[00:49:25] madhuvishy: yeah, it's all but merged at this point, and we can easily deploy. I think the only thing we're waiting for is the upstream pageviews.js to merge our PR so it accepts "all-projects", nuria submitted that yesterday [02:13:49] Analytics-EventLogging, MediaWiki-extensions-WikimediaEvents, Patch-For-Review, Performance: Convert WikimediaEvents statsv.js to use sendBeacon - https://phabricator.wikimedia.org/T112843#2082985 (Krinkle) p:Triage>Normal a:Krinkle [02:40:25] Could someone check or verify that webperf/deprecate.py is still running properly on hafnium? [02:40:30] It seems the metrics died on Feb 25 [02:40:52] last modified time for any property from mw.js.deprecate: Feb 25 05:35 [02:59:35] Analytics-EventLogging, MediaWiki-extensions-WikimediaEvents, Performance-Team, Patch-For-Review: Convert WikimediaEvents statsv.js to use sendBeacon - https://phabricator.wikimedia.org/T112843#2083106 (Krinkle) [07:15:24] Analytics-Kanban, Wikipedia-Android-App-Backlog, Mobile-App-Android-Sprint-77-Iridium: Count requests to RESTBase from the Android app - https://phabricator.wikimedia.org/T128612#2083330 (bearND) a:Milimetric>bearND [07:16:26] Analytics-Kanban, Wikipedia-Android-App-Backlog, Mobile-App-Android-Sprint-77-Iridium: Count requests to RESTBase from the Android app - https://phabricator.wikimedia.org/T128612#2080435 (bearND) [07:43:24] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2083401 (elukey) Finally we should have a good picture of what changes between 3.0 and 4.0 api from a general overview: * https://www.varnish-cache.... [07:51:49] (PS1) BearND: Use x-analytics pageview to count mobile apps pageviews [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274644 (https://phabricator.wikimedia.org/T128612) [09:46:00] Hi a-team :) [09:46:13] bd808_: If you wish I mat help for oozie stuff :) [09:56:24] (PS2) Joal: Add archive stage to last_access_uniques jobs [analytics/refinery] - https://gerrit.wikimedia.org/r/274187 (https://phabricator.wikimedia.org/T126767) [10:00:31] Analytics-Tech-community-metrics, Developer-Relations, DevRel-March-2016, Patch-For-Review, developer-notice: Check whether it is true that we have lost 40% of (Git) code contributors in the past 12 months - https://phabricator.wikimedia.org/T103292#1498055 (ori) Filtering at the repository leve... [10:01:36] hello joal! [10:05:40] Morning elukey :) [10:17:14] (PS1) Joal: Add information to oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274666 [10:40:02] Analytics: Compile a request data set for caching research and tuning - https://phabricator.wikimedia.org/T128132#2083680 (Danielsberger) Here's another idea for getting a smaller dataset. As the eventual goal is to reproduce the cache performance, we can focus on requests issued to just one or two caches (e... [13:13:03] (PS1) Joal: Improve webrequest oozie jobs emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274691 [13:30:43] (PS1) Joal: Improve pageview & projectview oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274693 [13:33:56] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084136 (elukey) Summary after today's brainstorming with @ema: * We progressed a bit the remaining unclear points about the parse* functions, some... [13:44:48] joal: question for you: what is the difference between page counts raw and page views? [13:46:53] ahhrgh I hate gmail [13:47:05] ? [13:48:29] elukey: http://article.gmane.org/gmane.org.wikimedia.analytics/3435 [13:49:28] joal: I'm just making sure you saw my file.encoding system setting thing [13:49:34] but I can't find the task [13:49:46] Hi milimetric: I saw tht yes [13:49:58] k, does it work? [13:50:04] in our case i mean [13:50:16] milimetric: The thing we need to double check with ottomata is that this setting is not overwritten by the jobs [13:50:23] I don't know :) [13:52:00] (CR) Elukey: [C: 1] Improve pageview & projectview oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274693 (owner: Joal) [13:52:23] (CR) Elukey: [C: 1] Add information to oozie error emails (1 comment) [analytics/refinery] - https://gerrit.wikimedia.org/r/274666 (owner: Joal) [13:54:58] ---^ I put a +1 in the wrong code review sorry [13:55:02] joal thanks :) [13:55:18] (PS1) Joal: Improve aqs, browser & restbase oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274698 [13:59:54] milimetric: When speaking of this setting, I mean the JAVA_TOOLS_OPTION one [14:00:03] yes [14:00:18] i hope the jobs don't override it in a way we can't control [14:00:20] that would be... insane [14:00:21] :) [14:01:54] Well, the thing is that the JVM has so many parameters that hadoop related stuff as script to set many of them. I don't know if those use JAVA_TOOLS_OPTION or not, and if they are correct enough not to blackly override if there's already something in it :) [14:02:15] We need to double check :) [14:08:39] (PS1) Joal: Improve pagecounts oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274700 [14:11:12] (PS2) Joal: Improve pageview & projectview oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274693 [14:21:25] (PS1) Joal: Improve mobile_apps oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274706 [14:30:02] (PS1) Joal: Improve mediacounts oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274707 [14:37:22] (PS1) Joal: Improve last_access_uniques oozie error emails [analytics/refinery] - https://gerrit.wikimedia.org/r/274710 [14:46:17] (PS1) Joal: Add error emails to cassandra loading oozie jobs [analytics/refinery] - https://gerrit.wikimedia.org/r/274714 [14:49:49] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084318 (Ottomata) > Would it be feasible to just reuse varnishncsa and pipe its output (apache like log format) to a piece of software that just par... [14:52:07] yarhhh need power! heading home for POWERRRR [15:06:17] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084349 (elukey) @Ottomata the code is really optimized as you were saying to map VSL structures to log tags in apache format, but it is essentiall... [15:08:01] https://github.com/xcir/python-varnishapi/blob/varnish40/varnishapi.py [15:08:05] love it :D [15:30:19] (CR) Milimetric: [C: 2 V: 2] Add hive queries for the traffic breakdown reports [analytics/reportupdater-queries] - https://gerrit.wikimedia.org/r/272635 (https://phabricator.wikimedia.org/T127326) (owner: Mforns) [15:37:28] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084445 (Ottomata) > the code is really optimized as you were saying to map VSL structures to log tags in apache format, but it is essentially what n... [15:40:00] git st [15:40:03] oops :) [15:42:05] :) [15:43:06] ottomata: the new varnish logging scheme is really awesome, if only it was documented properly :D [15:43:42] yeah, sounds awesome, grouped tags into transactions sounds waayyyy better [15:43:51] then tracking txids and grouping the together your self [15:43:54] them* [15:45:23] looks like we are getting closer to know what to do, definitely a big change [15:46:29] (PS1) Joal: Add destination overwriting capacity to camus [analytics/camus] - https://gerrit.wikimedia.org/r/274736 (https://phabricator.wikimedia.org/T128611) [15:46:52] Logging off for a bit to visit my grandmother, will be back for standup! [15:48:35] (CR) Nuria: [C: -1] Use x-analytics pageview to count mobile apps pageviews (1 comment) [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274644 (https://phabricator.wikimedia.org/T128612) (owner: BearND) [15:58:37] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084490 (faidon) >>! In T124278#2084445, @Ottomata wrote: >> the code is really optimized as you were saying to map VSL structures to log tags in apa... [15:58:53] Analytics, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2084491 (Ottomata) Ja! Am happy to help too! I’ve poked around in that code a bit too, so I might be