func (archiver *Archiver) CopyFileWithTar(src, dst string) (err error) {
// If `dst` ends with a trailing slash '/', the final destination path
// will be `dst/base(src)`.
func CopyFileWithTar(src, dst string) (err error) {
return defaultArchiver.CopyFileWithTar(src, dst)
utils.Debugf("CopyFileWithTar(%s, %s)", src, dst)
srcSt, err := os.Stat(src)
if err != nil {
return err
}
if srcSt.IsDir() {
return fmt.Errorf("Can't copy a directory")
}
// Clean up the trailing /
if dst[len(dst)-1] == '/' {
dst = path.Join(dst, filepath.Base(src))
}
// Create the holding directory if necessary
if err := os.MkdirAll(filepath.Dir(dst), 0700); err != nil && !os.IsExist(err) {
return err
}
r, w := io.Pipe()
errC := utils.Go(func() error {
defer w.Close()
srcF, err := os.Open(src)
if err != nil {
return err
}
defer srcF.Close()
tw := tar.NewWriter(w)
hdr, err := tar.FileInfoHeader(srcSt, "")
if err != nil {
return err
}
hdr.Name = filepath.Base(dst)
if err := tw.WriteHeader(hdr); err != nil {
return err
}
if _, err := io.Copy(tw, srcF); err != nil {
return err
}
tw.Close()
return nil
})
defer func() {
if er := <-errC; err != nil {
err = er
}
}()
return Untar(r, filepath.Dir(dst), nil)
}
// CmdStream executes a command, and returns its stdout as a stream.
@@ -787,33 +625,17 @@ func NewTempArchive(src Archive, dir string) (*TempArchive, error) {
return nil, err
}
size := st.Size()
return &TempArchive{File: f, Size: size}, nil
return &TempArchive{f, size}, nil
}
type TempArchive struct {
*os.File
Size int64 // Pre-computed from Stat().Size() as a convenience
read int64
closed bool
}
// Close closes the underlying file if it's still open, or does a no-op
// to allow callers to try to close the TempArchive multiple times safely.
func (archive *TempArchive) Close() error {
if archive.closed {
return nil
}
archive.closed = true
return archive.File.Close()
Size int64 // Pre-computed from Stat().Size() as a convenience
}
func (archive *TempArchive) Read(data []byte) (int, error) {
n, err := archive.File.Read(data)
archive.read += int64(n)
if err != nil || archive.read == archive.Size {
archive.Close()
if err != nil {
os.Remove(archive.File.Name())
}
return n, err
+139
View File
@@ -0,0 +1,139 @@
package archive
import (
"bytes"
"fmt"
"github.com/dotcloud/docker/vendor/src/code.google.com/p/go/src/pkg/archive/tar"
"io"
"io/ioutil"
"os"
"os/exec"
"path"
"testing"
"time"
)
func TestCmdStreamLargeStderr(t *testing.T) {
cmd := exec.Command("/bin/sh", "-c", "dd if=/dev/zero bs=1k count=1000 of=/dev/stderr; echo hello")
out, err := CmdStream(cmd, nil)
if err != nil {
t.Fatalf("Failed to start command: %s", err)
}
errCh := make(chan error)
go func() {
_, err := io.Copy(ioutil.Discard, out)
errCh <- err
}()
select {
case err := <-errCh:
if err != nil {
t.Fatalf("Command should not have failed (err=%.100s...)", err)
}
case <-time.After(5 * time.Second):
t.Fatalf("Command did not complete in 5 seconds; probable deadlock")
}
}
func TestCmdStreamBad(t *testing.T) {
badCmd := exec.Command("/bin/sh", "-c", "echo hello; echo >&2 error couldn\\'t reverse the phase pulser; exit 1")
out, err := CmdStream(badCmd, nil)
if err != nil {
t.Fatalf("Failed to start command: %s", err)
}
if output, err := ioutil.ReadAll(out); err == nil {
t.Fatalf("Command should have failed")
} else if err.Error() != "exit status 1: error couldn't reverse the phase pulser\n" {
t.Fatalf("Wrong error value (%s)", err)
} else if s := string(output); s != "hello\n" {
t.Fatalf("Command output should be '%s', not '%s'", "hello\\n", output)
}
}
func TestCmdStreamGood(t *testing.T) {
cmd := exec.Command("/bin/sh", "-c", "echo hello; exit 0")
out, err := CmdStream(cmd, nil)
if err != nil {
t.Fatal(err)
}
if output, err := ioutil.ReadAll(out); err != nil {
t.Fatalf("Command should not have failed (err=%s)", err)
} else if s := string(output); s != "hello\n" {
t.Fatalf("Command output should be '%s', not '%s'", "hello\\n", output)
}
}
func tarUntar(t *testing.T, origin string, compression Compression) error {
archive, err := Tar(origin, compression)
if err != nil {
t.Fatal(err)
}
defer archive.Close()
buf := make([]byte, 10)
if _, err := archive.Read(buf); err != nil {
return err
}
wrap := io.MultiReader(bytes.NewReader(buf), archive)
detectedCompression := DetectCompression(buf)
if detectedCompression.Extension() != compression.Extension() {
return fmt.Errorf("Wrong compression detected. Actual compression: %s, found %s", compression.Extension(), detectedCompression.Extension())
}
tmp, err := ioutil.TempDir("", "docker-test-untar")
if err != nil {
return err
}
defer os.RemoveAll(tmp)
if err := Untar(wrap, tmp, nil); err != nil {
return err
}
if _, err := os.Stat(tmp); err != nil {
return err
}
changes, err := ChangesDirs(origin, tmp)
if err != nil {
return err
}
if len(changes) != 0 {
t.Fatalf("Unexpected differences after tarUntar: %v", changes)
}
return nil
}
func TestTarUntar(t *testing.T) {
origin, err := ioutil.TempDir("", "docker-test-untar-origin")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(origin)
if err := ioutil.WriteFile(path.Join(origin, "1"), []byte("hello world"), 0700); err != nil {
t.Fatal(err)
}
if err := ioutil.WriteFile(path.Join(origin, "2"), []byte("welcome!"), 0700); err != nil {
t.Fatal(err)
}
for _, c := range []Compression{
Uncompressed,
Gzip,
} {
if err := tarUntar(t, origin, c); err != nil {
t.Fatalf("Error tar/untar for compression %s: %s", c.Extension(), err)
}
}
}
// Some tar archives such as http://haproxy.1wt.eu/download/1.5/src/devel/haproxy-1.5-dev21.tar.gz
// use PAX Global Extended Headers.
// Failing prevents the archives from being uncompressed during ADD
func TestTypeXGlobalHeaderDoesNotFail(t *testing.T) {
hdr := tar.Header{Typeflag: tar.TypeXGlobalHeader}
err := createTarFile("pax_global_header", "some_dir", &hdr, nil)
if err != nil {
t.Fatal(err)
}
}
+46 -89
View File
@@ -1,20 +1,17 @@
package archive
import (
"archive/tar"
"bytes"
"fmt"
"github.com/dotcloud/docker/pkg/system"
"github.com/dotcloud/docker/utils"
"github.com/dotcloud/docker/vendor/src/code.google.com/p/go/src/pkg/archive/tar"
"io"
"os"
"path/filepath"
"sort"
"strings"
"syscall"
"time"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/pkg/pools"
"github.com/docker/docker/pkg/system"
)
type ChangeType int
@@ -43,13 +40,6 @@ func (change *Change) String() string {
return fmt.Sprintf("%s %s", kind, change.Path)
}
// for sort.Sort
type changesByPath []Change
func (c changesByPath) Less(i, j int) bool { return c[i].Path < c[j].Path }
func (c changesByPath) Len() int { return len(c) }
func (c changesByPath) Swap(i, j int) { c[j], c[i] = c[i], c[j] }
// Gnu tar and the go tar writer don't have sub-second mtime
// precision, which is problematic when we apply changes via tar
// files, we handle this by comparing for exact times, *or* same
@@ -65,8 +55,6 @@ func sameFsTimeSpec(a, b syscall.Timespec) bool {
(a.Nsec == b.Nsec || a.Nsec == 0 || b.Nsec == 0)
}
// Changes walks the path rw and determines changes for the files in the path,
// with respect to the parent layers
func Changes(layers []string, rw string) ([]Change, error) {
var changes []Change
err := filepath.Walk(rw, func(path string, f os.FileInfo, err error) error {
@@ -142,10 +130,9 @@ func Changes(layers []string, rw string) ([]Change, error) {
type FileInfo struct {
parent *FileInfo
name string
stat *system.Stat_t
stat syscall.Stat_t
children map[string]*FileInfo
capability []byte
added bool
}
func (root *FileInfo) LookUp(path string) *FileInfo {
@@ -174,10 +161,11 @@ func (info *FileInfo) path() string {
return filepath.Join(info.parent.path(), info.name)
}
func (info *FileInfo) isDir() bool {
return info.parent == nil || info.stat.Mode&syscall.S_IFDIR == syscall.S_IFDIR
}
func (info *FileInfo) addChanges(oldInfo *FileInfo, changes *[]Change) {
sizeAtEntry := len(*changes)
if oldInfo == nil {
// add
change := Change{
@@ -185,7 +173,6 @@ func (info *FileInfo) addChanges(oldInfo *FileInfo, changes *[]Change) {
Kind: ChangeAdd,
}
*changes = append(*changes, change)
info.added = true
}
// We make a copy so we can modify it to detect additions
@@ -202,22 +189,27 @@ func (info *FileInfo) addChanges(oldInfo *FileInfo, changes *[]Change) {
oldChild, _ := oldChildren[name]
if oldChild != nil {
// change?
oldStat := oldChild.stat
newStat := newChild.stat
oldStat := &oldChild.stat
newStat := &newChild.stat
// Note: We can't compare inode or ctime or blocksize here, because these change
// when copying a file into a container. However, that is not generally a problem
// because any content change will change mtime, and any status change should
// be visible when actually comparing the stat fields. The only time this
// breaks down is if some code intentionally hides a change by setting
// back mtime
if statDifferent(oldStat, newStat) ||
if oldStat.Mode != newStat.Mode ||
oldStat.Uid != newStat.Uid ||
oldStat.Gid != newStat.Gid ||
oldStat.Rdev != newStat.Rdev ||
// Don't look at size for dirs, its not a good measure of change
(oldStat.Size != newStat.Size && oldStat.Mode&syscall.S_IFDIR != syscall.S_IFDIR) ||
!sameFsTimeSpec(system.GetLastModification(oldStat), system.GetLastModification(newStat)) ||
bytes.Compare(oldChild.capability, newChild.capability) != 0 {
change := Change{
Path: newChild.path(),
Kind: ChangeModify,
}
*changes = append(*changes, change)
newChild.added = true
}
// Remove from copy so we can detect deletions
@@ -235,19 +227,6 @@ func (info *FileInfo) addChanges(oldInfo *FileInfo, changes *[]Change) {
*changes = append(*changes, change)
}
// If there were changes inside this directory, we need to add it, even if the directory
// itself wasn't changed. This is needed to properly save and restore filesystem permissions.
if len(*changes) > sizeAtEntry && info.isDir() && !info.added && info.path() != "/" {
change := Change{
Path: info.path(),
Kind: ChangeModify,
}
// Let's insert the directory entry before the recently added entries located inside this dir
*changes = append(*changes, change) // just to resize the slice, will be overwritten
copy((*changes)[sizeAtEntry+1:], (*changes)[sizeAtEntry:])
(*changes)[sizeAtEntry] = change
}
}
func (info *FileInfo) Changes(oldInfo *FileInfo) []Change {
@@ -296,11 +275,9 @@ func collectFileInfo(sourceDir string) (*FileInfo, error) {
parent: parent,
}
s, err := system.Lstat(path)
if err != nil {
if err := syscall.Lstat(path, &info.stat); err != nil {
return err
}
info.stat = s
info.capability, _ = system.Lgetxattr(path, "security.capability")
@@ -314,36 +291,20 @@ func collectFileInfo(sourceDir string) (*FileInfo, error) {
return root, nil
}
// ChangesDirs compares two directories and generates an array of Change objects describing the changes.
// If oldDir is "", then all files in newDir will be Add-Changes.
// Compare two directories and generate an array of Change objects describing the changes
func ChangesDirs(newDir, oldDir string) ([]Change, error) {
var (
oldRoot, newRoot *FileInfo
err1, err2 error
errs = make(chan error, 2)
)
go func() {
if oldDir != "" {
oldRoot, err1 = collectFileInfo(oldDir)
}
errs <- err1
}()
go func() {
newRoot, err2 = collectFileInfo(newDir)
errs <- err2
}()
// block until both routines have returned
for i := 0; i < 2; i++ {
if err := <-errs; err != nil {
return nil, err
}
oldRoot, err := collectFileInfo(oldDir)
if err != nil {
return nil, err
}
newRoot, err := collectFileInfo(newDir)
if err != nil {
return nil, err
}
return newRoot.Changes(oldRoot), nil
}
// ChangesSize calculates the size in bytes of the provided changes, based on newDir.
func ChangesSize(newDir string, changes []Change) int64 {
var size int64
for _, change := range changes {
@@ -358,20 +319,19 @@ func ChangesSize(newDir string, changes []Change) int64 {
return size
}
// ExportChanges produces an Archive from the provided changes, relative to dir.
func major(device uint64) uint64 {
return (device >> 8) & 0xfff
}
func minor(device uint64) uint64 {
return (device & 0xff) | ((device >> 12) & 0xfff00)
}
func ExportChanges(dir string, changes []Change) (Archive, error) {
reader, writer := io.Pipe()
tw := tar.NewWriter(writer)
go func() {
ta := &tarAppender{
TarWriter: tar.NewWriter(writer),
Buffer: pools.BufioWriter32KPool.Get(nil),
SeenFiles: make(map[uint64]string),
}
// this buffer is needed for the duration of this piped stream
defer pools.BufioWriter32KPool.Put(ta.Buffer)
sort.Sort(changesByPath(changes))
// In general we log errors here but ignore them because
// during e.g. a diff operation the container can continue
// mutating the filesystem and we can see transient errors
@@ -381,32 +341,29 @@ func ExportChanges(dir string, changes []Change) (Archive, error) {
whiteOutDir := filepath.Dir(change.Path)
whiteOutBase := filepath.Base(change.Path)
whiteOut := filepath.Join(whiteOutDir, ".wh."+whiteOutBase)
timestamp := time.Now()
hdr := &tar.Header{
Name: whiteOut[1:],
Size: 0,
ModTime: timestamp,
AccessTime: timestamp,
ChangeTime: timestamp,
ModTime: time.Now(),
AccessTime: time.Now(),
ChangeTime: time.Now(),
}
if err := ta.TarWriter.WriteHeader(hdr); err != nil {
logrus.Debugf("Can't write whiteout header: %s", err)
if err := tw.WriteHeader(hdr); err != nil {
utils.Debugf("Can't write whiteout header: %s\n", err)
}
} else {
path := filepath.Join(dir, change.Path)
if err := ta.addTarFile(path, change.Path[1:]); err != nil {
logrus.Debugf("Can't add file %s to tar: %s", path, err)
if err := addTarFile(path, change.Path[1:], tw); err != nil {
utils.Debugf("Can't add file %s to tar: %s\n", path, err)
}
}
}
// Make sure to check the error on Close.
