[00:11:34] !log Rerun failed wikidata-json_entity-weekly-coord instances after having created the missing hive table [00:11:35] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [06:16:06] helloooo [06:59:32] hola! [07:04:27] 10Analytics, 10Gerrit, 10Gerrit-Privilege-Requests: Give access to Wikistats 2 to l10n-bot - https://phabricator.wikimedia.org/T245805 (10fdans) [07:04:42] 10Analytics, 10Analytics-Wikistats, 10translatewiki.net: Add stats.wikimedia.org/v2 as a TranslateWiki project - https://phabricator.wikimedia.org/T240621 (10fdans) The en.json strings have now been merged to master so they can be added to TranslateWiki: https://github.com/wikimedia/analytics-wikistats2/blo... [07:11:23] 10Analytics, 10Analytics-Wikistats, 10translatewiki.net: Add stats.wikimedia.org as a TranslateWiki project - https://phabricator.wikimedia.org/T240621 (10fdans) [07:25:14] (03PS1) 10Elukey: Fix wikidata/json_entity/weekly workflow's add_partition action [analytics/refinery] - 10https://gerrit.wikimedia.org/r/573874 [07:25:18] fdans: --^ [07:25:33] I think we'd need a quick refinery deploy + restart of coordinator [07:26:06] (03CR) 10Fdans: [V: 03+2 C: 03+2] "Nice catch, thank you Luca" [analytics/refinery] - 10https://gerrit.wikimedia.org/r/573874 (owner: 10Elukey) [07:26:37] (03PS2) 10Elukey: oozie/wikidata/json_entity/weekly: fix workflow's add_partition action [analytics/refinery] - 10https://gerrit.wikimedia.org/r/573874 [07:26:44] I was changing the name of the commit [07:26:51] ok [07:26:51] nevermind :) [07:26:58] elukey: feel free to merge [07:27:10] ah goood! [07:27:13] (03CR) 10Elukey: [V: 03+2] oozie/wikidata/json_entity/weekly: fix workflow's add_partition action [analytics/refinery] - 10https://gerrit.wikimedia.org/r/573874 (owner: 10Elukey) [07:27:21] all right I think we can deploy [07:27:31] elukey: will redeploy refinery as soon as I'm done with this rebase [07:27:52] yes yes nothing urgent [07:29:05] I have suspended the coordinator to stop spam :) [08:03:22] (03CR) 10Fdans: Add language selection functionality to Wikistats (032 comments) [analytics/wikistats2] - 10https://gerrit.wikimedia.org/r/564047 (https://phabricator.wikimedia.org/T238752) (owner: 10Fdans) [08:05:29] (03PS5) 10Fdans: Add language selection functionality to Wikistats [analytics/wikistats2] - 10https://gerrit.wikimedia.org/r/564047 (https://phabricator.wikimedia.org/T238752) [08:19:22] !log deploying refinery [08:19:23] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [08:24:02] brb [08:40:23] elukey: which coordinator do I have to restart? [08:45:49] it is called wikidata-json_entity-weekly [08:46:09] https://hue.wikimedia.org/oozie/list_oozie_coordinator/0021970-200203112045319-oozie-oozi-C/ [08:46:22] start time Mon, 06 Jan 2020 00:00:00 [08:51:36] elukey: thank you! [08:51:47] np! [09:05:26] 10Analytics, 10Analytics-Wikistats, 10translatewiki.net: Add stats.wikimedia.org to translatewiki.net - https://phabricator.wikimedia.org/T240621 (10Nikerabbit) [09:11:14] 10Analytics, 10Analytics-Wikistats, 10translatewiki.net: Add stats.wikimedia.org to translatewiki.net - https://phabricator.wikimedia.org/T240621 (10Nikerabbit) I assume this https://github.com/wikimedia/analytics-wikistats2/blob/master/src/i18n/aa.json will be removed? `aa` is a language code for Afar. I n... [09:18:37] 10Analytics, 10Analytics-Wikistats, 10translatewiki.net: Add stats.wikimedia.org to translatewiki.net - https://phabricator.wikimedia.org/T240621 (10fdans) @Nikerabbit thank you for the quick reply. I thought I had removed the "for" usage (I only used at the very beginning when learning about i18n templates... [09:26:38] (03PS1) 10Fdans: Remove aa.json, replace untranslatable string [analytics/wikistats2] - 10https://gerrit.wikimedia.org/r/573953 (https://phabricator.wikimedia.org/T240621) [09:46:59] 10Analytics: Unify puppet roles for stat and notebook hosts - https://phabricator.wikimedia.org/T243934 (10elukey) High level idea about a way to simplify https://wikitech.wikimedia.org/wiki/Analytics/Data_access#Access_Groups: 1) since we have kerberos authentication, I'd propose to just decide a set of groups... [10:09:03] 10Analytics, 10User-Elukey: Standard partman recipe for druid hosts - https://phabricator.wikimedia.org/T245810 (10fgiunchedi) [10:18:42] 10Analytics, 10User-Elukey: Standard partman recipe for druid hosts - https://phabricator.wikimedia.org/T245810 (10MoritzMuehlenhoff) One option would be to manually create /srv/druid symlinks on the existing installed base and then switch Puppet to use it, with buster reimages and hw refreshes, the underlying... [10:31:50] fdans: already deployed refinery to hdfs? I can take care of the coord restart [10:42:48] 10Analytics, 10Gerrit, 10Gerrit-Privilege-Requests, 10User-MarcoAurelio: Give access to Wikistats 2 to l10n-bot - https://phabricator.wikimedia.org/T245805 (10MarcoAurelio) a:03MarcoAurelio [10:43:05] already deployed and restarted [10:46:17] ah ok the old coord was not killed [10:46:19] lemme do it [10:47:21] fdans: honor is yours then to reply to me with the work done :) [10:47:32] (so we can keep track of it in alerts@ as well) [10:49:03] (03PS1) 10MarcoAurelio: Allow L10n-Bot to commit translation updates [analytics/wikistats2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/573970 (https://phabricator.wikimedia.org/T245805) [10:49:28] (03PS2) 10MarcoAurelio: Allow L10n-Bot to commit translation updates [analytics/wikistats2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/573970 (https://phabricator.wikimedia.org/T245805) [10:58:23] 10Analytics, 10Gerrit, 10Gerrit-Privilege-Requests, 10Patch-For-Review, 10User-MarcoAurelio: Give access to Wikistats 2 to l10n-bot - https://phabricator.wikimedia.org/T245805 (10MarcoAurelio) 05Open→03Stalled We'll need to add wikistats2 to translatewiki.net first otherwise the above patches won't d... [11:02:16] 10Analytics, 10Gerrit, 10Gerrit-Privilege-Requests, 10Patch-For-Review, 10User-MarcoAurelio: Give access to Wikistats 2 to l10n-bot - https://phabricator.wikimedia.org/T245805 (10MarcoAurelio) p:05Triage→03Low [11:19:24] 10Analytics, 10Operations, 10serviceops, 10vm-requests, 10Patch-For-Review: Create a ganeti VM in eqiad: an-launcher1001 - https://phabricator.wikimedia.org/T244717 (10elukey) [11:37:37] PROBLEM - cache_upload: Varnishkafka webrequest Delivery Errors per second -esams- on icinga1001 is CRITICAL: 3710 ge 5 https://wikitech.wikimedia.org/wiki/Analytics/Systems/Varnishkafka https://grafana.wikimedia.org/d/000000253/varnishkafka?panelId=20&fullscreen&orgId=1&var-datasource=esams+prometheus/ops&var-source=webrequest&var-cp_cluster=cache_upload&var-instance=All [11:38:37] PROBLEM - cache_text: Varnishkafka webrequest Delivery Errors per second -esams- on icinga1001 is CRITICAL: 6047 ge 5 https://wikitech.wikimedia.org/wiki/Analytics/Systems/Varnishkafka https://grafana.wikimedia.org/d/000000253/varnishkafka?panelId=20&fullscreen&orgId=1&var-datasource=esams+prometheus/ops&var-source=webrequest&var-cp_cluster=cache_text&var-instance=All [11:39:53] there is/was a problem in esams caching, SRE investigating [11:41:49] RECOVERY - cache_upload: Varnishkafka webrequest Delivery Errors per second -esams- on icinga1001 is OK: (C)5 ge (W)1 ge 0 https://wikitech.wikimedia.org/wiki/Analytics/Systems/Varnishkafka https://grafana.wikimedia.org/d/000000253/varnishkafka?panelId=20&fullscreen&orgId=1&var-datasource=esams+prometheus/ops&var-source=webrequest&var-cp_cluster=cache_upload&var-instance=All [11:41:50] !log restart varnishkafka on cp3057 (stuck in