able to offer some help. [16:00:19] haha joal you were working on the camus thing!? [16:00:19] haha [16:00:27] i started to yesterday, but then got got up in other stuff [16:00:28] nice! [16:00:35] Oh. sorry ottomata :( [16:00:49] Should have told you :) [16:00:51] hehe [16:00:53] no problem [16:00:55] i didn't start coding [16:01:00] was just getting set up with testing camus changes [16:01:16] today was 'fed-up-of-backfilling-day' [16:01:28] Currently testing the new camus version [16:01:31] Will let you know [16:01:46] nice!!! [16:01:48] this looks good [16:01:55] i was just going to change the main committer code, I like this subclass [16:01:57] less intrusive :) [16:02:25] I thought two times about it before doing it this way :) [16:03:41] milimetric: yt? [16:03:44] (CR) Ottomata: Add destination overwriting capacity to camus (1 comment) [analytics/camus] - https://gerrit.wikimedia.org/r/274736 (https://phabricator.wikimedia.org/T128611) (owner: Joal) [16:03:49] hi nuria [16:04:07] milimetric: should i rebase my dashiki patch or you have it rebased locally? [16:04:34] nuria: I pushed my patch to gerrit, so you can grab mine and work from there [16:05:41] milimetric: ok, did not see it, it is a different patch entirely, will abandon mine [16:05:50] it is?! [16:06:03] (CR) Nuria: "I think it looks good we need to create a ticket to surface this work, thanks for being so through." [analytics/refinery] - https://gerrit.wikimedia.org/r/274666 (owner: Joal) [16:06:06] that doesn't make sense... i just ammended your commit [16:06:27] nuria: no, it's on your change: https://gerrit.wikimedia.org/r/#/c/270867/ [16:06:30] I just added a new patch [16:06:34] milimetric: ah sorry [16:14:24] (PS2) Joal: Add destination overwriting capacity to camus [analytics/camus] - https://gerrit.wikimedia.org/r/274736 (https://phabricator.wikimedia.org/T128611) [16:14:50] (CR) Joal: Add destination overwriting capacity to camus (1 comment) [analytics/camus] - https://gerrit.wikimedia.org/r/274736 (https://phabricator.wikimedia.org/T128611) (owner: Joal) [16:17:34] (CR) Nuria: "Patch is missing tests." (1 comment) [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274307 (https://phabricator.wikimedia.org/T128530) (owner: EBernhardson) [16:18:31] (CR) Joal: Add information to oozie error emails (1 comment) [analytics/refinery] - https://gerrit.wikimedia.org/r/274666 (owner: Joal) [16:24:11] (CR) Ottomata: [C: 1] Add destination overwriting capacity to camus [analytics/camus] - https://gerrit.wikimedia.org/r/274736 (https://phabricator.wikimedia.org/T128611) (owner: Joal) [16:27:57] (CR) EBernhardson: [C: -1] "Also lots of mixed tabs and spaces" (1 comment) [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274307 (https://phabricator.wikimedia.org/T128530) (owner: EBernhardson) [16:28:29] joal: https://gerrit.wikimedia.org/r/#/c/274736/2 WOW [16:29:31] elukey: As I told ottomata, today was fed-up-of-backfilling day :) [16:29:45] :D [16:30:34] a-team: standddupppp [16:31:31] ottomata, madhuvishy : standupppp [16:31:32] madhuvishy, ottomata : standup ? [16:32:39] OOp [16:53:36] Analytics-Kanban: Add better error emails to oozie jobs - https://phabricator.wikimedia.org/T128749#2084684 (JAllemandou) a:JAllemandou [16:58:58] milimetric: are you coming to tasking? [16:59:03] yes [16:59:19] ok, 9:05 [16:59:22] cc milimetric [16:59:31] cool [17:20:15] (CR) BearND: Use x-analytics pageview to count mobile apps pageviews (1 comment) [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274644 (https://phabricator.wikimedia.org/T128612) (owner: BearND) [17:21:56] nuria about https://gerrit.wikimedia.org/r/#/c/274644/. Do you plan to create your own patch for what you suggested? [17:23:23] Analytics-Kanban, Patch-For-Review: Migrate limn-mobile-data/reportupdater reports to use standalone reportupdater - https://phabricator.wikimedia.org/T128375#2084771 (Milimetric) a:mforns>Milimetric [17:25:18] Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Create regular backups of Analytics MySQL Meta instance - https://phabricator.wikimedia.org/T127991#2084778 (Milimetric) a:Ottomata [17:28:18] Analytics-EventLogging, Analytics-Kanban, Patch-For-Review, codfw-rollout, codfw-rollout-Jan-Mar-2016: EventLogging needs to be ready for codfw failover - https://phabricator.wikimedia.org/T127209#2084788 (Milimetric) a:Nuria [17:30:37] Analytics-EventLogging, Analytics-Kanban, Patch-For-Review: Add IP field only to schemas that need it. Remove it from EL capsule and do not collect it by default {mole} - https://phabricator.wikimedia.org/T126366#2084819 (Milimetric) [17:30:39] Analytics-EventLogging, Analytics-Kanban, QuickSurveys: Add magic word for hashed client IP event log field for Quick Surveys - https://phabricator.wikimedia.org/T127502#2084817 (Milimetric) Open>declined getting rid of all IPs [17:42:16] Analytics-Kanban, Patch-For-Review: Clean up Client IP and hashing related code on Eventlogging {oryx} {mole} - https://phabricator.wikimedia.org/T128408#2085006 (Milimetric) [17:44:58] Analytics-Kanban: annotations not visible on dashiki - https://phabricator.wikimedia.org/T128494#2085021 (Milimetric) Open>declined it turns out rendering of the tooltips is really slow in some cases - we should look into why OS X is so slow... :P [17:58:07] (CR) Nuria: Use x-analytics pageview to count mobile apps pageviews (1 comment) [analytics/refinery/source] - https://gerrit.wikimedia.org/r/274644 (https://phabricator.wikimedia.org/T128612) (owner: BearND) [17:58:32] Analytics-Kanban: Add partitions to webrequest text and upload topics - https://phabricator.wikimedia.org/T127351#2085088 (Milimetric) a:JAllemandou>Ottomata [18:03:00] ottomata: so embarassing, i figured out why hue never loads for me [18:03:29] ottomata: i had a rule in etc/hosts since the dawn of the word ..I must have been testing something and i forgot [18:05:11] oh wow, I would have never found that [18:08:26] Analytics: Change the Pageview API's RESTBase docs for the top endpoint - https://phabricator.wikimedia.org/T120019#2085120 (Milimetric) p:High>Normal [18:09:57] Analytics-Kanban, Operations, Traffic, Patch-For-Review: varnishkafka integration with Varnish 4 for analytics - https://phabricator.wikimedia.org/T124278#2085127 (Milimetric) [18:15:22] Analytics: Invalid page titles are appearing in the top_articles data - https://phabricator.wikimedia.org/T117346#1771813 (Milimetric) We'll remove "-", as a special case, from the data that's computed for the top endpoint. That means we'll be removing the probably miniscule number of pageviews to the actua... [18:16:10] Analytics: Make visualization of last access data {mole} - https://phabricator.wikimedia.org/T126768#2085168 (Milimetric) [18:16:12] Analytics: Visualize unique devices data in dashiki - https://phabricator.wikimedia.org/T122533#2085169 (Milimetric) [18:17:06] Analytics: Create a central page in wikitech to act as a central hub so users know where to go for different types of data - https://phabricator.wikimedia.org/T122970#2085184 (Milimetric) [18:17:08] Analytics: Add help page in wikitech on what the analytics team can do for you similar to release engineering page - https://phabricator.wikimedia.org/T116188#2085185 (Milimetric) [18:25:42] ottomata: confirmation the camus overwrite works :) [18:26:12] ottomata: Also, do you have news on hardware (cassandra and/or Druid?) [18:27:23] nuria what's the timeline for the pageview check? Should we just do it for apps now, then refactor the code to also add to the place you want it to be? [18:29:20] bearND: no, focus on the apps sending the header , we shall implement the "recognition" for everyone but likely not this week. There are other things we had prioritized we are working on. You can also do code changes but should be done for all code paths, plus also they need tests. [18:30:39] PSHH no, things