if err := ta.TarWriter.Close(); err != nil {
logrus.Debugf("Can't close layer: %s", err)
}
if err := writer.Close(); err != nil {
logrus.Debugf("failed close Changes writer: %s", err)
if err := tw.Close(); err != nil {
utils.Debugf("Can't close layer: %s\n", err)
}
writer.Close()
}()
return reader, nil
}
@@ -6,7 +6,6 @@ import (
"os/exec"
"path"
"sort"
"syscall"
"testing"
"time"
)
@@ -26,6 +25,13 @@ func copyDir(src, dst string) error {
return nil
}
// Helper to sort []Change by path
type byPath struct{ changes []Change }
func (b byPath) Less(i, j int) bool { return b.changes[i].Path < b.changes[j].Path }
func (b byPath) Len() int { return len(b.changes) }
func (b byPath) Swap(i, j int) { b.changes[i], b.changes[j] = b.changes[j], b.changes[i] }
type FileType uint32
const (
@@ -92,130 +98,17 @@ func createSampleDir(t *testing.T, root string) {
}
}
func TestChangeString(t *testing.T) {
modifiyChange := Change{"change", ChangeModify}
toString := modifiyChange.String()
if toString != "C change" {
t.Fatalf("String() of a change with ChangeModifiy Kind should have been %s but was %s", "C change", toString)
}
addChange := Change{"change", ChangeAdd}
toString = addChange.String()
if toString != "A change" {
t.Fatalf("String() of a change with ChangeAdd Kind should have been %s but was %s", "A change", toString)
}
deleteChange := Change{"change", ChangeDelete}
toString = deleteChange.String()
if toString != "D change" {
t.Fatalf("String() of a change with ChangeDelete Kind should have been %s but was %s", "D change", toString)
}
}
func TestChangesWithNoChanges(t *testing.T) {
rwLayer, err := ioutil.TempDir("", "docker-changes-test")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(rwLayer)
layer, err := ioutil.TempDir("", "docker-changes-test-layer")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(layer)
createSampleDir(t, layer)
changes, err := Changes([]string{layer}, rwLayer)
if err != nil {
t.Fatal(err)
}
if len(changes) != 0 {
t.Fatalf("Changes with no difference should have detect no changes, but detected %d", len(changes))
}
}
func TestChangesWithChanges(t *testing.T) {
rwLayer, err := ioutil.TempDir("", "docker-changes-test")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(rwLayer)
// Create a folder
dir1 := path.Join(rwLayer, "dir1")
os.MkdirAll(dir1, 0740)
deletedFile := path.Join(dir1, ".wh.file1-2")
ioutil.WriteFile(deletedFile, []byte{}, 0600)
modifiedFile := path.Join(dir1, "file1-1")
ioutil.WriteFile(modifiedFile, []byte{0x00}, 01444)
// Let's add a subfolder for a newFile
subfolder := path.Join(dir1, "subfolder")
os.MkdirAll(subfolder, 0740)
newFile := path.Join(subfolder, "newFile")
ioutil.WriteFile(newFile, []byte{}, 0740)
// Let's create folders that with have the role of layers with the same data
layer, err := ioutil.TempDir("", "docker-changes-test-layer")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(layer)
createSampleDir(t, layer)
os.MkdirAll(path.Join(layer, "dir1/subfolder"), 0740)
// Let's modify modtime for dir1 to be sure it's the same for the two layer (to not having false positive)
fi, err := os.Stat(dir1)
if err != nil {
return
}
mtime := fi.ModTime()
stat := fi.Sys().(*syscall.Stat_t)
atime := time.Unix(int64(stat.Atim.Sec), int64(stat.Atim.Nsec))
layerDir1 := path.Join(layer, "dir1")
os.Chtimes(layerDir1, atime, mtime)
changes, err := Changes([]string{layer}, rwLayer)
if err != nil {
t.Fatal(err)
}
sort.Sort(changesByPath(changes))
expectedChanges := []Change{
{"/dir1/file1-1", ChangeModify},
{"/dir1/file1-2", ChangeDelete},
{"/dir1/subfolder", ChangeModify},
{"/dir1/subfolder/newFile", ChangeAdd},
}
for i := 0; i < max(len(changes), len(expectedChanges)); i++ {
if i >= len(expectedChanges) {
t.Fatalf("unexpected change %s\n", changes[i].String())
}
if i >= len(changes) {
t.Fatalf("no change for expected change %s\n", expectedChanges[i].String())
}
if changes[i].Path == expectedChanges[i].Path {
if changes[i] != expectedChanges[i] {
t.Fatalf("Wrong change for %s, expected %s, got %s\n", changes[i].Path, changes[i].String(), expectedChanges[i].String())
}
} else if changes[i].Path < expectedChanges[i].Path {
t.Fatalf("unexpected change %s\n", changes[i].String())
} else {
t.Fatalf("no change for expected change %s != %s\n", expectedChanges[i].String(), changes[i].String())
}
}
}
// Create an directory, copy it, make sure we report no changes between the two
func TestChangesDirsEmpty(t *testing.T) {
src, err := ioutil.TempDir("", "docker-changes-test")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(src)
createSampleDir(t, src)
dst := src + "-copy"
if err := copyDir(src, dst); err != nil {
t.Fatal(err)
}
defer os.RemoveAll(dst)
changes, err := ChangesDirs(dst, src)
if err != nil {
t.Fatal(err)
@@ -245,7 +138,7 @@ func mutateSampleDir(t *testing.T, root string) {
}
// Rewrite a file
if err := ioutil.WriteFile(path.Join(root, "file2"), []byte("fileNN\n"), 0777); err != nil {
if err := ioutil.WriteFile(path.Join(root, "file2"), []byte("fileN\n"), 0777); err != nil {
t.Fatal(err)
}
@@ -253,12 +146,12 @@ func mutateSampleDir(t *testing.T, root string) {
if err := os.RemoveAll(path.Join(root, "file3")); err != nil {
t.Fatal(err)
}
if err := ioutil.WriteFile(path.Join(root, "file3"), []byte("fileMM\n"), 0404); err != nil {
if err := ioutil.WriteFile(path.Join(root, "file3"), []byte("fileM\n"), 0404); err != nil {
t.Fatal(err)
}
// Touch file
if err := os.Chtimes(path.Join(root, "file4"), time.Now().Add(time.Second), time.Now().Add(time.Second)); err != nil {
if err := os.Chtimes(path.Join(root, "file4"), time.Now(), time.Now()); err != nil {
t.Fatal(err)
}
@@ -302,7 +195,7 @@ func mutateSampleDir(t *testing.T, root string) {
}
// Touch dir
if err := os.Chtimes(path.Join(root, "dir3"), time.Now().Add(time.Second), time.Now().Add(time.Second)); err != nil {
if err := os.Chtimes(path.Join(root, "dir3"), time.Now(), time.Now()); err != nil {
t.Fatal(err)
}
}
@@ -327,11 +220,12 @@ func TestChangesDirsMutated(t *testing.T) {
t.Fatal(err)
}
sort.Sort(changesByPath(changes))
sort.Sort(byPath{changes})
expectedChanges := []Change{
{"/dir1", ChangeDelete},
{"/dir2", ChangeModify},
{"/dir3", ChangeModify},
{"/dirnew", ChangeAdd},
{"/file1", ChangeDelete},
{"/file2", ChangeModify},
@@ -392,7 +286,7 @@ func TestApplyLayer(t *testing.T) {
t.Fatal(err)
}
if _, err := ApplyLayer(src, layerCopy); err != nil {
if err := ApplyLayer(src, layerCopy); err != nil {
t.Fatal(err)
}
@@ -405,41 +299,3 @@ func TestApplyLayer(t *testing.T) {
t.Fatalf("Unexpected differences after reapplying mutation: %v", changes2)
}
}
func TestChangesSizeWithNoChanges(t *testing.T) {
size := ChangesSize("/tmp", nil)
if size != 0 {
t.Fatalf("ChangesSizes with no changes should be 0, was %d", size)
}
}
func TestChangesSizeWithOnlyDeleteChanges(t *testing.T) {
changes := []Change{
{Path: "deletedPath", Kind: ChangeDelete},
}
size := ChangesSize("/tmp", changes)
if size != 0 {
t.Fatalf("ChangesSizes with only delete changes should be 0, was %d", size)
}
}
func TestChangesSize(t *testing.T) {
parentPath, err := ioutil.TempDir("", "docker-changes-test")
defer os.RemoveAll(parentPath)
addition := path.Join(parentPath, "addition")
if err := ioutil.WriteFile(addition, []byte{0x01, 0x01, 0x01}, 0744); err != nil {
t.Fatal(err)
}
modification := path.Join(parentPath, "modification")
if err = ioutil.WriteFile(modification, []byte{0x01, 0x01, 0x01}, 0744); err != nil {
t.Fatal(err)
}
changes := []Change{
{Path: "addition", Kind: ChangeAdd},
{Path: "modification", Kind: ChangeModify},
}
size := ChangesSize(parentPath, changes)
if size != 6 {
t.Fatalf("ChangesSizes with only delete changes should be 0, was %d", size)
}
}
+45 -53
View File
@@ -1,23 +1,46 @@
package archive
import (
"archive/tar"
"fmt"
"github.com/dotcloud/docker/vendor/src/code.google.com/p/go/src/pkg/archive/tar"
"io"
"io/ioutil"
"os"
"path/filepath"
"strings"
"syscall"
"github.com/docker/docker/pkg/pools"
"github.com/docker/docker/pkg/system"
"time"
)
func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
// Linux device nodes are a bit weird due to backwards compat with 16 bit device nodes.
// They are, from low to high: the lower 8 bits of the minor, then 12 bits of the major,
// then the top 12 bits of the minor
func mkdev(major int64, minor int64) uint32 {
return uint32(((minor & 0xfff00) << 12) | ((major & 0xfff) << 8) | (minor & 0xff))
}
func timeToTimespec(time time.Time) (ts syscall.Timespec) {
if time.IsZero() {
// Return UTIME_OMIT special value
ts.Sec = 0
ts.Nsec = ((1 << 30) - 2)
return
}
return syscall.NsecToTimespec(time.UnixNano())
}
// ApplyLayer parses a diff in the standard layer format from `layer`, and
// applies it to the directory `dest`.
func ApplyLayer(dest string, layer ArchiveReader) error {
// We need to be able to set any perms
oldmask := syscall.Umask(0)
defer syscall.Umask(oldmask)
layer, err := DecompressStream(layer)
if err != nil {
return err
}
tr := tar.NewReader(layer)
trBuf := pools.BufioReader32KPool.Get(tr)
defer pools.BufioReader32KPool.Put(trBuf)
var dirs []*tar.Header
@@ -32,11 +55,9 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
break
}
if err != nil {
return 0, err
return err
}
size += hdr.Size
// Normalize name, for safety and for a simple is-root check
hdr.Name = filepath.Clean(hdr.Name)
@@ -47,9 +68,9 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
parent := filepath.Dir(hdr.Name)
parentPath := filepath.Join(dest, parent)
if _, err := os.Lstat(parentPath); err != nil && os.IsNotExist(err) {
err = system.MkdirAll(parentPath, 0600)
err = os.MkdirAll(parentPath, 600)
if err != nil {
return 0, err
return err
}
}
}
@@ -64,32 +85,24 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
aufsHardlinks[basename] = hdr
if aufsTempdir == "" {
if aufsTempdir, err = ioutil.TempDir("", "dockerplnk"); err != nil {
return 0, err
return err
}
defer os.RemoveAll(aufsTempdir)
}
if err := createTarFile(filepath.Join(aufsTempdir, basename), dest, hdr, tr, true); err != nil {
return 0, err
if err := createTarFile(filepath.Join(aufsTempdir, basename), dest, hdr, tr); err != nil {
return err
}
}
continue
}
path := filepath.Join(dest, hdr.Name)
rel, err := filepath.Rel(dest, path)
if err != nil {
return 0, err
}
if strings.HasPrefix(rel, "../") {
return 0, breakoutError(fmt.Errorf("%q is outside of %q", hdr.Name, dest))
}
base := filepath.Base(path)
if strings.HasPrefix(base, ".wh.") {
originalBase := base[len(".wh."):]
originalPath := filepath.Join(filepath.Dir(path), originalBase)
if err := os.RemoveAll(originalPath); err != nil {
return 0, err
return err
}
} else {
// If path exits we almost always just want to remove and replace it.
