[00:05:18] (03PS1) 10Krinkle: doc: Add entry for Arc Lamp and Excimer [integration/docroot] - 10https://gerrit.wikimedia.org/r/593970 (https://phabricator.wikimedia.org/T251213) [00:05:47] (03CR) 10Krinkle: [C: 03+2] doc: Add entry for Arc Lamp and Excimer [integration/docroot] - 10https://gerrit.wikimedia.org/r/593970 (https://phabricator.wikimedia.org/T251213) (owner: 10Krinkle) [00:06:43] (03Merged) 10jenkins-bot: doc: Add entry for Arc Lamp and Excimer [integration/docroot] - 10https://gerrit.wikimedia.org/r/593970 (https://phabricator.wikimedia.org/T251213) (owner: 10Krinkle) [00:10:07] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO, 10Operations, 10Patch-For-Review: Migrate contint* hosts to Buster - https://phabricator.wikimedia.org/T224591 (10Krinkle) @hashar I noticed that the `zuul` com... [00:25:24] (03PS1) 10Krinkle: jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 [00:26:10] (03CR) 10jerkins-bot: [V: 04-1] jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [00:26:27] (03CR) 10Krinkle: "I noticed in some mw-config jobs that the Jenkins output mentioned a dirty 'composer.lock' file, which is due to the dev-requires being in" [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [06:27:08] (03CR) 10Thiemo Kreuz (WMDE): Full rewrite of the UnusedUseStatementSniff, 4x faster (037 comments) [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/592415 (owner: 10Thiemo Kreuz (WMDE)) [06:34:33] 10Beta-Cluster-Infrastructure, 10Wikidata, 10Wikidata-Campsite, 10Regression, and 2 others: A revision of https://wikidata.beta.wmflabs.org/wiki/Q64 consistently doesn't load on beta wikidata - https://phabricator.wikimedia.org/T249018 (10DannyS712) Now just gives `Service Temporarily Unavailable Our serve... [06:34:40] 10MediaWiki-Codesniffer, 10User-thiemowmde: Allow float|int documentation for strictly typed float parameters - https://phabricator.wikimedia.org/T251722 (10thiemowmde) [06:38:05] 10MediaWiki-Codesniffer, 10User-thiemowmde: Allow float|int documentation for strictly typed float parameters - https://phabricator.wikimedia.org/T251722 (10DannyS712) Should thit wait until after the cleanup of `FunctionCommentSniff` in {T250958}? [06:38:06] (03PS5) 10DannyS712: Move FunctionCommentSniff annotation checks to FunctionAnnotationSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593746 (https://phabricator.wikimedia.org/T250958) [06:38:43] 10Beta-Cluster-Infrastructure, 10Wikidata, 10Wikidata-Campsite, 10Regression, and 2 others: A revision of https://wikidata.beta.wmflabs.org/wiki/Q64 consistently doesn't load on beta wikidata - https://phabricator.wikimedia.org/T249018 (10RhinosF1) @DannyS712: loads fine here [06:39:31] 10Beta-Cluster-Infrastructure, 10Wikidata, 10Wikidata-Campsite, 10Regression, and 2 others: A revision of https://wikidata.beta.wmflabs.org/wiki/Q64 consistently doesn't load on beta wikidata - https://phabricator.wikimedia.org/T249018 (10DannyS712) >>! In T249018#6103373, @RhinosF1 wrote: > @DannyS712: lo... [06:41:52] 10Beta-Cluster-Infrastructure, 10Wikidata, 10Wikidata-Campsite, 10Regression, and 2 others: A revision of https://wikidata.beta.wmflabs.org/wiki/Q64 consistently doesn't load on beta wikidata - https://phabricator.wikimedia.org/T249018 (10RhinosF1) I know what I did. Sorry, still half alseep. > If you rep... [06:46:05] (03PS1) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeesSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:50:30] (03PS2) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:50:32] (03PS3) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:50:34] (03CR) 10jerkins-bot: [V: 04-1] Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) (owner: 10DannyS712) [06:50:36] 10MediaWiki-Codesniffer, 10Patch-For-Review, 10User-DannyS712: Split up FunctionCommentSniff - https://phabricator.wikimedia.org/T250958 (10thiemowmde) [06:51:01] (03PS4) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:51:49] 10MediaWiki-Codesniffer, 10User-thiemowmde: Allow float|int documentation for strictly typed float parameters - https://phabricator.wikimedia.org/T251722 (10thiemowmde) I don't think this will be among the things we refactor out of the sniff. So no, I don't think this needs to wait. [06:53:19] (03PS5) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:56:04] (03PS6) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [06:57:05] (03PS7) 10DannyS712: Move FunctionCommentSniff EmptySees check to new EmptySeeSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/593980 (https://phabricator.wikimedia.org/T250958) [07:51:12] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10brennen) a:05brennen→03LarsWirzenius This should, I believe, be clear to move forward on Monday morning once the [[... [10:05:56] 10Release-Engineering-Team-TODO, 10Packaging, 10Patch-For-Review, 10Upstream: gbp buildpackage with GIT_PBUILDER_AUTOCONF=no causes DIST to be ignored - https://phabricator.wikimedia.org/T233020 (10Volans) Thanks for the merge and the fix. It's now picking correctly stretch for me, I still had to change fr... [10:43:27] 10Release-Engineering-Team-TODO, 10Packaging, 10Patch-For-Review, 10Upstream: gbp buildpackage with GIT_PBUILDER_AUTOCONF=no causes DIST to be ignored - https://phabricator.wikimedia.org/T233020 (10Volans) I have spoken too soon, the support for components is also affected, as it depends on `DIST` too at t... [10:48:09] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10LarsWirzenius) That seems to have been merged, but I don't know if it's been deployed. [11:58:00] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)): Make deloy-promote more robust against failures - https://phabricator.wikimedia.org/T251750 (10LarsWirzenius) [12:06:06] (03CR) 10Gehel: [C: 03+1] "LGTM as far as I understand JJB :(" [integration/config] - 10https://gerrit.wikimedia.org/r/593300 (https://phabricator.wikimedia.org/T247123) (owner: 10Hashar) [12:54:00] PROBLEM - Free space - all mounts on deployment-snapshot01 is CRITICAL: CRITICAL: deployment-prep.deployment-snapshot01.diskspace._data.byte_percentfree (No valid datapoints found)deployment-prep.deployment-snapshot01.diskspace.root.byte_percentfree (<10.00%) [13:00:48] PROBLEM - Free space - all mounts on deployment-deploy01 is CRITICAL: CRITICAL: deployment-prep.deployment-deploy01.diskspace._srv.byte_percentfree (<10.00%) [13:03:07] 10Phabricator, 10Release-Engineering-Team (Development services), 10User-MModell: Make sure elasticsearch 6 is supported in phabricator - https://phabricator.wikimedia.org/T181393 (10Paladox) ` root@phabricator-prod-1001:/srv/phab/phabricator# bin/search init Initializing search service "Elasticsearch". Serv... [13:04:01] RECOVERY - Free space - all mounts on deployment-snapshot01 is OK: OK: deployment-prep.deployment-snapshot01.diskspace._data.byte_percentfree (No valid datapoints found) [13:09:59] 10Release-Engineering-Team-TODO, 10Packaging, 10Patch-For-Review, 10Upstream: gbp buildpackage with GIT_PBUILDER_AUTOCONF=no causes DIST to be ignored - https://phabricator.wikimedia.org/T233020 (10Volans) Tested @JMeybohm's fixes and all worked fine with my package. [13:10:01] PROBLEM - Free space - all mounts on deployment-snapshot01 is CRITICAL: CRITICAL: deployment-prep.deployment-snapshot01.diskspace._data.byte_percentfree (No valid datapoints found)deployment-prep.deployment-snapshot01.diskspace.root.byte_percentfree (<10.00%) [13:10:40] PROBLEM - Parsoid on deployment-parsoid11 is CRITICAL: connect to address 172.16.1.115 and port 8000: Connection refused [13:10:49] RECOVERY - Free space - all mounts on deployment-deploy01 is OK: OK: All targets OK [13:11:33] (03CR) 10Hashar: [C: 03+2] "The pom.xml now supports developerConnection" [integration/config] - 10https://gerrit.wikimedia.org/r/593300 (https://phabricator.wikimedia.org/T247123) (owner: 10Hashar) [13:12:36] (03Merged) 10jenkins-bot: Override scm url for wikidata/query/rdf [integration/config] - 10https://gerrit.wikimedia.org/r/593300 (https://phabricator.wikimedia.org/T247123) (owner: 10Hashar) [13:19:45] 10Release-Engineering-Team-TODO, 10Packaging, 10Patch-For-Review, 10Upstream: gbp buildpackage with GIT_PBUILDER_AUTOCONF=no causes DIST to be ignored - https://phabricator.wikimedia.org/T233020 (10JMeybohm) 05Open→03Resolved a:03JMeybohm I've also updated the docs on gbp buildpackage at: https://ger... [13:20:01] RECOVERY - Free space - all mounts on deployment-snapshot01 is OK: OK: deployment-prep.deployment-snapshot01.diskspace._data.byte_percentfree (No valid datapoints found) [13:20:04] (03PS1) 10Lars Wirzenius: Fix: scap.yaml to work with current Subplot [tools/scap] - 10https://gerrit.wikimedia.org/r/594166 [13:21:39] (03CR) 10Lars Wirzenius: "A tiny change to adapt to changes in Subplot. This change is needed to make the test suite work again in environments where Subplot is ins" [tools/scap] - 10https://gerrit.wikimedia.org/r/594166 (owner: 10Lars Wirzenius) [13:21:52] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10hashar) p:05Triage→03Medium I ran the job again https://int... [14:17:50] 10Release-Engineering-Team-TODO, 10Scap, 10MediaWiki-Internationalization, 10Performance-Team, 10Patch-For-Review: Use static php array files for l10n cache at WMF (instead of CDB) - https://phabricator.wikimedia.org/T99740 (10Krinkle) >>! In T99740#6089198, @Joe wrote: > * Rendering the enwiki Barack Ob... [14:34:20] 10Beta-Cluster-Infrastructure, 10Growth-Team, 10Core Platform Team Workboards (External Code Reviews), 10Patch-For-Review, and 2 others: Create beta viwiki - https://phabricator.wikimedia.org/T246511 (10daniel) Ping @Pchelolo [14:40:04] Project beta-scap-eqiad build #299105: 04FAILURE in 5 min 34 sec: https://integration.wikimedia.org/ci/job/beta-scap-eqiad/299105/ [14:41:51] PROBLEM - English Wikipedia Main page on beta-cluster is CRITICAL: CRITICAL - Socket timeout after 10 seconds [14:41:56] PROBLEM - App Server Main HTTP Response on deployment-mediawiki-07 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [14:46:46] RECOVERY - English Wikipedia Main page on beta-cluster is OK: HTTP OK: HTTP/1.1 200 OK - 93554 bytes in 2.138 second response time [14:48:50] Yippee, build fixed! [14:48:50] Project beta-scap-eqiad build #299106: 09FIXED in 4 min 18 sec: https://integration.wikimedia.org/ci/job/beta-scap-eqiad/299106/ [14:55:55] 10Beta-Cluster-Infrastructure, 10Growth-Team, 10Core Platform Team Workboards (External Code Reviews), 10User-Urbanecm, 10Wiki-Setup (Create): Create beta viwiki - https://phabricator.wikimedia.org/T246511 (10Pchelolo) RESTBase was enabled on the new domain: https://vi.wikipedia.beta.wmflabs.org/api/rest... [14:56:56] (03PS2) 10Thcipriani: Fix: scap.yaml to work with current Subplot [tools/scap] - 10https://gerrit.wikimedia.org/r/594166 (owner: 10Lars Wirzenius) [14:57:31] (03CR) 10Thcipriani: [C: 03+2] Fix: scap.yaml to work with current Subplot [tools/scap] - 10https://gerrit.wikimedia.org/r/594166 (owner: 10Lars Wirzenius) [15:01:19] (03Merged) 10jenkins-bot: Fix: scap.yaml to work with current Subplot [tools/scap] - 10https://gerrit.wikimedia.org/r/594166 (owner: 10Lars Wirzenius) [15:01:48] RECOVERY - App Server Main HTTP Response on deployment-mediawiki-07 is OK: HTTP OK: HTTP/1.1 200 OK - 93251 bytes in 0.872 second response time [15:25:43] 10Beta-Cluster-Infrastructure, 10Growth-Team, 10Core Platform Team Workboards (External Code Reviews), 10User-Urbanecm, 10Wiki-Setup (Create): Create beta viwiki - https://phabricator.wikimedia.org/T246511 (10Urbanecm) 05Open→03Resolved So seems everything is finally done, then [15:33:11] 10Continuous-Integration-Config, 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Core Platform Team, and 2 others: Move math related repositories from GitHub(?) to gerrit - https://phabricator.wikimedia.org/T249908 (10Jdforrester-WMF) >>! In... [15:33:19] 10Continuous-Integration-Config, 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Core Platform Team, and 2 others: Move math related repositories from GitHub(?) to gerrit - https://phabricator.wikimedia.org/T249908 (10Jdforrester-WMF) 05Open... [15:48:45] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10Jdforrester-WMF) [15:54:51] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10DannyS712) >>! In T249962#6104044, @LarsWirzenius wrote: > That seems to have been merged, but I don't know if it's bee... [16:01:10] (03CR) 10Hashar: "Apparently operations/mediawiki-config is the last use of composer-install-dev-only , though mediawiki/vendor was supposed to use it as we" (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:01:47] PROBLEM - Free space - all mounts on deployment-deploy01 is CRITICAL: CRITICAL: deployment-prep.deployment-deploy01.diskspace._srv.byte_percentfree (<10.00%) [16:04:52] (03PS2) 10Jforrester: jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:05:58] (03PS3) 10Jforrester: jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:06:12] 10Continuous-Integration-Config, 10Code-Health, 10Readers-Web-Backlog (Kanbanana-2019-20-Q4), 10Vue.js: Configure ESLint for Vue.js search development ahead of it being done for all repos - https://phabricator.wikimedia.org/T249304 (10Catrope) eslint-config-wikimedia now automatically lints Vue files, star... [16:06:19] (03PS4) 10Jforrester: jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:06:45] 10Continuous-Integration-Config, 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Core Platform Team, and 2 others: Move math related repositories from GitHub(?) to gerrit - https://phabricator.wikimedia.org/T249908 (10Physikerwelt) >>! In T24... [16:10:57] (03PS1) 10Jforrester: dockerfiles: [composer-php70] Drop composer-install-dev-only, unused [integration/config] - 10https://gerrit.wikimedia.org/r/594223 [16:11:13] (03CR) 10Jforrester: [C: 03+2] "Deployed." [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:12:16] (03Merged) 10jenkins-bot: jjb: Let mw-config jobs honour the composer.lock file [integration/config] - 10https://gerrit.wikimedia.org/r/593971 (owner: 10Krinkle) [16:13:22] !log Cleaning git repositories on deployment-deploy01 [16:13:23] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [16:14:02] Project beta-scap-eqiad build #299115: 04FAILURE in 4 min 33 sec: https://integration.wikimedia.org/ci/job/beta-scap-eqiad/299115/ [16:16:47] RECOVERY - Free space - all mounts on deployment-deploy01 is OK: OK: All targets OK [16:23:50] (03CR) 10Jforrester: [C: 03+2] dockerfiles: [composer-php70] Drop composer-install-dev-only, unused [integration/config] - 10https://gerrit.wikimedia.org/r/594223 (owner: 10Jforrester) [16:24:49] (03Merged) 10jenkins-bot: dockerfiles: [composer-php70] Drop composer-install-dev-only, unused [integration/config] - 10https://gerrit.wikimedia.org/r/594223 (owner: 10Jforrester) [16:25:17] PROBLEM - Free space - all mounts on deployment-deploy02 is CRITICAL: CRITICAL: deployment-prep.deployment-deploy02.diskspace._srv.byte_percentfree (<11.11%) [16:26:33] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10Mstyles) I know what that is, I forgot to add the archiva deplo... [16:26:36] Project beta-scap-eqiad build #299116: 04STILL FAILING in 11 min: https://integration.wikimedia.org/ci/job/beta-scap-eqiad/299116/ [16:27:29] 10MediaWiki-Codesniffer, 10User-DannyS712: Add `goto` to AlternativeSyntaxSniff - https://phabricator.wikimedia.org/T251783 (10DannyS712) [16:27:52] (03PS1) 10DannyS712: Add `goto` to AlternativeSyntaxSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) [16:28:28] (03PS2) 10DannyS712: Add `goto` to AlternativeSyntaxSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) [16:29:02] (03PS1) 10Mstyles: jjb: add archiva profile to wdqs deploy job [integration/config] - 10https://gerrit.wikimedia.org/r/594228 (https://phabricator.wikimedia.org/T247123) [16:31:56] (03PS1) 10Jforrester: jjb: Default 'logdir' for macros to /log [integration/config] - 10https://gerrit.wikimedia.org/r/594230 [16:31:58] (03PS1) 10Jforrester: jjb: Upgrade uses over to composer without composer-install-dev-only [integration/config] - 10https://gerrit.wikimedia.org/r/594231 [16:32:26] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10LarsWirzenius) Yeah, train is all at wmf.30 again. Now watching logs with extra nervosity. [16:33:34] Yippee, build fixed! [16:33:34] Project beta-scap-eqiad build #299117: 09FIXED in 5 min 21 sec: https://integration.wikimedia.org/ci/job/beta-scap-eqiad/299117/ [16:34:13] (03PS2) 10Jforrester: jjb: Default 'logdir' for macros to /log [integration/config] - 10https://gerrit.wikimedia.org/r/594230 [16:36:08] !log Docker: [composer-php70] Drop composer-install-dev-only, unused [16:36:09] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [16:36:22] maryum: hi, sounds like an easy patch ;) [16:37:04] (03CR) 10Hashar: [C: 03+2] "Deployed and building at https://integration.wikimedia.org/ci/job/wikidata-query-rdf-maven-release-docker-wdqs/7/console ;)" [integration/config] - 10https://gerrit.wikimedia.org/r/594228 (https://phabricator.wikimedia.org/T247123) (owner: 10Mstyles) [16:38:04] (03Merged) 10jenkins-bot: jjb: add archiva profile to wdqs deploy job [integration/config] - 10https://gerrit.wikimedia.org/r/594228 (https://phabricator.wikimedia.org/T247123) (owner: 10Mstyles) [16:39:17] hashar: Nice idea for defaulting to /log; see https://gerrit.wikimedia.org/r/594230 [16:40:36] ;) [16:40:37] 10MediaWiki-Codesniffer, 10User-DannyS712: Add a sniff to warn when public methods lack `@since` - https://phabricator.wikimedia.org/T251785 (10DannyS712) [16:41:05] James_F: if it is a noop, I will just +2 it ;) [16:41:17] hashar: I'm waiting for the docker build. [16:41:27] As after that I need to touch 200 jjb files anyway. :-( [16:41:48] hashar: Also, did you see puppet agent 1002? [16:41:55] (03CR) 10Hashar: [C: 03+2] "Noop indeed ;)" [integration/config] - 10https://gerrit.wikimedia.org/r/594230 (owner: 10Jforrester) [16:41:57] logdir: '{obj:logdir|/log}' [16:42:04] ^^ that earns you a jjb barnstar [16:42:05] ;D [16:42:10] * James_F grins. [16:42:13] puppet agent 1002? [16:42:26] I have lost track of a ton of things :/ [16:42:34] I made it on Friday. [16:42:55] https://integration.wikimedia.org/ci/computer/integration-agent-puppet-docker-1002/ [16:43:02] It's only labelled puppetTEST for nnow. [16:43:07] (03Merged) 10jenkins-bot: jjb: Default 'logdir' for macros to /log [integration/config] - 10https://gerrit.wikimedia.org/r/594230 (owner: 10Jforrester) [16:43:33] But even with manually installed docker it fails to be in the right group(?) [16:43:39] 00:00:00.454 docker: Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Post http://%2Fvar%2Frun%2Fdocker.sock/v1.40/containers/create: dial unix /var/run/docker.sock: connect: permission denied. [16:43:42] https://integration.wikimedia.org/ci/job/operations-puppet-tests-buster-docker-test/8/console [16:43:50] hashar: See https://integration.wikimedia.org/ci/job/operations-puppet-tests-buster-docker-test/8/console [16:43:53] Bah. [16:43:57] See https://gerrit.wikimedia.org/r/c/operations/puppet/+/593806, even [16:44:21] oh [16:44:43] so yeah hmm [16:44:49] apt.wm.o does not have the pinned docker version [16:44:51] If we can get that working, then it's just the cumin box and we're done with jessie. [16:44:54] Yeah. [16:44:57] and I have pinned it to avoid sudden upgrades [16:45:05] when some others magically want a different version [16:45:20] and because well, newer docker was terribly slow [16:45:20] It's still pinned with my patch, it's just pinned to a version that exists. [16:45:37] That was just the security supervisor slow down, right? [16:45:44] we just need to ensure we get the version from thirdparty/ci [16:46:08] yeah seccomp ( a linux kernel security feature) is what made it terribly slow [16:47:05] So the version doesn't matter much, now? [16:47:08] oh my god [16:47:43] ? [16:48:03] 8404d5cbef63a2c7f7c317f51ec2b0ee7e5f5bf1 [16:48:07] made it to use the Debian version [16:48:17] instead of the upstream one under thirdparty/ci [16:48:35] Yeah, I assumed you knew as you C+1'ed. [16:48:50] it is not like I have explicitly required for CI to have a specific version [16:48:51] It uses thirdparty-ci for stretch and jessie still. [16:49:34] yeah [16:49:38] then the devil is [16:49:48] we have apt::repository { 'thirdparty-ci' } [16:49:49] defined [16:50:07] but there is apparently no guarantee the package it provides will be picked over the onew provided by debian [16:50:12] though for jessie/stretch, that is the case [16:50:15] anyway I digress [16:50:20] This is on stretch. [16:50:30] I didn't want to try to get stuff working on buster right away. [16:50:35] Getting off jessie is more urgent. [16:50:51] yeah gotta reach out to Daniel to schedule the contint1001 upgrade [16:50:58] last week was too erratic familly wise [16:51:02] * James_F nods. [16:52:24] guess we can just cherry pick your patch on the integration puppetmaster [16:52:30] then mass upgrade docker [16:53:32] Let's get puppet-agent-1002 working first. [16:53:35] :-) [16:55:27] want me to guide you through the puppet cherry-pick or should I just do it? [16:56:14] oh it got +2ed [16:56:36] Yeah, easier to beg SRE. ;-) [16:57:11] ;D [16:57:50] Dear Sir Royal Enterprise, I could use a dime! [16:58:30] hashar: thanks for merging! [16:58:47] maryum: it fails with the same error though apparently -:( [16:59:16] hashar: seriously? maybe I put the profile in the wrong place. It comes before the release part when I run it on my machine [16:59:20] unrelated: git merge --squash does magic ;) [17:00:19] maryum: ahh might be [17:00:54] -Dresume=false -Duser.name=maven-release-user -DdeveloperConnection=scm:git:https://maven-release-user@gerrit.wikimedia.org/r/wikidata/query/rdf --settings /archiva-credentials.xml -B -Pdeploy-archiva release:prepare release:perform -Darguments=-Dgpg.skip=true [17:01:20] should -Pdeploy-archiva be passed after the goals? [17:01:29] * James_F manually triggers puppet agent on integration-agent-puppet-docker-1002 because he's impatient. [17:01:34] Success. [17:01:41] James_F: yum! [17:01:56] :-( [17:01:59] Still the job run fails. [17:02:03] 00:00:00.780 docker: Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Post http://%2Fvar%2Frun%2Fdocker.sock/v1.40/containers/create: dial unix /var/run/docker.sock: connect: permission denied. [17:02:22] Do we need to do something special? It runs as jenkins-deploy, right? [17:02:22] that would be docker refusing to start [17:02:26] That's in the docker group. [17:02:28] sudo systemctl status docker [17:02:42] Active: active (running) since Sat 2020-05-02 00:08:21 UTC; 2 days ago [17:02:45] ok [17:02:56] then check the jenkins-deploy user is actually a member of the group [17:02:59] I've restarted. [17:03:12] 10Phabricator (Upstream), 10Upstream: Restore highlighting of target comment in Phabricator - https://phabricator.wikimedia.org/T245089 (10epriestley) > It [the yellow highlight] was nice to bring the user's attention to the specific comment they were being linked to. I'm not entirely sure where in history th... [17:03:14] and last thing [17:03:16] jforrester@integration-agent-puppet-docker-1002:~$ sudo -u jenkins-deploy groups [17:03:16] wikidev docker project-bastion project-deployment-prep project-integration project-puppet-diffs [17:03:21] good [17:03:28] Last thing? [17:03:34] so potentially the jenkins agent has been added BEFORE the user got added to the group [17:03:45] so that the session is not aware of the docker group membership [17:03:56] so maybe try to disconnect the agent and connect it again [17:04:01] Ah, interesting. [17:04:02] Project mwcore-phpunit-coverage-master build #644: 04FAILURE in 2 hr 4 min: https://integration.wikimedia.org/ci/job/mwcore-phpunit-coverage-master/644/ [17:04:15] I think there is a way to reload the session somehow [17:04:26] but last time I have wasted a couple hours trying to find out how to reload the groups [17:04:37] Sure. :-) [17:04:40] and I am not even sure it is possible for an existing ssh session [17:04:41] so [17:04:44] RESTART! [17:04:45] ;) [17:04:53] Umm. [17:04:57] <===[JENKINS REMOTING CAPACITY]===>ERROR: Unexpected error in launching a slave. This is probably a bug in Jenkins. [17:04:58] java.lang.IllegalStateException: Already connected [17:05:02] What? [17:05:07] :D [17:05:11] Is that normal? [17:05:17] no idea [17:05:29] it is using UPPER CASE, so probably not that normal [17:05:34] Aha. [17:05:37] or the master magically reconnected it? [17:05:37] It's runnning! [17:05:41] https://integration.wikimedia.org/ci/job/operations-puppet-tests-buster-docker-test/10/console [17:05:54] * hashar brags and flexes [17:06:06] Assuming that works, I'll move puppet jobs over. [17:06:17] \o/ [17:06:39] integration-agent-puppet-docker-1001 is gone :( [17:06:42] ah no [17:06:44] Not yet! [17:06:51] It's the only puppet agent. [17:06:55] so the good news is: we get a stretch agent [17:06:58] the bad news is [17:07:04] * hashar rolls the drums [17:07:11] We need to move to buster. [17:07:14] 18012 ? S 10:07 \_ lldpd: connected to cloudvirt1005.eqiad.wmnet [17:07:27] Oh, it's on the "wrong" hardware? :-( [17:07:40] 10Continuous-Integration-Config, 10Scoring-platform-team: CI should check to see if our wheels are good - https://phabricator.wikimedia.org/T250746 (10Halfak) I think we'd just need to run the `pip install wheels/*.whl --no-deps` command and see if it exits successfully. [17:07:41] hashar: the command I run on my machine looks like `./mvnw -Pdeploy-archiva release:prepare` so that seems like the right thing. nothing about maven should be showing up in this job [17:07:43] yeah the hardware that has some faulty cpu governor which needs a tweak in the bios [17:07:45] .. [17:07:57] last update is that the hardware will eventually be decom at some point [17:07:58] so [17:08:07] we gotta delete it and rebuild it [17:08:08] ;( [17:08:25] and feel pain [17:08:32] But not today. [17:08:33] 10Continuous-Integration-Config, 10Scoring-platform-team: CI should check to see if our wheels are good - https://phabricator.wikimedia.org/T250746 (10Halfak) p:05Triage→03Medium [17:08:37] ? [17:09:58] maryum: ohhhh so actually the console log is different. The status code changed from some 5xx to a 401 Unauthorized [17:10:22] !log Pooling integration-agent-puppet-docker-1002 to run puppet CI T250502 [17:10:24] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [17:10:24] T250502: Rebuild integration-agent-puppet-docker-1001 to use Stretch - https://phabricator.wikimedia.org/T250502 [17:10:25] so maybe that is an issue with the archiva credentials [17:10:44] James_F: yes today, we should dish it out [17:10:55] James_F: those boxes are really way too slow [17:10:56] Oy. [17:11:01] :((( [17:11:14] hashar: most likely....different teams would have different credentials, I'm assuming [17:11:27] hashar: How do we get a new one to be on the right hardware? [17:11:30] or the same job couldn't access those stored credentials? [17:12:15] James_F: either get the instances moved to a new box which require WMCS team to do some magic command. Or delete it and create a new one and hope for the openstack scheduler to spawn it on a proper machine [17:12:28] Let's ask WMCS. [17:12:41] maryum: at least 00:00:01.856 + setfacl -m user:65534:r archiva-credentials.xml **** [17:12:53] hashar: Do you know what hosts are OK to use? [17:12:57] maryum: seems the credentials are there. But maybe those do not grant write access for wdqs? :( [17:13:39] hashar: maybe? I don't know anything about how the credentials work [17:15:01] maryum: I don't either :-( [17:15:21] but at least the -P parameter changed something :] [17:16:43] hashar: true, on to the next issue [17:17:03] James_F: I should have checked before :/ [17:17:14] * James_F blames sunspots. [17:17:53] maryum: I guess you will have to hunt someone familiar with archiva and or having access on it. Maybe ottomata would know [17:18:10] hashar: yes, someone on analytics would know I hope [17:18:22] 10Continuous-Integration-Config, 10Code-Health, 10Readers-Web-Backlog (Kanbanana-2019-20-Q4), 10Vue.js: Configure ESLint for Vue.js search development ahead of it being done for all repos - https://phabricator.wikimedia.org/T249304 (10Jdforrester-WMF) >>! In T249304#6105274, @Catrope wrote: > eslint-config... [17:19:47] hashar: I've pinged in -cloud, will see if they respond. [17:19:57] James_F: :) [17:20:26] maryum: https://wikitech.wikimedia.org/wiki/Archiva seems to suggest the user has to be in some LDAP. Though it is there for the refinery repository [17:23:02] hashar: right I know my personal user is....I don't know how that works for jenkins. this job is using the same maven-release-user information [17:23:07] as analytics [17:23:09] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10hashar) Moved to next error: `counterexample [INFO] [ERROR] Fai... [17:23:31] maryum: yeah so thoerically the credentials should be correct. But maybe we dont pass them rpoperly to maven [17:24:17] hashar: it seems pretty much the same between analytics....but something is missing somewhere. [17:24:22] (03CR) 10Thiemo Kreuz (WMDE): [C: 04-1] "How is "goto" an "alternative syntax"? "Alternative" for what? Sure, we should block it. But I think this is the wrong sniff to add this." [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [17:24:30] --settings /archiva-credentials.xml -B -Pdeploy-archiva [17:24:46] could it be that the -Pdeploy-archiva override what we have in the archiva-credentials.xml ? [17:25:13] (03CR) 10DannyS712: "> How is "goto" an "alternative syntax"? "Alternative" for what?" [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [17:25:16] 10Scap, 10ORES, 10Scoring-platform-team: Investigate: why are we getting overload errors during ORES deployments? - https://phabricator.wikimedia.org/T213116 (10Halfak) We are getting overload errors because the celery workers stop consuming tasks for a moment and it takes celery about 2 minutes to go from o... [17:25:18] hashar: maybe the jenkins stored credential isn't accessible? and that's possible as well [17:25:24] yeah [17:25:40] 10Scap, 10ORES, 10Scoring-platform-team: Document delay between canary and continued deployment to minimize overload during restarts - https://phabricator.wikimedia.org/T213116 (10Halfak) [17:26:58] hashar: 1002 was already on new hardware – 10*1*5, not 1005. [17:27:24] OHHH [17:27:38] !log Depooling integration-agent-puppet-docker-1001 from running puppet CI T250502 [17:27:41] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [17:27:41] T250502: Rebuild integration-agent-puppet-docker-1001 to use Stretch - https://phabricator.wikimedia.org/T250502 [17:27:42] (03CR) 10Thiemo Kreuz (WMDE): [C: 04-1] "I don't understand. "endif" is literally an alternative for a closing "}". Avoiding an "endif" is pretty straightforward because of this. " [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [17:27:43] I was looking at 1001 :-(( [17:27:46] I am sooo useless [17:27:48] Yes. [17:27:49] hashar: I don't think the archiva profile could override the credentials...it just overrides the release depository locations [17:27:52] I blame corona. [17:28:54] (03CR) 10DannyS712: "> I don't understand. "endif" is literally an alternative for a" [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [17:36:13] (03PS1) 10Jforrester: layout: [operations/puppet] Drop operations-puppet-tests-buster-docker-test [integration/config] - 10https://gerrit.wikimedia.org/r/594248 [17:36:15] (03PS1) 10Jforrester: jjb: Drop operations-puppet-tests-buster-docker-test [integration/config] - 10https://gerrit.wikimedia.org/r/594249 [17:36:27] maryum: from what I checked, it seems the credentials are correct. But maybe the user has to be granted write access to /org/wikidata/query/rdf or something like that [17:36:34] maryum: would be nice to be able to check archiva config [17:36:36] or its logs [17:36:41] (03CR) 10Jforrester: [C: 03+2] layout: [operations/puppet] Drop operations-puppet-tests-buster-docker-test [integration/config] - 10https://gerrit.wikimedia.org/r/594248 (owner: 10Jforrester) [17:37:39] (03CR) 10Jforrester: [C: 03+2] "Deployed." [integration/config] - 10https://gerrit.wikimedia.org/r/594249 (owner: 10Jforrester) [17:37:46] (03Merged) 10jenkins-bot: layout: [operations/puppet] Drop operations-puppet-tests-buster-docker-test [integration/config] - 10https://gerrit.wikimedia.org/r/594248 (owner: 10Jforrester) [17:38:42] (03Merged) 10jenkins-bot: jjb: Drop operations-puppet-tests-buster-docker-test [integration/config] - 10https://gerrit.wikimedia.org/r/594249 (owner: 10Jforrester) [17:41:33] !log Dropping integration-agent-puppet-docker-1001 from jenkins for T250502 [17:41:36] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [17:41:36] T250502: Rebuild integration-agent-puppet-docker-1001 to use Stretch - https://phabricator.wikimedia.org/T250502 [17:43:04] !log Shutting off integration-agent-puppet-docker-1001 in WMCS for T250502 [17:43:07] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [17:43:24] Let's wait a day or so just in case before deleting. [17:46:21] hashar: I'll see what I can figure out about the credentials.... [17:47:15] James_F: AbuseFilter is in gatedextensions. [17:47:46] aside from mediawiki-core and api-testing, no repos have 'sqlite' in a pipeline other than experimental. [17:47:46] Yes, that sounds right. [17:48:23] so it may've gotten lost for gated repos then. [17:51:00] (03CR) 10Thiemo Kreuz (WMDE): "Hm. I don't think this is a good reason. With this argument alone we could add it to any random sniff. The question is: does it make sense" [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [17:51:49] PROBLEM - Host integration-agent-puppet-docker-1001 is DOWN: CRITICAL - Host Unreachable (172.16.1.247) [17:55:06] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10LarsWirzenius) a:05LarsWirzenius→03brennen Over to Brennen again. [17:55:24] (03PS2) 10Jforrester: jjb: Upgrade uses over to composer without composer-install-dev-only [integration/config] - 10https://gerrit.wikimedia.org/r/594231 [17:55:36] (03CR) 10Jforrester: [C: 03+2] "Deployed." [integration/config] - 10https://gerrit.wikimedia.org/r/594231 (owner: 10Jforrester) [17:56:43] (03Merged) 10jenkins-bot: jjb: Upgrade uses over to composer without composer-install-dev-only [integration/config] - 10https://gerrit.wikimedia.org/r/594231 (owner: 10Jforrester) [17:57:49] 10Continuous-Integration-Config, 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)): operations-puppet-tests-buster-docker times out after 5 minutes - https://phabricator.wikimedia.org/T249726 (10Jdforrester-WMF) 05Stalled→03Resolved All puppet p... [17:58:14] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)): Rebuild integration-agent-puppet-docker-1001 to use Stretch - https://phabricator.wikimedia.org/T250502 (10Jdforrester-WMF) Will delete th... [17:58:28] 10Deployments, 10Phabricator, 10Developer Productivity, 10User-MModell, 10User-Majavah: Create a permalink which always redirects to the current week's train blocker task - https://phabricator.wikimedia.org/T207669 (10Majavah) 05Open→03Resolved [18:31:03] !log Un-stall beta cluster puppetmaster (100 patches behind due to merge conflict) [18:31:06] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [18:32:14] thcipriani: is this blocked due to bad or unknown prod would-be impact? https://gerrit.wikimedia.org/r/#/c/operations/puppet/+/361796/ [18:32:18] Or is it uncontroversial? [18:32:52] (03CR) 10Daimona Eaytoy: [C: 04-1] "I agree with Thiemo that 'goto' isn't really an alternative like, for instance, 'endif' is. Sometimes it's not even clear what the replace" (031 comment) [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [18:34:00] (03CR) 10DannyS712: Add `goto` to AlternativeSyntaxSniff (031 comment) [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [18:34:04] (03Abandoned) 10DannyS712: Add `goto` to AlternativeSyntaxSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/594227 (https://phabricator.wikimedia.org/T251783) (owner: 10DannyS712) [18:34:22] 10MediaWiki-Codesniffer, 10Patch-For-Review, 10User-DannyS712: Add `goto` to AlternativeSyntaxSniff - https://phabricator.wikimedia.org/T251783 (10DannyS712) 05Open→03Invalid Already have a rule that does this... [18:46:28] James_F: do you know who I would need to talk to for permission to view this page? https://integration.wikimedia.org/ci/credentials/ [18:46:43] trying to debug the credentials issue with WDQS [18:56:34] maryum: https://github.com/wikimedia/puppet/blob/e512a33d6dad4052efbe835b401727197b16712c/modules/admin/data/data.yaml#L175 [18:56:56] reedy: thanks!! [18:57:52] paladox: Is https://github.com/wikimedia/puppet/pull/5 a bot or a person or something inbetween? [18:58:08] Reedy that's a LTA [18:58:51] WM one? [18:59:05] Yeh, i think he's been banned from the WMF [18:59:11] lol [18:59:41] Do you have any more details? Cause we should probably ban them from our GH org if possible.. [19:00:26] From what i know, they are upsessed with toolforge, they came into the miraheze channel frequentl saying something along the lines of "will miraheze have toolforge"? [19:00:42] RhinosF1 you probably know more ^ [19:02:08] Reedy and they keep opening up stupid pull requests in our puppet repo (similar to the pull you've linked (he touches varnish)) [19:02:37] This feels dejavu [19:03:06] Didn't we have someone recently wanting rights to "upgrade varnish" in phab? [19:03:07] * Reedy looks [19:03:13] yeh [19:05:04] Can't find it offhand [19:05:06] * Reedy grumbles [19:09:12] phab search sucks [19:09:30] 10Continuous-Integration-Config, 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10doxygen, and 2 others: Update Doxygen in CI to 1.8.17 or greater - https://phabricator.wikimedia.org/T242155 (10hashar) 1.8.17 has been tagged for Debian unstabl... [19:10:50] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10Mstyles) Talked to @Gehel and the issue is probably that the ar... [19:18:17] https://phabricator.wikimedia.org/T250350 is one of them, email sounds like the LTA. [19:21:29] 10Diffusion, 10Tool-spacemedia, 10Wikimedia-GitHub: Diffusion mirroring to GitHub doesn't work for tool-spacemedia - https://phabricator.wikimedia.org/T251699 (10Don-vip) @Reedy thank for the advice. I have pushed a commit to the GitHub repository so it now exists, but I still can't see anything happening th... [19:22:26] 10Diffusion, 10Tool-spacemedia, 10Wikimedia-GitHub: Diffusion mirroring to GitHub doesn't work for tool-spacemedia - https://phabricator.wikimedia.org/T251699 (10Reedy) Have you done any write/push actions to your diffusion repo? [19:25:43] Reedy i think the LTA is https://www.mediawiki.org/w/index.php?title=User:Joaquinito01 (but RhinosF1 could probably confirm that) [19:26:08] Aha [19:26:55] Not the exact name I remember from phab... But it was similar [19:27:45] yeh [19:27:48] > heroku-miraheze has been blocked from the wikimedia organization. [19:28:15] heh [19:38:43] (03CR) 10Kosta Harlan: [C: 04-1] "> Patch Set 1:" [integration/quibble] - 10https://gerrit.wikimedia.org/r/591033 (https://phabricator.wikimedia.org/T248484) (owner: 10Kosta Harlan) [19:38:50] (03Abandoned) 10Kosta Harlan: Remove Vector from default setup [integration/quibble] - 10https://gerrit.wikimedia.org/r/591033 (https://phabricator.wikimedia.org/T248484) (owner: 10Kosta Harlan) [19:40:35] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing: Confirm 'vendor' job for MW installs dev deps correctly - https://phabricator.wikimedia.org/T251819 (10Krinkle) [19:40:45] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing: Confirm 'vendor' job for MW installs uses prod deps correctly - https://phabricator.wikimedia.org/T251819 (10Krinkle) [19:40:56] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing: Confirm 'vendor' job for MW installs uses prod deps correctly - https://phabricator.wikimedia.org/T251819 (10Krinkle) a:03Krinkle [19:41:08] Krinkle: thx ;) [19:41:25] which leads me to wonder what we run for mediawiki/vendor [19:41:30] OH [19:41:35] ported into quibble probably [19:48:05] Reedy, paladox: could be Joaquinito01 [19:48:16] Wmf office could confirm [19:49:23] it’s a bit polite for him [19:51:52] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing: Confirm 'vendor' job for MW installs uses prod deps correctly - https://phabricator.wikimedia.org/T251819 (10hashar) Seems like I got confused earlier wondering how we installed composer dev dependencies for mediawiki/vendor . Turns out we don't, Quibb... [20:01:14] 10Diffusion, 10Tool-spacemedia, 10Wikimedia-GitHub: Diffusion mirroring to GitHub doesn't work for tool-spacemedia - https://phabricator.wikimedia.org/T251699 (10bd808) We have not had good luck with ssh key based replication from Phabricator to GitHub in the past. I will setup the desired mirror and log wha... [20:14:23] 10Diffusion, 10Tool-spacemedia, 10Wikimedia-GitHub, 10cloud-services-team (Kanban): Diffusion mirroring to GitHub doesn't work for tool-spacemedia - https://phabricator.wikimedia.org/T251699 (10bd808) 05Open→03Resolved a:03bd808 Setting up a mirror to the https://github.com/toolforge org account: # C... [20:16:41] 10Continuous-Integration-Config, 10Librarization, 10MediaWiki-Vendor: Jenkins: Create job for verifying committed "vendor" directory from composer - https://phabricator.wikimedia.org/T74952 (10hashar) [20:18:04] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing: Confirm 'vendor' job for MW installs uses prod deps correctly - https://phabricator.wikimedia.org/T251819 (10Krinkle) 05Open→03Resolved https://github.com/wikimedia/quibble/blob/8191974690e7a84344664cfdc6b95864e473eeab/quibble/commands.py#L372-L383... [20:19:50] 10Continuous-Integration-Config, 10Librarization, 10MediaWiki-Vendor, 10Developer Productivity: Add Jenkins job to mediawiki/vendor repo to verify output matches clean build - https://phabricator.wikimedia.org/T74952 (10Krinkle) [20:19:53] 10Continuous-Integration-Config, 10Librarization, 10MediaWiki-Vendor, 10Developer Productivity: Add Jenkins job to mediawiki/vendor repo to verify output matches clean build - https://phabricator.wikimedia.org/T74952 (10Krinkle) [20:20:25] 10Continuous-Integration-Config, 10Librarization, 10MediaWiki-Vendor, 10Developer Productivity: Add Jenkins job to mediawiki/vendor repo to verify