[00:28:29] <wikibugs>	 10Analytics, 10DBA, 10Data-Services, 10Research, 10cloud-services-team (Kanban): Implement technical details and process for "datasets_p" on wikireplica hosts - https://phabricator.wikimedia.org/T173511#3530850 (10MZMcBride) This task is potentially of interest to me. I previously ran a dumb report with...
[00:53:34] <wikibugs>	 10Analytics, 10DBA, 10Data-Services, 10Research, 10cloud-services-team (Kanban): Implement technical details and process for "datasets_p" on wikireplica hosts - https://phabricator.wikimedia.org/T173511#3739451 (10bd808) >>! In T173511#3739381, @MZMcBride wrote: > If this task were implemented, under "Pr...
[02:22:41] <wikibugs>	 (03PS1) 10Milimetric: Add hif.wiktionary to whitelist [analytics/refinery] - 10https://gerrit.wikimedia.org/r/389658
[02:22:56] <wikibugs>	 (03CR) 10Milimetric: [V: 032 C: 032] Add hif.wiktionary to whitelist [analytics/refinery] - 10https://gerrit.wikimedia.org/r/389658 (owner: 10Milimetric)
[07:15:55] <wikibugs>	 10Analytics-Kanban, 10Analytics-Wikistats: Fonts from fonts.googleapis.com on wikistats - https://phabricator.wikimedia.org/T178317#3688129 (10Nemo_bis) > this might violate the privacy policy  Definitely yes. Thanks for removing them.
[07:30:41] <wikibugs>	 10Analytics, 10Analytics-Wikistats, 10I18n: Move non-SI prefixes to user- or locale-specific interface - https://phabricator.wikimedia.org/T179906#3739830 (10Nemo_bis)
[08:54:50] <elukey>	 hello team! eventlogging_cleaner just finished to sanitize data up to the middle of 2016 \o/
[09:31:49] <joal>	 Yay elukey !
[09:41:08] <elukey>	 it is difficult to find a table with data to sanitize
[10:23:52] <elukey>	 joal: would you mind to log to the staging database on db1108 and see if you can create a table?
[10:24:18] <joal>	 sure elukey 
[10:24:36] <joal>	 elukey: however I am not sure about user/procedure
[10:24:48] <joal>	 elukey: Should I use my ldap username, or research?
[10:25:13] <elukey>	 In theory both should work
[10:25:27] <joal>	 ok
[10:27:08] <joal>	 elukey: ldap username fails
[10:28:31] <elukey>	 joal: can you retry now?
[10:29:05] <joal>	 elukey: worked using research user
[10:29:49] <joal>	 elukey: not for my personal user nope
[10:31:11] <elukey>	 are you able to connect to anything on db1108? I checked and you don't seem to have perms for your username
[10:31:44] <elukey>	 (sorry for the annoying questions :( )
[10:32:15] <joal>	 elukey: elukey nope, can't even log in mysql
[10:32:34] <joal>	 elukey: don't worry, I'm glad I can help :)
[10:33:12] <elukey>	 ahhh okok so it makes sense
[10:33:17] <elukey>	 good all works as expected :)
[10:33:24] <joal>	 great !
[10:33:36] <elukey>	 so I created a empty staging db on db1108 so people will have a scratchpad
[10:33:54] <elukey>	 then we'll need to figure out how/where to migrate the one on db1047 to dbstore1002 or elsewhere
[10:34:05] <elukey>	 I'll write everything in the email to engineering
[10:34:10] <elukey>	 and ask people to chime in the task
[10:34:11] <joal>	 You rock elukey :)
[10:34:20] <elukey>	 \o/
[11:44:16] <wikibugs>	 10Analytics-Tech-community-metrics, 10Developer-Relations (Oct-Dec 2017): Empty or incorrect data on Kibana's "Git-Demographics" dashboard - https://phabricator.wikimedia.org/T171240#3740667 (10Aklapper)
[11:44:29] <wikibugs>	 10Analytics-Tech-community-metrics, 10Developer-Relations (Oct-Dec 2017): Empty or incorrect data on Kibana's "Git-Demographics" dashboard - https://phabricator.wikimedia.org/T171240#3458588 (10Aklapper) 05Open>03Resolved
[12:34:54] <joal>	 Taking a break a-team
[12:35:14] <fdans>	 elukey you're the beeeeest
[12:47:03] <elukey>	 fdans: hello Francisco, not sure why but thanks <3
[12:47:16] * elukey going to lunch + errand
[14:16:42] <elukey>	 mforns: o/
[14:18:16] <elukey>	 whenever you have time I'd like to do some vetting on el tables (db1108)
[14:18:20] <elukey>	 el cleaner completed!
