[00:01:50] 10LibUp, 10Patch-For-Review: libup committed and uploaded an upgrade twice for the same package on the same repo - https://phabricator.wikimedia.org/T280438 (10Legoktm) I tried to figure out how to introduce locks into the queue and eventually gave up. Instead I think we can avoid duplicate pushes by * Ensuri... [00:03:02] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10Krinkle) [00:11:19] ok, I tried to use https://wikitech.wikimedia.org/wiki/Blubber/Download#Blubber_as_a_(micro)Service instead of my local blubber version and that changed things [00:11:46] then I learned "v3 is unsupported" (still in examples though) and just tried v4 and that built a Dockerfile now, cool [01:21:36] mutante: fyi, dan is on parental leave for the month (just saw you ping him now) [01:34:37] greg-g: oh, ok. thanks. I figured it out (for now) [02:35:44] 10Beta-Cluster-Infrastructure, 10WikibaseMediaInfo, 10Beta-Cluster-reproducible: scap sync fails on beta with Error: Class 'MediaWiki\Extension\MediaSearch\I18nUtils' not found - https://phabricator.wikimedia.org/T282206 (10Urbanecm) [03:42:36] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10Krinkle) [03:42:42] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10Krinkle) [03:47:12] 10Beta-Cluster-Infrastructure, 10Structured-Data-Backlog, 10WikibaseMediaInfo, 10Beta-Cluster-reproducible: scap sync fails on beta with Error: Class 'MediaWiki\Extension\MediaSearch\I18nUtils' not found - https://phabricator.wikimedia.org/T282206 (10Krinkle) [03:55:40] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments: 1.37.0-wmf.5 deployment blockers - https://phabricator.wikimedia.org/T281146 (10Krinkle) [04:20:59] Krinkle: 01 is currently active and on stretch, 03 is buster and will take over once some blockers with jenkins are solved. if you need to, 02 could be decom'd [04:23:52] Cool, no rush. I've written it down this time. [04:53:51] 10Release-Engineering-Team (Doing), 10Patch-For-Review, 10User-brennen: logspam-watch broken - https://phabricator.wikimedia.org/T281121 (10thcipriani) p:05Triage→03Medium [05:09:05] 10Release-Engineering-Team, 10Code-Health-DevEd-Curriculum: The Train - the workshop - https://phabricator.wikimedia.org/T249152 (10thcipriani) [06:13:00] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10tstarling) [07:18:46] gehel: dcausse: bonjour, ça y est j'ai mon badge StackOverFlowError en java! Du coup j'ai une question de base, comment je peux avoir la stack entière ? A priori je n'ai que les 1025 premieres lignes :] [07:19:14] :) [07:19:16] j'imagine que il y a un parametre dans la jvm pour limiter la taille max de la stack [07:19:18] seulement 1025? de quoi tu te pleins ? [07:19:28] Sans doute dans le framework the logging [07:19:31] ou alors le log coupe arbitrairement [07:19:40] la trace est la https://phabricator.wikimedia.org/T281737#7054076 [07:20:01] et c'est dans Jenkins qui utilise euh log4j? ;D [07:20:45] avec ca tu devrais trouver où tu rentres dans cette recursion non? [07:20:46] ça ressemble méchamment à un problème de récursion infinie [07:21:35] il y a un truc avec deux systemes de gestion de droit (spring et acegis) qui se tire la bourre. Et bien sur ça n'arrive qu'une fois de temps en temps [07:22:04] s/tire/tirent/ [07:22:15] mais ça je vais laisser le soin à upstream de me filer des billes [07:24:18] peut-être: https://logging.apache.org/log4j/2.x/log4j-core/apidocs/org/apache/logging/log4j/core/util/Constants.html#MAX_REUSABLE_MESSAGE_SIZE [07:24:49] "The default value is , ..." [07:24:52] nice doc :] [07:25:05] yep :( [07:25:24] Mais tu peux peut-être expérimenter avec `log4j.maxReusableMsgSize` [07:25:31] au moins j'ai patché des méthodes dépréciées dans le code, si ça se trouve ça regle le probleme [07:25:34] Pas sûr que ce soit effectivement là que ça coince [07:26:00] J'imagine que tu utilise un PatternLayout pour formater les logs? [07:26:37] whatever is shipped by Jenkins :D [07:27:13] il y a un truc très basique dans un jenkins.log (juste une ligne, pas de trace), sinon c'est via le web UI et je n'ai pas trouvé comment configurer ça [07:27:31] mais je vais trouver! Au moins maintenant j'ai une piste: regarder la config du logging [07:27:40] on peut ajouter un modifier sur le message