@@ -99,13 +112,12 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
if fi, err := os.Lstat(path); err == nil {
if !(fi.IsDir() && hdr.Typeflag == tar.TypeDir) {
if err := os.RemoveAll(path); err != nil {
return 0, err
return err
}
}
}
trBuf.Reset(tr)
srcData := io.Reader(trBuf)
srcData := io.Reader(tr)
srcHdr := hdr
// Hard links into /.wh..wh.plnk don't work, as we don't extract that directory, so
@@ -114,18 +126,18 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
linkBasename := filepath.Base(hdr.Linkname)
srcHdr = aufsHardlinks[linkBasename]
if srcHdr == nil {
return 0, fmt.Errorf("Invalid aufs hardlink")
return fmt.Errorf("Invalid aufs hardlink")
}
tmpFile, err := os.Open(filepath.Join(aufsTempdir, linkBasename))
if err != nil {
return 0, err
return err
}
defer tmpFile.Close()
srcData = tmpFile
}
if err := createTarFile(path, dest, srcHdr, srcData, true); err != nil {
return 0, err
if err := createTarFile(path, dest, srcHdr, srcData); err != nil {
return err
}
// Directory mtimes must be handled at the end to avoid further
@@ -140,29 +152,9 @@ func UnpackLayer(dest string, layer ArchiveReader) (size int64, err error) {
path := filepath.Join(dest, hdr.Name)
ts := []syscall.Timespec{timeToTimespec(hdr.AccessTime), timeToTimespec(hdr.ModTime)}
if err := syscall.UtimesNano(path, ts); err != nil {
return 0, err
return err
}
}
return size, nil
}
// ApplyLayer parses a diff in the standard layer format from `layer`, and
// applies it to the directory `dest`. Returns the size in bytes of the
// contents of the layer.
func ApplyLayer(dest string, layer ArchiveReader) (int64, error) {
dest = filepath.Clean(dest)
// We need to be able to set any perms
oldmask, err := system.Umask(0)
if err != nil {
return 0, err
}
defer system.Umask(oldmask) // ignore err, ErrNotSupportedPlatform
layer, err = DecompressStream(layer)
if err != nil {
return 0, err
}
return UnpackLayer(dest, layer)
return nil
}
+1 -1
View File
@@ -1,8 +1,8 @@
package archive
import (
"archive/tar"
"bytes"
"github.com/dotcloud/docker/vendor/src/code.google.com/p/go/src/pkg/archive/tar"
"io/ioutil"
)
+3
View File
@@ -0,0 +1,3 @@
Sam Alba <sam@dotcloud.com> (@samalba)
Joffrey Fuhrer <joffrey@dotcloud.com> (@shin-)
Ken Cochrane <ken@dotcloud.com> (@kencochrane)
+290
View File
@@ -0,0 +1,290 @@
package auth
import (
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"github.com/dotcloud/docker/utils"
"io/ioutil"
"net/http"
"os"
"path"
"strings"
)
// Where we store the config file
const CONFIGFILE = ".dockercfg"
// Only used for user auth + account creation
const INDEXSERVER = "https://index.docker.io/v1/"
//const INDEXSERVER = "https://indexstaging-docker.dotcloud.com/v1/"
var (
ErrConfigFileMissing = errors.New("The Auth config file is missing")
)
type AuthConfig struct {
Username string `json:"username,omitempty"`
Password string `json:"password,omitempty"`
Auth string `json:"auth"`
Email string `json:"email"`
ServerAddress string `json:"serveraddress,omitempty"`
}
type ConfigFile struct {
Configs map[string]AuthConfig `json:"configs,omitempty"`
rootPath string
}
func IndexServerAddress() string {
return INDEXSERVER
}
// create a base64 encoded auth string to store in config
func encodeAuth(authConfig *AuthConfig) string {
authStr := authConfig.Username + ":" + authConfig.Password
msg := []byte(authStr)
encoded := make([]byte, base64.StdEncoding.EncodedLen(len(msg)))
base64.StdEncoding.Encode(encoded, msg)
return string(encoded)
}
// decode the auth string
func decodeAuth(authStr string) (string, string, error) {
decLen := base64.StdEncoding.DecodedLen(len(authStr))
decoded := make([]byte, decLen)
authByte := []byte(authStr)
n, err := base64.StdEncoding.Decode(decoded, authByte)
if err != nil {
return "", "", err
}
if n > decLen {
return "", "", fmt.Errorf("Something went wrong decoding auth config")
}
arr := strings.SplitN(string(decoded), ":", 2)
if len(arr) != 2 {
return "", "", fmt.Errorf("Invalid auth configuration file")
}
password := strings.Trim(arr[1], "\x00")
return arr[0], password, nil
}
// load up the auth config information and return values
// FIXME: use the internal golang config parser
func LoadConfig(rootPath string) (*ConfigFile, error) {
configFile := ConfigFile{Configs: make(map[string]AuthConfig), rootPath: rootPath}
confFile := path.Join(rootPath, CONFIGFILE)
if _, err := os.Stat(confFile); err != nil {
return &configFile, nil //missing file is not an error
}
b, err := ioutil.ReadFile(confFile)
if err != nil {
return &configFile, err
}
if err := json.Unmarshal(b, &configFile.Configs); err != nil {
arr := strings.Split(string(b), "\n")
if len(arr) < 2 {
return &configFile, fmt.Errorf("The Auth config file is empty")
}
authConfig := AuthConfig{}
origAuth := strings.Split(arr[0], " = ")
if len(origAuth) != 2 {
return &configFile, fmt.Errorf("Invalid Auth config file")
}
authConfig.Username, authConfig.Password, err = decodeAuth(origAuth[1])
if err != nil {
return &configFile, err
}
origEmail := strings.Split(arr[1], " = ")
if len(origEmail) != 2 {
return &configFile, fmt.Errorf("Invalid Auth config file")
}
authConfig.Email = origEmail[1]
authConfig.ServerAddress = IndexServerAddress()
configFile.Configs[IndexServerAddress()] = authConfig
} else {
for k, authConfig := range configFile.Configs {
authConfig.Username, authConfig.Password, err = decodeAuth(authConfig.Auth)
if err != nil {
return &configFile, err
}
authConfig.Auth = ""
configFile.Configs[k] = authConfig
authConfig.ServerAddress = k
}
}
return &configFile, nil
}
// save the auth config
func SaveConfig(configFile *ConfigFile) error {
confFile := path.Join(configFile.rootPath, CONFIGFILE)
if len(configFile.Configs) == 0 {
os.Remove(confFile)
return nil
}
configs := make(map[string]AuthConfig, len(configFile.Configs))
for k, authConfig := range configFile.Configs {
authCopy := authConfig
authCopy.Auth = encodeAuth(&authCopy)
authCopy.Username = ""
authCopy.Password = ""
authCopy.ServerAddress = ""
configs[k] = authCopy
}
b, err := json.Marshal(configs)
if err != nil {
return err
}
err = ioutil.WriteFile(confFile, b, 0600)
if err != nil {
return err
}
return nil
}
// try to register/login to the registry server
func Login(authConfig *AuthConfig, factory *utils.HTTPRequestFactory) (string, error) {
var (
status string
reqBody []byte
err error
client = &http.Client{}
reqStatusCode = 0
serverAddress = authConfig.ServerAddress
)
if serverAddress == "" {
serverAddress = IndexServerAddress()
}
loginAgainstOfficialIndex := serverAddress == IndexServerAddress()
// to avoid sending the server address to the server it should be removed before being marshalled
authCopy := *authConfig
authCopy.ServerAddress = ""
jsonBody, err := json.Marshal(authCopy)
if err != nil {
return "", fmt.Errorf("Config Error: %s", err)
}
// using `bytes.NewReader(jsonBody)` here causes the server to respond with a 411 status.
b := strings.NewReader(string(jsonBody))
req1, err := http.Post(serverAddress+"users/", "application/json; charset=utf-8", b)
if err != nil {
return "", fmt.Errorf("Server Error: %s", err)
}
reqStatusCode = req1.StatusCode
defer req1.Body.Close()
reqBody, err = ioutil.ReadAll(req1.Body)
if err != nil {
return "", fmt.Errorf("Server Error: [%#v] %s", reqStatusCode, err)
}
if reqStatusCode == 201 {
if loginAgainstOfficialIndex {
status = "Account created. Please use the confirmation link we sent" +
" to your e-mail to activate it."
} else {
status = "Account created. Please see the documentation of the registry " + serverAddress + " for instructions how to activate it."
}
} else if reqStatusCode == 400 {
if string(reqBody) == "\"Username or email already exists\"" {
req, err := factory.NewRequest("GET", serverAddress+"users/", nil)
req.SetBasicAuth(authConfig.Username, authConfig.Password)
resp, err := client.Do(req)
if err != nil {
return "", err
}
defer resp.Body.Close()
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return "", err
}
if resp.StatusCode == 200 {
status = "Login Succeeded"
} else if resp.StatusCode == 401 {
return "", fmt.Errorf("Wrong login/password, please try again")
} else if resp.StatusCode == 403 {
if loginAgainstOfficialIndex {
return "", fmt.Errorf("Login: Account is not Active. Please check your e-mail for a confirmation link.")
}
return "", fmt.Errorf("Login: Account is not Active. Please see the documentation of the registry %s for instructions how to activate it.", serverAddress)
} else {
return "", fmt.Errorf("Login: %s (Code: %d; Headers: %s)", body, resp.StatusCode, resp.Header)
}
} else {
return "", fmt.Errorf("Registration: %s", reqBody)
}
} else if reqStatusCode == 401 {
// This case would happen with private registries where /v1/users is
// protected, so people can use `docker login` as an auth check.
req, err := factory.NewRequest("GET", serverAddress+"users/", nil)
req.SetBasicAuth(authConfig.Username, authConfig.Password)
resp, err := client.Do(req)
if err != nil {
return "", err
}
defer resp.Body.Close()
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return "", err
}
if resp.StatusCode == 200 {
status = "Login Succeeded"
} else if resp.StatusCode == 401 {
return "", fmt.Errorf("Wrong login/password, please try again")
} else {
return "", fmt.Errorf("Login: %s (Code: %d; Headers: %s)", body,
resp.StatusCode, resp.Header)
}
} else {
return "", fmt.Errorf("Unexpected status code [%d] : %s", reqStatusCode, reqBody)
}
return status, nil
}
// this method matches a auth configuration to a server address or a url
func (config *ConfigFile) ResolveAuthConfig(hostname string) AuthConfig {
if hostname == IndexServerAddress() || len(hostname) == 0 {
// default to the index server
return config.Configs[IndexServerAddress()]
}
// First try the happy case
if c, found := config.Configs[hostname]; found {
return c
}
convertToHostname := func(url string) string {
stripped := url
if strings.HasPrefix(url, "http://") {
stripped = strings.Replace(url, "http://", "", 1)
} else if strings.HasPrefix(url, "https://") {
stripped = strings.Replace(url, "https://", "", 1)
}
nameParts := strings.SplitN(stripped, "/", 2)
return nameParts[0]
}
// Maybe they have a legacy config file, we will iterate the keys converting
// them to the new format and testing
normalizedHostename := convertToHostname(hostname)
for registry, config := range config.Configs {
if registryHostname := convertToHostname(registry); registryHostname == normalizedHostename {
return config
}
}
// When all else fails, return an empty auth config
return AuthConfig{}
}
+36 -60
View File
@@ -1,20 +1,17 @@
package registry
package auth
import (
"io/ioutil"
"os"
"path/filepath"
"testing"
"github.com/docker/docker/cliconfig"
)
func TestEncodeAuth(t *testing.T) {
newAuthConfig := &cliconfig.AuthConfig{Username: "ken", Password: "test", Email: "test@example.com"}
authStr := cliconfig.EncodeAuth(newAuthConfig)
decAuthConfig := &cliconfig.AuthConfig{}
newAuthConfig := &AuthConfig{Username: "ken", Password: "test", Email: "test@example.com"}
authStr := encodeAuth(newAuthConfig)
decAuthConfig := &AuthConfig{}
var err error
decAuthConfig.Username, decAuthConfig.Password, err = cliconfig.DecodeAuth(authStr)
decAuthConfig.Username, decAuthConfig.Password, err = decodeAuth(authStr)
if err != nil {
t.Fatal(err)
}
@@ -29,16 +26,18 @@ func TestEncodeAuth(t *testing.T) {
}
}
func setupTempConfigFile() (*cliconfig.ConfigFile, error) {
func setupTempConfigFile() (*ConfigFile, error) {
root, err := ioutil.TempDir("", "docker-test-auth")
if err != nil {
return nil, err
}
root = filepath.Join(root, cliconfig.CONFIGFILE)
configFile := cliconfig.NewConfigFile(root)
configFile := &ConfigFile{
rootPath: root,
Configs: make(map[string]AuthConfig),
}
for _, registry := range []string{"testIndex", IndexServerAddress()} {
configFile.AuthConfigs[registry] = cliconfig.AuthConfig{
configFile.Configs[registry] = AuthConfig{
Username: "docker-user",
Password: "docker-pass",
Email: "docker@docker.io",
@@ -53,14 +52,14 @@ func TestSameAuthDataPostSave(t *testing.T) {
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(configFile.Filename())
defer os.RemoveAll(configFile.rootPath)
err = configFile.Save()
err = SaveConfig(configFile)
if err != nil {
t.Fatal(err)
}
authConfig := configFile.AuthConfigs["testIndex"]
authConfig := configFile.Configs["testIndex"]
if authConfig.Username != "docker-user" {
t.Fail()
}
@@ -80,22 +79,14 @@ func TestResolveAuthConfigIndexServer(t *testing.T) {
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(configFile.Filename())
defer os.RemoveAll(configFile.rootPath)
indexConfig := configFile.AuthConfigs[IndexServerAddress()]
officialIndex := &IndexInfo{
Official: true,
for _, registry := range []string{"", IndexServerAddress()} {
resolved := configFile.ResolveAuthConfig(registry)
if resolved != configFile.Configs[IndexServerAddress()] {
t.Fail()
}
}
privateIndex := &IndexInfo{
Official: false,
}
resolved := ResolveAuthConfig(configFile, officialIndex)
assertEqual(t, resolved, indexConfig, "Expected ResolveAuthConfig to return IndexServerAddress()")
resolved = ResolveAuthConfig(configFile, privateIndex)
assertNotEqual(t, resolved, indexConfig, "Expected ResolveAuthConfig to not return IndexServerAddress()")
}
func TestResolveAuthConfigFullURL(t *testing.T) {
@@ -103,39 +94,30 @@ func TestResolveAuthConfigFullURL(t *testing.T) {
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(configFile.Filename())
defer os.RemoveAll(configFile.rootPath)
registryAuth := cliconfig.AuthConfig{
registryAuth := AuthConfig{
Username: "foo-user",
Password: "foo-pass",
Email: "foo@example.com",
}
localAuth := cliconfig.AuthConfig{
localAuth := AuthConfig{
Username: "bar-user",
Password: "bar-pass",
Email: "bar@example.com",
}
officialAuth := cliconfig.AuthConfig{
Username: "baz-user",
Password: "baz-pass",
Email: "baz@example.com",
}
configFile.AuthConfigs[IndexServerAddress()] = officialAuth
expectedAuths := map[string]cliconfig.AuthConfig{
"registry.example.com": registryAuth,
"localhost:8000": localAuth,
"registry.com": localAuth,
}
configFile.Configs["https://registry.example.com/v1/"] = registryAuth
configFile.Configs["http://localhost:8000/v1/"] = localAuth
configFile.Configs["registry.com"] = registryAuth
validRegistries := map[string][]string{
"registry.example.com": {
"https://registry.example.com/v1/": {
"https://registry.example.com/v1/",
"http://registry.example.com/v1/",
"registry.example.com",
"registry.example.com/v1/",
},
"localhost:8000": {
"http://localhost:8000/v1/": {
"https://localhost:8000/v1/",
"http://localhost:8000/v1/",
"localhost:8000",
@@ -150,24 +132,18 @@ func TestResolveAuthConfigFullURL(t *testing.T) {
}
for configKey, registries := range validRegistries {
configured, ok := expectedAuths[configKey]
if !ok || configured.Email == "" {
t.Fail()
}
index := &IndexInfo{
Name: configKey,
}
for _, registry := range registries {
configFile.AuthConfigs[registry] = configured
resolved := ResolveAuthConfig(configFile, index)
var (
configured AuthConfig
ok bool
)
resolved := configFile.ResolveAuthConfig(registry)
if configured, ok = configFile.Configs[configKey]; !ok {
t.Fail()
}
if resolved.Email != configured.Email {
t.Errorf("%s -> %q != %q\n", registry, resolved.Email, configured.Email)
}
delete(configFile.AuthConfigs, registry)
resolved = ResolveAuthConfig(configFile, index)
if resolved.Email == configured.Email {
t.Errorf("%s -> %q == %q\n", registry, resolved.Email, configured.Email)
}
}
}
}
-155
View File
@@ -1,155 +0,0 @@
package builder
import (
"fmt"
"strings"
)
type FlagType int
const (
boolType FlagType = iota
stringType
)
type BuilderFlags struct {
Args []string // actual flags/args from cmd line
flags map[string]*Flag
used map[string]*Flag
Err error
}
type Flag struct {
bf *BuilderFlags
name string
flagType FlagType
Value string
}
func NewBuilderFlags() *BuilderFlags {
return &BuilderFlags{
flags: make(map[string]*Flag),
used: make(map[string]*Flag),
}
}
func (bf *BuilderFlags) AddBool(name string, def bool) *Flag {
flag := bf.addFlag(name, boolType)
if flag == nil {
return nil
}
if def {
flag.Value = "true"
} else {
flag.Value = "false"
}
return flag
}
func (bf *BuilderFlags) AddString(name string, def string) *Flag {
flag := bf.addFlag(name, stringType)
if flag == nil {
return nil
}
flag.Value = def
return flag
}
func (bf *BuilderFlags) addFlag(name string, flagType FlagType) *Flag {
if _, ok := bf.flags[name]; ok {
bf.Err = fmt.Errorf("Duplicate flag defined: %s", name)
return nil
}
newFlag := &Flag{
bf: bf,
name: name,
flagType: flagType,
}
bf.flags[name] = newFlag
return newFlag
}
func (fl *Flag) IsUsed() bool {
if _, ok := fl.bf.used[fl.name]; ok {
return true
}
return false
}
func (fl *Flag) IsTrue() bool {
if fl.flagType != boolType {
// Should never get here
panic(fmt.Errorf("Trying to use IsTrue on a non-boolean: %s", fl.name))
}
return fl.Value == "true"
}
func (bf *BuilderFlags) Parse() error {
// If there was an error while defining the possible flags
// go ahead and bubble it back up here since we didn't do it
// earlier in the processing
if bf.Err != nil {
return fmt.Errorf("Error setting up flags: %s", bf.Err)
}
for _, arg := range bf.Args {
if !strings.HasPrefix(arg, "--") {
return fmt.Errorf("Arg should start with -- : %s", arg)
}
if arg == "--" {
return nil
}
arg = arg[2:]
value := ""
index := strings.Index(arg, "=")
if index >= 0 {
value = arg[index+1:]
arg = arg[:index]
}
flag, ok := bf.flags[arg]
if !ok {
return fmt.Errorf("Unknown flag: %s", arg)
}
if _, ok = bf.used[arg]; ok {
return fmt.Errorf("Duplicate flag specified: %s", arg)
}
bf.used[arg] = flag
switch flag.flagType {
case boolType:
// value == "" is only ok if no "=" was specified
if index >= 0 && value == "" {
return fmt.Errorf("Missing a value on flag: %s", arg)
}
lower := strings.ToLower(value)
if lower == "" {
flag.Value = "true"
} else if lower == "true" || lower == "false" {
flag.Value = lower
} else {
return fmt.Errorf("Expecting boolean value for flag %s, not: %s", arg, value)
}
case stringType:
if index < 0 {
return fmt.Errorf("Missing a value on flag: %s", arg)
}
flag.Value = value
default:
panic(fmt.Errorf("No idea what kind of flag we have! Should never get here!"))
