[09:16:45] 10Gerrit, 10Proton, 10Readers-Web-Backlog, 10Services: Rename chromium-render gerrit project to Proton - https://phabricator.wikimedia.org/T211375 (10Peachey88) [10:12:11] 10Gerrit, 10Proton, 10Readers-Web-Backlog, 10Core Platform Team Backlog (Attic), 10Services (attic): Rename chromium-render gerrit project to Proton - https://phabricator.wikimedia.org/T211375 (10mobrovac) [11:54:05] ! [remote rejected] HEAD -> refs/for/master (the number of pushed changes in a batch exceeds the max limit 10) [11:54:13] any way to bypass that? [12:31:25] 10Gerrit, 10Proton, 10Readers-Web-Backlog, 10Core Platform Team Backlog (Attic), 10Services (attic): Rename chromium-render gerrit project to Proton - https://phabricator.wikimedia.org/T211375 (10ovasileva) p:05Triage>03High [12:49:33] 10Gerrit: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) [12:53:28] 10Gerrit: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) p:05Triage>03Unbreak! [13:00:14] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Aklapper) (If you have pushed changes you want to revert, why don't you revert the changes locally and then push again?) [13:02:05] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) >>! In T211420#4805831, @Aklapper wrote: > (If you have pushed changes you want to revert, why don't you revert the changes locally and then push again?) Because I don't want to... [13:24:23] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) I've created revert patches for each commit accidentally pushed. [13:39:05] Reedy: ping [13:39:12] hi [13:39:20] Reedy: can I PM? [13:39:23] yeah [13:39:28] okay thanks [13:42:46] Hauskatze: git push origin HEAD:refs/for/ [13:42:53] Is what you wanted :) [13:43:11] git push puts it straight into the branch without a review [13:43:23] now I know :| [13:43:49] I've created revert patches for all of them and git review-R'd them all [13:43:58] Heh [13:44:12] Hauskatze: you can also revoke the push right for your self too :) [13:44:23] To prevent pushing directly into the repo [13:44:23] I'm not using git push anymore [13:44:38] that was an idiocy of me, to bypass ! [remote rejected] HEAD -> refs/for/master (the number of pushed changes in a batch exceeds the max limit 10) [13:44:51] Yeh that is done on purpose [13:44:53] I though they'd end all for review [13:44:55] That limit [13:45:04] but NO, they had to go straight to the repo [13:45:11] Heh [13:45:24] I've slapped myself with a big codfish, no need for you to do it again [13:46:13] Heh [13:51:56] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Reedy) As discussed on IRC.. If the tests are passing, at this point, it might be less disruptive to just leave the commits, not revert them, don't change the HEAD pointer.... And then at wors... [13:59:43] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) I'd prefer to revert because those patches should get a proper review IMHO. The revert patches are all now uploaded to gerrit and are waiting for review (they all are submitted t... [14:14:26] 10Gerrit, 10Release-Engineering-Team (Backlog), 10Patch-For-Review: Upgrade to Gerrit 2.16 - https://phabricator.wikimedia.org/T200739 (10Paladox) looks like gerrit 2.16.1 will be out soon https://gerrit-review.googlesource.com/c/gerrit/+/206717 [14:24:18] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Reedy) >>! In T211420#4805905, @MarcoAurelio wrote: > I'd prefer to revert because those patches should get a proper review IMHO. The revert patches are all now uploaded to gerrit and are wait... [15:12:41] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) I'm reverting them now. I know I won't have peace of mind if I don't undo this. I'll try to submit those for review later today or tomorrow. [15:31:08] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) Okay I think that while less elegant, the individual reverts restored the repo to its previous status. Please review. Thanks. [16:21:31] 10Gerrit, 10Patch-For-Review: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Jdforrester-WMF) Yup, LGTM: ` lang=bash ~/D/c/a/UploadLocal   master  git log -n 1 Fri 7 Dec 08:19:39 2018 commit 08c2f76ff87466bd3cc63d38... [16:21:40] 10Gerrit, 10Patch-For-Review, 10User-MarcoAurelio: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Jdforrester-WMF) 05Open>03Resolved a:03MarcoAurelio [17:22:02] 10Gerrit, 10Patch-For-Review, 10User-MarcoAurelio: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) Thank you for checking @Jdforrester-WMF .- I'll have some work down the line soon today I hope. Re. git push: it was my fault. I enabled it because git-re... [17:29:04] 10Gerrit, 10Patch-For-Review, 10User-MarcoAurelio: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Jdforrester-WMF) >>! In T211420#4806434, @MarcoAurelio wrote: > Thank you for checking @Jdforrester-WMF .