timeouts to kafka, analytics alarms raised) [11:41:51] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [11:41:54] ok all good now [11:42:41] not really, text still messed up [11:49:08] !log restart varnishkafka on various cp30xx nodes [11:49:09] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [11:51:11] RECOVERY - cache_text: Varnishkafka webrequest Delivery Errors per second -esams- on icinga1001 is OK: (C)5 ge (W)1 ge 0 https://wikitech.wikimedia.org/wiki/Analytics/Systems/Varnishkafka https://grafana.wikimedia.org/d/000000253/varnishkafka?panelId=20&fullscreen&orgId=1&var-datasource=esams+prometheus/ops&var-source=webrequest&var-cp_cluster=cache_text&var-instance=All [11:51:59] yesss [11:52:42] * elukey lunch + errand! [13:28:15] fdans, elukey - Thanks a lot for fixing my mess - I had foud that bug when testing and supposedly corrected it :S Please excuse me [13:54:58] (03PS1) 10Fdans: Remove check for forward slash in per file mediarequests [analytics/aqs] - 10https://gerrit.wikimedia.org/r/573998 [13:55:29] joal: np! [13:55:32] it was a easy fix :) [14:10:53] 10Analytics: Unify puppet roles for stat and notebook hosts - https://phabricator.wikimedia.org/T243934 (10Ottomata) > should we just deprecate research and add the missing members to statistics-privatedata-users > analytics-users seems not valid anymore, very few users, I'd just propose to follow up with them a... [14:13:07] 10Analytics: Unify puppet roles for stat and notebook hosts - https://phabricator.wikimedia.org/T243934 (10elukey) Even better yes, I thought that some use cases were still to be supported (I always misremember stat-related stuff). Two groups will be way better! [14:22:13] oo/ [14:22:19] aaaah I have two heads [14:37:25] (03CR) 10Milimetric: [C: 03+2] Remove check for forward slash in per file mediarequests [analytics/aqs] - 10https://gerrit.wikimedia.org/r/573998 (owner: 10Fdans) [14:37:50] ottomata: o/ https://gerrit.wikimedia.org/r/#/c/operations/dns/+/571479/ when you have a moment [15:05:52] (03CR) 10Milimetric: [C: 04-1] "I missed this in the parent change but were you going to add the language switching to dev mode too?" (032 comments) [analytics/wikistats2] - 10https://gerrit.wikimedia.org/r/564047 (https://phabricator.wikimedia.org/T238752) (owner: 10Fdans) [15:20:21] 10Analytics, 10Better Use Of Data, 10Desktop Improvements, 10Product-Infrastructure-Team-Backlog, and 6 others: Client side error logging production launch - https://phabricator.wikimedia.org/T226986 (10jlinehan) [15:20:23] 10Analytics, 10Better Use Of Data, 10Epic, 10MW-1.35-notes (1.35.0-wmf.21; 2020-02-25), and 3 others: Prototype client to log errors in vagrant - https://phabricator.wikimedia.org/T235189 (10jlinehan) 05Open→03Resolved Thanks to @Krinkle and @Ottomata and @phuedx for all the timely code review, this is... [15:22:28] 10Analytics, 10Better Use Of Data, 10Epic, 10MW-1.35-notes (1.35.0-wmf.21; 2020-02-25), and 3 others: Develop small client-side error logger for MediaWiki - https://phabricator.wikimedia.org/T235189 (10jlinehan) [15:44:03] milimetric: o/ [15:44:10] hey elukey [15:44:28] one question if you have time - I am building an-launcher1001, and I'd love to move the stat1006 RU jobs to it [15:44:50] what things do I need to do, besides moving puppet config? [15:45:03] notify people, copy files around, etc.. [15:54:10] mforns: how goes and can I help :) [15:54:12] ? [15:54:33] heya ottomata it's going :] [15:54:50] I changed back to configure + produce [15:54:54] aye k [15:54:56] I'm testing [15:55:00] just globals? makes sense I think. [15:55:16] well, properties [15:55:30] mw.eventProducer.configure(), etc? [15:57:09] 10Analytics, 10Operations, 10serviceops, 10vm-requests: Create a ganeti VM in eqiad: an-launcher1001 - https://phabricator.wikimedia.org/T244717 (10elukey) 05Stalled→03Open ` Creating new VM named an-launcher1001.eqiad.wmnet in eqiad with row=C vcpu=4 memory=8 gigabytes disk=100 gigabytes link=analytic... [15:57:12] 10Analytics: Unify puppet roles for stat and notebook hosts - https://phabricator.wikimedia.org/T243934 (10elukey) [16:04:28] !log altered event.CentralNoticeImpression table column event.campaignStatuses to type string, will backfill data - T244771 [16:04:30] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [16:04:30] T244771: Refining is failing to refine centranoticeimpression events - https://phabricator.wikimedia.org/T244771 [16:07:14] ottomata, yes [16:07:36] coo [16:07:44] ottomata, I'm getting the same problem as yesterday with the JSON error :C this time with firefox [16:08:09] the empty body? [16:08:11] elukey: you'd also move the stat1007 RU jobs to it, right? [16:08:24] ottomata, yes empty body [16:08:29] very weird [16:08:39] there should be nothing that needs to happen, I just looked over all the jobs to see if any of them have weird permissions indicating people mess with the files manually [16:09:22] there's nothing, so we can just send out an email internally that says we moved all RU jobs to an-launcher1001 and if people need access to logs they can contact us [16:09:27] milimetric: yes all of them [16:09:30] but honestly even that's redundant because they'd contact us anyway [16:09:43] ottomata, btw, I held back with changing client to producer... [16:09:45] mforns: and in dev console you can see the request payload with the event? [16:09:51] so I'd do nothing, I'm kind of curious who comes asking questions, maybe we can help them with their workflows [16:09:59] mforns: ya? because jason? [16:10:08] yes, request payload is visible in the console [16:10:25] ottomata, yes also, but also because I thought client was ok [16:10:38] is it not a client? [16:11:21] yes but a client of... event? [16:11:45] oh I see [16:11:57] eventLogClient? [16:12:30] asking jason to joiin our bikeshed [16:12:34] k [16:13:15] now that we are thinking about it, we might need somehting better than prooducer, since he intends for more stuff to go in this code, right? [16:13:16] e.g. sampling, etc. [16:13:17] ? [16:13:33] aha [16:14:18] the only place where it is called eventClient is mw.eventClient, usually it's ext.eventLogging.client [16:14:50] we could just change that [16:15:23] mw.elClient? [16:16:24] 10Analytics: Unable to access SWAP notebooks using LDAP - https://phabricator.wikimedia.org/T230627 (10Fsalutari) Hi, I am experiencing the same error, but I am not able to log in neither into https://turnilo.wikimedia.org/ nor https://yarn.wikimedia.org/. I have also tried different browsers and changed my pa... [16:16:42] well, whatever we attach to mw. will be the main api [16:16:46] right? [16:16:53] the module is called ext.eventLogging.client [16:17:16] which may or may not be the right thing to do [16:17:24] we _could_ put this into core module? [16:17:30] if we felt it was better there [16:17:31] not sure [16:17:37] kinda nice to have it separate [16:17:49] not sure where jason wants to put the extra code [16:17:59] aha [16:18:04] more RL modules may not be the best idea, so maybe this is better in core if it is always needed? [16:18:07] what extra code? [16:18:14] all the other phases [16:19:02] oh ok [16:19:19] jason said he'd be back on in afternoon [16:19:25] mforns: let's defer til then, will post on ticket. [16:19:29] ok [16:24:01] mforns: i'm looking at EL RL modules now, i wonder if there is a reason to have a new one after all. i think we originally wanted it that way for separation of concerns, but maybe just a new file in the main ext.eventLogging module is finee [16:24:09] that would make it load on