[16:21:27] <wikibugs>	 Analytics-Backlog, Analytics-Cluster: Test storage strategy 1 - Cassandra {slug} [5 pts] - https://phabricator.wikimedia.org/T101786#1382847 (JAllemandou)
[16:21:33] <kevinator>	 ottomata: thanks... just replied and added Ariel to the thread
[16:21:43] <wikibugs>	 Analytics-Backlog, Analytics-Cluster: Test PostgreSQL as a storage strategy {slug} [5 pts] - https://phabricator.wikimedia.org/T101787#1382850 (Milimetric)
[16:22:01] <wikibugs>	 Analytics-Backlog, Analytics-Cluster: Test Cassandra as a storage strategy {slug} [5 pts] - https://phabricator.wikimedia.org/T101786#1382853 (Milimetric)
[16:22:56] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Write a new Camus consumer and store the data for Event Logging {stag} [21 pts] - https://phabricator.wikimedia.org/T98784#1382855 (JAllemandou)
[16:28:34] <wikibugs>	 Analytics-Cluster, Analytics-Kanban: {wren} PV Aggregates - https://phabricator.wikimedia.org/T96314#1382868 (Milimetric)
[16:30:09] <wikibugs>	 Analytics-Backlog: Make EL On Call calendar for Q4 - https://phabricator.wikimedia.org/T95298#1382871 (kevinator) Open>Resolved the analytics calendar shows who's on-call
[16:30:38] <wikibugs>	 Analytics-Backlog: Change mediawiki-storage api queries to adapt to the api changes [? pts] {crow} - https://phabricator.wikimedia.org/T101539#1382875 (Milimetric) p:Triage>High
[16:33:28] <wikibugs>	 Analytics-Backlog, Analytics-Cluster: Epic: qchris transition - https://phabricator.wikimedia.org/T86135#1382883 (kevinator) Let's look at this again after wikimania when we have time to "clean house"
[16:41:36] <kevinator>	 joal: ut?
[16:41:53] <joal>	 I am :(
[16:42:09] <kevinator>	 I'm in the hangout for our meeting.... not the batcave
[16:42:18] <joal>	 same for e :(
[16:42:24] <joal>	 alternate one probably
[16:42:40] <joal>	 hm
[16:42:57] <kevinator>	 do you want to try the batcave... it was working earlier
[16:43:07] <joal>	 sent you an invite
[16:43:21] <joal>	 Do you have it ?
[16:43:32] <kevinator>	 no email?  through hangout?
[16:43:37] <joal>	 hangout
[16:43:38] <kevinator>	 nothing is ringing
[16:43:43] <joal>	 Let's batcave
[16:44:04] <joal>	 I'm in
[16:44:56] <joal>	 kevinator: :(
[16:45:03] <kevinator>	 omw
[17:08:12] <Ironholds>	 where does vitalsigns live again?
[17:19:12] <milimetric>	 Ironholds: https://vital-signs.wmflabs.org/
[17:19:39] <milimetric>	 ottomata: ok, I wanna look at beta now, I'm gonna check out the puppet version, let me know if you found anything
[17:20:32] <milimetric>	 uh... where's puppet on there?
[17:20:32] <Ironholds>	 milimetric, ta!
[17:21:00] <Ironholds>	 milimetric, how do I get to non-pageviews?
[17:21:18] <YuviPanda>	 milimetric: deployment-salt.eqiad.wmflabs /var/lib/git/operations/puppet
[17:21:18] <milimetric>	 Ironholds: the metrics are filtered to ones that run somewhat reliably
[17:21:25] <milimetric>	 but basically just click Add Metric
[17:21:30] <milimetric>	 YuviPanda: thx!
[17:21:43] <Ironholds>	 aha
[17:21:49] <milimetric>	 YuviPanda: no, that's not on there, must not be self-hosted
[17:21:53] <Ironholds>	 I can only see 3 metrics. Ah well
[17:21:58] <milimetric>	 Ironholds: yeah, that's right
[17:22:07] <Ironholds>	 ...ah.
[17:22:10] <YuviPanda>	 milimetric: ? on beta? everything points to the puppet repo on deployment-salt.eqiad.wmflabs instance
[17:22:24] <milimetric>	 Ironholds: https://metrics.wmflabs.org/static/public/dash/ has all of them but you can see how gappy some of them are
[17:22:43] <milimetric>	 YuviPanda: my bad, misunderstood, thx
[17:22:57] <YuviPanda>	 milimetric: np
[17:24:41] <milimetric>	 ottomata: I guess I need to wait for when you come back, I don't understand how to play with just the EL puppet setup here
[17:25:42] <ottomata>	 milimetric: i'm here
[17:25:53] <ottomata>	 just being bugge dabout varnishkafka error alerts
[17:25:57] <ottomata>	 trying to make opsens happy w me :p
[17:26:08] <ottomata>	 what's up?