have been sitting with that hardware stuff, eh? [18:30:42] i'll look through those and poke them [18:30:50] even the ones i have poked have just been sitting [18:32:02] ottomata: retro? [18:32:04] thanks ottomata :) [18:32:05] ggellerman: retro? [18:32:30] ottomata: retro or orter ? [18:37:09] Analytics: Invalid page titles are appearing in the top_articles data - https://phabricator.wikimedia.org/T117346#2085344 (Ironholds) Huh. The pageID approach won't work? :( [18:43:09] must I retro :D [18:43:10] ? [18:43:17] just got some lunch, joining [18:43:33] no 'need', just a ping saying we started [18:55:03] Analytics-Kanban, Wikipedia-Android-App-Backlog, Mobile-App-Android-Sprint-77-Iridium, Patch-For-Review: Count requests to RESTBase from the Android app - https://phabricator.wikimedia.org/T128612#2085441 (bearND) [19:00:51] Analytics-Kanban, Wikipedia-Android-App-Backlog, Mobile-App-Android-Sprint-77-Iridium, Patch-For-Review: Count requests to RESTBase from the Android app - https://phabricator.wikimedia.org/T128612#2080435 (MBinder_WMF) @dbrant If this is really an #analytics task (evidenced by @Milimetric story-poi... [19:01:37] going offline a-team!! byyeee [19:01:44] Bye elukey :) [19:05:09] byeee [19:06:49] ottomata: I think the sendBeacon patch is good to go [19:07:18] ori: we need to test it on beta labs right? to see how UA transfers [19:07:48] yeah, I meant good to go as in good to be merged into master, so it can roll out to labs [19:07:54] I did not mean deploy to prod :) [19:08:37] ori +1, [19:08:42] i just dunno how it works with deployment to beta [19:08:47] if it gets merged in master, i can go to beta? [19:08:50] without going to prod? [19:08:55] it* [19:11:13] Hello, we would like to access (one-time) wmf.webrequest data. In my research group, we want to analyse the data, and based on them get the most executed code on wikimedia, which is the basis for further research. What is the correct way of accessing the data in our case? [19:13:19] a-team I'm for today [19:13:25] Bye ! [19:14:03] ottomata: it goes into beta first, and eventually into prod [19:15:35] it will be live on beta within an hour of being merged. It will hit production on March 8: https://wikitech.wikimedia.org/wiki/Deployments#deploycal-item-20160308T2000 [19:15:37] cool, should I merge it? [19:16:01] ori i'll let you merge [19:16:09] I think it's good to go, but if you want to sanity-check first with nuria / madhuvishy that's cool [19:16:15] dash_2: good question, I'm not really sure how to answer [19:16:22] you might want to email the analytics maillin list [19:16:34] https://lists.wikimedia.org/mailman/listinfo/analytics [19:18:21] ottomata: Thank you, I will. [19:25:50] Analytics-Kanban, user-notice: Pageviews API reporting inaccurate data for pages titles containing special characters - https://phabricator.wikimedia.org/T128295#2085593 (Johan) [19:52:44] ottomata: I tried to use notify to print some debug messages in puppet, didn't work, it said "Could not find class notify for reportcard-puppet-testing.reportcard.eqiad.wmflabs" [19:56:19] HMmMm [19:56:30] you you did like notify { " wooo": }? [19:56:37] i think you can also try notice('woowoo') [19:56:55] not sure though [20:12:25] milimetric: I need to test build but dashiki's pageview patch is ready to go with teh last fix you did, correct? [20:12:56] nuria: I built a dashboard for each main layout and tested them, it was good [20:13:10] the only issue left is updating the pageviews.js reference after they merge your PR [20:13:22] milimetric: it is set to latest [20:13:35] and he fixed it on master [20:13:40] rather than merging my PR [20:13:50] which hd a bug [20:13:53] *had [20:14:00] ->shame [20:14:25] aha, ok [20:14:35] good, so then the all-projects should work [20:14:43] and that means everything I know is good [20:17:12] k, let me do another pass for triple checking and we can merge [20:18:20] ottomata: i guess they