output matches clean build - https://phabricator.wikimedia.org/T74952 (10Krinkle) Tagging as productivity. This would significantly reduce CR cost for PHP vendo... [20:28:27] 10Continuous-Integration-Config, 10Librarization, 10MediaWiki-Vendor, 10Developer Productivity: Add Jenkins job to mediawiki/vendor repo to verify output matches clean build - https://phabricator.wikimedia.org/T74952 (10hashar) It seems we can have a job using the composer CI Docker container and just run:... [20:30:09] Reedy, paladox: he’s stupider than I thought https://tools.wmflabs.org/ldap/user/XPvistaseven2020 [20:30:20] lol [20:30:29] i linked above :) [20:30:30] I'll pass it onto cloud [20:30:43] Reedy: can you block him on wikitech? And maybe we can blacklist the email [20:31:39] Done [20:33:07] Reedy: which part [20:33:17] blocking [20:33:36] Reedy: what about an email blacklist? [20:35:34] hashar: I want to do a quibble release this week, BTW; 0.0.42 was broken and we only found out after the release, so I want to get 0.0.43 out with the fixes and the good stuff in 0.0.42. If you have anything you wanted in… [20:36:46] ah yeah quibble [20:36:54] 10Continuous-Integration-Config, 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10MediaWiki-Core-Testing, 10Vue.js: Use Jest in MediaWiki CI to test Vue.js components and plugins - https://phabricator.wikimedia.org/T251790 (10Jdforrester-WMF) This sounds like a reasonable plan (though people wi... [20:36:58] But buster migration is more urgent. :-) [20:37:14] maryum: hashar knows more than me, sorry. [20:37:19] awight has sent a lot of patches but I have troubles to find the couple hours needed to review each of them ;D [20:37:27] * James_F grins. [20:37:33] The parallel execution stuff looks good. [20:37:39] yeah then [20:39:02] (03CR) 10Hashar: "I do all my development in 80 columns terminals, maybe I am a Dinosaur. Regardless, I find it distracting to have over long lines and rea" [integration/quibble] - 10https://gerrit.wikimedia.org/r/591007 (owner: 10Awight) [20:39:14] hashar: 80! Wow. [20:39:20] yeah [20:39:23] single screen [20:39:26] hashar: That's not a dinosaur. That's much older. [20:39:30] with low resolution so there is not a lot of space [20:40:08] I usually have vertically split screen, one side with vim the other with gerrit/phabricator/irc or whatever relevant [20:40:21] or I go with a full screen terminal with full screen vim [20:40:29] which is like hmm ~ 170 columns [20:40:39] so split that gives me two vim each with ~ 85 colums [20:40:43] Right. [20:40:53] and I can't use high density screens anymore [20:41:01] my vision has degraded [20:41:07] My third-width terminal screen is 126 chars wide. [20:41:08] (I am past 40 ...) [20:41:47] https://en.wikipedia.org/wiki/Presbyopia .. [20:42:05] Ack. [20:43:11] so eg, I gotta get my laptop sufficiently far away [20:43:17] or drop my glasses ;D [20:43:23] anyway .. I am ranting [20:44:06] so for Quibble release [20:44:21] the open changes are mostly refactors of some sort or some new exciting features [20:44:40] and all the parallelization stuff, I haven't even looked at it yet ;( [20:46:21] RhinosF1: You missed a few it seems :P [20:46:39] Reedy: missed a few what? [20:46:47] Accounts on wikitech [20:46:52] https://wikitech.wikimedia.org/wiki/Special:Log/Reedy [20:47:19] Reedy: I’ll let the office know. Did he use his email on all those accounts? [20:47:39] Variations of joaquinito [20:47:51] James_F: so I guess yes +1 on cutting a new release. The only thing to watch is the parsoid job that will need to be updated but I think we already have a pending patch for it in integration/config [20:48:09] maryum: I had trouble trying to reproduce an upload. Guess I should try again ;) [20:48:33] hashar: did you see the latest comments on the phab ticket? I'm not sure what you're referring to [20:48:46] Reedy: I have a theory, maybe he wants to be caught. Why else would you deliberately place such incriminating evidence? [20:48:51] maryum: the archiva credentials are stored in a XML file which is read by maven as a user setting. It has been uploaded to the Jenkins master but can not be viewed through the Jenkins web interface [20:48:55] Laziness? [20:49:04] Incompetence? [20:49:05] hashar: We do, yes. [20:49:07] * hashar crawls through phabricator mails [20:49:19] Reedy: Is it not polite to assume brains first? [20:49:29] James_F: so be bold ;] I am hoping to have more bandwith to process awight patches [20:49:36] * James_F grins. [20:49:41] Cool. But no rush. :-) [20:49:43] RhinosF1: For people wasting other peoples times? [20:50:07] 10Continuous-Integration-Config, 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10MediaWiki-Core-Testing, 10Vue.js: Use Jest in MediaWiki CI to test Vue.js components and plugins - https://phabricator.wikimedia.org/T251790 (10egardner) @Jdforrester-WMF great, I can start a patch sometime this w... [20:50:08] hashar: that's fine. I think the issue should be sorted soon, see updates on phab ticket, https://phabricator.wikimedia.org/T247123 [20:50:13] maryum: AHHHHHHH [20:50:53] Reedy: True, but I somewhat doubt he’s stupid. He’s got method sometimes. Even his drugged up pseduo-identity has method. [20:50:56] maryum: I would never have figured that out ;) We can also store in jenkins a different setting file / different user [20:51:26] hashar: me neither! we already put the change in for the parent pom rename, so it should just work.....theoretically [20:53:08] maryum: let me retrieve the credential file somehow [20:53:29] hashar: sounds good, would be nice to verify [20:59:43] ah yeah got them [21:04:03] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10hashar) >>! In T247123#6106324, @Mstyles wrote: > Talked to @Ge... [21:05:08] maryum: fun thing, in your comment you linked to github resources refering to the 'master' branch [21:05:15] so by the time I looked at the url, your patch already got merged [21:05:23] and thus the serverId looked exactly the same ;] [21:05:38] hashar: ugh didn't think about that [21:05:40] (03CR) 10Jforrester: [C: 03+1] "I'm OK with this, but not fanatical. ;-)" [integration/quibble] - 10https://gerrit.wikimedia.org/r/591007 (owner: 10Awight) [21:06:05] maryum: yeah it is common tar pit (or whatever is the proper english idiom to say that it is a trap) [21:06:07] anyway [21:06:20] hashar: noted [21:06:41] does that mean we can try running the job again? [21:06:58] or does some magic has to happen in the wdqs repo to use the new discovery-parent-pom? [21:07:54] hashar: magic still....I have to get someone to approve and merge https://gerrit.wikimedia.org/r/c/wikidata/query/rdf/+/594304 [21:07:57] so not quite magic [21:08:00] but one more step [21:09:22] maryum: I guess I can do it [21:11:33] hashar: hmm why not? I don't think it's controversial [21:13:55] maryum: +2ed ;) [21:14:24] maryum: also do you have the proper permissions to trigger a build of the jenkins job? ( https://integration.wikimedia.org/ci/job/wikidata-query-rdf-maven-release-docker-wdqs/ ) [21:14:45] hashar: yep I