[17:33:06] <milimetric>	 ottomata: so I don't see how puppet is set up in beta cluster
[17:33:12] <milimetric>	 for EL
[17:33:36] <milimetric>	 panda was saying it's on deployment-salt
[17:33:44] <milimetric>	 so is that what you reverted yesterday?
[17:34:12] <ottomata>	 milimetric: deployment-salt is the puppet master i think
[17:34:13] <ottomata>	 yes
[17:34:17] <ottomata>	 no, i reverted a code deploy
[17:34:31] <ottomata>	 but, soething was messed up with puppet not related to my deploy
[17:34:53] <ottomata>	 the variable for the log directory is not rendering properly in the log consumer template configs
[17:36:22] <milimetric>	 yeah, I saw /etc/eventlogging.d/*/ configs were weird, had like "file:///all-events.log" and stuff
[17:36:39] <milimetric>	 so... if puppet is up to date, doesn't this mean this would be happening in prod too?
[17:39:24] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Update Schema Talk pages - https://phabricator.wikimedia.org/T103133#1383127 (kevinator) NEW
[17:41:07] <ottomata>	 dunno milimetric um, sorry, will try to be with you asap, want to make this commit
[17:41:10] <wikibugs>	 Analytics-Kanban: enforce policy for each Schema [8 pts] {tick} - https://phabricator.wikimedia.org/T102518#1383150 (kevinator)
[17:41:12] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Update Schema Talk pages - https://phabricator.wikimedia.org/T103133#1383149 (kevinator)
[17:41:16] <ottomata>	 check to see what it is being set to in the beta puppetmaster
[17:41:20] <ottomata>	 on deployment-salt
[17:41:23] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Update Schema Talk pages {tick} - https://phabricator.wikimedia.org/T103133#1383152 (kevinator)
[17:41:26] <ottomata>	 in /var/lib/git/operations/puppet
[17:41:35] <wikibugs>	 Analytics-Kanban, Analytics-Wikimetrics, Community-Wikimetrics, Patch-For-Review: Give the option of using the same parameters for all reports for a given cohort {dove} [21 pts] - https://phabricator.wikimedia.org/T74117#1383157 (Milimetric) >>! In T74117#1371105, @Capt_Swing wrote: > I don't quite...
[17:41:46] <kevinator>	 mforns: regarding your question on the eatherpad:
[17:41:55] <mforns>	 kevinator, yep
[17:42:03] <kevinator>	 I think you can track both... what it says in the talk page and what it should say
[17:42:34] <mforns>	 kevinator, ok, for each field that can have inconsistencies? like owner, or purging status?
[17:43:35] <mforns>	 kevinator, makes sense, this way we can also see in the spreadsheet-dashboard if the talk page is up-to-date
[17:45:34] <kevinator>	 mforns: I just had a new throught...
[17:45:37] <kevinator>	 er thought
[17:45:43] <milimetric>	 https://www.irccloud.com/pastebin/CpqrZ6aU/
[17:46:10] <kevinator>	 mforns: do we really need to capture both what the talk pages say and what it should be?
[17:46:27] <kevinator>	 maybe we only need to track what it should be.
[17:46:52] <kevinator>	 We know there are mistakes in the talk pages.  We'll fix that when we get to task T103133
[17:47:16] <mforns>	 kevinator, the only advantage I see is to know if the talk page is up-to-date, but we can add a column "talk page updated" maybe
[17:47:25] <mforns>	 instead of having lots of columns
[17:47:34] <kevinator>	 I like that!
[17:47:46] <mforns>	 this will make the spreadsheet editing easier and faster
[17:47:57] <mforns>	 ok
[17:47:58] <ottomata>	 yeah milimetric, check to see what output is being set to in puppet on deployment-salt's repo
[17:48:03] <mforns>	 let's do that then
[17:48:40] <mforns>	 kevinator, can I remove line 16 from the definition of done?
[17:48:51] <kevinator>	 yes
[17:50:09] <mforns>	 kevinator, ok will do, thanks
[17:50:45] <kevinator>	 what's the URL for hte pages... I can't find it.  Can you post it in the etherpad?
[17:51:19] <milimetric>	 ottomata: it's doing ${log_dir}/all-events.log and before that $log_dir = $::eventlogging::log_dir so I guess $::eventlogging::log_dir is empty, where's that coming from?