every page load, even if we aren't using the new code yet [16:24:10] not sure. [16:24:22] then again, we will probably load t his in WikimediaEvents anyway [16:24:26] which is basically every page load [16:24:28] so maybe it doesn't matter [16:24:48] also, if we put it in ext.eventLogging [16:24:57] we can attach it to mw in subscriber.js [16:25:15] hm [16:25:25] we could even attach it as functions directly to mw.eventLog [16:25:31] mw.eventLog.configure, mw.eventLog.producer [16:25:33] produce [16:25:33] ? [16:25:52] aha [16:25:55] will post on ticket, see what jason says [16:26:15] not against it, the only thing I'd say is have separate files, otherwise code will be too long [16:26:21] yes separate files for sure [16:28:36] ottomata, re "we can attach it to mw in subscriber.js": the last version has the mw.eventClient = ... in the same file as the client itself, because it's a singleton. [16:28:57] aye ok [16:29:05] but we could do it from subscriber anyway, makes sense [16:29:10] or wherever not sure where best [16:29:24] but maybe attaching to mw.eventLog is good? or mw.eventLog.client ? [16:29:25] hm [16:30:09] is the target to replace eventLog? [16:30:20] in the end? [16:30:30] like after everything is migrated? [16:30:47] or are we going to support old style forever through the wrapper? [16:30:53] (03CR) 10Milimetric: [C: 03+2] Remove aa.json, replace untranslatable string [analytics/wikistats2] - 10https://gerrit.wikimedia.org/r/573953 (https://phabricator.wikimedia.org/T240621) (owner: 10Fdans) [16:30:56] i thnk we ultimately can't. too much code that will call mw.eventLog.logEvent. we aree going to make logEvent() call the functions you are writing now [16:31:12] i dunno, maybe eventLog can be consider the legacy way? [16:31:12] ok [16:31:13] hm [16:31:32] probalby forever throgh wrapper, but we should probably encourage folks to use new API directly? [16:31:34] not 100% [16:31:37] man, I think eventClient is not too bad! [16:31:40] haha [16:31:47] hehehe [16:32:33] although.. hm the word event has collisions with JS... :( [16:32:41] heh [16:32:50] eventStreamClient? hm [16:32:57] mepClient [16:32:59] no way [16:33:07] mep is a program name [16:33:09] heheh [16:33:21] made up for annual planning purposes [16:33:29] eventStreamClient is good! just a bit longer [16:33:36] smarkToolsForBetterDataClient [16:33:37] heheheh [16:33:39] smart* [16:33:45] heheheh [16:34:29] wait [16:34:41] how about mw.eventStream [16:34:56] mw.eventStream.configure [16:35:00] mw.eventStream.produce [16:35:45] (03CR) 10Milimetric: [C: 04-1] "good thoughts here from Timo: https://phabricator.wikimedia.org/T244712#5883902" [analytics/aqs] - 10https://gerrit.wikimedia.org/r/571968 (https://phabricator.wikimedia.org/T244712) (owner: 10Fdans) [16:37:12] hm [16:37:20] sounds lik you are configuring a single stream [16:37:20] gm [16:37:21] hm [16:37:32] mw.eventStreams.configure + produce [16:37:34] plural>? [16:37:44] not bad [16:39:04] yea! [16:42:15] 10Analytics: Unify puppet roles for stat and notebook hosts - https://phabricator.wikimedia.org/T243934 (10elukey) [16:42:18] 10Analytics, 10Operations, 10serviceops, 10vm-requests, 10Patch-For-Review: Create a ganeti VM in eqiad: an-launcher1001 - https://phabricator.wikimedia.org/T244717 (10elukey) 05Open→03Resolved a:03elukey [16:42:19] ottomata, I pushed the last version to https://gerrit.wikimedia.org/r/#/c/mediawiki/extensions/EventLogging/+/573677/ [16:42:21] 10Analytics, 10Analytics-Kanban, 10Better Use Of Data, 10Product-Infrastructure-Team-Backlog, 10Patch-For-Review: EventLogging MEP Upgrade Phase 1 - https://phabricator.wikimedia.org/T244521 (10Ottomata) OK @jlinehan @mforns, it's bikeshed time. ===== First: name of file & class & singleton I think `mw... [16:42:43] ottomata, can