e17837a9b0490f1a86ac1dab8d0a8ed9df4e1233: mozreview: switch import style (Bug 1244448). r=mdoglio r=gps
Steven MacLeod <smacleod@mozilla.com> - Fri, 29 Jan 2016 10:16:30 -0500 - rev 362189
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
mozreview: switch import style (Bug 1244448). r=mdoglio r=gps
1ff0c6cfaa6ea77dc38d0ad773d71b1d20e55ff0: docs: note that bundleclone isn't required if running Mercurial 3.7+
Gregory Szorc <gps@mozilla.com> - Wed, 03 Feb 2016 15:24:40 -0800 - rev 362188
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
docs: note that bundleclone isn't required if running Mercurial 3.7+
056012f6f31d61b207ff531f8e0b4ac11081675b: hghooks: update commit SHA-1 for failing test
Gregory Szorc <gps@mozilla.com> - Wed, 03 Feb 2016 14:31:00 -0800 - rev 362187
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
hghooks: update commit SHA-1 for failing test This is an accidental regression from 22c12053bfd3.
efa891880f4a5594a5b3ad950f09aec731f54164: testing: run MySQL within bmoweb container
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 20:23:00 -0800 - rev 362186
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: run MySQL within bmoweb container When the bmoweb and bmodb containers were initially created, we were adhering to Docker's best practices for one process per container. As the number of containers in our testing environment has grown, it has become obvious that Docker doesn't scale well to spawning several containers simultaneously. Furthermore, we're seeing weird issues with the MySQL containers not running properly. The MySQL server used by bmoweb isn't used by anything else. There is no good reason to not have it embedded within the bmoweb container. This commit moves the MySQL server from a standalone container to inside bmoweb.
6d78c3c8da9f0b8cbac11dbc8d2d6c7adc06ec79: testing: support "sha256:" prefixes on images
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 20:02:46 -0800 - rev 362185
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: support "sha256:" prefixes on images Docker 1.10 appears to add "sha256:" on the prefix to images. Teach get_full_image() how to expand a partial image fragment to a full one when the full strings have "sha256:" prefixes.
69bd60d95e86bb0fcfb261dadd5eecc9a743eb10: Backed out changeset 2a64294718d1
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 13:00:16 -0800 - rev 362184
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
Backed out changeset 2a64294718d1
dd85192c07046824ec2b05c51e7a976c867ca424: testing: upgrade MySQL to 5.6.28 from 5.6.20
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 12:23:54 -0800 - rev 362183
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: upgrade MySQL to 5.6.28 from 5.6.20 Should pull in various bug fixes. Hopefully this fixes the issue with MySQL randomly hanging inside Docker.
243e49ba9f1b26978053e0f6ddc8da7fe228accd: testing: remove bmodb builder
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 12:23:10 -0800 - rev 362182
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: remove bmodb builder Only the bmodb-volatile builder has been used for over a year.
2a64294718d14b85691e80eeb5d7504b4a151e1b: testing: check bmofetch clobber file when determining if bootstrap needed
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 12:05:33 -0800 - rev 362181
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: check bmofetch clobber file when determining if bootstrap needed Otherwise we won't rebuild the BMO image unless the bootstrap clobber file is touched.
111ec2bb061d31ed5295a45c211d0152de6bca2f: mozreviewbots: update test output
Gregory Szorc <gps@mozilla.com> - Tue, 02 Feb 2016 10:50:24 -0800 - rev 362180
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
mozreviewbots: update test output It looks like we didn't update these files when we added diff output to review request dumping. It also appears we missed the update when upgrading to Review Board 2.5, as "calculated_trophies" was missing from 1 test.
77860597587a6b2cd5b961657b0e75f2befce065: git-mozreview: git command for interacting with MozReview (bug 1153053); r=glandium, dminor
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 16:31:59 -0800 - rev 362179
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
git-mozreview: git command for interacting with MozReview (bug 1153053); r=glandium, dminor This commit establishes the `git mozreview` command which supports submitting commits to MozReview for review. The command consists of the "configure" and "push" sub-commands. "configure" launches a wizard to configure the local Git repo to use MozReview. It is a rough equivalent of `mach mercurial-setup`. As part of configuration, we establish a "commit-msg" hook that adds a unique identifier to commit messages. The intent of this identifier is to track commits on MozReview. This will help MozReview intelligently react to history rewrites. Unfortunately, we don't yet use this identifier to track commits. But we have plans to improve MozReview's tracking support in the weeks or months ahead, so putting the tracking infrastructure in place now should make the transition easier. `git mozreview push` is the important command: it submits commits to MozReview for review. The command first pushes commits to MozReview using git-cinnabar then it makes a call to the appropriate web service API to turn pushed commits (actually Mercurial changesets) into review requests on MozReview. The workflow is modeled after the existing Mercurial workflow. MozReview-Commit-ID: 47EedQkLjeV
6b2aaa096dc5a0c9bc17f64d454be2b13068db94: reviewboard: allow fake ids file path to be passed in; r=dminor
Gregory Szorc <gps@mozilla.com> - Thu, 28 Jan 2016 11:57:26 -0800 - rev 362178
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
reviewboard: allow fake ids file path to be passed in; r=dminor So we can eventually re-use this function from Git.