[17:52:03] <ottomata>	 that is from the eventlogging module it hink
[17:52:12] <ottomata>	 in modules/eventlogging/manifests/init.pp
[17:52:13] <ottomata>	 likley
[17:57:12] <wikibugs>	 Analytics-Kanban: Gather information on all the schemas [13 pts] {tick} - https://phabricator.wikimedia.org/T102515#1383237 (kevinator)
[18:02:32] <wikibugs>	 Analytics-Kanban: Gather information on all the schemas [13 pts] {tick} - https://phabricator.wikimedia.org/T102515#1383262 (kevinator)
[18:03:13] <ottomata>	 milimetric: how goes?
[18:03:49] <milimetric>	 looking in init.pp now
[18:05:02] <milimetric>	 ottomata weird: $log_dir = '/srv/log/eventlogging'
[18:05:03] <wikibugs>	 Analytics-Kanban: Gather information on all the schemas [13 pts] {tick} - https://phabricator.wikimedia.org/T102515#1383273 (kevinator) I refactored the title and description a bit.  Initially we thought we could use the talk pages to track our progress in this project.. but it's apparent that using a spreads...
[18:05:19] <ottomata>	 yeah, hm, i'm trying a notify {
[18:05:28] <ottomata>	    notify { "eventlogging log_dir is '${log_dir}, and from the module: '${::eventlogging::log_dir}'": }
[18:05:47] <milimetric>	 k
[18:09:18] <wikibugs>	 Analytics-Backlog: Stand up piwik in a permanent and privacy-sensitive way - https://phabricator.wikimedia.org/T98058#1383299 (Milimetric) When you say "at our scale" are you thinking english wikipedia sending pageview data to it?  The use cases are much smaller than that.  Some WMF chapters, the annual repo...
[18:09:50] <ottomata>	 HM, weird, milimetric i think i fixed it, but i don't know why this fixes it, commit coming atcha.
[18:10:15] <ottomata>	 i think it has somethign to do with how the eventlogging moodule class is included, and the fact the the role class has a similar name as the module, and stupid puppet scoping
[18:10:26] <wikibugs>	 Analytics-Kanban: Android debug builds event logging not recorded {oryx} [3 pts] - https://phabricator.wikimedia.org/T102881#1383301 (Milimetric) yep, that's what we see too, we're actively debugging this now, we don't know what's going on.
[18:10:29] <wikibugs>	 Analytics-Backlog: Stand up piwik in a permanent and privacy-sensitive way - https://phabricator.wikimedia.org/T98058#1383302 (yuvipanda) Yes, even the mobile site would be way too big.
[18:11:17] <wikibugs>	 Analytics-Backlog: Stand up piwik in a permanent and privacy-sensitive way - https://phabricator.wikimedia.org/T98058#1383308 (yuvipanda) Either way, it needs a much bigger discussion with ops as well and I don't think I can be the point person for that.
[18:11:23] <milimetric>	 ottomata: it's weird that it wasn't broken before though, those names were always the same right?
[18:11:30] <YuviPanda>	 milimetric: ^ this basically became too big for me, sorry
[18:11:35] <mforns>	 madhuvishy, when filling in the "contact assumed" column in tick's spreadsheet, did you contact any of the owners to ask? I mean, are some of those already confirmed?
[18:11:53] <milimetric>	 YuviPanda: EL is in prod and it literally chokes after like 600 events per second or so
[18:12:04] <milimetric>	 I don't really see the scale issue here...
[18:12:17] <milimetric>	 our existing tool that's already in prod seems worse
[18:12:26] <YuviPanda>	 milimetric: sure, but this is a conversation I'm not capable of having, I think :) I was interested in setting it up for a  very small set of use cases, and it just now became way too big and now I'm uninterested...
[18:12:33] <ottomata>	 milimetric: yes i agree it is weird
[18:12:39] <milimetric>	 YuviPanda: ok :) fair enough
[18:12:40] <ottomata>	 milimetric: https://gerrit.wikimedia.org/r/#/c/219412/
[18:13:05] <ottomata>	 the one that solves it is the first one
[18:13:28] <ottomata>	 going to merge
[18:13:33] <YuviPanda>	 milimetric: :) it also made it move from something that'll just be on labs (a weeklong project) to something that's much bigger as well
[18:13:44] <madhuvishy>	 mforns: no.. Kevin said he would  help vet it though
[18:13:45] <milimetric>	 ottomata: ok, it just looks like puppet voodoo to me :)
[18:14:44] <madhuvishy>	 mforns: I put that based on schema page history and some other assumptions.
[18:14:52] <milimetric>	 YuviPanda: I don't see it the same way, but it's possible I'm missing some nuance.  I basically see - puppet code makes piwik run in labs and prod, with a very normal puppet setup.  But there are way too many use cases for us to ignore this, so I'll have to work on it I guess
[18:15:27] <YuviPanda>	 milimetric: I think you'll get more objections from other opsen for including another big third party PHP application in prod, and it'll basically drag out.