you please test it in your vagrant, to see it it works like yesterday? [16:43:14] k [16:44:18] it works mforns [17:00:15] 10Analytics, 10Analytics-Kanban, 10Better Use Of Data, 10Product-Infrastructure-Team-Backlog, 10Patch-For-Review: EventLogging MEP Upgrade Phase 1 - https://phabricator.wikimedia.org/T244521 (10mforns) > First: name of file & class & singleton I think mw.eventStreams is cool! > Second: RL package module... [17:01:20] ping standup [17:01:48] coming [17:02:25] 10Analytics, 10Analytics-Kanban, 10Better Use Of Data, 10Product-Infrastructure-Team-Backlog, 10Patch-For-Review: EventLogging MEP Upgrade Phase 1 - https://phabricator.wikimedia.org/T244521 (10Ottomata) > If not, and you think the name EventLogging will prevail forever EventLogging will prevail forever,... [17:02:43] 10Analytics, 10Better Use Of Data, 10Epic, 10MW-1.35-notes (1.35.0-wmf.21; 2020-02-25), and 3 others: Develop small client-side error logger for MediaWiki - https://phabricator.wikimedia.org/T235189 (10Ottomata) YEEhaw thank you alL! [18:04:49] an-launcher1001 is alive! :) [18:06:58] O.o [18:15:36] all right going off [18:15:39] have a good weekend! [18:38:36] awesome, have a great one [18:45:54] PROBLEM - Check if the Hadoop HDFS Fuse mountpoint is readable on an-launcher1001 is CRITICAL: CRITICAL https://wikitech.wikimedia.org/wiki/Analytics/Systems/Cluster/Hadoop/Administration%23Fixing_HDFS_mount_at_/mnt/hdfs [18:52:15] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf) [18:59:04] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf) [19:11:35] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf) [19:12:35] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf) [19:15:20] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf) As I continue to encounter this problem in my work, I'm continuing to investigate and develop the description. I would appreciate any additions, suggestions, or c... [19:17:46] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10Ottomata) Is there any pattern to the types of queries that cause this, or does it just happen randomly? Is there a query you can always reproduce this with? Does the same prob... [19:19:45] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10Ottomata) Also, I see in one of your screenshots that you have a toPandas call. (I guess this is what wmfdata.hive does). This will bring down the entire result from the spark... [19:29:56] 10Analytics, 10Analytics-Kanban, 10Event-Platform, 10Services (watching): jsonschema-tools should add a 'latest' symlink - https://phabricator.wikimedia.org/T245859 (10Ottomata) [20:11:55] nuria: question -- do you know how many pages is WikimediaEvents loaded on? how does it not pull EventLogging in, doesn't it have a dependency? or is that dependency only for some of the resource modules? [20:12:12] I guess anybody can answer this if they know [20:17:46] RECOVERY - Check if the Hadoop HDFS Fuse mountpoint is readable on an-launcher1001 is OK: OK https://wikitech.wikimedia.org/wiki/Analytics/Systems/Cluster/Hadoop/Administration%23Fixing_HDFS_mount_at_/mnt/hdfs [20:20:05] it does pull EL in for sure [20:21:02] hip: nuria is not working today [20:21:15] ottomata: gotcha [20:21:17] so for error logging though [20:21:26] which is in the module that has a dependency on EL [20:21:37] and error logging is supposed to be on every page (I assume) doesn't that just put EL on every page too? [20:21:57] not that it's an issue, just that isn't that a key reason why EL is this distinct extension is to save on page budget etc. ? [20:22:42] what I mean is, is it just that we need to re-organize the errorlogging bit in WME so that it doesn't need to have that dependency, or is it the case that EL is actually going out on almost every