[14:36:32] <mforns>	 elukey, coool, OK will do today
[14:36:45] <mforns>	 elukey, completed completed? until today?
[14:37:10] <elukey>	 mforns: nope, until 2016-06-21
[14:37:23] <mforns>	 ok, great :]
[14:37:33] <elukey>	 I tried to check this morning some tables but most of the times all fields were already nulled
[14:37:40] <mforns>	 this way I can test before and after
[14:37:45] <mforns>	 aha
[14:38:07] <elukey>	 so I didnt' find some proof that all is working :D
[14:39:25] <wikibugs>	 10Analytics-EventLogging, 10Analytics-Kanban, 10Patch-For-Review: Resolve EventCapsule / MySQL / Hive schema discrepancies - https://phabricator.wikimedia.org/T179625#3741250 (10Ottomata) @krinkle, I just submitted https://gerrit.wikimedia.org/r/#/c/389713/, let me know what you think.  Also, do you use the...
[14:40:07] <elukey>	 mforns: but it is impressive that in 3/4 days db1108 crunched this amount of data
[14:40:17] <elukey>	 true that we don't have anymore some huge tables
[14:40:33] <mforns>	 aha
[14:40:34] <elukey>	 but it runs fast anyway :D
[14:40:49] <elukey>	 100k updates with sleep 2 seconds
[14:41:12] <mforns>	 yea, that's awesome
[14:43:51] <elukey>	 ottomata: o/
[14:44:02] <elukey>	 I am rolling restart jumbo for jvm updates
[14:45:10] <ottomata>	 hiii ok!
[14:50:12] <wikibugs>	 10Analytics-EventLogging, 10Analytics-Kanban: Timestamp format in Hive-refined EventLogging tables is incompatible with MySQL version - https://phabricator.wikimedia.org/T179540#3741320 (10Ottomata) > Aye! but MySQL users are not the only user of this data! The performance team folks use it build Grafana dashb...
[14:58:54] <wikibugs>	 10Analytics-Kanban, 10User-Elukey: Restart Analytics JVM daemons for open-jdk security updates - https://phabricator.wikimedia.org/T179943#3741372 (10elukey)
[15:06:14] <elukey>	 joal: --^
[15:23:54] <joal>	 elukey: not in the list: NameNode and ResourceManager (analytics100[12] - On purpose? already done?)
[15:26:10] <elukey>	 nono I forgot, adding them
[15:27:00] <wikibugs>	 10Analytics-Kanban, 10User-Elukey: Restart Analytics JVM daemons for open-jdk security updates - https://phabricator.wikimedia.org/T179943#3741501 (10elukey)
[15:36:32] <wikibugs>	 10Analytics-Kanban, 10User-Elukey: Restart Analytics JVM daemons for open-jdk security updates - https://phabricator.wikimedia.org/T179943#3741522 (10elukey)
[15:37:17] <elukey>	 ottomata: qq whenever you have time - I'd need to create a druid cluster (even one node) in labs to test my prometheus stuff
[15:37:49] <elukey>	 is there anything running by any chance?
[15:37:52] <ottomata>	 naw
[15:37:56] <ottomata>	 but should be easy enough
[15:38:21] <ottomata>	 have fun copy and pasting all those hiera yamls into the right place because I know you love those profile params without defaults so miuch nannanananabooboooooo :p :)
[15:38:48] * elukey <_<
[15:39:04] <elukey>	 :D
[15:39:13] <elukey>	 what deep storage should I configure?
[15:39:20] <ottomata>	 do you need to test hdfs stuff?  role::druid::worker
[15:39:23] <ottomata>	 oops sorry
[15:39:25] <ottomata>	 first q:
[15:39:29] <ottomata>	 do you need to test hdfs stuff?
[15:39:30] <ottomata>	 ?