[18:16:08] <YuviPanda>	 milimetric: either way, it doesn't feel fun to me anymore, and I was offering to help because I found it an interesting project of an interesting size...
[18:16:43] <milimetric>	 no, I'm fully behind you doing this only if you find it fun
[18:17:24] <milimetric>	 on an amusing side note, I think "Yuvi having fun" ~ 1 / "Dan having fun"
[18:18:08] <milimetric>	 I'll try to raise this as a top priority for us and maybe use it as an opportunity to learn puppet better
[18:18:18] <YuviPanda>	 (i just referred the ticket to the other opsen for general opinions)
[18:18:29] <milimetric>	 cool, thx
[18:18:51] <YuviPanda>	 milimetric: yw! and +1 to milimetric learning puppet
[18:20:32] <milimetric>	 I'm going to step out for 30 minutes
[18:23:29] <mforns>	 madhuvishy, ok! thanks! I will start reaching out to them to ask then
[18:23:45] <mforns>	 madhuvishy, what do you think of marking green the ones that are confirmed?
[18:31:27] <grrrit-wm>	 (PS1) Ricordisamoa: Load jQuery and Bootstrap from cdnjs [analytics/quarry/web] - https://gerrit.wikimedia.org/r/219417
[18:33:05] <grrrit-wm>	 (CR) Yuvipanda: [C: 2] "Thank you!" [analytics/quarry/web] - https://gerrit.wikimedia.org/r/219417 (owner: Ricordisamoa)
[18:33:25] <grrrit-wm>	 (Merged) jenkins-bot: Load jQuery and Bootstrap from cdnjs [analytics/quarry/web] - https://gerrit.wikimedia.org/r/219417 (owner: Ricordisamoa)
[18:34:42] <madhuvishy>	 mforns: cool that works. We'll be adding team info too right?
[18:43:27] <mforns>	 madhuvishy, right!
[18:43:28] <grrrit-wm>	 (PS1) Ricordisamoa: Load CodeMirror from cdnjs, using minified files [analytics/quarry/web] - https://gerrit.wikimedia.org/r/219422
[18:44:51] <milimetric>	 back
[18:47:50] <milimetric>	 ottomata: I see the logs are being written again
[18:48:18] <ottomata>	 cool!
[18:53:19] <wikibugs>	 Analytics-Kanban: Android debug builds event logging not recorded {oryx} [3 pts] - https://phabricator.wikimedia.org/T102881#1383543 (Milimetric) @bearND & @Niedzielski - how do you generate events exactly?  Andrew got the logs working again and I'd like to troubleshoot further.
[18:53:58] <grrrit-wm>	 (CR) Ricordisamoa: "Other libraries:" [analytics/quarry/web] - https://gerrit.wikimedia.org/r/219422 (owner: Ricordisamoa)
[19:07:34] <joal>	 Guys, have a good weekend !
[19:08:41] <milimetric>	 have a good one joal
[19:09:46] <ottomata>	 laterrrs!
[19:10:31] <mforns>	 bye!
[19:34:24] <madhuvishy>	 mforns: i'll work through the list from the bottom?
[19:34:51] <madhuvishy>	 hmmm, wonder if that's a good ide
[19:35:47] <madhuvishy>	 mforns: i sorted my sheet by contact. so its easy to send emails/ping
[19:36:22] <mforns>	 madhuvishy, I had created a copy to another tab
[19:36:25] <mforns>	 and sorted that one
[19:36:35] <madhuvishy>	 mforns: oh cool, sorry i dint notice
[19:37:02] <mforns>	 madhuvishy, but we can work on the main one if you want, and resort it again after finished?
[19:37:15] <madhuvishy>	 mforns: no this is cool :)
[19:37:22] <mforns>	 ok :]
[19:37:53] <mforns>	 madhuvishy, so yea! I started at the top of the spreadsheet
[19:38:14] <mforns>	 I'm writing to adam baso
[19:38:35] <madhuvishy>	 mforns: cool.
[19:38:45] <mforns>	 madhuvishy, do you want me to cc you if I write emails?
