[00:20:58] <milimetric>	 madhuvishy: separate vital signs role makes lots of sense, but it's ok to do it one step at a time
[00:21:09] <madhuvishy>	 milimetric: yeah okay
[00:21:10] <milimetric>	 this is just a migration, we can refactor in the next step
[00:21:15] <madhuvishy>	 i fixed everything else
[00:21:22] <milimetric>	 and it's ok to call this not done until that is all clean
[00:21:27] <milimetric>	 sweet
[00:21:58] <madhuvishy>	 the caching headers are automatically set to do the right thing - there are minor differences - and we can look at it more detail on monday
[00:23:39] <milimetric>	 madhuvishy: interesting, directory browsing doesn't work anymore, but the files still serve: https://metrics.wmflabs.org/static/public/datafiles/
[00:23:54] <milimetric>	 it's ok, I don't mind, but I'm curious we can talk monday
[00:24:11] <milimetric>	 are you going to clone those pageviews things now or should i do it manually?
[00:24:21] <milimetric>	 after that I wanna make sure the scheduler and all other things are working properly
[00:24:38] <madhuvishy>	 milimetric: I'll do it
[00:24:48] <milimetric>	 k
[00:24:55] <madhuvishy>	 is vital signs on limn1?
[00:25:08] <madhuvishy>	 what would it take to get it out of there?
[00:25:31] <milimetric>	 just the dashiki puppet thing you were working on
[00:25:40] <milimetric>	 that was almost done, right?
[00:25:48] <madhuvishy>	 milimetric: that was merged
[00:25:59] <milimetric>	 oh, then we can get it out anytime really
[00:26:09] <madhuvishy>	 milimetric: then i can do it now
[00:26:11] <madhuvishy>	 might as well
[00:26:17] <milimetric>	 well, wait, let's get the other stuff working
[00:26:29] <milimetric>	 the Pageviews, LegacyPageviews, etc.
[00:26:47] <madhuvishy>	 milimetric: that is used by vitalsigns right?
[00:27:02] <madhuvishy>	 i will git clone them in the new instance
[00:27:10] <milimetric>	 ?
[00:27:22] <madhuvishy>	 ummm
[00:27:36] <madhuvishy>	 so vitalsigns is looking for the files in these directories right?
[00:27:37] <milimetric>	 there's some misunderstanding there, vital signs has nothing to do with the data for those metrics
[00:27:43] <madhuvishy>	 oh
[00:27:51] <madhuvishy>	 okayy
[00:27:57] <milimetric>	 vital signs is looking for them on metrics.wmflabs.org
[00:27:58] <madhuvishy>	 sorry then i dint understand correctly
[00:28:04] <milimetric>	 so moving vital signs won't help
[00:28:21] <madhuvishy>	 ummm, what if it dint look for them on metrics.wmflabs.org?
[00:28:40] <madhuvishy>	 and looked for them in a more relevant place.
[00:28:52] <milimetric>	 then we'd have to set up some other server where we symlink those.
[00:29:13] <milimetric>	 i agreed with what you said before, to make a separate role for the data for those metrics
[00:29:35] <milimetric>	 like some static-file-server thing that could be configured to serve different repos
[00:29:45] <elukey>	 if anybody is interested: https://wikitech.wikimedia.org/wiki/File:Infrastructure_overview.png (still veeeeery high level and probably imprecise, I need to merge stuff from https://wikitech.wikimedia.org/wiki/Clusters)
[00:30:15] <elukey>	 no more excuses to slack now
[00:30:16] <elukey>	 :)
[00:30:21] <milimetric>	 madhuvishy: but i feel like that kind of thing is better talked about Monday in tasking than Friday when we're trying to fix things :)
[00:31:18] <milimetric>	 madhuvishy: but if this bothers you so much, we can just git clone those things manually in wikimetrics-01
[00:31:22] <milimetric>	 I have no problem with that
[00:31:27] <madhuvishy>	 milimetric: okay, so the only thing that's broken now is vital signs right? because of the 2 folders?
[00:31:32] <madhuvishy>	 nooo
[00:31:43] <milimetric>	 it's not broken, it's just not showing those two metrics, it's showing the rest
[00:31:48] <madhuvishy>	 okay
[00:32:05] <madhuvishy>	 so you can verify everything else now?
[00:32:20] <milimetric>	 i haven't tried making reports with wikimetrics and stuff, but i'm sure i can fix that myself
[00:32:22] <madhuvishy>	 i guess it was never updating since puppet broke weeks ago?
[00:33:38] <milimetric>	 well, I'm not sure.  Because something else went wrong and those repositories haven't seen updates *themselves* (which means there's a problem on the cluster somewhere where they're computed)
[00:37:36] <milimetric>	 elukey: make sure you mention that at standup on Monday, I wanna take a look but I don't have time tonight
[00:38:17] <elukey>	 sure!
[00:40:02] <madhuvishy>	 milimetric: but puppet was doing the updates no?
[00:40:16] <madhuvishy>	 https://www.irccloud.com/pastebin/0HuAP7FM/
[00:40:31] <milimetric>	 madhuvishy: that was pulling the data from the repo, yes
[00:40:48] <milimetric>	 madhuvishy: but a separate cron is comitting and pushing that data
[00:41:14] <milimetric>	 *committing? (/me thinks double letters in English are meant to drive /me insane)
[00:41:31] <madhuvishy>	 :P
[00:41:47] <milimetric>	 i'm not sure where the commit / push cron is, stat1002?
[00:41:52] <madhuvishy>	 milimetric: anyway, it's super late on a friday for you. i'll add a temporary role to fix those reports if it's super important
[00:41:59] <madhuvishy>	 if not i can fix it the right way
[00:42:15] <madhuvishy>	 is vital signs on dashiki?