page? [20:22:44] https://gerrit.wikimedia.org/r/plugins/gitiles/mediawiki/extensions/WikimediaEvents/+/master/extension.json#132 [20:23:02] ah [20:23:07] i seee [20:23:27] not totally sure but [20:23:36] i think its because the extension has several RL module packages [20:23:42] and those are where the deps are specified [20:24:02] so ext.wikimediaEvents depends on ext.EventLogging [20:24:13] yeah, so I'm thinking this is just misconfigured in the sense that it doesn't actually depend on EL. Honestly I have no idea how this is organized in terms of its RL modules. [20:24:16] ext.wikimediaEvents.loggedin does not depend on ext.EventLogging [20:24:17] I couldn't make much sense of it [20:24:31] i think your current error logging code does not depend on EL, right? [20:24:31] well does ext.wikimediaEvents.statsd.js ? [20:24:36] (but it might in the future?) [20:24:57] ext.wikimediaEvents/statsd.js does, just because it is in the ext.wikimediaEvents package module [20:25:03] but, maybe it doesn' thave to be? [20:25:04] yeah it doesn't ... for the future, I don't know where we want to go but I think a key objection was that error logging would be desired for every page, and introducing an EL dep would require EL on every page [20:25:06] oh clientError.js is in there too [20:25:07] ya [20:25:14] ¯\_(ツ)_/¯ [20:25:43] hm, wasn't the objection about making that requirement happen from mw core? [20:26:01] i dnnoo! [20:26:05] yes that was the objection, but I think like in that meeting I said it was not really a hard dependency... I mean [20:26:47] I don't object to it being in WME, but I am surprised if things in WME go out on every page kinda [20:27:11] yeah, i dunno, maybe we need Krinklle for this discussion? [20:27:40] yeah [20:28:03] whatever the answer is....he approved your code so woohoo! [20:28:04] hahah [20:28:45] hip: just saw your gchat response, would love if we can quickly bikeshed this one today [20:28:49] we can hjop in a hangout with mforns, he's here now [20:28:53] if we need to [20:29:02] oh okay that'd be great [20:29:06] am still hoping we might be able to get this into next weeks train! [20:29:11] mforns: back to the bc? [20:29:16] I only have until 4:00 EST becaus I have to give an interview but then I'm free after that as well [20:29:26] ottomata, yea! [20:29:38] ottomata: get the EL patch? wow you're really gunning [20:29:43] YEAH [20:29:49] hip, https://meet.google.com/rxb-bjxn-nip [20:29:49] well let's try it, I'm ready to talk if y'all are [20:29:50] lets' go! [23:10:35] (03CR) 10Jforrester: [V: 03+2 C: 03+2] Allow L10n-Bot to commit translation updates [analytics/wikistats2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/573970 (https://phabricator.wikimedia.org/T245805) (owner: 10MarcoAurelio) [23:11:50] (03CR) 10Jforrester: [V: 03+2 C: 03+2] "Unfortunately only Analytics have submit rights here, sorry." [analytics/wikistats2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/573970 (https://phabricator.wikimedia.org/T245805) (owner: 10MarcoAurelio) [23:51:46] 10Analytics, 10Product-Analytics, 10Tracking-Neverending: Spark query issues - https://phabricator.wikimedia.org/T245891 (10nshahquinn-wmf) [23:52:24] 10Analytics, 10Product-Analytics, 10Tracking-Neverending: Spark query issues - https://phabricator.wikimedia.org/T245891 (10nshahquinn-wmf) [23:52:40] 10Analytics, 10Product-Analytics, 10Tracking-Neverending: Spark query issues - https://phabricator.wikimedia.org/T245891 (10nshahquinn-wmf) [23:52:42] 10Analytics, 10Product-Analytics (Kanban): Update wmfdata to use sensible Spark settings - https://phabricator.wikimedia.org/T245097 (10nshahquinn-wmf) [23:52:44] 10Analytics, 10Product-Analytics: wmfdata cannot recover from a crashed Spark session - https://phabricator.wikimedia.org/T245713 (10nshahquinn-wmf)