}
}
return nil
}
-187
View File
@@ -1,187 +0,0 @@
package builder
import (
"testing"
)
func TestBuilderFlags(t *testing.T) {
var expected string
var err error
// ---
bf := NewBuilderFlags()
bf.Args = []string{}
if err := bf.Parse(); err != nil {
t.Fatalf("Test1 of %q was supposed to work: %s", bf.Args, err)
}
// ---
bf = NewBuilderFlags()
bf.Args = []string{"--"}
if err := bf.Parse(); err != nil {
t.Fatalf("Test2 of %q was supposed to work: %s", bf.Args, err)
}
// ---
bf = NewBuilderFlags()
flStr1 := bf.AddString("str1", "")
flBool1 := bf.AddBool("bool1", false)
bf.Args = []string{}
if err = bf.Parse(); err != nil {
t.Fatalf("Test3 of %q was supposed to work: %s", bf.Args, err)
}
if flStr1.IsUsed() == true {
t.Fatalf("Test3 - str1 was not used!")
}
if flBool1.IsUsed() == true {
t.Fatalf("Test3 - bool1 was not used!")
}
// ---
bf = NewBuilderFlags()
flStr1 = bf.AddString("str1", "HI")
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{}
if err = bf.Parse(); err != nil {
t.Fatalf("Test4 of %q was supposed to work: %s", bf.Args, err)
}
if flStr1.Value != "HI" {
t.Fatalf("Str1 was supposed to default to: HI")
}
if flBool1.IsTrue() {
t.Fatalf("Bool1 was supposed to default to: false")
}
if flStr1.IsUsed() == true {
t.Fatalf("Str1 was not used!")
}
if flBool1.IsUsed() == true {
t.Fatalf("Bool1 was not used!")
}
// ---
bf = NewBuilderFlags()
flStr1 = bf.AddString("str1", "HI")
bf.Args = []string{"--str1"}
if err = bf.Parse(); err == nil {
t.Fatalf("Test %q was supposed to fail", bf.Args)
}
// ---
bf = NewBuilderFlags()
flStr1 = bf.AddString("str1", "HI")
bf.Args = []string{"--str1="}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
expected = ""
if flStr1.Value != expected {
t.Fatalf("Str1 (%q) should be: %q", flStr1.Value, expected)
}
// ---
bf = NewBuilderFlags()
flStr1 = bf.AddString("str1", "HI")
bf.Args = []string{"--str1=BYE"}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
expected = "BYE"
if flStr1.Value != expected {
t.Fatalf("Str1 (%q) should be: %q", flStr1.Value, expected)
}
// ---
bf = NewBuilderFlags()
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool1"}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
if !flBool1.IsTrue() {
t.Fatalf("Test-b1 Bool1 was supposed to be true")
}
// ---
bf = NewBuilderFlags()
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool1=true"}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
if !flBool1.IsTrue() {
t.Fatalf("Test-b2 Bool1 was supposed to be true")
}
// ---
bf = NewBuilderFlags()
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool1=false"}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
if flBool1.IsTrue() {
t.Fatalf("Test-b3 Bool1 was supposed to be false")
}
// ---
bf = NewBuilderFlags()
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool1=false1"}
if err = bf.Parse(); err == nil {
t.Fatalf("Test %q was supposed to fail", bf.Args)
}
// ---
bf = NewBuilderFlags()
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool2"}
if err = bf.Parse(); err == nil {
t.Fatalf("Test %q was supposed to fail", bf.Args)
}
// ---
bf = NewBuilderFlags()
flStr1 = bf.AddString("str1", "HI")
flBool1 = bf.AddBool("bool1", false)
bf.Args = []string{"--bool1", "--str1=BYE"}
if err = bf.Parse(); err != nil {
t.Fatalf("Test %q was supposed to work: %s", bf.Args, err)
}
if flStr1.Value != "BYE" {
t.Fatalf("Teset %s, str1 should be BYE", bf.Args)
}
if !flBool1.IsTrue() {
t.Fatalf("Teset %s, bool1 should be true", bf.Args)
}
}
-37
View File
@@ -1,37 +0,0 @@
// This package contains the set of Dockerfile commands.
package command
const (
Env = "env"
Label = "label"
Maintainer = "maintainer"
Add = "add"
Copy = "copy"
From = "from"
Onbuild = "onbuild"
Workdir = "workdir"
Run = "run"
Cmd = "cmd"
Entrypoint = "entrypoint"
Expose = "expose"
Volume = "volume"
User = "user"
)
// Commands is list of all Dockerfile commands
var Commands = map[string]struct{}{
Env: {},
Label: {},
Maintainer: {},
Add: {},
Copy: {},
From: {},
Onbuild: {},
Workdir: {},
Run: {},
Cmd: {},
Entrypoint: {},
Expose: {},
Volume: {},
User: {},
}
-538
View File
@@ -1,538 +0,0 @@
package builder
// This file contains the dispatchers for each command. Note that
// `nullDispatch` is not actually a command, but support for commands we parse
// but do nothing with.
//
// See evaluator.go for a higher level discussion of the whole evaluator
// package.
import (
"fmt"
"io/ioutil"
"path/filepath"
"regexp"
"runtime"
"sort"
"strings"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/nat"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/runconfig"
)
const (
// NoBaseImageSpecifier is the symbol used by the FROM
// command to specify that no base image is to be used.
NoBaseImageSpecifier string = "scratch"
)
// dispatch with no layer / parsing. This is effectively not a command.
func nullDispatch(b *Builder, args []string, attributes map[string]bool, original string) error {
return nil
}
// ENV foo bar
//
// Sets the environment variable foo to bar, also makes interpolation
// in the dockerfile available from the next statement on via ${foo}.
//
func env(b *Builder, args []string, attributes map[string]bool, original string) error {
if runtime.GOOS == "windows" {
return fmt.Errorf("ENV is not supported on Windows.")
}
if len(args) == 0 {
return fmt.Errorf("ENV requires at least one argument")
}
if len(args)%2 != 0 {
// should never get here, but just in case
return fmt.Errorf("Bad input to ENV, too many args")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
// TODO/FIXME/NOT USED
// Just here to show how to use the builder flags stuff within the
// context of a builder command. Will remove once we actually add
// a builder command to something!
/*
flBool1 := b.BuilderFlags.AddBool("bool1", false)
flStr1 := b.BuilderFlags.AddString("str1", "HI")
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
fmt.Printf("Bool1:%v\n", flBool1)
fmt.Printf("Str1:%v\n", flStr1)
*/
commitStr := "ENV"
for j := 0; j < len(args); j++ {
// name ==> args[j]
// value ==> args[j+1]
newVar := args[j] + "=" + args[j+1] + ""
commitStr += " " + newVar
gotOne := false
for i, envVar := range b.Config.Env {
envParts := strings.SplitN(envVar, "=", 2)
if envParts[0] == args[j] {
b.Config.Env[i] = newVar
gotOne = true
break
}
}
if !gotOne {
b.Config.Env = append(b.Config.Env, newVar)
}
j++
}
return b.commit("", b.Config.Cmd, commitStr)
}
// MAINTAINER some text <maybe@an.email.address>
//
// Sets the maintainer metadata.
func maintainer(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) != 1 {
return fmt.Errorf("MAINTAINER requires exactly one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
b.maintainer = args[0]
return b.commit("", b.Config.Cmd, fmt.Sprintf("MAINTAINER %s", b.maintainer))
}
// LABEL some json data describing the image
//
// Sets the Label variable foo to bar,
//
func label(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) == 0 {
return fmt.Errorf("LABEL requires at least one argument")
}
if len(args)%2 != 0 {
// should never get here, but just in case
return fmt.Errorf("Bad input to LABEL, too many args")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
commitStr := "LABEL"
if b.Config.Labels == nil {
b.Config.Labels = map[string]string{}
}
for j := 0; j < len(args); j++ {
// name ==> args[j]
// value ==> args[j+1]
newVar := args[j] + "=" + args[j+1] + ""
commitStr += " " + newVar
b.Config.Labels[args[j]] = args[j+1]
j++
}
return b.commit("", b.Config.Cmd, commitStr)
}
// ADD foo /path
//
// Add the file 'foo' to '/path'. Tarball and Remote URL (git, http) handling
// exist here. If you do not wish to have this automatic handling, use COPY.
//
func add(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) < 2 {
return fmt.Errorf("ADD requires at least two arguments")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
return b.runContextCommand(args, true, true, "ADD")
}
// COPY foo /path
//
// Same as 'ADD' but without the tar and remote url handling.
//
func dispatchCopy(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) < 2 {
return fmt.Errorf("COPY requires at least two arguments")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
return b.runContextCommand(args, false, false, "COPY")
}
// FROM imagename
//
// This sets the image the dockerfile will build on top of.
//
func from(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) != 1 {
return fmt.Errorf("FROM requires one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
name := args[0]
if name == NoBaseImageSpecifier {
b.image = ""
b.noBaseImage = true
return nil
}
image, err := b.Daemon.Repositories().LookupImage(name)
if b.Pull {
image, err = b.pullImage(name)
if err != nil {
return err
}
}
if err != nil {
if b.Daemon.Graph().IsNotExist(err, name) {
image, err = b.pullImage(name)
}
// note that the top level err will still be !nil here if IsNotExist is
// not the error. This approach just simplifies the logic a bit.
if err != nil {
return err
}
}
return b.processImageFrom(image)
}
// ONBUILD RUN echo yo
//
// ONBUILD triggers run when the image is used in a FROM statement.
//
// ONBUILD handling has a lot of special-case functionality, the heading in
// evaluator.go and comments around dispatch() in the same file explain the
// special cases. search for 'OnBuild' in internals.go for additional special
// cases.
//
func onbuild(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) == 0 {
return fmt.Errorf("ONBUILD requires at least one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
triggerInstruction := strings.ToUpper(strings.TrimSpace(args[0]))
switch triggerInstruction {
case "ONBUILD":
return fmt.Errorf("Chaining ONBUILD via `ONBUILD ONBUILD` isn't allowed")
case "MAINTAINER", "FROM":
return fmt.Errorf("%s isn't allowed as an ONBUILD trigger", triggerInstruction)
}
original = regexp.MustCompile(`(?i)^\s*ONBUILD\s*`).ReplaceAllString(original, "")
b.Config.OnBuild = append(b.Config.OnBuild, original)
return b.commit("", b.Config.Cmd, fmt.Sprintf("ONBUILD %s", original))
}
// WORKDIR /tmp
//
// Set the working directory for future RUN/CMD/etc statements.
//
func workdir(b *Builder, args []string, attributes map[string]bool, original string) error {
if len(args) != 1 {
return fmt.Errorf("WORKDIR requires exactly one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
workdir := args[0]
if !filepath.IsAbs(workdir) {
workdir = filepath.Join("/", b.Config.WorkingDir, workdir)
}
b.Config.WorkingDir = workdir
return b.commit("", b.Config.Cmd, fmt.Sprintf("WORKDIR %v", workdir))
}
// RUN some command yo
//
// run a command and commit the image. Args are automatically prepended with
// 'sh -c' under linux or 'cmd /S /C' under Windows, in the event there is
// only one argument. The difference in processing:
//
// RUN echo hi # sh -c echo hi (Linux)
// RUN echo hi # cmd /S /C echo hi (Windows)
// RUN [ "echo", "hi" ] # echo hi
//
func run(b *Builder, args []string, attributes map[string]bool, original string) error {
if b.image == "" && !b.noBaseImage {
return fmt.Errorf("Please provide a source image with `from` prior to run")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
args = handleJsonArgs(args, attributes)
if !attributes["json"] {
if runtime.GOOS != "windows" {
args = append([]string{"/bin/sh", "-c"}, args...)
} else {
args = append([]string{"cmd", "/S /C"}, args...)