[00:42:19] <milimetric>	 uh... not sure how important it is, I don't like having things not working
[00:42:37] <milimetric>	 dashiki was used to build vital signs, but vital signs is running on limn1 not the new dashiki instance
[00:42:46] <madhuvishy>	 ya that's fine
[00:45:49] <madhuvishy>	 milimetric: can i build it from scratch if i cloned dashiki?
[00:46:45] <milimetric>	 madhuvishy: yes, no problems with that
[00:47:10] <milimetric>	 madhuvishy: this is the config: https://meta.wikimedia.org/wiki/Config:VitalSigns
[00:47:40] <madhuvishy>	 milimetric: gulp --layout <> --config VitalSigns?
[00:47:49] <madhuvishy>	 what would the layout be?
[00:48:00] <milimetric>	 metrics-by-project
[00:48:03] <madhuvishy>	 okay
[00:48:50] <milimetric>	 but I'm really not sure why you're conflating that with the pageviews data
[00:50:05] <madhuvishy>	 milimetric: https://github.com/wikimedia/analytics-dashiki/blob/14044aec79e49fe8cf797817f6a0854bcae15b01/src/app/config.js#L33
[00:50:50] <milimetric>	 madhuvishy: that may be old, I *think* the config is just up to the metric, but that's a separate thing that we need to change
[00:50:53] <madhuvishy>	 the pageviews data doesn't belong in wikimetrics too in my opinion - so i'm wondering if we can have a dashiki instance that will have a puppet role that does the cloning
[00:51:15] <madhuvishy>	 and vitalsigns will pick it up from there
[00:51:18] <milimetric>	 madhuvishy: but the data *definitely* doesn't belong in a dashiki instance
[00:51:43] <madhuvishy>	 milimetric: okayyy - so we need to figure out a place for it
[00:52:21] <madhuvishy>	 milimetric: may be a vitalsigns instance
[00:52:31] <milimetric>	 again, Nuria, etc. all decided metrics.wmflabs.org was a good place to serve it.  I have no opinion but I suggest not changing that right at this very moment :)
[00:55:28] <madhuvishy>	 okay, i'm gonna git clone them then. and we will figure it out on monday
[00:55:40] <milimetric>	 k
[00:57:12] <milimetric>	 I'll test the reports then
[00:58:27] <milimetric>	 sweet! that was easy :)
[01:00:00] <milimetric>	 madhuvishy: you said the scheduler wasn't working in the SAL?
[01:00:19] <milimetric>	 it looks started, I can check on it tonight after it should be running if you want to be sure
[01:00:31] <madhuvishy>	 milimetric: yeah but I fixed it
[01:00:35] <milimetric>	 ah, cool
[01:00:36] <madhuvishy>	 thought i logged that too
[01:01:20] <milimetric>	 ah, you did, I missed it
[01:12:06] <milimetric>	 yay, pageviews work :)
[01:12:18] <milimetric>	 k, calling it a night, I think everything's working fine
[01:12:49] <milimetric>	 nice job, Madhu :)
[01:13:35] <madhuvishy>	 milimetric: cool! there is some folder called /srv/aggregator-data that's not being git cloned from anywhere in old puppet, but exists in old prod server
[01:13:39] <madhuvishy>	 dont know about that
[01:13:46] <madhuvishy>	 but everything else should be fine
[01:13:56] <madhuvishy>	 thanks for helping out :) good night!
[01:14:38] <milimetric>	 madhuvishy: I think that was some old relic, it's not needed as far as I can tel
[01:14:41] <milimetric>	 *tell
[01:14:46] <milimetric>	 good night, have a nice weekend
[01:14:58] <madhuvishy>	 okay
[01:24:59] <elukey>	 milimetric: o/
[06:31:50] <wikibugs>	 Analytics-Tech-community-metrics: SCM project 'core' contains Pywikibot and MediaWiki - https://phabricator.wikimedia.org/T123808#1938890 (jayvdb)
[12:15:18] <wikibugs>	 Analytics-Kanban: Aggregator for projectviews stuck in 2015 {hawk} [3pts] - https://phabricator.wikimedia.org/T123832#1938992 (JAllemandou) NEW a:JAllemandou
[15:18:14] <wikibugs>	 Analytics-General-or-Unknown, Community-Advocacy, Wikimedia-Extension-setup, Wikipedia-iOS-App-Product-Backlog: enable Piwik on ru.wikimedia.org - https://phabricator.wikimedia.org/T91963#1939084 (Milimetric) To keep the archives happy:  We've deployed piwik to production and are testing it out wi...
[19:29:04] <wikibugs>	 Analytics-EventLogging: Message 'jsonschema-invalid-notinenum' is being passed wrong parameters - https://phabricator.wikimedia.org/T123842#1939202 (matmarex) NEW
[19:43:33] <wikibugs>	 Analytics-EventLogging: Message 'jsonschema-invalid-notinenum' is being passed wrong parameters - https://phabricator.wikimedia.org/T123842#1939218 (matmarex) a:matmarex
[22:06:01] <wikibugs>	 Analytics-Wikistats: Discrepancies in historical total active editor numbers - https://phabricator.wikimedia.org/T87738#1939307 (Nemo_bis) >>! In T87738#1366152, @Nemo_bis wrote: > I see DragonBot~pmswiki in https://stats.wikimedia.org/EN/TablesWikipediaPMS.htm#sleeping  > Are more bots being counted as user...
[23:01:01] <wikibugs>	 Analytics-Tech-community-metrics: scr-backlog.html should also exclude V-2 changesets - https://phabricator.wikimedia.org/T123848#1939341 (Aklapper) NEW