[15:39:56] <elukey>	 nono i need something that pushes metrics via http
[15:40:10] <ottomata>	 ok, i think if you don't set that stuff
[15:40:16] <ottomata>	 it'll pick sane defaults in the module
[15:40:20] <elukey>	 super
[15:40:28] <elukey>	 and I guess I can run a single node right?
[15:40:40] <ottomata>	 yeah
[15:40:45] <ottomata>	 you'll need zk
[15:40:47] <ottomata>	 somewhere
[15:41:07] <elukey>	 yep yep
[15:41:12] <ottomata>	 i think you can leave out setting all the *::properties hiera profile params to empty
[15:41:27] <ottomata>	 profile::druid::common::use_cdh: false
[15:41:36] <elukey>	 ack!
[15:41:56] <elukey>	 going to do some unit testing and then I'll try to work on druid in labs
[15:42:13] <ottomata>	 oo, if you get a zk runnin glocally on the same host
[15:42:19] <ottomata>	 the druid will default to using localhost:2181
[15:49:26] <nuria_>	 ottomata: hola
[15:49:52] <nuria_>	 ottomata: the db on  beta cluster  might be horked , i ma going to reboot mysql
[15:51:12] <ottomata>	 waait
[15:51:14] <ottomata>	 i'm in it now
[15:51:41] <ottomata>	 nuria_:  what's the problem?
[15:51:50] <nuria_>	 ottomata: it does not create tables
[15:51:59] <nuria_>	 ottomata: there is no popup table
[15:52:11] <ottomata>	 i deleted the log databse yesterday
[15:52:15] <ottomata>	 because it had totally filled up disk
[15:52:22] <ottomata>	 do we know there are popup events in beta right now?
[15:52:40] <nuria_>	 ottomata: i send several yesterday
[15:52:44] <nuria_>	 ottomata: and you can too
[15:52:46] <nuria_>	 https://wikitech.wikimedia.org/wiki/Analytics/Systems/EventLogging/TestingOnBetaCluster#How_to_log_a_client-side_event_to_Beta_Cluster_directly
[15:53:00] <nuria_>	 ottomata: let me see those are valid ones
[15:57:18] <ottomata>	 i see them coming in, butnot going to valid-mixed topic
[15:59:53] <wikibugs>	 10Analytics-Kanban, 10User-Elukey: Restart Analytics JVM daemons for open-jdk security updates - https://phabricator.wikimedia.org/T179943#3741623 (10MoritzMuehlenhoff) Note that the hadoop clusters and kafka* are running Java 7 and there hasn't been an openjdk-7 release yet (so also no update in Debian), so a...
[16:00:43] <nuria_>	 ping milimetric ottomata 
[16:01:22] <ottomata>	 oh
[16:01:47] <joal>	 milimetric: Can you hear us?
[16:01:50] <wikibugs>	 10Analytics-Kanban, 10User-Elukey: Restart Analytics JVM daemons for open-jdk security updates - https://phabricator.wikimedia.org/T179943#3741633 (10elukey) Yep, I was planning to keep this open until the new updates will arrive for jdk 7, but if this is too far in the future I'll open a new phab task later o...
[16:06:51] <ottomata>	  oh nuria duh.  Popups is blacklisted from MySQL
[16:29:21] <wikibugs>	 10Analytics-Kanban: Geowiki stopped updating on October 24th - https://phabricator.wikimedia.org/T179952#3741723 (10Milimetric)
[16:31:27] <nuria_>	 ottomata: i am a dummy . twice
[17:10:34] <joal>	 Gone for dinner a-team, back after
[17:29:03] <wikibugs>	 10Analytics, 10EventBus, 10Services (next): Clean up retry-retry Kafka topics - https://phabricator.wikimedia.org/T179958#3741948 (10Pchelolo)
[17:49:27] * elukey off!
[17:50:25] <elukey>	 a-team let me know if you want me to make adjustments to the email for engineering@ (About the db refactor), if I don't hear anything I'll send it tomorrow morning :)
[17:50:28] <elukey>	 byyyee
[18:20:35] <nuria_>	 milimetric: i had put the caching dataset here: https://analytics.wikimedia.org/datasets/archive/limn-public-data/caching/
[18:20:54] <nuria_>	 milimetric: but i think things got moved arround, do you know where it could be now?