qui limite la taille du message, mais je doute que ce soit ça [07:28:47] méthode un peu bourrin, réduire la taille de la stack, pour faire planter la récursion plus tôt (`-Xss`) [07:28:56] mais ça a des effets secondaites [07:31:40] je fouillerai plus tard, je vais déjà tester mon code ;) [07:32:26] et j'ai aussi "appris" à poster un formulaire en java avec org.apache.http https://gerrit-review.googlesource.com/c/plugins/its-phabricator/+/305522/2/src/main/java/com/googlesource/gerrit/plugins/its/phabricator/conduit/ConduitConnection.java ! ;D [07:32:32] bref c'est tout merci gehel ! [07:33:17] bon courage! [08:17:12] c'est dur l'open source :D [08:17:26] je me retrouver à tenter d'ajouter un bill of materials [08:17:35] et bien sur ensuite à fixer la CI [08:29:56] 10LibUp: LibUp hits 30 minute timeout when processing Wikibase - https://phabricator.wikimedia.org/T282217 (10Legoktm) [08:36:57] 10LibUp, 10wdwb-tech: LibUp hits 30 minute timeout when processing Wikibase - https://phabricator.wikimedia.org/T282217 (10Addshore) [09:29:17] gehel: mon premier bom maven! https://github.com/jenkinsci/gearman-plugin/pull/11/commits/61a9c982649c3f16be5ce3c231e03e5ef3b66a98 :) [11:52:57] hashar: Bravo [14:14:19] (03CR) 10Daimona Eaytoy: "> Patch Set 2:" [integration/config] - 10https://gerrit.wikimedia.org/r/684019 (https://phabricator.wikimedia.org/T257301) (owner: 10Daimona Eaytoy) [15:19:58] (03CR) 10Jforrester: "> Patch Set 2:" [integration/config] - 10https://gerrit.wikimedia.org/r/684019 (https://phabricator.wikimedia.org/T257301) (owner: 10Daimona Eaytoy) [16:20:43] (03PS2) 10Jforrester: Zuul: [operations/software/mailman-templates] Add CI of debian-glue [integration/config] - 10https://gerrit.wikimedia.org/r/685938 (https://phabricator.wikimedia.org/T282018) (owner: 10Legoktm) [16:20:51] (03CR) 10Jforrester: [C: 03+2] Zuul: [operations/software/mailman-templates] Add CI of debian-glue [integration/config] - 10https://gerrit.wikimedia.org/r/685938 (https://phabricator.wikimedia.org/T282018) (owner: 10Legoktm) [16:21:04] (03CR) 10Jforrester: [C: 03+2] "> Patch Set 1:" [integration/config] - 10https://gerrit.wikimedia.org/r/685938 (https://phabricator.wikimedia.org/T282018) (owner: 10Legoktm) [16:23:36] (03Merged) 10jenkins-bot: Zuul: [operations/software/mailman-templates] Add CI of debian-glue [integration/config] - 10https://gerrit.wikimedia.org/r/685938 (https://phabricator.wikimedia.org/T282018) (owner: 10Legoktm) [16:37:55] !log Zuul: [operations/software/mailman-templates] Add CI of debian-glue T282018 [16:37:58] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [16:37:58] T282018: Make customized Mailman3 templates translatable - https://phabricator.wikimedia.org/T282018 [16:38:00] Oops. [17:08:56] (03PS8) 10Legoktm: Add dependencies for social tools and some other things that I maintain [integration/config] - 10https://gerrit.wikimedia.org/r/684455 (owner: 10Jack Phoenix) [17:09:34] James_F: oops? but thank you :D [17:10:37] legoktm: Oops because I merged and then forgot to deploy for 15 minutes. :-) [17:10:46] :P [17:11:18] legoktm: Also, do you ever get `Build failed: unable to find user root: no matching entries in passwd file (image.py:176)` when docker-pkg'ing? Context is https://gerrit.wikimedia.org/r/c/integration/config/+/684019/ [17:12:13] I haven't actually run docker-pkg locally in a while, but let me see [17:12:35] Oh, spoiler, running docker-pkg locally as of https://gerrit.wikimedia.org/r/q/639277b3433b3322a210959f70449bc0f8ab7078 fails for me. :-( [17:12:47] So use the commit before that if updating. [17:15:58] https://forums.docker.com/t/unable-to-find-user-root-no-matching-entries-in-passwd-file/26545/11 [17:16:22] "The only solution is to restart docker service - which is an interruption." [17:16:28] Wow, really? [17:16:31] * James_F tries that. [17:21:09] yeah, it seems like this is just some random docker bug that they haven't gotten around to fixing [17:21:24] https://stackoverflow.com/questions/41676835/unable-to-find-user-root-no-matching-entries-in-passwd-file-in-docker "This issue is caused by a docker engine bug but which is tracked privately, Docker is asking users to restart the engine!" [17:21:36] (03CR) 10Legoktm: [C: 03+2] Add dependencies for social