[19:38:51] <mforns>	 for that
[19:42:24] <milimetric>	 ottomata: something really strange is up
[19:42:30] <milimetric>	 I'm sending events like this:
[19:42:33] <milimetric>	 https://meta.wikimedia.org/beacon/event?%7B%22event%22%3A%7B%22version%22%3A%221.5%22%2C%22eventId%22%3A189837%2C%22notificationType%22%3A%22flow-mention%22%2C%22notificationGroup%22%3A%22other%22%2C%22sender%22%3A%223902%22%2C%22recipientUserId%22%3A3867%2C%22recipientEditCount%22%3A502%2C%22deliveryMethod%22%3A%22web%22%7D%2C%22revision%22%3A-1%2C%22schema
[19:42:33] <milimetric>	 %22%3A%22Echo%22%2C%22webHost%22%3A%22meta.wikimedia.org%22%2C%22wiki%22%3A%22metawiki%22%7D;
[19:42:34] <ottomata>	 good timing, my things are looking better atm!
[19:42:38] <milimetric>	 cool :)
[19:42:53] <milimetric>	 so, I logged what URL sendBeacon is seeing on the JS side
[19:42:55] <milimetric>	 and its that ^
[19:43:16] <ottomata>	 ok
[19:43:25] <milimetric>	 but I don't see any activity anywhere in the logs
[19:43:35] <milimetric>	 not /var/log/upstart/event*log, not /var/log/eventlogging/*
[19:43:55] <milimetric>	 yet I get a 204 as expected
[19:44:06] <milimetric>	 is something not working with the new /beacon/event endpoint in beta?
[19:44:14] <ottomata>	 oh in beta?
[19:44:20] <ottomata>	 meta.wikimedia.org is beta?
[19:44:31] <milimetric>	 oh wait... lol
[19:44:52] <ottomata>	 hehe
[19:44:54] <milimetric>	 no!... uh... but that's what mw.eventLog logged as the URL
[19:45:12] <ottomata>	 i don't htink meta.wm answers to /beacon/event, does it?
[19:45:20] <ottomata>	 oh you got a 204?
[19:45:22] <milimetric>	 it tells me 204
[19:45:27] <ottomata>	 maybe it has changed bc bits move
[19:45:33] <milimetric>	 yeah, uh... this is bad... doesn't this mean all beta events are ending up in prod right now?
[19:45:35] <milimetric>	 yeah... it does
[19:46:02] <ottomata>	 dunno where mw.eventLog is configured
[19:46:31] <milimetric>	 not sure either, tracking now
[19:46:47] <milimetric>	 	baseUrl = mw.config.get( 'wgEventLoggingBaseUri' );
[19:47:56] <madhuvishy>	 mforns: yeah, that'd be great. I'll do the same - unless i'm pinging them on irc/asking at office
[19:48:06] <mforns>	 madhuvishy, ok
[19:50:20] <milimetric>	 uh, never mind :) that was me being dumb ottomata
[19:50:29] <milimetric>	 so confusing debugging this
[19:53:06] <ottomata>	 haha, ok cool :)
[19:53:11] <ottomata>	 milimetric: will you be back on east coast monday?
[19:53:25] <milimetric>	 yes
[19:53:29] <milimetric>	 i'm here now
[19:53:50] <ottomata>	 oh ok, cool, so you'll be up our morning time and we can work on eventlogging deploy
[19:55:34] <milimetric>	 ottomata: yeah, but this issue is kinda gnawing at me
[19:55:41] <milimetric>	 I still can't get events through to the logs
[19:55:44] <milimetric>	 /en.wikipedia.beta.wmflabs.org/beacon/event?%7B%22event%22%3A%7B%22version%22%3A%221.5%22%2C%22eventId%22%3A189837%2C%22notificationType%22%3A%22flow-mention%22%2C%22notificationGroup%22%3A%22other%22%2C%22sender%22%3A%223902%22%2C%22recipientUserId%22%3A3867%2C%22recipientEditCount%22%3A502%2C%22deliveryMethod%22%3A%22web%22%7D%2C%22revision%22%3A7731316%2C
[19:55:45] <milimetric>	 %22schema%22%3A%22Echo%22%2C%22webHost%22%3A%22en.wikipedia.beta.wmflabs.org%22%2C%22wiki%22%3A%22enwiki%22%7D;
[19:56:09] <ottomata>	 ok, hm
[19:56:11] <ottomata>	 let's seeeeeeee
[19:56:46] <ottomata>	 milimetric: do you know what labs host serves that?
[19:56:58] <ottomata>	 deployment-cache-bits01? deployment-cache-text02?
[19:56:58] <milimetric>	 mmmm no
[19:57:15] <milimetric>	 i'd guess the text?  'cause it's being served from the sites themselves now right?
[19:57:23] <ottomata>	 ya hm. probably
[19:57:54] <ottomata>	 hm i can't log in there!
[19:57:57] <milimetric>	 i've got a meeting now :( but let me know if you can figure out what's going on
[19:58:51] <ottomata>	 PSSH
[19:58:54] <ottomata>	 i can't log in there! why!?