[18:30:00] <milimetric>	 nuria_: https://analytics.wikimedia.org/datasets/archive/public-datasets/analytics/caching/
[18:30:02] <milimetric>	 is that it?
[18:31:03] <nuria_>	 milimetric: YES!
[18:31:12] <milimetric>	 :) yay
[18:31:21] <wikibugs>	 10Analytics-Kanban: Compile a request data set for caching research and tuning - https://phabricator.wikimedia.org/T128132#3742235 (10Nuria) Updated link to data: https://analytics.wikimedia.org/datasets/archive/public-datasets/analytics/caching/
[18:31:27] <mforns_>	 hey team, anyone is up for some navigationtiming-druid brainbounce? there's some data already in pivot: https://tinyurl.com/ybs7u4z5
[18:32:17] <nuria_>	 mforns_: i can , let me move places and find a source of power
[18:32:28] <mforns_>	 k!
[18:36:44] <nuria_>	 mforns: omw to cave
[18:40:48] <ebernhardson>	 anyone know how to fail a hive job in oozie if some partition doesn't exist? I re-deployed one of our repos yesterday but forgot to set the partition for wmf_raw.mediawiki_project_namespace_map appropriately, so while the jobs finished their result was partitions with 0 rows. Ideally i'd like to figure out how to make the hive script fail if the partition doesn't exist
[18:42:59] <ottomata>	 hmmm, good q ebernhardson not sure hm.
[18:43:54] <ottomata>	 basically you want fail if 0 results?
[18:43:58] <ottomata>	 hm
[18:44:03] <ebernhardson>	 ottomata: i suppose that would work too
[18:44:06] <ottomata>	 ebernhardson:  not sure, i think maybe
[18:44:13] <ottomata>	 you'd need to explicitly check if partition exists somehow
[18:44:17] <ottomata>	 as a separate oozie step
[18:44:21] <ottomata>	 you can
[18:44:25] <ottomata>	 show partitions tableName
[18:44:26] <ottomata>	 in hive
[18:44:27] <ottomata>	 but
[18:44:28] <ottomata>	 hm
[18:44:46] <ottomata>	 ebernhardson:  are your oozie jobs launching on existence of partitions afte rload?
[18:45:16] <ebernhardson>	 ottomata: ooh, yea i suppose i could add a dataset requirement
[18:45:27] <ebernhardson>	 ottomata: it currently only depends on wmf_raw.cirrussearchrequestset and wmf.webrequest partitions
[18:46:41] <ottomata>	 ebernhardson:  ya that would probably help, if you need partitions to exist before job runs
[18:46:55] <ottomata>	 probably better to make explicit dataset dependency rather than failling if not exists
[18:53:00] <joal>	 +1 for that approach ebernhardson :)
[18:56:29] <ebernhardson>	 what creates wmf_raw.mediawiki_project_namespace_map ? I don't see it in analytics/refinery so not sure what to set for instance variables in oozie. It looks like i need roughly month-1, but not quite exactly
[18:57:12] <ebernhardson>	 for example oct was created nov 3, sept was created oct 3, aug was created sept 13
[18:59:01] <joal>	 ebernhardson: wmf_raw.mediawiki_project_namespace_map data gets created from a cron runnin
[18:59:26] <joal>	 refinery/bin/wmf_raw.mediawiki_project_namespace_map
[18:59:47] <ebernhardson>	 joal: ahh ok, thanks!
