[00:05:43] <wikibugs>	 Analytics-EventLogging, Analytics-Kanban: Move Eventlogging Kafka writer to use pykafka's Producer instead of python-kafka {stag} [8 pts] - https://phabricator.wikimedia.org/T109244#1547838 (madhuvishy)
[00:53:00] <wikibugs>	 Analytics-EventLogging, MobileFrontend, Technical-Debt: MobileFrontend's schema code should be upstreamed to EventLogging - https://phabricator.wikimedia.org/T109398#1547889 (Jdlrobson) NEW a:Jdlrobson
[03:44:07] <wikibugs>	 Analytics-EventLogging, Database: Add index on event_action, event_isAnon and event_namespaceId to NavigationTiming tables - https://phabricator.wikimedia.org/T70396#1548019 (Krenair)
[03:50:52] <wikibugs>	 Analytics-Tech-community-metrics: Closed tickets in Bugzilla migrated without closing event? - https://phabricator.wikimedia.org/T107254#1548023 (RobLa-WMF) I had this on my (admittedly not high priority) wishlist of things to fix up.  Lemme copy over the email I sent to Greg last week:  > I saw the recent d...
[13:02:48] <halfak>	 o/ joal & milimetric
[13:02:59] <milimetric>	 hey halfak
[13:03:10] <halfak>	 I'll be in there in  minute
[13:03:23] <milimetric>	 no worries, I'm in the meeting, jo's not here
[14:14:22] <joal>	 hey halfak, milimetric
[14:14:28] <joal>	 Sorry for not having shown up :(
[14:14:53] * halfak growls like a warewolf
[14:14:55] <halfak>	 :P
[14:15:11] * joal hides away as far as I can !
[14:15:17] <joal>	 :)
[14:15:25] <halfak>	 We talked about building up the event stream we'd like to start experimenting with in kafka.
[14:15:48] <joal>	 I had nothing new on my side (mostly kafka upgrade this week)
[14:16:01] <joal>	 halfak: cool :)
[14:16:06] <halfak>	 I have some code for turning MediaWiki's various datasources into events.  I'd like to run that against the DB/dumps in order to get some preliminary datasets.
[14:16:19] <halfak>	 With these datasets, we can test out some of the wikistats metrics
[14:16:31] <halfak>	 and evaluate this event feed for use in wikistats 2.0
[14:16:53] <joal>	 GReat :)
[14:17:41] <joal>	 If I can help, you know where to find me (usually :S)
[14:18:18] <halfak>	 We will likely need to perform a giant sort of events after I generate them.
[14:18:32] <halfak>	 If we're doing this for wikistat2.0 that sort could involve text -- not just metadata.
[14:18:44] <joal>	 halfak: I think I get the point :)
[14:23:06] <halfak>	 Just wanted you to know what's coming
[14:24:18] <joal>	 halfak: Thanks :)
[14:46:44] <wikibugs>	 Analytics-Kanban, Patch-For-Review, WMF-deploy-2015-08-11_(1.26wmf18): Event Logging sends mysql consumer stats to statsd [8 pts] {oryx} - https://phabricator.wikimedia.org/T105935#1549083 (kevinator) Open>Resolved
[14:46:58] <wikibugs>	 Analytics-EventLogging, Analytics-Kanban, Patch-For-Review: EventLogging Icinga Alerts should look at a longer period of time to prevent false positives {stag} [5 pts] - https://phabricator.wikimedia.org/T108339#1549087 (kevinator) Open>Resolved
[14:47:14] <wikibugs>	 Analytics-Kanban, Patch-For-Review: Check and potentially timebox limn-flow-data reports {tick} [5 pts] - https://phabricator.wikimedia.org/T107502#1549089 (kevinator) Open>Resolved
[14:56:47] <joal>	 hey ottomata
[14:56:50] <ottomata>	 hiya!
[14:56:57] <ottomata>	 late morning for me :)  how's it/
[14:56:57] <ottomata>	 ?
[14:57:12] <joal>	 backfilled a lot from this morning :)
[14:57:21] <joal>	 But I wonder if this as not put camus late :(
[14:58:51] <ottomata>	 oh wha?
[15:00:04] <joal>	 The proportion of wrong load job is pretty high :(
[15:00:35] <ottomata>	 ?
[15:00:36] <ottomata>	 oh
[15:00:45] <ottomata>	 oh, hm, loking now, camus looks good
[15:01:01] <ottomata>	 it isn't lagging, but maybe load jobs on new data is lagging
[15:01:03] <ottomata>	 that's ok though
[15:01:09] <ottomata>	 i don't mind jobs in hadoop lagging
[15:01:31] <ottomata>	 joal: run stat1002:/home/otto/bin/camus_lag
[15:01:31] <ottomata>	 :)
[15:01:40] <ottomata>	 it isn't very accurate, as it only looks at the last history file
[15:01:45] <joal>	 it's not jobs lagging that concerns me, it's the wrong text load jobs
[15:01:51] <ottomata>	 ?
[15:02:12] <ottomata>	 you mean if the load starts before camus is  done?
[15:02:15] <ottomata>	 that would be camus lagging
[15:02:46] <joal>	 load_text is wrong since 04 this morning
[15:02:54] <ottomata>	 still don't understand
[15:03:07] <ottomata>	 what do you mean wrong?
[15:03:22] <joal>	 The outcome of the load job is "killed"
[15:03:25] <ottomata>	 ohhhh
[15:03:29] <ottomata>	 looking
[15:03:37] <joal>	 meaning that some stats are probably wrong :(
[15:03:54] <joal>	 I need to learn how to double check for missing data in the sequence stats
[15:04:15] <ottomata>	 hm
[15:04:20] <ottomata>	 i see.
[15:04:20] <ottomata>	 hm
[15:04:29] <ottomata>	 you can just run the query from the job that creates them
[15:04:38] <ottomata>	 i'll check one now
[15:04:44] <ottomata>	 grr that will be so annoying if that is what is happening