[20:00:09] <milimetric>	 did you try ssh instead of pssh? :P
[20:10:18] <ottomata>	 haha
[20:10:46] <ottomata>	 milimetric: can you log in there?
[20:12:51] <milimetric>	 no
[20:17:27] <ottomata>	 Yuvi   Panda: tells me it is beacuse pupet is dead there
[20:17:28] <ottomata>	 or something
[20:17:36] <ottomata>	 can't help much if i can't see what varnish is configured to do there
[20:21:31] <wikibugs>	 Analytics-Kanban: Android debug builds event logging not recorded {oryx} [3 pts] - https://phabricator.wikimedia.org/T102881#1383861 (Niedzielski) @Milimetric, thanks for the update! I [[ https://git.wikimedia.org/blob/apps%2Fandroid%2Fwikipedia/a1658435a6dbf8242e3879da963617557df35c17/wikipedia%2Fsrc%2Fmain%...
[20:22:01] <milimetric>	 thanks ottomata, looks like we're stuck - i'll update after my meeting
[20:28:39] <ottomata>	 milimetric: it looks to me like beta EL is broken due to https://phabricator.wikimedia.org/T102570 + recent outage nastiness + who knows what else
[20:28:44] <ottomata>	 but, i can't log into verify :/
[20:29:20] <milimetric>	 k, makes sense
[20:29:38] <milimetric>	 well... but isn't the beta cluster separate from "labs"?
[20:30:02] <ottomata>	 iunnnoooo
[20:31:38] <ottomata>	 oo i'm in
[20:39:08] <ottomata>	 milimetric:
[20:39:21] <ottomata>	 otto@deployment-cache-text02:/etc/varnish$ sudo service varnishncsa-eventlogging status
[20:39:21] <ottomata>	  * varnishncsa-eventlogging is not running
[20:39:21] <ottomata>	 otto@deployment-cache-text02:/etc/varnish$ sudo service varnishncsa-eventlogging start
[20:39:21] <ottomata>	  * Starting HTTP accelerator log deamon varnishncsa-eventlogging                                                                                                                                                                                                      [ OK ]
[20:39:21] <ottomata>	 otto@deployment-cache-text02:/etc/varnish$ sudo service varnishncsa-eventlogging status
[20:39:21] <ottomata>	  * varnishncsa-eventlogging is running
[20:39:27] <ottomata>	 varnisncsa was not running
[20:39:36] <ottomata>	 so it wasn't sending anything to eventlogging
[20:39:39] <ottomata>	 on udp
[20:41:05] <ottomata>	 probalby it died for some reason (NFS stuff too?) and puppet never ran so it never started it
[20:45:09] <milimetric>	 ottomata: yay! I'm seeing client side events
[20:45:12] <milimetric>	 you da MAN
[20:45:17] <milimetric>	 I'll update the bug after the meeting
[20:47:48] <wikibugs>	 Analytics-Cluster, operations: Build Kafka 0.8.1.1 package for Jessie  and upgrade Brokers to Jessie. - https://phabricator.wikimedia.org/T98161#1383923 (Ottomata)
[20:47:51] <wikibugs>	 Analytics-Cluster, operations: Create jmxtrans Jessie package - https://phabricator.wikimedia.org/T103106#1383921 (Ottomata) Open>Resolved Imported v250 from: http://central.maven.org/maven2/org/jmxtrans/jmxtrans/250/
[20:50:12] <mforns>	 madhuvishy, got lost there
[20:50:12] <madhuvishy>	 mforns: :D
[20:50:33] <madhuvishy>	 nah its okay :) great, I'll add those when I talk to James
[20:59:48] <wikibugs>	 Analytics-Kanban: Android debug builds event logging not recorded {oryx} [3 pts] - https://phabricator.wikimedia.org/T102881#1383945 (Milimetric) All - Andrew figured this out.  varnisncsa was not running in beta, so none of the client events were making it through.  He restarted it and I definitely see event...
[21:01:37] <milimetric>	 bye guys - have a nice weekend
[21:03:39] <ottomata>	 me to bYYYe
[21:22:20] <madhuvishy>	 mforns: are you writing to Prateek?
[21:22:42] <mforns>	 madhuvishy, not yet
[21:22:50] <mforns>	 I was looking at that row though
[21:23:09] <madhuvishy>	 mforns: okay, i almost finished writing email so will send :)
[21:23:20] <mforns>	 madhuvishy, ok then go ahead
[21:23:53] <mforns>	 madhuvishy, I'll write to Deskana
[21:24:04] <madhuvishy>	 mforns: okay perfect
[21:33:09] <wikibugs>	 Analytics-Kanban: Android debug builds event logging not recorded {oryx} [3 pts] - https://phabricator.wikimedia.org/T102881#1384048 (Niedzielski) @Milimetric, great news! I confirm that some events appear to be logging properly! I'm not sure this is yet working for new table events. I added a new event revis...
