[01:45:45] <springle>	 halfak: yt?
[01:46:01] <halfak>	 unofficially, yes.  What's up?
[01:46:36] <springle>	 datasets/wikimedia_wikis.tsv --query=sql/editor_months.sql
[01:46:47] <springle>	 how many concurrenct connections is that likely to use?
[01:47:26] <springle>	 or rather, whatever launches those on stat1003
[01:48:11] <halfak>	 I only needs to use one connection at a time.  It should create one connection per slice and re-use it.
[01:48:22] <springle>	 there are currently ~250 sleeping research connections to -store
[01:48:46] <halfak>	 Oh.  that's not me.  that script is reading the s[1-7] slaves because I need that fancy archive index.
[01:48:54] <springle>	 i've raised max_connections temporarily
[01:49:16] <springle>	 hmm
[01:49:23] <springle>	 python     3241      halfak  257u  IPv4 48131274      0t0  TCP stat1003.wikimedia.org:52352->dbstore1002.eqiad.wmnet:mysql (ESTABLISHED)
[01:49:30] <springle>	  3241 pts/6    S+    41:33 /home/halfak/env/3/bin/python -tt -c from mae import query_wikis; query_wikis.main(); datasets/wikimedia_wikis.tsv --query=sql/editor_months.sql
[01:49:35] <springle>	 sure?
[01:49:37] * halfak is confirming. 
[01:49:49] <halfak>	 Oh wait.  Did someone change the DNS?
[01:50:02] <springle>	 s3-analytics-slave is switched over, yes
[01:50:06] <springle>	 (and the indexes exist)
[01:50:08] <halfak>	 There was an email about that I responded to right away with the updated index ticket.
[01:50:13] <springle>	 ^
[01:50:15] <halfak>	 Oh.  Should be OK then.
[01:50:19] <springle>	 yep
[01:50:33] <springle>	 so is that many concurrent connections expeted? i should plan for it?
[01:50:53] <halfak>	 Nope
[01:51:06] <halfak>	 Are you sure that all of the connections are from that one process?
[01:52:14] <springle>	 yes
[01:52:33] <springle>	 on stat1003, run: lsof -i tcp | grep mysql
[01:52:52] <springle>	 all from process 3241
[01:53:08] <springle>	 maybe not closing connections properly?
[01:53:37] <halfak>	 It's not supposed to close 'em.  Looks like it's not sharing them properly though.
[01:53:55] <springle>	 :)
[01:54:15] <halfak>	 So, this has been running for a few days without issue.  Anything happen recently?
[01:54:25] <springle>	 the DNS switch
[01:54:33] <springle>	 -store had a lower max_connections
[01:54:41] <springle>	 only reason i noticed
[01:54:53] <halfak>	 Cleared.
[01:54:56] <springle>	 it's not a showstopper
[01:55:02] <springle>	 jusy would be nice to know why
[01:57:29] <springle>	 halfak: actually, i lied. s2-analytics-slave switched recently; s3 was switch a while ago because the old slave was crashing regularly
[01:57:58] <springle>	 may have nothing to do with this
[01:59:42] <halfak>	 Yeah.  Not 100% on the issue.  My best guess is that I have a bug in memoization for keeping the connection.  Regardless, I don't need to hold onto the connection, so I'm just going to close it after every query.
[01:59:59] <halfak>	 Once dbstore is ready, I can do this all from one connection.
[02:00:05] <springle>	 cool
[02:00:08] <springle>	 thanks! ;)
[02:00:17] <halfak>	 Yeah.  Thanks for letting me know.
[02:01:07] <halfak>	 springle, Say, did you ever get the little wooden robot from andrewbogott?
[02:01:22] <springle>	 hehe yeah
[02:01:39] <halfak>	 great :)
[02:01:47] <springle>	 it was odd to be handed that in the middle of athens without warning
[02:02:47] <halfak>	 It's not often you get to connect with someone on the other side of the planet.
[02:02:59] <halfak>	 If you were on this content, I would have bought you a beer instead.