server side change for EL is merged now [20:19:37] ja [20:19:44] ori is going going to deploy it to beta [20:22:27] (CR) Nuria: [C: 2] "Merging." [analytics/dashiki] - https://gerrit.wikimedia.org/r/270867 (https://phabricator.wikimedia.org/T124063) (owner: Nuria) [20:22:33] ottomata: I did class { 'reportupdater': [20:22:33] user => 'milimetric', [20:22:33] base_path => '/srv', [20:22:33] } [20:22:48] base_path /srv/reportupdater [20:22:48] no? [20:22:52] and it says: Could not find dependency User[milimetric] for Git::Clone[analytics/reportupdater] [20:22:56] oh ja [20:22:56] for sure [20:23:01] pupet is not managing users like that [20:23:03] (just doing that on labs to test) [20:23:12] you could create some test user [20:23:13] or just say root [20:23:15] eh [20:23:17] root isn't there [20:23:18] MBMMm [20:23:19] actually [20:23:24] we should remove that dependency. [20:23:28] it shouldn't be there if it isn't being managed [20:23:31] patch coming at you... [20:23:38] k :) [20:24:43] milimetric: pushed [20:35:47] (CR) Milimetric: [C: 2 V: 2] Add all reportupdater files to this dedicated repo [analytics/reportupdater] - https://gerrit.wikimedia.org/r/272712 (https://phabricator.wikimedia.org/T127327) (owner: Mforns) [20:37:27] milimetric: I am going to deploy dashiki to staging, will carry more code than pageview-Api changes as there were other changes on master [20:38:22] nuria: sure, but those changes were deployed [20:38:29] milimetric: ah , ok [20:38:30] do you know how to do the new deploy? [20:38:32] with fab? [20:40:24] itssooeasyy [20:40:26] :D [20:41:47] madhuvishy: I know, but I wasn't sure if nuria ever done it [20:42:03] or ever even installed fab (which can be weird if you get the wrong paramiko and all that) [20:42:09] ya i don't think she has, and some explanation of the set up may be warranted too [20:42:12] milimetric: it is real easy [20:42:21] I am a nijaaa [20:42:25] :D [20:43:09] milimetric: you have to change hostnames on the cli when deploying to the prod one right? that's the only gotcha I know [20:43:22] madhuvishy: kudos on making it easy [20:44:31] madhuvishy: yeah, kudos to me for making that hard because otherwise I'd accidentally deploy to prod all the time [20:44:57] madhuvishy: do we need to be added to project to ssh to instance? [20:45:01] dashiki-staging-01.dashiki.eqiad.wmflabs [20:45:15] madhuvishy: cause it doesn't look like i can ssh, can you? [20:45:22] nuria yeah, lemme add you [20:46:21] k nuria you're an admin [20:46:57] (i added marcel too) [20:47:08] milimetric: k, let me see if i can ssh [20:47:16] nuria: milimetric feel free to !log here or on #wikimedia-labs [20:47:16] milimetric: i bet puppet needs to run [20:47:29] madhuvishy: I will just deploy to staging [20:47:36] nuria: sure [20:47:48] are you making puppet changes? [20:47:55] ah yes [20:47:57] Hiera [20:48:00] but [20:48:11] hmmm vital-signs already exists? [20:48:42] madhuvishy: no, i was thinking puppet needs to run after me being an admin [20:48:49] madhuvishy: so i can ssh into machine [20:49:02] madhuvishy: i just made that up, but it seemed logical [20:49:03] huh you can't get in? [20:49:25] madhuvishy: now i can [20:49:29] ok cool [20:49:31] madhuvishy: i just had to wait a few mins [20:49:38] ottomata: everything looks good, I fixed a couple of typos, can you verify [20:50:06] ottomata: the only weird thing is that /srv/reportupdater is owned by milimetric:root and all the others are owned by milimetric:wikidev [20:50:11] ottomata: howdy! are you going to share the password to analytics-search with Ironholds and me? we can't `su analytics-search` yet [20:50:15] I didn't see how that was happening, it seems fine, just wondering if it was on purpose [20:51:01] hm, milimetric that is strange [20:51:07] bearloga: there is no pw [20:51:14] you need to be in the analytics-search-users group [20:51:16] madhuvishy: i guess