6e94ce75c07f112ddc605e8f7a8fcab114b24fd9: testing: install git-cinnabar in testing environment; r=smacleod
Gregory Szorc <gps@mozilla.com> - Tue, 19 Jan 2016 13:52:57 -0800 - rev 362177
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: install git-cinnabar in testing environment; r=smacleod We'll need git-cinnabar present to test Git push to MozReview.
e6b7002791d62f96d7cbc50a6c24795aaae1891c: mozreview: more logging for batch review request API (bug 1243630); r=dminor
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 16:33:52 -0800 - rev 362176
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
mozreview: more logging for batch review request API (bug 1243630); r=dminor This should help us establish a better audit trail and will help us debug issues, including basic timing information.
258e7ae1a885f29417f9496b577eb5657e5323fd: Bug 1244738: Upgrade Bugsy to v0.7.0 r=gps
David Burns <dburns@mozilla.com> - Mon, 01 Feb 2016 15:36:27 +0000 - rev 362175
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
Bug 1244738: Upgrade Bugsy to v0.7.0 r=gps Change log: * Correct pyflakes error * Add error codes returned from Bugzilla to errror messages * Handle when Bugzilla returns errors with a 200 status code * Correct test when ting into an existing bug * Handle the case where isnt on the * Add the ability to get the blockers and dependent bugs * Pop messages on the bug that were added for creation otherwise you get an error * Get the ability to get keywords associated to the bug. * Add the ability to get the cc list of a bug
82c4760e61a825796657af0af9e93d9f2c9174de: Backed out changeset 79284f382e6b (bug 1243501)
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 16:04:03 -0800 - rev 362174
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
Backed out changeset 79284f382e6b (bug 1243501) It broke fresh image building.
006ab4f3e6d25218a2d9bf8c9ab03fcb0bd35c5b: mozext: use bootstrap.py (bug 1244786); r=dminor
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 12:17:30 -0800 - rev 362173
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
mozext: use bootstrap.py (bug 1244786); r=dminor bootstrap.py exists to configure sys.path with every vendored Python package path. mozext wasn't using this script - instead it was manually configuring sys.path. A recent change to mozautomation introduced requests as a dependency. Since mozext wasn't inserting pylib/requests into sys.path, this could result in an import failure due to requests not being found. Using bootstrap.py adds pylib/requests and makes the ImportError go away.
79284f382e6bb2a2558a22338a1b23f02e158475: ansible/hg: use pip 8 (bug 1243501); r=fubar
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 09:31:07 -0800 - rev 362172
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
ansible/hg: use pip 8 (bug 1243501); r=fubar pip 8 contains peep's hash verification functionality. As part of this commit, we introduce a standalone task for configuring a virtualenv. We replace the virtualenv muckery with it. And we update the requirements.txt files to pip 8's syntax.
a03c4d9228d8e3f6c948f0c5e2819ebd46172518: testing: replace Docker IP in test output with DOCKER_HOSTNAME (bug 1244588); r=dminor
Gregory Szorc <gps@mozilla.com> - Mon, 01 Feb 2016 09:21:40 -0800 - rev 362171
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
testing: replace Docker IP in test output with DOCKER_HOSTNAME (bug 1244588); r=dminor The Mercurial test harness supports replacing certain string patterns with other patterns. This helps mask over differences between test invocations. e.g. port ":35232" will get replaced with ":$HGPORT." Up until this patch, we have been globbing over IP addresses of services running in Docker. When test output changes, the test harness's diffing mechanism isn't smart enough to preserve the "(glob)" and the "*" over the IP address is replaced with the raw IP. After this patch, the IP address for the Docker host will be replaced with "$DOCKER_HOSTNAME." When test output changes, lines containing URLs/globs should no longer be reported as changed and require manual correction (unless something outside the URL changes, of course). This will make future mass updates of test output much simpler, since in many cases you can simply `mv` the .t.err file over the .t file to reflect new output. MozReview-Commit-ID: 6syJDfc2lJk
2c487b8be40d4d5454ba620637eea68fb9d24bd4: autoland: prevent partial and incorrect landings (bug 1243276) r=gps
Dan Minor <dminor@mozilla.com> - Wed, 27 Jan 2016 14:15:31 -0500 - rev 362170
Push 16998 by rwood@mozilla.com at Mon, 02 May 2016 19:42:03 +0000
autoland: prevent partial and incorrect landings (bug 1243276) r=gps In Bug 1243276 we ended up with a partial landing of the wrong commits. The landing tree was closed, but due to a bug with the treestatus code we did not find out about this until push time, which was causing us to skip the clean up code for non-public commits. I think this missing clean up along with the fact that we pulled and rebased the same commits multiple times got us in to a state where 'tip' no longer referred to the commits we wanted to push. This patch adds a strip of non-public commits to the beginning of any transplant attempt. In addition, we add a check to make sure that the set of incoming commit descriptions matches the set of outgoing commit descriptions. This should prevent us from landing incorrect commits unless we have somehow associated commit descriptions with the wrong commits. I tested the new code manually by changing the handling of the encoding of commit descriptions so they did not match after rewriting, but I not sure we can test this automatically without having races between the test code and autoland.
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 +300000 tip