[19:00:13] <joal>	 This cron is supposed to run every month, adding a new snapshot - it runs the 3rd of the month, to prevent having big uniques job in parallel with buig sqoop jobs
[19:00:18] <joal>	 ebernhardson: --^
[19:00:36] <joal>	 I think for aug we had an issue and reissue it manually :)
[19:01:20] <joal>	 also ebernhardson - Once data is downloaded, we have an oozie job rebuilding the hive table (refinery/oozie/mediawiki/history/load)
[19:01:59] <ebernhardson>	 ok i'll figure out how to get oozie + hive to use the new snapshot after the 3rd
[19:02:12] <joal>	 ebernhardson: The above job just run REPAIR on all tables we update (sqooped + namespace_map) and flags them with a _PARTITIONED for later oozie dependencies
[19:02:51] <joal>	 You have refinery/oozie/mediawiki/history/datasets_raw.xml to help with that: )
[19:02:55] <joal>	 ebernhardson: --^
[19:03:46] <DarTar>	 hey analytics
[19:03:47] <DarTar>	 I was checking out the status of wikistats 2.0, beautiful work
[19:03:50] <ebernhardson>	 ooh, yes! somehow my grep missed that one
[19:03:56] <DarTar>	 one thing I noticed, IDK if you’re working on it:
[19:04:27] <DarTar>	 timestamps in the human-readable table view should probably be formatted 
[19:04:59] <DarTar>	 e.g. Unique Devices, Monthly, enwiki – gives
[19:05:00] <DarTar>	 Month	Total
[19:05:01] <DarTar>	 "2016-01-01T00:00:00.000Z"	661172290
[19:07:20] <joal>	 Hi DarTar, thanks for the feedback :)
[19:07:38] <DarTar>	 hey joal 
[19:07:50] <DarTar>	 love the overall UX and usability
[19:07:54] <joal>	 DarTar: Would you mind creating a task about that ? fdans and milimetric currently work the UI, so it could be picked up pretty fast :)
[19:08:01] <DarTar>	 will do
[19:08:07] <joal>	 DarTar: Some effort has been put onto that, yes ;)
[19:08:34] <DarTar>	 the other thing I noticed is the metric definition, currntly linking out to full-text on wiki. Any plans to have tooltips?
[19:08:54] <joal>	 DarTar: I have no clue :)
[19:09:20] <DarTar>	 joal: I’ll file tasks, you guys can close them as invalid as appropriate
[19:09:35] <ottomata>	 mforns:  nuria_, what happens if i merge the mariadb el whitelist change?
[19:09:42] <ottomata>	 is anything else needed?
[19:09:46] <ottomata>	 its totally safe to merge?
[19:09:48] <mforns>	 ottomata, no
[19:09:49] <joal>	 I'm pretty sure they'll be given some thoughts DarTar :) Many thanks for that !
[19:10:20] <mforns>	 ottomata, yes, it means the next time we execute the purging script, the fields that were removed from the white-list will be also purged!
[19:10:34] <ottomata>	 ok cool
[19:10:37] <ottomata>	 merging then
[19:10:40] <mforns>	 :]
[19:13:15] <DarTar>	 joal: is #Analytics-wikistats the right tag?
[19:13:50] <joal>	 DarTar: #Analytics, and #analytics-wikistats please :)
[19:14:12] <DarTar>	 joal: cool
[19:15:02] <nuria_>	 DarTar: probably not tooltips for the near term
[19:15:22] <milimetric>	 DarTar (cc joal): analytics-anything automatically tags analytics, btw, so you can tag intuitively
[19:15:32] <nuria_>	 DarTar: but yes to date formatting for humans 
[19:15:44] <joal>	 Oh I didn't know !!! Thanks milimetric :)
[19:15:52] <milimetric>	 nuria_ / DarTar we do have a tooltip task that I'd like to get to sooner than later: https://phabricator.wikimedia.org/T177950
[19:17:10] <wikibugs>	 10Analytics, 10Analytics-Wikistats: Format timestamps as human-readable datetimes in Wikistats 2.0 table view - https://phabricator.wikimedia.org/T179971#3742368 (10DarTar)
[19:18:58] <wikibugs>	 10Analytics, 10Analytics-Wikistats: Add tooltips for metric definitions in Wikistats 2.0 - https://phabricator.wikimedia.org/T179973#3742396 (10DarTar)
[19:19:12] <DarTar>	 milimetric: oh cool
[19:19:29] <DarTar>	 I just posted a phab task :) 
[19:19:51] <nuria_>	 ottomata: let me know if you want my very not helpful self helping you test EL chnages in beta
[19:19:59] <nuria_>	 ottomata: after my stellar performance yesterday
[19:25:45] <wikibugs>	 10Analytics, 10Analytics-Wikistats: How to link/cite to Wikistats 2.0 - https://phabricator.wikimedia.org/T179975#3742432 (10DarTar)
[19:26:06] <DarTar>	 alright, done filing wikistats suggestions. ttyl folks
[19:26:25] <wikibugs>	 10Analytics, 10Analytics-Wikistats: How to link to/cite Wikistats 2.0 - https://phabricator.wikimedia.org/T179975#3742445 (10DarTar)
[19:26:46] <joal>	 milimetric: I think I have encountered a new edge case for deletions :)
[19:26:50] <joal>	 milimetric: wanna see?