https://vital-signs-test.wmflabs.org/#projects=ruwiki,itwiki,dewiki,frwiki,enwiki,eswiki,jawiki/metrics=Pageviews doesn't exist? [20:51:26] sorry https://vital-signs-test.wmflabs.org [20:51:39] nuria: looking [20:51:42] milimetric: puppet looks like it should ensure group wikideb [20:51:44] but, its ok i guess [20:51:46] madhuvishy: dns exists looks like but no code in instance [20:51:54] ottomata: can you please add me to that group? [20:52:07] ottomata: and also Ironholds (please) [20:52:40] milimetric: i did job_repositories_path plural intentionally [20:52:41] nuria: ya - did you deploy vitalsigns? [20:52:45] there are multiple repositories in that path [20:52:51] madhuvishy: ya but i had no permits [20:52:52] nuria: to staging? [20:52:55] madhuvishy: doing it again [20:53:09] OH [20:53:09] whoops [20:53:11] sorry i see [20:53:11] :) [20:53:13] you fixed [20:53:15] just a typo, yea [20:53:28] the owner => $user is right too, right? [20:53:31] yeah [20:53:32] sorry [20:53:34] I can do group => 'wikidev', should I? [20:53:41] just fixed [20:53:41] madhuvishy: ja, i bet i do not have sudo, lemmme see [20:54:00] k, sweet, then you can merge this whenever, everything else looked fine, cronjob and all [20:54:08] oh wait! [20:54:14] ja we gotta be careful! [20:54:14] no merge yet ... hahah, sorry [20:54:14] :) [20:54:15] yeah [20:54:19] gotta make the plan [20:54:21] yup [20:54:23] maybe tomorrow [20:54:29] * milimetric puts on thinkin' cap and makes plan [20:54:32] k :) [20:58:32] madhuvishy: your labs password is your wikitech password correct? [21:00:13] nuria: password? [21:00:44] madhuvishy: mmm.. maybe i have to configure fab not to ask me for a pasword [21:00:50] uh oh [21:01:01] nuria: hold on let me verify something [21:01:09] also run pip install --upgrade fabric [21:02:14] madhuvishy: trying again [21:02:26] madhuvishy: looks like now it is working [21:02:38] ok good [21:06:44] nuria: the proxy was pointing to the wrong instance - i just fixed it [21:06:53] vital-signs is up now [21:07:03] madhuvishy: i see [21:08:45] https://vital-signs-test.wmflabs.org/#projects=all,ruwiki,itwiki,dewiki,frwiki,enwiki,eswiki,jawiki/metrics=Pageviews [21:09:15] madhuvishy: worked great! one question: [21:09:45] all apache configuration is on puppet right? [21:10:51] nuria: yes [21:11:03] puppet + hiera [21:11:06] madhuvishy: excellent, works great [21:11:53] nuria: :) [21:12:51] (PS1) Nuria: Removing console.log [analytics/dashiki] - https://gerrit.wikimedia.org/r/274811 [21:18:24] milimetric: deployed to staging, one small fix but take a look to see if I missed anything: https://vital-signs-test.wmflabs.org/#projects=enwiki/metrics=Pageviews [21:19:31] milimetric: also 1 reason why i dislike "all" too is that it only works for pageview metric, errors on others and that looks not so hot [21:19:45] milimetric: but nothing we have to fix now [21:31:06] nuria: yeah, i agree, that's not awesome [21:31:10] it looks good [21:31:20] i got obsessed with why some of the numbers are slightly different [21:31:39] I guess the pageview API way takes advantage of any data that's re-computed [21:31:46] which is great [21:31:51] I'll merge [21:32:04] (CR) Milimetric: [C: 2 V: 2] Fetch Pageview Data from Pageview API [analytics/dashiki] - https://gerrit.wikimedia.org/r/270867 (https://phabricator.wikimedia.org/T124063) (owner: Nuria) [21:32:46] sweeet, I'm gonna go get lunch (it's almost 5pm !!!) but let me know if you wanna deploy nuria, or just go ahead and deploy only vital-signs, that's totally cool [21:46:31] Analytics-Cluster, Operations, hardware-requests: eqiad: New Hive / Oozie server node in eqiad Analytics VLAN - https://phabricator.wikimedia.org/T124945#2086189 (Ottomata) Bump! I suppose we are still waiting for these nodes though? [22:18:32] Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Enable purging of old jobs and coordinators in oozie - https://phabricator.wikimedia.org/T127988#2086427 (Ottomata) [22:19:09] Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Create regular backups of Analytics MySQL Meta instance - https://phabricator.wikimedia.org/T127991#2086428 (Ottomata) Should do https://phabricator.wikimedia.org/T127988 first so we prune MySQL db before attempting backup. [22:19:20] Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Enable purging of old jobs and coordinators in oozie - https://phabricator.wikimedia.org/T127988#2060457 (Ottomata) [22:19:22] Analytics-Cluster, Analytics-Kanban, Patch-For-Review: Create regular backups of Analytics MySQL Meta instance - https://phabricator.wikimedia.org/T127991#2086429 (Ottomata) [22:42:37] hey people!!!!!!!!!11111111 [22:42:50] madhuvishy: milimetric can I kill wikimetrics1 / wikimetrics1-staging instance? or at least shut them down? [22:42:52] these are the old ones [22:42:57] * yuvipanda ist rying to reduce total number of NFS clients [22:43:11] yuvipanda: aah - I think both can die [22:43:20] madhuvishy: would you like to do the honors? [22:43:38] madhuvishy: you can just do a shutdown -h now on both and we can delete them if nobody complains in a week [22:44:19] yuvipanda: :) There's nothing on them that people can access - so it might as well die [22:44:26] i'll let milimetric confirm this [22:44:31] ok [22:44:32] and then lets kill it with joy [22:44:36] ok [22:45:36] madhuvishy: https://phabricator.wikimedia.org/T128804 [22:46:05] yuvipanda: cool - only limn1 needs it now? [22:46:37] i dont know if it needs nfs and for what [22:47:04] yuvipanda: I need to make sure of a couple of things, one bit [22:47:13] milimetric: limn1 hasn't had a puppet run in forever [22:47:16] so it doesn't actually matter :D [22:47:22] if I don't explicitly kill it [22:47:24] oh [22:47:25] Analytics, Labs: Kill all unnecessary NFS usage in the analytics project - https://phabricator.wikimedia.org/T128804#2086561 (yuvipanda) https://wikitech.wikimedia.org/w/index.php?title=Hiera%3AAnalytics&type=revision&diff=344048&oldid=310235 turns off new NFS mounts for new instances! @madhuvishy is loo... [22:47:26] yuvipanda: limn1 doesn't need nfs [22:47:28] I keep confusing madhuvishy and milimetric bah [22:47:30] Analytics, Labs: Kill all unnecessary NFS usage in the analytics project - https://phabricator.wikimedia.org/T128804#2086564 (yuvipanda) [22:47:52] cool so i think only wikimetrics1 and wikimetrics-staging [22:47:52] milimetric: oh, so nothing in the analytics project needs NFS anymore? can I kill it for all of the instances? :D [22:48:45] I think the new wikimetrics is not yet backing up like the old one [22:48:52] indeed [22:48:55] it isn't [22:49:00] that's the only hesitance [22:49:13] milimetric: backing up what? [22:49:23] but the old instance odesn't actually have anything in it anymore [22:49:25] files? [22:49:29] and the new instances arne't in the same project [22:49:31] so [22:49:32] yes [22:49:37] madhuvishy: so the old backup would do this: [22:49:49] dump the db every hour, zip it [22:49:58] db is being backed up already milimetric [22:50:01] zip all the recurrent report runs, zip it [22:50:27] yes those aren't being backed up [22:50:28] Copy 24 hours worth of backup and keep it around [22:50:43] then take the latest and make it a daily backup, keep 7 days or so of daily backups around [22:50:43] but db is, labsdb automatically does it [22:51:01] thing is, we literally never used it once in two years [22:51:07] even then, wikimetrics1 can't save us [22:51:09] even if we lost stuff, people didn't care [22:51:22] the only thing that's somewhat valuable is the vital signs recurrent reports [22:51:30] because those are *very* expensive to generate [22:51:37] this is lal unrelated to the analytics project :) [22:51:42] but we're moving to Druid anyway, and they'll be fairly cheap there [22:51:43] yes [22:51:47] since the new wikimetrics isn't even in the same project [22:51:50] it's unrelated [22:51:52] i agree [22:52:03]