}
}
runCmd := flag.NewFlagSet("run", flag.ContinueOnError)
runCmd.SetOutput(ioutil.Discard)
runCmd.Usage = nil
config, _, _, err := runconfig.Parse(runCmd, append([]string{b.image}, args...))
if err != nil {
return err
}
cmd := b.Config.Cmd
// set Cmd manually, this is special case only for Dockerfiles
b.Config.Cmd = config.Cmd
runconfig.Merge(b.Config, config)
defer func(cmd *runconfig.Command) { b.Config.Cmd = cmd }(cmd)
logrus.Debugf("[BUILDER] Command to be executed: %v", b.Config.Cmd)
hit, err := b.probeCache()
if err != nil {
return err
}
if hit {
return nil
}
c, err := b.create()
if err != nil {
return err
}
// Ensure that we keep the container mounted until the commit
// to avoid unmounting and then mounting directly again
c.Mount()
defer c.Unmount()
err = b.run(c)
if err != nil {
return err
}
if err := b.commit(c.ID, cmd, "run"); err != nil {
return err
}
return nil
}
// CMD foo
//
// Set the default command to run in the container (which may be empty).
// Argument handling is the same as RUN.
//
func cmd(b *Builder, args []string, attributes map[string]bool, original string) error {
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
cmdSlice := handleJsonArgs(args, attributes)
if !attributes["json"] {
if runtime.GOOS != "windows" {
cmdSlice = append([]string{"/bin/sh", "-c"}, cmdSlice...)
} else {
cmdSlice = append([]string{"cmd", "/S /C"}, cmdSlice...)
}
}
b.Config.Cmd = runconfig.NewCommand(cmdSlice...)
if err := b.commit("", b.Config.Cmd, fmt.Sprintf("CMD %q", cmdSlice)); err != nil {
return err
}
if len(args) != 0 {
b.cmdSet = true
}
return nil
}
// ENTRYPOINT /usr/sbin/nginx
//
// Set the entrypoint (which defaults to sh -c on linux, or cmd /S /C on Windows) to
// /usr/sbin/nginx. Will accept the CMD as the arguments to /usr/sbin/nginx.
//
// Handles command processing similar to CMD and RUN, only b.Config.Entrypoint
// is initialized at NewBuilder time instead of through argument parsing.
//
func entrypoint(b *Builder, args []string, attributes map[string]bool, original string) error {
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
parsed := handleJsonArgs(args, attributes)
switch {
case attributes["json"]:
// ENTRYPOINT ["echo", "hi"]
b.Config.Entrypoint = runconfig.NewEntrypoint(parsed...)
case len(parsed) == 0:
// ENTRYPOINT []
b.Config.Entrypoint = nil
default:
// ENTRYPOINT echo hi
if runtime.GOOS != "windows" {
b.Config.Entrypoint = runconfig.NewEntrypoint("/bin/sh", "-c", parsed[0])
} else {
b.Config.Entrypoint = runconfig.NewEntrypoint("cmd", "/S /C", parsed[0])
}
}
// when setting the entrypoint if a CMD was not explicitly set then
// set the command to nil
if !b.cmdSet {
b.Config.Cmd = nil
}
if err := b.commit("", b.Config.Cmd, fmt.Sprintf("ENTRYPOINT %q", b.Config.Entrypoint)); err != nil {
return err
}
return nil
}
// EXPOSE 6667/tcp 7000/tcp
//
// Expose ports for links and port mappings. This all ends up in
// b.Config.ExposedPorts for runconfig.
//
func expose(b *Builder, args []string, attributes map[string]bool, original string) error {
portsTab := args
if len(args) == 0 {
return fmt.Errorf("EXPOSE requires at least one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
if b.Config.ExposedPorts == nil {
b.Config.ExposedPorts = make(nat.PortSet)
}
ports, bindingMap, err := nat.ParsePortSpecs(append(portsTab, b.Config.PortSpecs...))
if err != nil {
return err
}
for _, bindings := range bindingMap {
if bindings[0].HostIp != "" || bindings[0].HostPort != "" {
fmt.Fprintf(b.ErrStream, " ---> Using Dockerfile's EXPOSE instruction"+
" to map host ports to container ports (ip:hostPort:containerPort) is deprecated.\n"+
" Please use -p to publish the ports.\n")
}
}
// instead of using ports directly, we build a list of ports and sort it so
// the order is consistent. This prevents cache burst where map ordering
// changes between builds
portList := make([]string, len(ports))
var i int
for port := range ports {
if _, exists := b.Config.ExposedPorts[port]; !exists {
b.Config.ExposedPorts[port] = struct{}{}
}
portList[i] = string(port)
i++
}
sort.Strings(portList)
b.Config.PortSpecs = nil
return b.commit("", b.Config.Cmd, fmt.Sprintf("EXPOSE %s", strings.Join(portList, " ")))
}
// USER foo
//
// Set the user to 'foo' for future commands and when running the
// ENTRYPOINT/CMD at container run time.
//
func user(b *Builder, args []string, attributes map[string]bool, original string) error {
if runtime.GOOS == "windows" {
return fmt.Errorf("USER is not supported on Windows.")
}
if len(args) != 1 {
return fmt.Errorf("USER requires exactly one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
b.Config.User = args[0]
return b.commit("", b.Config.Cmd, fmt.Sprintf("USER %v", args))
}
// VOLUME /foo
//
// Expose the volume /foo for use. Will also accept the JSON array form.
//
func volume(b *Builder, args []string, attributes map[string]bool, original string) error {
if runtime.GOOS == "windows" {
return fmt.Errorf("VOLUME is not supported on Windows.")
}
if len(args) == 0 {
return fmt.Errorf("VOLUME requires at least one argument")
}
if err := b.BuilderFlags.Parse(); err != nil {
return err
}
if b.Config.Volumes == nil {
b.Config.Volumes = map[string]struct{}{}
}
for _, v := range args {
v = strings.TrimSpace(v)
if v == "" {
return fmt.Errorf("Volume specified can not be an empty string")
}
b.Config.Volumes[v] = struct{}{}
}
if err := b.commit("", b.Config.Cmd, fmt.Sprintf("VOLUME %v", args)); err != nil {
return err
}
return nil
}
-346
View File
@@ -1,346 +0,0 @@
// Package builder is the evaluation step in the Dockerfile parse/evaluate pipeline.
//
// It incorporates a dispatch table based on the parser.Node values (see the
// parser package for more information) that are yielded from the parser itself.
// Calling NewBuilder with the BuildOpts struct can be used to customize the
// experience for execution purposes only. Parsing is controlled in the parser
// package, and this division of resposibility should be respected.
//
// Please see the jump table targets for the actual invocations, most of which
// will call out to the functions in internals.go to deal with their tasks.
//
// ONBUILD is a special case, which is covered in the onbuild() func in
// dispatchers.go.
//
// The evaluator uses the concept of "steps", which are usually each processable
// line in the Dockerfile. Each step is numbered and certain actions are taken
// before and after each step, such as creating an image ID and removing temporary
// containers and images. Note that ONBUILD creates a kinda-sorta "sub run" which
// includes its own set of steps (usually only one of them).
package builder
import (
"fmt"
"io"
"os"
"path/filepath"
"strings"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/api"
"github.com/docker/docker/builder/command"
"github.com/docker/docker/builder/parser"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/daemon"
"github.com/docker/docker/pkg/fileutils"
"github.com/docker/docker/pkg/streamformatter"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/symlink"
"github.com/docker/docker/pkg/tarsum"
"github.com/docker/docker/runconfig"
"github.com/docker/docker/utils"
)
// Environment variable interpolation will happen on these statements only.
var replaceEnvAllowed = map[string]struct{}{
command.Env: {},
command.Label: {},
command.Add: {},
command.Copy: {},
command.Workdir: {},
command.Expose: {},
command.Volume: {},
command.User: {},
}
var evaluateTable map[string]func(*Builder, []string, map[string]bool, string) error
func init() {
evaluateTable = map[string]func(*Builder, []string, map[string]bool, string) error{
command.Env: env,
command.Label: label,
command.Maintainer: maintainer,
command.Add: add,
command.Copy: dispatchCopy, // copy() is a go builtin
command.From: from,
command.Onbuild: onbuild,
command.Workdir: workdir,
command.Run: run,
command.Cmd: cmd,
command.Entrypoint: entrypoint,
command.Expose: expose,
command.Volume: volume,
command.User: user,
}
}
// internal struct, used to maintain configuration of the Dockerfile's
// processing as it evaluates the parsing result.
type Builder struct {
Daemon *daemon.Daemon
// effectively stdio for the run. Because it is not stdio, I said
// "Effectively". Do not use stdio anywhere in this package for any reason.
OutStream io.Writer
ErrStream io.Writer
Verbose bool
UtilizeCache bool
cacheBusted bool
// controls how images and containers are handled between steps.
Remove bool
ForceRemove bool
Pull bool
// set this to true if we want the builder to not commit between steps.
// This is useful when we only want to use the evaluator table to generate
// the final configs of the Dockerfile but dont want the layers
disableCommit bool
AuthConfig *cliconfig.AuthConfig
ConfigFile *cliconfig.ConfigFile
// Deprecated, original writer used for ImagePull. To be removed.
OutOld io.Writer
StreamFormatter *streamformatter.StreamFormatter
Config *runconfig.Config // runconfig for cmd, run, entrypoint etc.
// both of these are controlled by the Remove and ForceRemove options in BuildOpts
TmpContainers map[string]struct{} // a map of containers used for removes
dockerfileName string // name of Dockerfile
dockerfile *parser.Node // the syntax tree of the dockerfile
image string // image name for commit processing
maintainer string // maintainer name. could probably be removed.
cmdSet bool // indicates is CMD was set in current Dockerfile
BuilderFlags *BuilderFlags // current cmd's BuilderFlags - temporary
context tarsum.TarSum // the context is a tarball that is uploaded by the client
contextPath string // the path of the temporary directory the local context is unpacked to (server side)
noBaseImage bool // indicates that this build does not start from any base image, but is being built from an empty file system.
// Set resource restrictions for build containers
cpuSetCpus string
cpuSetMems string
cpuShares int64
cpuPeriod int64
cpuQuota int64
cgroupParent string
memory int64
memorySwap int64
cancelled <-chan struct{} // When closed, job was cancelled.
}
// Run the builder with the context. This is the lynchpin of this package. This
// will (barring errors):
//
// * call readContext() which will set up the temporary directory and unpack
// the context into it.
// * read the dockerfile
// * parse the dockerfile
// * walk the parse tree and execute it by dispatching to handlers. If Remove
// or ForceRemove is set, additional cleanup around containers happens after
// processing.
// * Print a happy message and return the image ID.
//
func (b *Builder) Run(context io.Reader) (string, error) {
if err := b.readContext(context); err != nil {
return "", err
}
defer func() {
if err := os.RemoveAll(b.contextPath); err != nil {
logrus.Debugf("[BUILDER] failed to remove temporary context: %s", err)
}
}()
if err := b.readDockerfile(); err != nil {
return "", err
}
// some initializations that would not have been supplied by the caller.
b.Config = &runconfig.Config{}
b.TmpContainers = map[string]struct{}{}
for i, n := range b.dockerfile.Children {
select {
case <-b.cancelled:
logrus.Debug("Builder: build cancelled!")
fmt.Fprintf(b.OutStream, "Build cancelled")
return "", fmt.Errorf("Build cancelled")
default:
// Not cancelled yet, keep going...
}
if err := b.dispatch(i, n); err != nil {
if b.ForceRemove {
b.clearTmp()
}
return "", err
}
fmt.Fprintf(b.OutStream, " ---> %s\n", stringid.TruncateID(b.image))
if b.Remove {
b.clearTmp()
}
}
if b.image == "" {
return "", fmt.Errorf("No image was generated. Is your Dockerfile empty?")
}
fmt.Fprintf(b.OutStream, "Successfully built %s\n", stringid.TruncateID(b.image))
return b.image, nil
}
// Reads a Dockerfile from the current context. It assumes that the
// 'filename' is a relative path from the root of the context
func (b *Builder) readDockerfile() error {
// If no -f was specified then look for 'Dockerfile'. If we can't find
// that then look for 'dockerfile'. If neither are found then default
// back to 'Dockerfile' and use that in the error message.
if b.dockerfileName == "" {
b.dockerfileName = api.DefaultDockerfileName
tmpFN := filepath.Join(b.contextPath, api.DefaultDockerfileName)
if _, err := os.Lstat(tmpFN); err != nil {
tmpFN = filepath.Join(b.contextPath, strings.ToLower(api.DefaultDockerfileName))
if _, err := os.Lstat(tmpFN); err == nil {
b.dockerfileName = strings.ToLower(api.DefaultDockerfileName)
}
}
}
origFile := b.dockerfileName
filename, err := symlink.FollowSymlinkInScope(filepath.Join(b.contextPath, origFile), b.contextPath)
if err != nil {
return fmt.Errorf("The Dockerfile (%s) must be within the build context", origFile)
}
fi, err := os.Lstat(filename)
if os.IsNotExist(err) {
return fmt.Errorf("Cannot locate specified Dockerfile: %s", origFile)
}
if fi.Size() == 0 {
return fmt.Errorf("The Dockerfile (%s) cannot be empty", origFile)
}
f, err := os.Open(filename)
if err != nil {
return err
}
b.dockerfile, err = parser.Parse(f)
f.Close()
if err != nil {
return err
}
// After the Dockerfile has been parsed, we need to check the .dockerignore
// file for either "Dockerfile" or ".dockerignore", and if either are
// present then erase them from the build context. These files should never
// have been sent from the client but we did send them to make sure that
// we had the Dockerfile to actually parse, and then we also need the
// .dockerignore file to know whether either file should be removed.
// Note that this assumes the Dockerfile has been read into memory and
// is now safe to be removed.
excludes, _ := utils.ReadDockerIgnore(filepath.Join(b.contextPath, ".dockerignore"))
if rm, _ := fileutils.Matches(".dockerignore", excludes); rm == true {
os.Remove(filepath.Join(b.contextPath, ".dockerignore"))
b.context.(tarsum.BuilderContext).Remove(".dockerignore")
}
if rm, _ := fileutils.Matches(b.dockerfileName, excludes); rm == true {
os.Remove(filepath.Join(b.contextPath, b.dockerfileName))
b.context.(tarsum.BuilderContext).Remove(b.dockerfileName)
}
return nil
}
// This method is the entrypoint to all statement handling routines.
//
// Almost all nodes will have this structure:
// Child[Node, Node, Node] where Child is from parser.Node.Children and each
// node comes from parser.Node.Next. This forms a "line" with a statement and
// arguments and we process them in this normalized form by hitting
// evaluateTable with the leaf nodes of the command and the Builder object.