[19:27:49] <milimetric>	 on my way joal
[19:30:57] <wikibugs>	 10Analytics, 10Analytics-Wikistats: Beta: How to link to/cite Wikistats 2.0 - https://phabricator.wikimedia.org/T179975#3742449 (10Nuria)
[19:54:46] <wikibugs>	 10Analytics-Kanban, 10Performance-Team, 10Patch-For-Review: Explore NavigationTiming by faceted properties - EventLogging refine - https://phabricator.wikimedia.org/T166414#3742488 (10mforns) We encountered a couple difficulties in the way Pivot works versus the nature of NavigationTiming measures: - Navigat...
[19:58:18] <wikibugs>	 10Analytics-Kanban: Create scala-spark job to ingest simple data sets from Hive-EventLogging to Druid to Pivot - https://phabricator.wikimedia.org/T179976#3742504 (10mforns)
[20:10:58] <milimetric>	 back is killing me, gonna go see a chiropractor
[20:11:00] <milimetric>	 oof, can't focus
[20:24:57] <chelsyx>	 Hello! Do we have the application data in hadoop? I was running a query in commonswiki database and used `INSTR()` on a large table. It took more than a week to run and the query got killed finnaly... :(
[20:27:42] <ottomata>	 what does applicatino data mean?
[20:29:09] <chelsyx>	 like the data we have in mysql for enwiki, commonswiki, etc, which contains tables like user, page, logging, etc
[20:29:58] <chelsyx>	 https://upload.wikimedia.org/wikipedia/commons/9/94/MediaWiki_1.28.0_database_schema.svg
[20:30:31] <wikibugs>	 10Analytics-EventLogging, 10Analytics-Kanban: Timestamp format in Hive-refined EventLogging tables is incompatible with MySQL version - https://phabricator.wikimedia.org/T179540#3742635 (10Ottomata) I talked with Timo and with the A-team, and we decided it'd be best to remove `timestamp` altogether in favor of...
[20:31:06] <ottomata>	 chelsyx: https://wikitech.wikimedia.org/wiki/Analytics/Data_Lake/Edits
[20:32:16] <chelsyx>	 But not all tables are there, right? I need the tables like  image, categorylinks...
[20:36:41] <nuria_>	 chelsyx: no, not all tables are scooped
[20:37:02] <nuria_>	 chelsyx: the ones that are are the ones needed  to rebuild history
[20:37:17] <nuria_>	 chelsyx: and image is not one of them
[20:37:36] <nuria_>	 chelsyx: you can scoop data though to your user space
[20:38:36] <nuria_>	 chelsyx: tables scooped for mw are here: https://github.com/wikimedia/analytics-refinery/blob/master/bin/sqoop-mediawiki-tables
[20:39:10] <nuria_>	 chelsyx: but scooping might not be teh solution here, what data are you interested in?
[20:41:14] <nuria_>	 fdans, milimetric : iam going to split off js loading and css loading in webpack ok?
[20:42:36] <chelsyx>	 nuria_: I need to count the number of files by number of categories on commons wiki. This is the query that got killed: https://phabricator.wikimedia.org/P6283
[20:44:32] <nuria_>	 chelsyx: and this data is not on dumps at all (no idea, just asking)
[20:46:08] <chelsyx>	 nuria_: I'm not sure if it is in dumps... I had a hard time parsing dumps data... scooping tables to my user space looks like a good solution for me
[20:46:46] <chelsyx>	 nuria_: Thanks!
[20:46:47] <nuria_>	 chelsyx:  scooping is also not free of hardships so you know
[20:47:02] <nuria_>	 chelsyx:  so i woudl not rule out dumps
[20:47:45] <chelsyx>	 nuria_: Got it. I will try. Thanks! :)
[21:03:26] <wikibugs>	 10Analytics-EventLogging, 10Analytics-Kanban, 10Patch-For-Review: Resolve EventCapsule / MySQL / Hive schema discrepancies - https://phabricator.wikimedia.org/T179625#3742753 (10Ottomata)
[21:03:34] <wikibugs>	 10Analytics-EventLogging, 10Analytics-Kanban, 10Patch-For-Review: Resolve EventCapsule / MySQL / Hive schema discrepancies - https://phabricator.wikimedia.org/T179625#3731162 (10Ottomata)
[21:10:46] <ottomata>	 nuria_:  kinda want to deploy https://gerrit.wikimedia.org/r/#/c/388255/ tomorrow morning.