[21:34:45] <madhuvishy>	 mforns: so are we almost done?
[21:35:08] <mforns>	 madhuvishy, I think so! I'm looking at the last 3 schemas
[21:35:20] <mforns>	 I think the last 2 are junk
[21:35:47] <mforns>	 madhuvishy, they were probably tests, because they last 1 or 2 days
[21:36:03] <madhuvishy>	 mforns: ah okay - we should just purge them then?
[21:36:25] <mforns>	 madhuvishy, I would mark them as with no owner / not used / just tests
[21:36:29] <mforns>	 I dunno
[21:37:07] <mforns>	 madhuvishy, but third-last lasted for more than 1 year, and I think it belongs to Matt Flaschen
[21:37:22] <mforns>	 because it has a similar name to other schemas of his
[21:37:29] <madhuvishy>	 mforns: aah, let me add that to email I sent to Matt
[21:37:37] <mforns>	 OK
[21:39:25] <mforns>	 madhuvishy, you discovered the owner of BannerImpression?
[21:39:49] <madhuvishy>	 No, that, and campaigns is pending
[21:40:26] <madhuvishy>	 mforns: https://meta.wikimedia.org/wiki/Schema:Campaigns_obsolete
[21:41:23] <madhuvishy>	 mforns: the talk page says "We reduced the scope to just setting a cookie, so no need to log anything. We'll just log the campaign value in Schema:ServerSideAccountCreation.", and it only has 8 rows
[21:41:51] <madhuvishy>	 so I'm not sure if it's worth reaching out
[21:42:20] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Can Search up sampling 5%? - https://phabricator.wikimedia.org/T103186#1384061 (kevinator) NEW
[21:42:34] <mforns>	 madhuvishy, aha. yea maybe we can mark it as no owner?
[21:42:51] <madhuvishy>	 mforns: the talk page says spagewmf is the owner
[21:42:57] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Can Search up sampling to 5%? - https://phabricator.wikimedia.org/T103186#1384073 (kevinator)
[21:43:03] <mforns>	 madhuvishy, yes, OK
[21:43:23] <madhuvishy>	 mforns: i just think we can delete it anyway.
[21:43:24] <mforns>	 madhuvishy, maybe we can add a value to status column for those cases: abandoned (or sth like this)
[21:43:32] <madhuvishy>	 it's clear they are not using it
[21:43:36] <mforns>	 yes
[21:44:02] <madhuvishy>	 mforns: yeah - i just feel like difference between inactive and abandoned will become blurry
[21:44:12] <mforns>	 yes...
[21:44:31] <madhuvishy>	 hmmm
[21:44:48] <mforns>	 in this case, we can keep spage as the owner
[21:45:00] <mforns>	 because it was him at the time the schema was active...
[21:45:13] <madhuvishy>	 mforns: but i think we should add a field to our sheet to say - delete schema and everything in it
[21:45:34] <madhuvishy>	 mforns: yup
[21:45:54] <mforns>	 madhuvishy, yes, or simply set "approved for auto-purging" = true
[21:46:15] <madhuvishy>	 mforns: hmmm not the same if we are deleting the table itself no?
[21:46:15] <mforns>	 this will mean -> deletion of old data (and probably the table, too)
[21:46:38] <mforns>	 madhuvishy, yes, it depends on what we do with empty tables
[21:46:43] <madhuvishy>	 mforns: hmmm, i thought they were different things.
[21:47:17] <mforns>	 mmm
[21:47:29] <madhuvishy>	 some cases - we may get requests saying, keep the table - we may use in future. auto purging may empty it - but schema is still in use
[21:47:42] <mforns>	 madhuvishy, yes makes sense
[21:47:52] <madhuvishy>	 some cases - it is, we dont need it anymore, purge everything
[21:49:01] <madhuvishy>	 mforns: hmm we need to work on defining the states it could be in
[21:49:01] <mforns>	 madhuvishy, maybe this could be the difference between inactive(no events for now) and terminated(the table was deleted by request of its owner)
[21:49:12] <madhuvishy>	 mforns: yeah i was thinking the same
[21:49:16] <mforns>	 ok
[21:49:24] <madhuvishy>	 so confusing :)
[21:49:28] <mforns>	 hehe
[21:50:05] <madhuvishy>	 mforns: cool. so last one - bannerimpression
[21:50:06] <mforns>	 madhuvishy, ok so only 1 missing?