//
// ONBUILD is a special case; in this case the parser will emit:
// Child[Node, Child[Node, Node...]] where the first node is the literal
// "onbuild" and the child entrypoint is the command of the ONBUILD statmeent,
// such as `RUN` in ONBUILD RUN foo. There is special case logic in here to
// deal with that, at least until it becomes more of a general concern with new
// features.
func (b *Builder) dispatch(stepN int, ast *parser.Node) error {
cmd := ast.Value
attrs := ast.Attributes
original := ast.Original
flags := ast.Flags
strs := []string{}
msg := fmt.Sprintf("Step %d : %s", stepN, strings.ToUpper(cmd))
if len(ast.Flags) > 0 {
msg += " " + strings.Join(ast.Flags, " ")
}
if cmd == "onbuild" {
if ast.Next == nil {
return fmt.Errorf("ONBUILD requires at least one argument")
}
ast = ast.Next.Children[0]
strs = append(strs, ast.Value)
msg += " " + ast.Value
if len(ast.Flags) > 0 {
msg += " " + strings.Join(ast.Flags, " ")
}
}
// count the number of nodes that we are going to traverse first
// so we can pre-create the argument and message array. This speeds up the
// allocation of those list a lot when they have a lot of arguments
cursor := ast
var n int
for cursor.Next != nil {
cursor = cursor.Next
n++
}
l := len(strs)
strList := make([]string, n+l)
copy(strList, strs)
msgList := make([]string, n)
var i int
for ast.Next != nil {
ast = ast.Next
var str string
str = ast.Value
if _, ok := replaceEnvAllowed[cmd]; ok {
var err error
str, err = ProcessWord(ast.Value, b.Config.Env)
if err != nil {
return err
}
}
strList[i+l] = str
msgList[i] = ast.Value
i++
}
msg += " " + strings.Join(msgList, " ")
fmt.Fprintln(b.OutStream, msg)
// XXX yes, we skip any cmds that are not valid; the parser should have
// picked these out already.
if f, ok := evaluateTable[cmd]; ok {
b.BuilderFlags = NewBuilderFlags()
b.BuilderFlags.Args = flags
return f(b, strList, attrs, original)
}
return fmt.Errorf("Unknown instruction: %s", strings.ToUpper(cmd))
}
-778
View File
@@ -1,778 +0,0 @@
package builder
// internals for handling commands. Covers many areas and a lot of
// non-contiguous functionality. Please read the comments.
import (
"crypto/sha256"
"encoding/hex"
"fmt"
"io"
"io/ioutil"
"net/http"
"net/url"
"os"
"path"
"path/filepath"
"sort"
"strings"
"syscall"
"time"
"github.com/Sirupsen/logrus"
"github.com/docker/docker/builder/parser"
"github.com/docker/docker/daemon"
"github.com/docker/docker/graph"
imagepkg "github.com/docker/docker/image"
"github.com/docker/docker/pkg/archive"
"github.com/docker/docker/pkg/chrootarchive"
"github.com/docker/docker/pkg/httputils"
"github.com/docker/docker/pkg/ioutils"
"github.com/docker/docker/pkg/jsonmessage"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/progressreader"
"github.com/docker/docker/pkg/stringid"
"github.com/docker/docker/pkg/system"
"github.com/docker/docker/pkg/tarsum"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/registry"
"github.com/docker/docker/runconfig"
)
func (b *Builder) readContext(context io.Reader) error {
tmpdirPath, err := ioutil.TempDir("", "docker-build")
if err != nil {
return err
}
decompressedStream, err := archive.DecompressStream(context)
if err != nil {
return err
}
if b.context, err = tarsum.NewTarSum(decompressedStream, true, tarsum.Version0); err != nil {
return err
}
if err := chrootarchive.Untar(b.context, tmpdirPath, nil); err != nil {
return err
}
b.contextPath = tmpdirPath
return nil
}
func (b *Builder) commit(id string, autoCmd *runconfig.Command, comment string) error {
if b.disableCommit {
return nil
}
if b.image == "" && !b.noBaseImage {
return fmt.Errorf("Please provide a source image with `from` prior to commit")
}
b.Config.Image = b.image
if id == "" {
cmd := b.Config.Cmd
b.Config.Cmd = runconfig.NewCommand("/bin/sh", "-c", "#(nop) "+comment)
defer func(cmd *runconfig.Command) { b.Config.Cmd = cmd }(cmd)
hit, err := b.probeCache()
if err != nil {
return err
}
if hit {
return nil
}
container, err := b.create()
if err != nil {
return err
}
id = container.ID
if err := container.Mount(); err != nil {
return err
}
defer container.Unmount()
}
container, err := b.Daemon.Get(id)
if err != nil {
return err
}
// Note: Actually copy the struct
autoConfig := *b.Config
autoConfig.Cmd = autoCmd
// Commit the container
image, err := b.Daemon.Commit(container, "", "", "", b.maintainer, true, &autoConfig)
if err != nil {
return err
}
b.image = image.ID
return nil
}
type copyInfo struct {
origPath string
destPath string
hash string
decompress bool
tmpDir string
}
func (b *Builder) runContextCommand(args []string, allowRemote bool, allowDecompression bool, cmdName string) error {
if b.context == nil {
return fmt.Errorf("No context given. Impossible to use %s", cmdName)
}
if len(args) < 2 {
return fmt.Errorf("Invalid %s format - at least two arguments required", cmdName)
}
dest := args[len(args)-1] // last one is always the dest
copyInfos := []*copyInfo{}
b.Config.Image = b.image
defer func() {
for _, ci := range copyInfos {
if ci.tmpDir != "" {
os.RemoveAll(ci.tmpDir)
}
}
}()
// Loop through each src file and calculate the info we need to
// do the copy (e.g. hash value if cached). Don't actually do
// the copy until we've looked at all src files
for _, orig := range args[0 : len(args)-1] {
if err := calcCopyInfo(
b,
cmdName,
&copyInfos,
orig,
dest,
allowRemote,
allowDecompression,
true,
); err != nil {
return err
}
}
if len(copyInfos) == 0 {
return fmt.Errorf("No source files were specified")
}
if len(copyInfos) > 1 && !strings.HasSuffix(dest, "/") {
return fmt.Errorf("When using %s with more than one source file, the destination must be a directory and end with a /", cmdName)
}
// For backwards compat, if there's just one CI then use it as the
// cache look-up string, otherwise hash 'em all into one
var srcHash string
var origPaths string
if len(copyInfos) == 1 {
srcHash = copyInfos[0].hash
origPaths = copyInfos[0].origPath
} else {
var hashs []string
var origs []string
for _, ci := range copyInfos {
hashs = append(hashs, ci.hash)
origs = append(origs, ci.origPath)
}
hasher := sha256.New()
hasher.Write([]byte(strings.Join(hashs, ",")))
srcHash = "multi:" + hex.EncodeToString(hasher.Sum(nil))
origPaths = strings.Join(origs, " ")
}
cmd := b.Config.Cmd
b.Config.Cmd = runconfig.NewCommand("/bin/sh", "-c", fmt.Sprintf("#(nop) %s %s in %s", cmdName, srcHash, dest))
defer func(cmd *runconfig.Command) { b.Config.Cmd = cmd }(cmd)
hit, err := b.probeCache()
if err != nil {
return err
}
if hit {
return nil
}
container, _, err := b.Daemon.Create(b.Config, nil, "")
if err != nil {
return err
}
b.TmpContainers[container.ID] = struct{}{}
if err := container.Mount(); err != nil {
return err
}
defer container.Unmount()
for _, ci := range copyInfos {
if err := b.addContext(container, ci.origPath, ci.destPath, ci.decompress); err != nil {
return err
}
}
if err := b.commit(container.ID, cmd, fmt.Sprintf("%s %s in %s", cmdName, origPaths, dest)); err != nil {
return err
}
return nil
}
func calcCopyInfo(b *Builder, cmdName string, cInfos *[]*copyInfo, origPath string, destPath string, allowRemote bool, allowDecompression bool, allowWildcards bool) error {
if origPath != "" && origPath[0] == '/' && len(origPath) > 1 {
origPath = origPath[1:]
}
origPath = strings.TrimPrefix(origPath, "./")
// Twiddle the destPath when its a relative path - meaning, make it
// relative to the WORKINGDIR
if !filepath.IsAbs(destPath) {
hasSlash := strings.HasSuffix(destPath, "/")
destPath = filepath.Join("/", b.Config.WorkingDir, destPath)
// Make sure we preserve any trailing slash
if hasSlash {
destPath += "/"
}
}
// In the remote/URL case, download it and gen its hashcode
if urlutil.IsURL(origPath) {
if !allowRemote {
return fmt.Errorf("Source can't be a URL for %s", cmdName)
}
ci := copyInfo{}
ci.origPath = origPath
ci.hash = origPath // default to this but can change
ci.destPath = destPath
ci.decompress = false
*cInfos = append(*cInfos, &ci)
// Initiate the download
resp, err := httputils.Download(ci.origPath)
if err != nil {
return err
}
// Create a tmp dir
tmpDirName, err := ioutil.TempDir(b.contextPath, "docker-remote")
if err != nil {
return err
}
ci.tmpDir = tmpDirName
// Create a tmp file within our tmp dir
tmpFileName := path.Join(tmpDirName, "tmp")
tmpFile, err := os.OpenFile(tmpFileName, os.O_RDWR|os.O_CREATE|os.O_EXCL, 0600)
if err != nil {
return err
}
// Download and dump result to tmp file
if _, err := io.Copy(tmpFile, progressreader.New(progressreader.Config{
In: resp.Body,
Out: b.OutOld,
Formatter: b.StreamFormatter,
Size: int(resp.ContentLength),
NewLines: true,
ID: "",
Action: "Downloading",
})); err != nil {
tmpFile.Close()
return err
}
fmt.Fprintf(b.OutStream, "\n")
tmpFile.Close()
// Set the mtime to the Last-Modified header value if present
// Otherwise just remove atime and mtime
times := make([]syscall.Timespec, 2)
lastMod := resp.Header.Get("Last-Modified")
if lastMod != "" {
mTime, err := http.ParseTime(lastMod)
// If we can't parse it then just let it default to 'zero'
// otherwise use the parsed time value
if err == nil {
times[1] = syscall.NsecToTimespec(mTime.UnixNano())
}
}
if err := system.UtimesNano(tmpFileName, times); err != nil {
return err
}
ci.origPath = path.Join(filepath.Base(tmpDirName), filepath.Base(tmpFileName))
// If the destination is a directory, figure out the filename.
if strings.HasSuffix(ci.destPath, "/") {
u, err := url.Parse(origPath)
if err != nil {
return err
}
path := u.Path
if strings.HasSuffix(path, "/") {
path = path[:len(path)-1]
}
parts := strings.Split(path, "/")
filename := parts[len(parts)-1]
if filename == "" {
return fmt.Errorf("cannot determine filename from url: %s", u)
}
ci.destPath = ci.destPath + filename
}
// Calc the checksum, even if we're using the cache
r, err := archive.Tar(tmpFileName, archive.Uncompressed)
if err != nil {
return err
}
tarSum, err := tarsum.NewTarSum(r, true, tarsum.Version0)
if err != nil {
return err
}
if _, err := io.Copy(ioutil.Discard, tarSum); err != nil {
return err
}
ci.hash = tarSum.Sum(nil)
r.Close()
return nil
}
// Deal with wildcards
if allowWildcards && ContainsWildcards(origPath) {
for _, fileInfo := range b.context.GetSums() {
if fileInfo.Name() == "" {
continue
}
match, _ := path.Match(origPath, fileInfo.Name())
if !match {
continue
}
// Note we set allowWildcards to false in case the name has
// a * in it
calcCopyInfo(b, cmdName, cInfos, fileInfo.Name(), destPath, allowRemote, allowDecompression, false)
}
return nil
}
// Must be a dir or a file
if err := b.checkPathForAddition(origPath); err != nil {
return err
}
fi, _ := os.Stat(path.Join(b.contextPath, origPath))
ci := copyInfo{}
ci.origPath = origPath
ci.hash = origPath
ci.destPath = destPath
ci.decompress = allowDecompression
*cInfos = append(*cInfos, &ci)
// Deal with the single file case
if !fi.IsDir() {
// This will match first file in sums of the archive
fis := b.context.GetSums().GetFile(ci.origPath)
if fis != nil {
ci.hash = "file:" + fis.Sum()
}
return nil
}
// Must be a dir
var subfiles []string
absOrigPath := path.Join(b.contextPath, ci.origPath)
// Add a trailing / to make sure we only pick up nested files under
// the dir and not sibling files of the dir that just happen to
// start with the same chars
if !strings.HasSuffix(absOrigPath, "/") {
absOrigPath += "/"
}
// Need path w/o / too to find matching dir w/o trailing /
absOrigPathNoSlash := absOrigPath[:len(absOrigPath)-1]
for _, fileInfo := range b.context.GetSums() {
absFile := path.Join(b.contextPath, fileInfo.Name())
// Any file in the context that starts with the given path will be
// picked up and its hashcode used. However, we'll exclude the
// root dir itself. We do this for a coupel of reasons:
// 1 - ADD/COPY will not copy the dir itself, just its children
// so there's no reason to include it in the hash calc
// 2 - the metadata on the dir will change when any child file
// changes. This will lead to a miss in the cache check if that
// child file is in the .dockerignore list.
if strings.HasPrefix(absFile, absOrigPath) && absFile != absOrigPathNoSlash {
subfiles = append(subfiles, fileInfo.Sum())
}
}
sort.Strings(subfiles)
hasher := sha256.New()
hasher.Write([]byte(strings.Join(subfiles, ",")))
ci.hash = "dir:" + hex.EncodeToString(hasher.Sum(nil))
return nil
}
func ContainsWildcards(name string) bool {
for i := 0; i < len(name); i++ {
ch := name[i]
if ch == '\\' {
i++
} else if ch == '*' || ch == '?' || ch == '[' {
return true
}
}
return false
}
func (b *Builder) pullImage(name string) (*imagepkg.Image, error) {
remote, tag := parsers.ParseRepositoryTag(name)
if tag == "" {
tag = "latest"
}
pullRegistryAuth := b.AuthConfig
if len(b.ConfigFile.AuthConfigs) > 0 {
// The request came with a full auth config file, we prefer to use that
repoInfo, err := b.Daemon.RegistryService.ResolveRepository(remote)
if err != nil {
return nil, err
}
resolvedAuth := registry.ResolveAuthConfig(b.ConfigFile, repoInfo.Index)
pullRegistryAuth = &resolvedAuth
}
imagePullConfig := &graph.ImagePullConfig{
AuthConfig: pullRegistryAuth,
OutStream: ioutils.NopWriteCloser(b.OutOld),
}
if err := b.Daemon.Repositories().Pull(remote, tag, imagePullConfig); err != nil {
return nil, err
}
image, err := b.Daemon.Repositories().LookupImage(name)
if err != nil {
return nil, err
}
return image, nil
}
func (b *Builder) processImageFrom(img *imagepkg.Image) error {
b.image = img.ID
if img.Config != nil {
b.Config = img.Config
}
if len(b.Config.Env) == 0 {
b.Config.Env = append(b.Config.Env, "PATH="+daemon.DefaultPathEnv)
}
// Process ONBUILD triggers if they exist
if nTriggers := len(b.Config.OnBuild); nTriggers != 0 {
fmt.Fprintf(b.ErrStream, "# Executing %d build triggers\n", nTriggers)
}
// Copy the ONBUILD triggers, and remove them from the config, since the config will be committed.