[21:10:50] <ottomata>	 got a +1 for me :D
[21:10:51] <ottomata>	 ?
[21:20:29] <nuria_>	 ottomata: sounds good, it being a no-op
[21:20:45] <nuria_>	 ottomata: besides my mostly failed attempts to look at beta i did not do anything
[21:21:03] <ottomata>	 its running in beta now, i'm going to delploy one more version there that updates the capsule scid
[21:21:05] <ottomata>	 with a new capsule version
[21:21:12] <nuria_>	 ottomata: I checked there were no errors in mysql consumer
[21:21:13] <ottomata>	 that version will be backwards compatible too
[21:21:20] <nuria_>	 ottomata: but did not looked at records on db
[21:21:23] <nuria_>	 ottomata: did you?
[21:21:41] <ottomata>	 i did yesterday, but not today
[21:21:53] <ottomata>	 actually yeah, i'm looking at one right now :)
[21:26:38] <nuria_>	 ottomata: ok , is UA string like it was before ( i imgine so or inserts will fail)
[21:26:42] <nuria_>	 *imagine
[21:27:16] <nuria_>	 fdans: do you have a .jscsrc for wikistats you want to share?
[21:28:34] <ottomata>	 nuria_:  it is currently string like before.
[21:28:50] <ottomata>	 i'm (trying?) to make userAgent in capsule schema a union type of ['object', 'string']
[21:28:52] <nuria_>	 ottomata: right, just triple checking
[21:29:18] <nuria_>	 ottomata: that seems like it should not work, does it?
[21:29:35] <ottomata>	 yeah it does, wiki not letting me save schema atm...
[21:29:40] <ottomata>	 Value [ "object", "string" ] not in enum for property Schema -> Properties (properties) -> userAgent -> Type (type)
[21:29:42] <ottomata>	 whatever that means!
[21:30:06] <ottomata>	 schema is valid though.
[21:37:22] <ottomata>	 yargh, nuria_ eventlogging extension validates that the schemas conform to a schema
[21:37:29] <ottomata>	 which is not just the draft 03 jsonschema
[21:37:53] <ottomata>	 https://github.com/wikimedia/mediawiki-extensions-EventLogging/blob/master/schemas/schemaschema.json
[21:39:16] <ottomata>	 rats, i think i gotta use type: any
[21:40:24] <nuria_>	 ottomata: so many things , i actually think that any  seems better here
[21:40:57] <nuria_>	 ottomata: cause object, string  and null seems a bit a random bag of types
[21:52:06] <ottomata>	 ok
[21:52:08] <nuria_>	 fdans: so once started npm run dev should pick up changes and deploy them to local dist-dev directory right?
[21:52:09] <ottomata>	 new capsule version
[21:52:09] <ottomata>	 https://meta.wikimedia.org/wiki/Schema:EventCapsule
[21:52:26] <ottomata>	 https://meta.wikimedia.org/w/index.php?title=Schema%3AEventCapsule&type=revision&diff=17397982&oldid=16619630
[21:52:46] <nuria_>	 ottomata: ok, if jsonschema swallows that...
[21:53:04] <fdans>	 nuria_: yes, it should regenerate bundle.js
[21:53:21] <ottomata>	 also, EL in beta is running with latest patch and latest schema
[21:53:33] <nuria_>	 fdans: i bet it does it when you change js files but not teh webpack config
[21:53:50] <fdans>	 that’s right!
[21:54:19] <nuria_>	 fdans: ya, i see
[23:30:56] <wikibugs>	 10Analytics-Kanban, 10Patch-For-Review: Remove AppInstallIId from EventLogging purging white-list - https://phabricator.wikimedia.org/T178174#3743156 (10JMinor) We don't consent to this change. Before you merge a change which will irrevocably delete data that could be vital to the long term understanding of ou...