- I'll have some work down the line soon today I hope. > > Re.... [17:29:10] 10Gerrit, 10Patch-For-Review, 10User-MarcoAurelio: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10Reedy) >>! In T211420#4806434, @MarcoAurelio wrote: > Do we have a way to commit more than 10 patches at once with git review? If you say had 15 commits to send... ` gi... [17:35:06] 10Gerrit, 10Patch-For-Review, 10User-MarcoAurelio: Rollback extension-UploadLocal to 4bade11 - https://phabricator.wikimedia.org/T211420 (10MarcoAurelio) Thanks. I'll try that. In other occasions I've used git push origin... sometimes the commit was rejected by the remote because it lacked a Change-Id, etc.... [18:08:36] PROBLEM - Host integration-slave-docker-1040 is DOWN: CRITICAL - Host Unreachable (10.68.19.167) [18:13:42] PROBLEM - Host integration-slave-docker-1037 is DOWN: CRITICAL - Host Unreachable (10.68.21.155) [19:43:14] PROBLEM - Host integration-puppetmaster01 is DOWN: CRITICAL - Host Unreachable (10.68.22.41) [20:02:06] How do I get the unit tests for an extension to run in database mode? Specifically, for WikibaseMediaInfo we have an integration test we want to put in to stop Wikibase breaking changes not being spotted. T211133 -> https://gerrit.wikimedia.org/r/c/478226 -> "exception 'Wikimedia\Rdbms\DBAccessError' with message 'Database access has been disabled.'" [20:02:07] T211133: WikibaseMediaInfo needs a basic integration test suite asap - https://phabricator.wikimedia.org/T211133 [20:12:46] 10Beta-Cluster-Infrastructure, 10ContentTranslation, 10Wikimedia-Site-requests, 10Patch-For-Review, 10User-Luke081515: Put beta eswiki to read-only mode - https://phabricator.wikimedia.org/T109157 (10bmansurov) Can we unlock this wiki for registered users? I need to create some pages for testing the reco... [20:40:22] 10Gerrit: Rename Gerrit repository "LdapGroups" to "LDAPGroups" - https://phabricator.wikimedia.org/T200736 (10Umherirrender) A repo LDAPGroups exists now and conflicts with the lowercase name on windows. Was the creation part of the rename? [20:40:40] PROBLEM - Host webperformance is DOWN: CRITICAL - Host Unreachable (10.68.20.166) [20:40:51] PROBLEM - Host integration-r-lang-01 is DOWN: CRITICAL - Host Unreachable (10.68.20.232) [21:12:58] 10Release-Engineering-Team (Kanban), 10Cloud-Services, 10Epic, 10Patch-For-Review: Migrate the Integration cloud project to eqiad1-r - https://phabricator.wikimedia.org/T208803 (10Andrew) I've now moved everything in this project except for jenkinstest.integration.eqiad.wmflabs. I can't tell what that is... [22:04:46] 10Gerrit, 10Release-Engineering-Team (Backlog), 10Patch-For-Review: Upgrade to Gerrit 2.16.1 - https://phabricator.wikimedia.org/T200739 (10Paladox) [22:05:04] 10Gerrit, 10Release-Engineering-Team (Backlog), 10Patch-For-Review: Upgrade to Gerrit 2.16.1 - https://phabricator.wikimedia.org/T200739 (10Paladox) [22:39:11] delete-project and webhooks are now core gerrit plugins [22:39:26] (as of gerrit 3.0) [22:52:47] James_F: run it with something like mwscript tests/phpunit/phpunit.php /path/to/test (php instead of mwscript probably works if you are not on vagrant) [22:53:08] the test class must extend MediaWikiTestCase and have an @group Database annotation [22:53:38] tgr: Yes, it already does. The problem is how do we get CI's job runner to run the @group database ones correctly? [22:54:25] (Or is it already doing so but not working for some reason?) [22:54:51] in theory it does so [22:55:59] not through the composer entry point though, so I assume the CI config must include the right job [22:57:09] Maybe the DB tests only run for core and not extensions or something magical? [23:00:14] they definitely run for extensions [23:00:30] if there is any magic involved, it is probably in the integration/config repo [23:01:56] No-doubt. :-) [23:02:01] but then again the core DB tests do run for that patch so that cannot be the problem [23:02:11] all tests are run by the same job [23:03:09] Hmm, yes. [23:03:14] * James_F sighs. [23:03:23] and the same test failed in PS4 so testfile autodiscovery is also working fine [23:04:02] PS 4 ran 17673 tests and PS5 only 4001 [23:04:16] Yes, I can run it locally (but not successfully, because vagrant and writing files aren't a good mix.) [23:04:36] Does PS5 abort before it gets to the remaining 13k tests? [23:04:56] maybe that warning is actually a fatal and phpunit is being stupid about it? [23:05:09] Yeah. [23:07:49] per T201290 "This exception from CentralAuth code sometimes causes Jenkins jobs for MediaWiki core commits to fail" [23:07:50] T201290: CentralAuth causes DBAccessError in non-database phpunit suite - https://phabricator.wikimedia.org/T201290 [23:07:58] so I guess just recheck [23:09:36] no, that doesn't make sense [23:09:48] "Sometimes"? [23:10:05] phpunit calculates the number of tests before starting to execute them [23:10:06] * James_F sighs. [23:10:21] Well, I'm sure it'll become obvious over the weekend. [23:10:52] so this was caused by some test code failing horribly, that must have been in a data provider or testcase constructor or some such [23:46:29] James_F: https://integration.wikimedia.org/ci/job/quibble-vendor-mysql-hhvm-docker/27079/console for https://gerrit.wikimedia.org/r/c/mediawiki/extensions/WikibaseMediaInfo/+/477820 also has 4001 patches (and the same warning, so that's apparently unrelated - also that warning seems to be triggered after all tests have run) [23:46:38] something is really weird there [23:47:48] Hmm. [23:52:42] PS4 ran phpunit with --testsuite extensions --exclude-group Broken,ParserFuzz,Stub,Database and that resulted in 17673 tests. PS5 ran it with --testsuite extensions --group Database --exclude-group Broken,ParserFuzz,Stub and that resulted in 4001 tests. [23:53:38] also, that's a crazy amount of tests, looking test results for other repos the extension testsuite does not seem to run core tests