onBuildTriggers := b.Config.OnBuild
b.Config.OnBuild = []string{}
// parse the ONBUILD triggers by invoking the parser
for stepN, step := range onBuildTriggers {
ast, err := parser.Parse(strings.NewReader(step))
if err != nil {
return err
}
for i, n := range ast.Children {
switch strings.ToUpper(n.Value) {
case "ONBUILD":
return fmt.Errorf("Chaining ONBUILD via `ONBUILD ONBUILD` isn't allowed")
case "MAINTAINER", "FROM":
return fmt.Errorf("%s isn't allowed as an ONBUILD trigger", n.Value)
}
fmt.Fprintf(b.OutStream, "Trigger %d, %s\n", stepN, step)
if err := b.dispatch(i, n); err != nil {
return err
}
}
}
return nil
}
// probeCache checks to see if image-caching is enabled (`b.UtilizeCache`)
// and if so attempts to look up the current `b.image` and `b.Config` pair
// in the current server `b.Daemon`. If an image is found, probeCache returns
// `(true, nil)`. If no image is found, it returns `(false, nil)`. If there
// is any error, it returns `(false, err)`.
func (b *Builder) probeCache() (bool, error) {
if !b.UtilizeCache || b.cacheBusted {
return false, nil
}
cache, err := b.Daemon.ImageGetCached(b.image, b.Config)
if err != nil {
return false, err
}
if cache == nil {
logrus.Debugf("[BUILDER] Cache miss")
b.cacheBusted = true
return false, nil
}
fmt.Fprintf(b.OutStream, " ---> Using cache\n")
logrus.Debugf("[BUILDER] Use cached version")
b.image = cache.ID
return true, nil
}
func (b *Builder) create() (*daemon.Container, error) {
if b.image == "" && !b.noBaseImage {
return nil, fmt.Errorf("Please provide a source image with `from` prior to run")
}
b.Config.Image = b.image
hostConfig := &runconfig.HostConfig{
CpuShares: b.cpuShares,
CpuPeriod: b.cpuPeriod,
CpuQuota: b.cpuQuota,
CpusetCpus: b.cpuSetCpus,
CpusetMems: b.cpuSetMems,
CgroupParent: b.cgroupParent,
Memory: b.memory,
MemorySwap: b.memorySwap,
NetworkMode: "bridge",
}
config := *b.Config
// Create the container
c, warnings, err := b.Daemon.Create(b.Config, hostConfig, "")
if err != nil {
return nil, err
}
for _, warning := range warnings {
fmt.Fprintf(b.OutStream, " ---> [Warning] %s\n", warning)
}
b.TmpContainers[c.ID] = struct{}{}
fmt.Fprintf(b.OutStream, " ---> Running in %s\n", stringid.TruncateID(c.ID))
if config.Cmd.Len() > 0 {
// override the entry point that may have been picked up from the base image
s := config.Cmd.Slice()
c.Path = s[0]
c.Args = s[1:]
} else {
config.Cmd = runconfig.NewCommand()
}
return c, nil
}
func (b *Builder) run(c *daemon.Container) error {
var errCh chan error
if b.Verbose {
errCh = c.Attach(nil, b.OutStream, b.ErrStream)
}
//start the container
if err := c.Start(); err != nil {
return err
}
finished := make(chan struct{})
defer close(finished)
go func() {
select {
case <-b.cancelled:
logrus.Debugln("Build cancelled, killing container:", c.ID)
c.Kill()
case <-finished:
}
}()
if b.Verbose {
// Block on reading output from container, stop on err or chan closed
if err := <-errCh; err != nil {
return err
}
}
// Wait for it to finish
if ret, _ := c.WaitStop(-1 * time.Second); ret != 0 {
return &jsonmessage.JSONError{
Message: fmt.Sprintf("The command '%s' returned a non-zero code: %d", b.Config.Cmd.ToString(), ret),
Code: ret,
}
}
return nil
}
func (b *Builder) checkPathForAddition(orig string) error {
origPath := path.Join(b.contextPath, orig)
origPath, err := filepath.EvalSymlinks(origPath)
if err != nil {
if os.IsNotExist(err) {
return fmt.Errorf("%s: no such file or directory", orig)
}
return err
}
if !strings.HasPrefix(origPath, b.contextPath) {
return fmt.Errorf("Forbidden path outside the build context: %s (%s)", orig, origPath)
}
if _, err := os.Stat(origPath); err != nil {
if os.IsNotExist(err) {
return fmt.Errorf("%s: no such file or directory", orig)
}
return err
}
return nil
}
func (b *Builder) addContext(container *daemon.Container, orig, dest string, decompress bool) error {
var (
err error
destExists = true
origPath = path.Join(b.contextPath, orig)
destPath string
)
destPath, err = container.GetResourcePath(dest)
if err != nil {
return err
}
// Preserve the trailing '/'
if strings.HasSuffix(dest, "/") || dest == "." {
destPath = destPath + "/"
}
destStat, err := os.Stat(destPath)
if err != nil {
if !os.IsNotExist(err) {
return err
}
destExists = false
}
fi, err := os.Stat(origPath)
if err != nil {
if os.IsNotExist(err) {
return fmt.Errorf("%s: no such file or directory", orig)
}
return err
}
if fi.IsDir() {
return copyAsDirectory(origPath, destPath, destExists)
}
// If we are adding a remote file (or we've been told not to decompress), do not try to untar it
if decompress {
// First try to unpack the source as an archive
// to support the untar feature we need to clean up the path a little bit
// because tar is very forgiving. First we need to strip off the archive's
// filename from the path but this is only added if it does not end in / .
tarDest := destPath
if strings.HasSuffix(tarDest, "/") {
tarDest = filepath.Dir(destPath)
}
// try to successfully untar the orig
if err := chrootarchive.UntarPath(origPath, tarDest); err == nil {
return nil
} else if err != io.EOF {
logrus.Debugf("Couldn't untar %s to %s: %s", origPath, tarDest, err)
}
}
if err := os.MkdirAll(path.Dir(destPath), 0755); err != nil {
return err
}
if err := chrootarchive.CopyWithTar(origPath, destPath); err != nil {
return err
}
resPath := destPath
if destExists && destStat.IsDir() {
resPath = path.Join(destPath, path.Base(origPath))
}
return fixPermissions(origPath, resPath, 0, 0, destExists)
}
func copyAsDirectory(source, destination string, destExisted bool) error {
if err := chrootarchive.CopyWithTar(source, destination); err != nil {
return err
}
return fixPermissions(source, destination, 0, 0, destExisted)
}
func fixPermissions(source, destination string, uid, gid int, destExisted bool) error {
// If the destination didn't already exist, or the destination isn't a
// directory, then we should Lchown the destination. Otherwise, we shouldn't
// Lchown the destination.
destStat, err := os.Stat(destination)
if err != nil {
// This should *never* be reached, because the destination must've already
// been created while untar-ing the context.
return err
}
doChownDestination := !destExisted || !destStat.IsDir()
// We Walk on the source rather than on the destination because we don't
// want to change permissions on things we haven't created or modified.
return filepath.Walk(source, func(fullpath string, info os.FileInfo, err error) error {
// Do not alter the walk root iff. it existed before, as it doesn't fall under
// the domain of "things we should chown".
if !doChownDestination && (source == fullpath) {
return nil
}
// Path is prefixed by source: substitute with destination instead.
cleaned, err := filepath.Rel(source, fullpath)
if err != nil {
return err
}
fullpath = path.Join(destination, cleaned)
return os.Lchown(fullpath, uid, gid)
})
}
func (b *Builder) clearTmp() {
for c := range b.TmpContainers {
rmConfig := &daemon.ContainerRmConfig{
ForceRemove: true,
RemoveVolume: true,
}
if err := b.Daemon.ContainerRm(c, rmConfig); err != nil {
fmt.Fprintf(b.OutStream, "Error removing intermediate container %s: %v\n", stringid.TruncateID(c), err)
return
}
delete(b.TmpContainers, c)
fmt.Fprintf(b.OutStream, "Removing intermediate container %s\n", stringid.TruncateID(c))
}
}
-239
View File
@@ -1,239 +0,0 @@
package builder
import (
"bytes"
"fmt"
"io"
"io/ioutil"
"os"
"strings"
"sync"
"github.com/docker/docker/api"
"github.com/docker/docker/builder/parser"
"github.com/docker/docker/cliconfig"
"github.com/docker/docker/daemon"
"github.com/docker/docker/graph/tags"
"github.com/docker/docker/pkg/archive"
"github.com/docker/docker/pkg/httputils"
"github.com/docker/docker/pkg/parsers"
"github.com/docker/docker/pkg/streamformatter"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/registry"
"github.com/docker/docker/runconfig"
"github.com/docker/docker/utils"
)
// whitelist of commands allowed for a commit/import
var validCommitCommands = map[string]bool{
"entrypoint": true,
"cmd": true,
"user": true,
"workdir": true,
"env": true,
"volume": true,
"expose": true,
"onbuild": true,
}
type Config struct {
DockerfileName string
RemoteURL string
RepoName string
SuppressOutput bool
NoCache bool
Remove bool
ForceRemove bool
Pull bool
Memory int64
MemorySwap int64
CpuShares int64
CpuPeriod int64
CpuQuota int64
CpuSetCpus string
CpuSetMems string
CgroupParent string
AuthConfig *cliconfig.AuthConfig
ConfigFile *cliconfig.ConfigFile
Stdout io.Writer
Context io.ReadCloser
// When closed, the job has been cancelled.
// Note: not all jobs implement cancellation.
// See Job.Cancel() and Job.WaitCancelled()
cancelled chan struct{}
cancelOnce sync.Once
}
// When called, causes the Job.WaitCancelled channel to unblock.
func (b *Config) Cancel() {
b.cancelOnce.Do(func() {
close(b.cancelled)
})
}
// Returns a channel which is closed ("never blocks") when the job is cancelled.
func (b *Config) WaitCancelled() <-chan struct{} {
return b.cancelled
}
func NewBuildConfig() *Config {
return &Config{
AuthConfig: &cliconfig.AuthConfig{},
ConfigFile: &cliconfig.ConfigFile{},
cancelled: make(chan struct{}),
}
}
func Build(d *daemon.Daemon, buildConfig *Config) error {
var (
repoName string
tag string
context io.ReadCloser
)
repoName, tag = parsers.ParseRepositoryTag(buildConfig.RepoName)
if repoName != "" {
if err := registry.ValidateRepositoryName(repoName); err != nil {
return err
}
if len(tag) > 0 {
if err := tags.ValidateTagName(tag); err != nil {
return err
}
}
}
if buildConfig.RemoteURL == "" {
context = ioutil.NopCloser(buildConfig.Context)
} else if urlutil.IsGitURL(buildConfig.RemoteURL) {
root, err := utils.GitClone(buildConfig.RemoteURL)
if err != nil {
return err
}
defer os.RemoveAll(root)
c, err := archive.Tar(root, archive.Uncompressed)
if err != nil {
return err
}
context = c
} else if urlutil.IsURL(buildConfig.RemoteURL) {
f, err := httputils.Download(buildConfig.RemoteURL)
if err != nil {
return err
}
defer f.Body.Close()
dockerFile, err := ioutil.ReadAll(f.Body)
if err != nil {
return err
}
// When we're downloading just a Dockerfile put it in
// the default name - don't allow the client to move/specify it
buildConfig.DockerfileName = api.DefaultDockerfileName
c, err := archive.Generate(buildConfig.DockerfileName, string(dockerFile))
if err != nil {
return err
}
context = c
}
defer context.Close()
sf := streamformatter.NewJSONStreamFormatter()
builder := &Builder{
Daemon: d,
OutStream: &streamformatter.StdoutFormater{
Writer: buildConfig.Stdout,
StreamFormatter: sf,
},
ErrStream: &streamformatter.StderrFormater{
Writer: buildConfig.Stdout,
StreamFormatter: sf,
},
Verbose: !buildConfig.SuppressOutput,
UtilizeCache: !buildConfig.NoCache,
Remove: buildConfig.Remove,
ForceRemove: buildConfig.ForceRemove,
Pull: buildConfig.Pull,
OutOld: buildConfig.Stdout,
StreamFormatter: sf,
AuthConfig: buildConfig.AuthConfig,
ConfigFile: buildConfig.ConfigFile,
dockerfileName: buildConfig.DockerfileName,
cpuShares: buildConfig.CpuShares,
cpuPeriod: buildConfig.CpuPeriod,
cpuQuota: buildConfig.CpuQuota,
cpuSetCpus: buildConfig.CpuSetCpus,
cpuSetMems: buildConfig.CpuSetMems,
cgroupParent: buildConfig.CgroupParent,
memory: buildConfig.Memory,
memorySwap: buildConfig.MemorySwap,
cancelled: buildConfig.WaitCancelled(),
}
id, err := builder.Run(context)
if err != nil {
return err
}
if repoName != "" {
return d.Repositories().Tag(repoName, tag, id, true)
}
return nil
}
func BuildFromConfig(d *daemon.Daemon, c *runconfig.Config, changes []string) (*runconfig.Config, error) {
ast, err := parser.Parse(bytes.NewBufferString(strings.Join(changes, "\n")))
if err != nil {
return nil, err
}
// ensure that the commands are valid
for _, n := range ast.Children {
if !validCommitCommands[n.Value] {
return nil, fmt.Errorf("%s is not a valid change command", n.Value)
}
}
builder := &Builder{
Daemon: d,
Config: c,
OutStream: ioutil.Discard,
ErrStream: ioutil.Discard,
disableCommit: true,
}
for i, n := range ast.Children {
if err := builder.dispatch(i, n); err != nil {
return nil, err
}
}
return builder.Config, nil
}
func Commit(d *daemon.Daemon, name string, c *daemon.ContainerCommitConfig) (string, error) {
container, err := d.Get(name)
if err != nil {
return "", err
}
newConfig, err := BuildFromConfig(d, c.Config, c.Changes)
if err != nil {
return "", err
}
if err := runconfig.Merge(newConfig, container.Config); err != nil {
return "", err
}
img, err := d.Commit(container, c.Repo, c.Tag, c.Comment, c.Author, c.Pause, newConfig)
if err != nil {
return "", err
}
return img.ID, nil
}
-32
View File
@@ -1,32 +0,0 @@
package main
import (
"fmt"
"os"
"github.com/docker/docker/builder/parser"
)
func main() {
var f *os.File
var err error
if len(os.Args) < 2 {
fmt.Println("please supply filename(s)")
os.Exit(1)
}
for _, fn := range os.Args[1:] {
f, err = os.Open(fn)
if err != nil {
panic(err)
}
ast, err := parser.Parse(f)
if err != nil {
panic(err)
} else {
fmt.Println(ast.Dump())
}
}
}
-55
View File
@@ -1,55 +0,0 @@
package parser
import (
"testing"
)
var invalidJSONArraysOfStrings = []string{
`["a",42,"b"]`,
`["a",123.456,"b"]`,
`["a",{},"b"]`,
`["a",{"c": "d"},"b"]`,
`["a",["c"],"b"]`,
`["a",true,"b"]`,
`["a",false,"b"]`,
`["a",null,"b"]`,
}
var validJSONArraysOfStrings = map[string][]string{
`[]`: {},
`[""]`: {""},
`["a"]`: {"a"},
`["a","b"]`: {"a", "b"},
`[ "a", "b" ]`: {"a", "b"},
`[ "a", "b" ]`: {"a", "b"},
` [ "a", "b" ] `: {"a", "b"},
`["abc 123", "♥", "☃", "\" \\ \/ \b \f \n \r \t \u0000"]`: {"abc 123", "♥", "☃", "\" \\ / \b \f \n \r \t \u0000"},
}
func TestJSONArraysOfStrings(t *testing.T) {
for json, expected := range validJSONArraysOfStrings {
if node, _, err := parseJSON(json); err != nil {
t.Fatalf("%q should be a valid JSON array of strings, but wasn't! (err: %q)", json, err)
} else {
i := 0
for node != nil {
if i >= len(expected) {
t.Fatalf("expected result is shorter than parsed result (%d vs %d+) in %q", len(expected), i+1, json)
}
if node.Value != expected[i] {
t.Fatalf("expected %q (not %q) in %q at pos %d", expected[i], node.Value, json, i)
}
node = node.Next
i++
}
if i != len(expected) {
t.Fatalf("expected result is longer than parsed result (%d vs %d) in %q", len(expected), i+1, json)
}
}
}
for _, json := range invalidJSONArraysOfStrings {
if _, _, err := parseJSON(json); err != errDockerfileNotStringArray {
t.Fatalf("%q should be an invalid JSON array of strings, but wasn't!", json)
}
}
}
-295
View File
@@ -1,295 +0,0 @@
package parser
// line parsers are dispatch calls that parse a single unit of text into a
// Node object which contains the whole statement. Dockerfiles have varied
// (but not usually unique, see ONBUILD for a unique example) parsing rules
// per-command, and these unify the processing in a way that makes it
// manageable.