tools and some other things that I maintain [integration/config] - 10https://gerrit.wikimedia.org/r/684455 (owner: 10Jack Phoenix) [17:22:18] Tsk, should be prefixed with `Zuul:` [17:22:23] And not use "I'. [17:23:06] (03Merged) 10jenkins-bot: Add dependencies for social tools and some other things that I maintain [integration/config] - 10https://gerrit.wikimedia.org/r/684455 (owner: 10Jack Phoenix) [17:24:04] oh, I'll edit it next time [17:24:08] !logs reloaded zuul for https://gerrit.wikimedia.org/r/684455 [17:24:17] Thanks. [17:30:40] legoktm: Hmm. I restarted docker and I still get the issue. Will restart my whole machine just in case… [17:40:44] … no. [17:40:45] Ah well. [18:20:10] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10Jdforrester-WMF) [18:20:34] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10Jdforrester-WMF) [18:26:13] (03PS3) 10Jforrester: dockerfiles: [mediawiki-phan-taint-check-demo] Provide composer [integration/config] - 10https://gerrit.wikimedia.org/r/684019 (https://phabricator.wikimedia.org/T257301) (owner: 10Daimona Eaytoy) [18:29:35] 10Beta-Cluster-Infrastructure, 10Beta-Cluster-reproducible, 10User-Majavah, 10User-Ryasmeen, 10User-zeljkofilipin: Cannot login to beta cluster: "There seems to be a problem with your login session..." - https://phabricator.wikimedia.org/T263617 (10Majavah) 05Open→03Resolved Hopefully fixed for good... [18:50:51] legoktm: Also, is there any way to check if Debian's php-redis suddenly dropped IGBINARY support? Context is T282133#7066508 [18:50:51] T282133: PhanUndeclaredConstantOfClass Reference to undeclared class constant \Redis::SERIALIZER_IGBINARY - https://phabricator.wikimedia.org/T282133 [18:52:32] https://packages.debian.org/buster/php-redis says it depends on php-igbinary which makes me assume it does have that support. [18:56:06] 10Release-Engineering-Team (Doing), 10Release, 10Train Deployments, 10User-brennen: 1.37.0-wmf.4 deployment blockers - https://phabricator.wikimedia.org/T281145 (10brennen) 05Open→03Resolved Looking stable on all wikis. Thanks all. [18:58:49] James_F: where do we get the packages for php73, though? [18:58:56] Krinkle: Sury. [18:59:06] Oh, hmm, or do we just use the basic buster ones? [18:59:21] usually php-foobar is versioned along with php itself [18:59:28] Only sometimes. [18:59:30] so for the php73 jobs specifically, I think maybe php-redis isn't from debian [19:00:41] Sury is the same as Debian for all practicality [19:00:58] Also I'm wrong, this whole set is still based off stretch. [19:01:14] Yeah. [19:02:48] https://github.com/oerdnj/deb.sury.org/issues/1590 https://github.com/oerdnj/deb.sury.org/issues/1566 did we customize memcached.ini or redis.ini or something? [19:05:23] I note we also have php8.0-redis commented out as it's not yet available (or wasn't). [19:05:38] No, I don't think so,. [19:06:13] We inherit from https://gerrit.wikimedia.org/r/plugins/gitiles/integration/config/+/refs/heads/master/dockerfiles/php73/Dockerfile.template [19:21:49] Reedy: How'd you feel about me temporarily suspending phan on release branches so that you can merge again? [19:25:06] James_F: Not ideal, but not the end of the world either [19:26:06] Sure, I don't want to block any emergency releases or whatever, but I'd also worry about us never getting it fixed. [19:27:35] can you just disable the broken checks? [19:27:40] instead of throwing out everything? [19:28:05] The PhanUndeclaredConstantOfClass in RedisConnectionPool? [19:28:11] Sure, I suppose. [20:11:51] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Umherirrender) [20:23:56] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Legoktm) The only thing I can think of is what we're running a different npm version? We're currently on 6.14.5. Latest 6.x is 6.14.13. [20:36:57] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Umherirrender) On AbuseFilter is now twiced updated for `hosted-git-info` possible by this problem - Second - https://gerrit.wikimedia.org/r/c/mediawiki/extensions/AbuseFilter/+/68675... [20:51:33] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Legoktm) After checking out https://gerrit.wikimedia.org/r/c/mediawiki/extensions/AbuseFilter/+/686753 npm audit says I still have an issue: ` km@9683552ac492:/AbuseFilter$ npm audit... [20:55:48] 10Deployments, 10Jouncebot: Jouncebot: `jouncebot thisweek` functionality - https://phabricator.wikimedia.org/T187229 (10bd808) p:05Medium→03Lowest The "ongoing" tracking seems to have vanished over the years. [21:19:04] 10Beta-Cluster-Infrastructure, 10User-Majavah, 10User-Zabe: Global developer for Zabe on beta - https://phabricator.wikimedia.org/T281239 (10Zabe) [21:19:06] 10Project-Admins, 10User-Kizule, 10User-Zabe: Create User-Zabe project - https://phabricator.wikimedia.org/T276719 (10Zabe) [21:19:08] 10phabricator maintenance bot, 10User-Ladsgroup, 10User-Zabe: Remove #Patch-For-Review when patch is abandoned in Gerrit - https://phabricator.wikimedia.org/T276390 (10Zabe) [21:19:12] 10Release-Engineering-Team (Development services), 10Release-Engineering-Team-TODO, 10phabricator maintenance bot, 10GerritBot, and 2 others: GerritBot doesn't add #Patch-for-review to restored patches - https://phabricator.wikimedia.org/T277597 (10Zabe) [21:30:42] (03CR) 10Umherirrender: [C: 03+2] Minor clean up to LowerCamelFunctionsNameSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/681374 (owner: 10DannyS712) [21:33:05] (03Merged) 10jenkins-bot: Minor clean up to LowerCamelFunctionsNameSniff [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/681374 (owner: 10DannyS712) [21:52:56] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Legoktm) Sigh, now we have https://gerrit.wikimedia.org/r/c/mediawiki/extensions/GlobalWatchlist/+/686779 and https://gerrit.wikimedia.org/r/c/mediawiki/extensions/AbuseFilter/+/686777 [21:53:56] npm makes my head hurt [21:59:38] github's dependabot is draining my travis ci minutes [21:59:48] I thought i had disbled it eveyrwhere, now it came back on my pesonal repos after doing nothing for years [22:00:29] something something convenient mistakes something something anti-competetive corporate fuckery [22:03:56] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Legoktm) @Umherirrender Should we just run `npm audit fix --only=dev` in a loop like 2 or 3 times? I don't fully understand what's going wrong here. [22:04:25] Krinkle: I got emails yesterday saying they were redoing how security notifications and dependabot works [22:04:56] also they (npm/GitHub) removed the ability to request a npm security advisory [22:05:19] and they never followed-up on my support ticket asking how to do so [22:10:53] 10LibUp: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Umherirrender) I have no idea how to handle this at best. The loop would be a workaround, but that sounds okay. Is npm audit giving an return code when there is still something to fix... [22:47:08] 10Phabricator: Request for Edit Form Configuration permissions - https://phabricator.wikimedia.org/T282292 (10jbolorinos-ctr) [22:48:23] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Radar), 10Abstract Wikipedia team: Provide a CI mechanism by which we can assert that front-end components can run independently of the MW back-end on which they rely - https://phabricator.wikimedia.org/T277720 (10Jdforrester-WMF) >>! In T2... [22:48:28] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Radar): Provide a CI mechanism by which we can assert that front-end components can run independently of the MW back-end on which they rely - https://phabricator.wikimedia.org/T277720 (10Jdforrester-WMF) We don't need to tag this task, ours i... [23:23:53] 10LibUp, 10Patch-For-Review: libup possible has issues with npm audit to get all vulnerability removed - https://phabricator.wikimedia.org/T282278 (10Legoktm) >>! In T282278#7070906, @Umherirrender wrote: > Is npm audit giving an return code when there is still something to fix? `npm audit` yes, `npm audit fi... [23:31:38] TIL we have php-based selenium tests [23:31:53] 10Phabricator: Request for Edit Form Configuration permissions - https://phabricator.wikimedia.org/T282292 (10jbolorinos-ctr) Here are the details I wanted to include in a form: = Overview: = **Summary:** A brief description of the issue including any details that might be helpful for resolving the issue. D... [23:31:55] https://gerrit.wikimedia.org/r/c/mediawiki/extensions/WikimediaEvents/+/686799/1/tests/browser/SearchSatisfactionTests.php [23:32:19] Facebook\WebDriver