do! [21:14:53] ouf [21:14:55] er [21:16:22] what would be fun is to convert all maven based repo to use similar jobs [21:16:56] hashar: yes that really should be the goal. Ideally our other maven projects could do the same thing [21:17:13] hashar: also it looks like this can't be merged? https://gerrit.wikimedia.org/r/c/wikidata/query/rdf/+/594304 due to the merge conflict I see [21:17:15] and I guess if one send a gpg signed tag, we might even have CI to react on it and automagically do the release [21:17:27] so git tag -s && git push --tags . Done! [21:18:25] maryum: that change is in CI and being processed [21:18:47] hashar: ah okay, I thought something different had to happen [21:19:04] hashar: that would be great with the gpg signed tags [21:19:16] yeah ;/ [21:26:38] maryum: change merged. I guess you can try running the job [21:45:18] I was distracted [21:45:20] running at https://integration.wikimedia.org/ci/job/wikidata-query-rdf-maven-release-docker-wdqs/8/console [21:48:47] hashar: fingers crossed! [21:54:03] oh my [21:54:06] SUCCESS!!!! [21:54:10] maryum: REJOICE! [21:54:11] ahah [21:54:17] well I guess it is week-end time now. [21:54:18] ! [21:54:35] hashar: wow that's amazing. thanks so much for your help!! [21:56:14] maryum: and congratulations on sorting out the dark arcanes of jjb [21:56:15] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10hashar) Eventually @Mstyles and I managed to get a build passin... [21:56:28] hashar: I'm two for two so far [22:01:44] * hashar learned another idiom [22:02:09] maryum: and I guess we can probably reuse the same logic for a bunch of other maven repos [22:02:26] gehel probably suggested it at some point [22:03:45] he did, I think there should be a way to standardize the job and pass different maven options as needed [22:14:15] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10Mstyles) Docs are updated -> https://wikitech.wikimedia.org/wik... [22:14:39] 10Release-Engineering-Team (Pipeline), 10Release-Engineering-Team-TODO, 10Operations, 10Prod-Kubernetes, and 2 others: TEC3:O6:O:6.1:Q3: Deployment Pipeline Documentation - https://phabricator.wikimedia.org/T213090 (10thcipriani) 05Open→03Resolved a:03jeena @LarsWirzenius and @Jdforrester-WMF work... [22:14:44] 10Release-Engineering-Team (Pipeline), 10Release-Engineering-Team-TODO, 10Operations, 10Release Pipeline, and 2 others: Migrate production services to kubernetes using the pipeline - https://phabricator.wikimedia.org/T198901 (10thcipriani) [22:17:02] 10Diffusion, 10Tool-spacemedia, 10Wikimedia-GitHub, 10cloud-services-team (Kanban): Diffusion mirroring to GitHub doesn't work for tool-spacemedia - https://phabricator.wikimedia.org/T251699 (10Don-vip) Thank you @bd808 ! :) [22:21:39] 10Release-Engineering-Team (Pipeline), 10Release-Engineering-Team-TODO, 10Operations, 10Release Pipeline, and 2 others: Migrate production services to kubernetes using the pipeline - https://phabricator.wikimedia.org/T198901 (10thcipriani) [22:22:10] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)): Delete Jenkins label DebianJessie - https://phabricator.wikimedia.org/T239981 (10Mstyles) [22:22:13] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Wikidata, 10Wikidata-Campsite, 10Wikidata-Query-Service, and 2 others: Migrate wikidata-query-rdf-release-silent release job to Docker - https://phabricator.wikimedia.org/T247123 (10Mstyles) 05Open→03Resolved [22:24:28] 10Release-Engineering-Team (Pipeline), 10Release-Engineering-Team-TODO, 10Release Pipeline, 10serviceops, 10Services (watching): TEC3:O3:O3.1:Q3 Goal - Move cxserver, citoid, changeprop, eventgate (new service) and ORES (partially) through the production CD Pipel... - https://phabricator.wikimedia.org/T212801 [22:24:32] 10Release-Engineering-Team (Pipeline), 10Release-Engineering-Team-TODO, 10Operations, 10Release Pipeline, and 2 others: Migrate production services to kubernetes using the pipeline - https://phabricator.wikimedia.org/T198901 (10thcipriani) [22:29:29] * hashar vanishes [22:40:20] 10Continuous-Integration-Config, 10MediaWiki-extensions-General: Add phan to MediaWiki extensions and skins for static analysis [cloneable] - https://phabricator.wikimedia.org/T179554 (10hashar) 05Open→03Resolved As I understand it, this task was for Google Summer of Code 2017 and intended to be cloneable... [22:42:01] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Cloud-VPS (Debian Jessie Deprecation): "integration" Cloud VPS project jessie deprecation - https://phabricator.wikimedia.org/T236576 (... [22:42:34] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (CI & Testing services), 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Cloud-VPS (Debian Jessie Deprecation): "integration" Cloud VPS project jessie deprecation - https://phabricator.wikimedia.org/T236576 (... [22:42:44] 10Release-Engineering-Team-TODO (2020-04 to 2020-06 (Q4)), 10Release, 10Train Deployments, 10User-brennen: 1.35.0-wmf.30 deployment blockers - https://phabricator.wikimedia.org/T249962 (10brennen) 05Open→03Resolved Has been stable since handed off from Lars. [22:51:35] James_F, just cumin left running in integration then? [22:51:43] in terms of jessie instances [22:59:42] Yeah. [23:00:01] I guess I need to sweet-talk SRE. ;-) [23:00:16] Thanks for all your work on this. [23:00:51] Of course, as soon as we’re done with Jessie we should start a Stretch to Buster task. [23:04:22] James_F: too true about Stretch -> Buster being needed sooner than we all expect. Moritz has already been warning the WMCS team about that. :/ [23:04:54] Maybe we should just run CI on Sid. ;-) [23:05:36] I was working on a buster variant of quibble but the changes to things like chrome -> chromium are a bit fiddly. [23:05:58] James_F, So I could build a cumin instance on stretch [23:06:02] but like [23:06:08] as you say we should probably get moving on that migration next [23:06:43] so it feels like it'd be of limited value [23:08:27] Maybe we should do that for now just to finish the Stretch stuff? [23:08:49] We’re already over four months late on bd808’s deadline. [23:09:19] I knew that CI and Beta would be the long tail [23:09:53] mostly now you are at the mercy of the core SREs deciding to start ripping code out of Puppet [23:10:09] but I think there are still some Jessie things lurking in prod [23:10:22] * Krinkle runs away [23:10:45] don't pay attention to tungsten. [23:11:32] bd808, yeah there's definitely a couple of things lurking on jessie simply because prod is still jessie for that thing [23:11:47] does citoid still run on Trusty? [23:11:55] we don't run any trusty boxes [23:12:02] anywhere [23:12:03] AFAIK [23:12:05] no it runs precise :P [23:12:09] lol [23:12:28] built on dead upstream tech :/ [23:12:36] * Krinkle is kidding [23:12:44] I think it moved to docker or something [23:13:10] yeah we have an instance configured to run it through docker in deployment-prep [23:14:16] James_F: aye, regarding chromium, does that mean quibble jobs are still on a fixed/stale version of chromium now? [23:14:31] for qunit and selenium