import (
"encoding/json"
"errors"
"fmt"
"strings"
"unicode"
)
var (
errDockerfileNotStringArray = errors.New("When using JSON array syntax, arrays must be comprised of strings only.")
)
// ignore the current argument. This will still leave a command parsed, but
// will not incorporate the arguments into the ast.
func parseIgnore(rest string) (*Node, map[string]bool, error) {
return &Node{}, nil, nil
}
// used for onbuild. Could potentially be used for anything that represents a
// statement with sub-statements.
//
// ONBUILD RUN foo bar -> (onbuild (run foo bar))
//
func parseSubCommand(rest string) (*Node, map[string]bool, error) {
if rest == "" {
return nil, nil, nil
}
_, child, err := parseLine(rest)
if err != nil {
return nil, nil, err
}
return &Node{Children: []*Node{child}}, nil, nil
}
// parse environment like statements. Note that this does *not* handle
// variable interpolation, which will be handled in the evaluator.
func parseNameVal(rest string, key string) (*Node, map[string]bool, error) {
// This is kind of tricky because we need to support the old
// variant: KEY name value
// as well as the new one: KEY name=value ...
// The trigger to know which one is being used will be whether we hit
// a space or = first. space ==> old, "=" ==> new
const (
inSpaces = iota // looking for start of a word
inWord
inQuote
)
words := []string{}
phase := inSpaces
word := ""
quote := '\000'
blankOK := false
var ch rune
for pos := 0; pos <= len(rest); pos++ {
if pos != len(rest) {
ch = rune(rest[pos])
}
if phase == inSpaces { // Looking for start of word
if pos == len(rest) { // end of input
break
}
if unicode.IsSpace(ch) { // skip spaces
continue
}
phase = inWord // found it, fall thru
}
if (phase == inWord || phase == inQuote) && (pos == len(rest)) {
if blankOK || len(word) > 0 {
words = append(words, word)
}
break
}
if phase == inWord {
if unicode.IsSpace(ch) {
phase = inSpaces
if blankOK || len(word) > 0 {
words = append(words, word)
// Look for = and if not there assume
// we're doing the old stuff and
// just read the rest of the line
if !strings.Contains(word, "=") {
word = strings.TrimSpace(rest[pos:])
words = append(words, word)
break
}
}
word = ""
blankOK = false
continue
}
if ch == '\'' || ch == '"' {
quote = ch
blankOK = true
phase = inQuote
}
if ch == '\\' {
if pos+1 == len(rest) {
continue // just skip \ at end
}
// If we're not quoted and we see a \, then always just
// add \ plus the char to the word, even if the char
// is a quote.
word += string(ch)
pos++
ch = rune(rest[pos])
}
word += string(ch)
continue
}
if phase == inQuote {
if ch == quote {
phase = inWord
}
// \ is special except for ' quotes - can't escape anything for '
if ch == '\\' && quote != '\'' {
if pos+1 == len(rest) {
phase = inWord
continue // just skip \ at end
}
pos++
nextCh := rune(rest[pos])
word += string(ch)
ch = nextCh
}
word += string(ch)
}
}
if len(words) == 0 {
return nil, nil, nil
}
// Old format (KEY name value)
var rootnode *Node
if !strings.Contains(words[0], "=") {
node := &Node{}
rootnode = node
strs := TOKEN_WHITESPACE.Split(rest, 2)
if len(strs) < 2 {
return nil, nil, fmt.Errorf(key + " must have two arguments")
}
node.Value = strs[0]
node.Next = &Node{}
node.Next.Value = strs[1]
} else {
var prevNode *Node
for i, word := range words {
if !strings.Contains(word, "=") {
return nil, nil, fmt.Errorf("Syntax error - can't find = in %q. Must be of the form: name=value", word)
}
parts := strings.SplitN(word, "=", 2)
name := &Node{}
value := &Node{}
name.Next = value
name.Value = parts[0]
value.Value = parts[1]
if i == 0 {
rootnode = name
} else {
prevNode.Next = name
}
prevNode = value
}
}
return rootnode, nil, nil
}
func parseEnv(rest string) (*Node, map[string]bool, error) {
return parseNameVal(rest, "ENV")
}
func parseLabel(rest string) (*Node, map[string]bool, error) {
return parseNameVal(rest, "LABEL")
}
// parses a whitespace-delimited set of arguments. The result is effectively a
// linked list of string arguments.
func parseStringsWhitespaceDelimited(rest string) (*Node, map[string]bool, error) {
if rest == "" {
return nil, nil, nil
}
node := &Node{}
rootnode := node
prevnode := node
for _, str := range TOKEN_WHITESPACE.Split(rest, -1) { // use regexp
prevnode = node
node.Value = str
node.Next = &Node{}
node = node.Next
}
// XXX to get around regexp.Split *always* providing an empty string at the
// end due to how our loop is constructed, nil out the last node in the
// chain.
prevnode.Next = nil
return rootnode, nil, nil
}
// parsestring just wraps the string in quotes and returns a working node.
func parseString(rest string) (*Node, map[string]bool, error) {
if rest == "" {
return nil, nil, nil
}
n := &Node{}
n.Value = rest
return n, nil, nil
}
// parseJSON converts JSON arrays to an AST.
func parseJSON(rest string) (*Node, map[string]bool, error) {
var myJson []interface{}
if err := json.NewDecoder(strings.NewReader(rest)).Decode(&myJson); err != nil {
return nil, nil, err
}
var top, prev *Node
for _, str := range myJson {
s, ok := str.(string)
if !ok {
return nil, nil, errDockerfileNotStringArray
}
node := &Node{Value: s}
if prev == nil {
top = node
} else {
prev.Next = node
}
prev = node
}
return top, map[string]bool{"json": true}, nil
}
// parseMaybeJSON determines if the argument appears to be a JSON array. If
// so, passes to parseJSON; if not, quotes the result and returns a single
// node.
func parseMaybeJSON(rest string) (*Node, map[string]bool, error) {
if rest == "" {
return nil, nil, nil
}
node, attrs, err := parseJSON(rest)
if err == nil {
return node, attrs, nil
}
if err == errDockerfileNotStringArray {
return nil, nil, err
}
node = &Node{}
node.Value = rest
return node, nil, nil
}
// parseMaybeJSONToList determines if the argument appears to be a JSON array. If
// so, passes to parseJSON; if not, attempts to parse it as a whitespace
// delimited string.
func parseMaybeJSONToList(rest string) (*Node, map[string]bool, error) {
node, attrs, err := parseJSON(rest)
if err == nil {
return node, attrs, nil
}
if err == errDockerfileNotStringArray {
return nil, nil, err
}
return parseStringsWhitespaceDelimited(rest)
}
-143
View File
@@ -1,143 +0,0 @@
// This package implements a parser and parse tree dumper for Dockerfiles.
package parser
import (
"bufio"
"io"
"regexp"
"strings"
"unicode"
"github.com/docker/docker/builder/command"
)
// Node is a structure used to represent a parse tree.
//
// In the node there are three fields, Value, Next, and Children. Value is the
// current token's string value. Next is always the next non-child token, and
// children contains all the children. Here's an example:
//
// (value next (child child-next child-next-next) next-next)
//
// This data structure is frankly pretty lousy for handling complex languages,
// but lucky for us the Dockerfile isn't very complicated. This structure
// works a little more effectively than a "proper" parse tree for our needs.
//
type Node struct {
Value string // actual content
Next *Node // the next item in the current sexp
Children []*Node // the children of this sexp
Attributes map[string]bool // special attributes for this node
Original string // original line used before parsing
Flags []string // only top Node should have this set
}
var (
dispatch map[string]func(string) (*Node, map[string]bool, error)
TOKEN_WHITESPACE = regexp.MustCompile(`[\t\v\f\r ]+`)
TOKEN_LINE_CONTINUATION = regexp.MustCompile(`\\[ \t]*$`)
TOKEN_COMMENT = regexp.MustCompile(`^#.*$`)
)
func init() {
// Dispatch Table. see line_parsers.go for the parse functions.
// The command is parsed and mapped to the line parser. The line parser
// recieves the arguments but not the command, and returns an AST after
// reformulating the arguments according to the rules in the parser
// functions. Errors are propagated up by Parse() and the resulting AST can
// be incorporated directly into the existing AST as a next.
dispatch = map[string]func(string) (*Node, map[string]bool, error){
command.User: parseString,
command.Onbuild: parseSubCommand,
command.Workdir: parseString,
command.Env: parseEnv,
command.Label: parseLabel,
command.Maintainer: parseString,
command.From: parseString,
command.Add: parseMaybeJSONToList,
command.Copy: parseMaybeJSONToList,
command.Run: parseMaybeJSON,
command.Cmd: parseMaybeJSON,
command.Entrypoint: parseMaybeJSON,
command.Expose: parseStringsWhitespaceDelimited,
command.Volume: parseMaybeJSONToList,
}
}
// parse a line and return the remainder.
func parseLine(line string) (string, *Node, error) {
if line = stripComments(line); line == "" {
return "", nil, nil
}
if TOKEN_LINE_CONTINUATION.MatchString(line) {
line = TOKEN_LINE_CONTINUATION.ReplaceAllString(line, "")
return line, nil, nil
}
cmd, flags, args, err := splitCommand(line)
if err != nil {
return "", nil, err
}
node := &Node{}
node.Value = cmd
sexp, attrs, err := fullDispatch(cmd, args)
if err != nil {
return "", nil, err
}
node.Next = sexp
node.Attributes = attrs
node.Original = line
node.Flags = flags
return "", node, nil
}
// The main parse routine. Handles an io.ReadWriteCloser and returns the root
// of the AST.
func Parse(rwc io.Reader) (*Node, error) {
root := &Node{}
scanner := bufio.NewScanner(rwc)
for scanner.Scan() {
scannedLine := strings.TrimLeftFunc(scanner.Text(), unicode.IsSpace)
line, child, err := parseLine(scannedLine)
if err != nil {
return nil, err
}
if line != "" && child == nil {
for scanner.Scan() {
newline := scanner.Text()
if stripComments(strings.TrimSpace(newline)) == "" {
continue
}
line, child, err = parseLine(line + newline)
if err != nil {
return nil, err
}
if child != nil {
break
}
}
if child == nil && line != "" {
line, child, err = parseLine(line)
if err != nil {
return nil, err
}
}
}
if child != nil {
root.Children = append(root.Children, child)
}
}
return root, nil
}
-75
View File
@@ -1,75 +0,0 @@
package parser
import (
"fmt"
"io/ioutil"
"os"
"path/filepath"
"testing"
)
const testDir = "testfiles"
const negativeTestDir = "testfiles-negative"
func getDirs(t *testing.T, dir string) []string {
f, err := os.Open(dir)
if err != nil {
t.Fatal(err)
}
defer f.Close()
dirs, err := f.Readdirnames(0)
if err != nil {
t.Fatal(err)
}
return dirs
}
func TestTestNegative(t *testing.T) {
for _, dir := range getDirs(t, negativeTestDir) {
dockerfile := filepath.Join(negativeTestDir, dir, "Dockerfile")
df, err := os.Open(dockerfile)
if err != nil {
t.Fatalf("Dockerfile missing for %s: %v", dir, err)
}
_, err = Parse(df)
if err == nil {
t.Fatalf("No error parsing broken dockerfile for %s", dir)
}
df.Close()
}
}
func TestTestData(t *testing.T) {
for _, dir := range getDirs(t, testDir) {
dockerfile := filepath.Join(testDir, dir, "Dockerfile")
resultfile := filepath.Join(testDir, dir, "result")
df, err := os.Open(dockerfile)
if err != nil {
t.Fatalf("Dockerfile missing for %s: %v", dir, err)
}
defer df.Close()
ast, err := Parse(df)
if err != nil {
t.Fatalf("Error parsing %s's dockerfile: %v", dir, err)
}
content, err := ioutil.ReadFile(resultfile)
if err != nil {
t.Fatalf("Error reading %s's result file: %v", dir, err)
}
if ast.Dump()+"\n" != string(content) {
fmt.Fprintln(os.Stderr, "Result:\n"+ast.Dump())
fmt.Fprintln(os.Stderr, "Expected:\n"+string(content))
t.Fatalf("%s: AST dump of dockerfile does not match result", dir)
}
}
}
@@ -1,3 +0,0 @@
FROM busybox
ENV PATH
@@ -1 +0,0 @@
CMD [ "echo", [ "nested json" ] ]

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