[21:50:09] <mforns>	 ok
[21:50:24] <madhuvishy>	 i'll write to anne gomez
[21:50:55] <mforns>	 madhuvishy, how did you know it was anne's
[21:51:07] <mforns>	 ok
[21:51:17] <madhuvishy>	 mforns: actually, i dont know
[21:51:35] <madhuvishy>	 i dint put that in. May be Kevin?
[21:51:39] <mforns>	 madhuvishy, did you put that email there?
[21:51:42] <mforns>	 oh1
[21:51:46] <mforns>	 ok, makes sense
[21:52:31] <mforns>	 fundraising -- banners
[21:54:07] <madhuvishy>	 mforns: yeah
[21:54:37] <mforns>	 madhuvishy, I was going to add s page as owner of campaigns
[21:54:41] <mforns>	 is it ok?
[21:54:54] <madhuvishy>	 mforns: i think i just did :)
[21:55:06] <madhuvishy>	 ya cool
[22:00:54] <mforns>	 madhuvishy, was thinking what team to asign to page's schema?
[22:01:22] <madhuvishy>	 mforns: hmmm not sure. I'll ask Kevin/ spage if I see him at the office
[22:01:39] <mforns>	 the campaign extension the schema refers to is written by several people, legoktm, ori, ...
[22:02:36] <mforns>	 ok
[22:03:12] <mforns>	 madhuvishy, should I write to agomez?
[22:03:19] <mforns>	 oh, sorry,
[22:04:25] <madhuvishy>	 mforns: I just spoke to Katie Horn in fundraising tech. She pointed me to couple other engineers in their team who'd know. I'm writing to them. There are only 12 rows in this schema - i'm guessing it can probably go - but emailing.
[22:05:09] <mforns>	 madhuvishy, oh, sorry I had momentarily forgotten what was its context, and that you said you were going to solve that
[22:05:47] <mforns>	 cool!
[22:06:39] <mforns>	 madhuvishy, it's getting too late for me here... going to sign off
[22:07:17] <madhuvishy>	 mforns: alright :) I sent the email. So all done I guess :) Have a great weekend!
[22:07:25] <mforns>	 have a nice weekend! see you on monday
[22:07:31] <mforns>	 bye :]
[22:42:34] <wikibugs>	 Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Bug in hostNormalizer function {hawk} - https://phabricator.wikimedia.org/T102233#1384261 (kevinator) Open>Resolved
[22:43:02] <wikibugs>	 Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Ooziefy and parquetize pageview intermediate aggregation using refined table fields {wren} [13 pts] - https://phabricator.wikimedia.org/T99931#1384262 (kevinator) Open>Resolved
[22:43:43] <wikibugs>	 Analytics-Kanban, Patch-For-Review: Add piwik instrumentation back to visual editor dashboard {lion} [5 pts] - https://phabricator.wikimedia.org/T102251#1384265 (kevinator) Open>Resolved
[22:44:22] <wikibugs>	 Analytics-Kanban: vital-signs.wmflabs.org has a race condition that causes metrics to not load {crow} [5 pts] - https://phabricator.wikimedia.org/T102630#1384268 (kevinator) Open>Resolved
[22:45:37] <wikibugs>	 Analytics-Kanban, Patch-For-Review: Add cache headers to the datasets.wikimedia.org/limn-public-data/metrics folder {lion} [5 pts] - https://phabricator.wikimedia.org/T101125#1384272 (kevinator) Open>Resolved
[22:46:26] <wikibugs>	 Analytics-Kanban: [Spike] Research best methods to track and update EL schema talk pages {tick} - https://phabricator.wikimedia.org/T102712#1384274 (kevinator) Open>Resolved
[22:47:01] <wikibugs>	 Analytics-Kanban, Patch-For-Review: Failure Types by User Type {lion} - https://phabricator.wikimedia.org/T91123#1384276 (kevinator) Open>Resolved
[22:48:58] <wikibugs>	 Analytics-Backlog, Analytics-EventLogging: Can Search up sampling to 5%? {oryx} - https://phabricator.wikimedia.org/T103186#1384279 (kevinator)
[22:58:23] <wikibugs>	 Analytics-General-or-Unknown, Community-Advocacy, Community-Liaison, Wikimedia-Extension-setup, Wikipedia-iOS-App: enable Piwik on ru.wikimedia.org - https://phabricator.wikimedia.org/T91963#1384287 (Fjalapeno)
[23:07:01] <wikibugs>	 Analytics, Engineering-Community, Research-and-Data, ECT-July-2015: Metrics about the use of the Wikimedia web APIs - https://phabricator.wikimedia.org/T102079#1384304 (GWicke) > This distinction might be trickier for our RESTBase APIs, since they can be used internally and externally  The same is t...