[02:03:36] <halfak>	 Or maybe both.  :D
[02:03:36] <springle>	 :D
[02:04:11] <springle>	 it's in cube form on my desk, when not involved in some dolls game in the kids' room
[02:05:31] <halfak>	 Alright.  I'm off.  First thing tomorrow: figure out how to re-run this script so I only gather what I don't have already.  Have a good one. :)
[02:05:39] <springle>	 gn
[13:34:33] <travis-ci>	 [travis-ci] wikimedia/mediawiki-extensions-EventLogging#208 (wmf/1.24wmf8 - 3d0a016 : Reedy): The build passed.
[13:34:33] <travis-ci>	 [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-EventLogging/commit/3d0a016a4c8e
[13:34:33] <travis-ci>	 [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-EventLogging/builds/26853826
[16:07:06] <grrrit-wm>	 (PS3) Nuria: Fix invalid users display for invalid cohort [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136434 (owner: Milimetric)
[16:12:00] <grrrit-wm>	 (CR) Nuria: [C: 2] Fix invalid users display for invalid cohort [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136434 (owner: Milimetric)
[16:12:06] <grrrit-wm>	 (Merged) jenkins-bot: Fix invalid users display for invalid cohort [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136434 (owner: Milimetric)
[16:38:31] <ori>	 ottomata: https://gerrit.wikimedia.org/r/#/c/137696/ ? (1-line typo fix)
[16:38:59] <ottomata>	 +1ed
[16:39:07] <ori>	 ottomata: danke
[16:39:13] <ottomata>	 yup
[17:13:53] <grrrit-wm>	 (PS2) Nuria: Reorganize UI for report creation [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/137306 (https://bugzilla.wikimedia.org/66017) (owner: Milimetric)
[17:15:43] <grrrit-wm>	 (CR) Nuria: "Looks much better, I have changed screen a bit so message fits better" [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/137306 (https://bugzilla.wikimedia.org/66017) (owner: Milimetric)
[17:18:22] <grrrit-wm>	 (CR) Nuria: [C: 2] Reorganize UI for report creation [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/137306 (https://bugzilla.wikimedia.org/66017) (owner: Milimetric)
[17:18:29] <grrrit-wm>	 (Merged) jenkins-bot: Reorganize UI for report creation [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/137306 (https://bugzilla.wikimedia.org/66017) (owner: Milimetric)
[17:24:16] <grrrit-wm>	 (PS5) Nuria: Add Newly Registered Users metric [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136431 (https://bugzilla.wikimedia.org/65944) (owner: Milimetric)
[17:46:56] <grrrit-wm>	 (PS6) Nuria: Add Newly Registered Users metric [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136431 (https://bugzilla.wikimedia.org/65944) (owner: Milimetric)
[17:47:17] <grrrit-wm>	 (CR) Nuria: [C: 2] Add Newly Registered Users metric [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136431 (https://bugzilla.wikimedia.org/65944) (owner: Milimetric)
[17:47:55] <grrrit-wm>	 (Merged) jenkins-bot: Add Newly Registered Users metric [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/136431 (https://bugzilla.wikimedia.org/65944) (owner: Milimetric)
[19:36:21] <grrrit-wm>	 (PS3) AndyRussG: WIP Create a cohort from campaign membership [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/126927 (owner: Awight)
[19:36:29] <grrrit-wm>	 (CR) jenkins-bot: [V: -1] WIP Create a cohort from campaign membership [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/126927 (owner: Awight)
[19:48:44] <halfak>	 Ack!  Permissions just went crazy on stat3.
[19:48:52] <halfak>	 ottomata, something going on?
[19:50:02] <halfak>	 hmm... seems to be OK now.
[19:50:20] <ottomata>	 uhh?
[19:50:22] <ottomata>	 ha, ok...?
[19:50:37] <halfak>	 Yeah.  I just had all of my open buffers notify me of a change on disk.
[19:50:59] <halfak>	 When I did ls -l, I saw userID and groupID rather than names.
[19:51:02] <halfak>	 Now I see names again.
[19:51:43] <halfak>	 Anyway.  Seems to be fine now.  -.o.-
[19:56:18] <halfak>	 Hmm... I just had a file I was editing zeroed.  I'm going to switch editors and see what happens.
[20:37:31] <tgr>	 where do I find the mapping between eventlogging types and mysql types?
[21:00:54] <tgr>	 specifically, I am trying to find out what type I should be using to log large integers
[21:01:02] <tgr>	 like a uuid