[01:23:43] 10Analytics, 10Analytics-Wikistats, 10Operations, 10Wikidata, and 6 others: Create Wikiversity Hindi - https://phabricator.wikimedia.org/T168765#3514829 (10Dcljr) @Reedy Should the To-Do list be completed before this task is closed? [02:06:31] 10Analytics, 10Android-app-feature-Feeds, 10Mobile-Content-Service, 10Pageviews-API, and 4 others: Why top views data of different sources is not the same? - https://phabricator.wikimedia.org/T172379#3514854 (10Shizhao) >>! In T172379#3514125, @Nuria wrote: > I doubt the feed algorithm considers pageviews... [04:44:23] 10Analytics, 10Analytics-Wikistats, 10Operations, 10Wikidata, and 6 others: Create Wikiversity Hindi - https://phabricator.wikimedia.org/T168765#3514911 (10Jayprakash12345) 05Resolved>03Open The namespace are not redirecting respectively Portal: प्रवेशद्वार: School: विद्यालय: Collection: संग्रह: And... [04:55:27] 10Analytics, 10Android-app-feature-Feeds, 10Mobile-Content-Service, 10Pageviews-API, and 4 others: Why top views data of different sources is not the same? - https://phabricator.wikimedia.org/T172379#3514914 (10Nuria) Sorry but I do not understand what is the question, I think we have provided quite a bit... [04:55:38] 10Analytics, 10Android-app-feature-Feeds, 10Mobile-Content-Service, 10Pageviews-API, and 4 others: Why top views data of different sources is not the same? - https://phabricator.wikimedia.org/T172379#3514915 (10Nuria) 05Open>03Resolved [06:07:22] 10Analytics-Data-Quality, 10Analytics-Kanban, 10Readers-Web-Backlog (Tracking): Pageview drop in ro.wikipedia hu.wikipedia and fr.wikipedia - https://phabricator.wikimedia.org/T170845#3514959 (10Nuria) @Strainu: Sorry for my deletions that make this ticket harder to read. Let me try to summarize. What I see... [06:11:57] 10Analytics, 10Analytics-Wikistats, 10Operations, 10Wikidata, and 6 others: Create Wikiversity Hindi - https://phabricator.wikimedia.org/T168765#3514961 (10Dereckson) Actually, we want to keep a standard common set of extensions on wikis, so forget the `Topic:` namespace and find a new name. You don't wan... [06:35:43] 10Analytics-Kanban, 10User-Elukey: Calculate how much Popups events EL databases can host - https://phabricator.wikimedia.org/T172322#3514992 (10elukey) >>! In T172322#3514505, @Tbayer wrote: > And a related question: https://wikitech.wikimedia.org/wiki/Analytics/Systems/EventLogging#Hadoop currently says: >... [06:41:05] 10Analytics, 10Analytics-Wikistats, 10Operations, 10Wikidata, and 6 others: Create Wikiversity Hindi - https://phabricator.wikimedia.org/T168765#3515000 (10Dereckson) [06:59:16] 10Analytics, 10Analytics-Wikistats, 10Operations, 10Wikidata, and 6 others: Create Wikiversity Hindi - https://phabricator.wikimedia.org/T168765#3477623 (10Anooprao) Wouldn't talk pages should named चर्चा instead of वार्ता in hindi [07:28:13] 10Analytics-Data-Quality, 10Analytics-Kanban, 10Readers-Web-Backlog (Tracking): Pageview drop in ro.wikipedia hu.wikipedia and fr.wikipedia - https://phabricator.wikimedia.org/T170845#3515031 (10Strainu) I don't have the time or resources needed to do the investigation myself. Thanks for your help anyway, I... [07:29:14] 10Analytics-Kanban, 10User-Elukey: Calculate how much Popups events EL databases can host - https://phabricator.wikimedia.org/T172322#3515032 (10Tbayer) OK, so for the avoidance of confusion, I assume that this archiving process with Scoop will be separate from the ongoing hourly import via Camus as described... [08:09:53] 10Analytics-Kanban, 10User-Elukey: Calculate how much Popups events EL databases can host - https://phabricator.wikimedia.org/T172322#3515042 (10elukey) I am not a big expert in this part of the analytics infrastructure so this answer is not authoritative, but IIUC from https://phabricator.wikimedia.org/T16261... [08:30:03] 10Analytics-Kanban, 10User-Elukey: Calculate how much Popups events EL databases can host - https://phabricator.wikimedia.org/T172322#3515044 (10elukey) Forgot to mention that we are going to document everything once we archive the table, so it will not be a guessing game to find it on HDFS. We'll also make cl... [08:33:31] so my fix for eventstreams didn't really fix the problem [08:33:57] now the other candidate is eventlogging, since we have been importing data from the same kafka topics overloaded by eventstreams [08:34:07] buuut eventlogging still does not have the last librdkafka [08:34:15] that limits the damage of the API mismatch/timeouts [08:34:26] so my next target is that one [08:59:17] !log updated librdkafka1 to 0.9.4.1 on eventlog1001 [08:59:18] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [09:31:03] nope no changes [09:31:04] sigh [09:31:30] as FYI puppet was broken on notebook*, just fixed it (it has been in this state for days :( ) [11:14:02] hellooooooo team [11:17:29] o/ [12:08:46] 10Analytics-Kanban, 10Operations, 10Patch-For-Review, 10User-Elukey: Analytics Kafka cluster causing timeouts to Varnishkafka since July 28th - https://phabricator.wikimedia.org/T172681#3515219 (10elukey) Upgraded librdkafka to 0.9.4.1 on eventlog1001 but no real changes registered. I tried to set again d... [12:26:34] 10Analytics-Kanban, 10Operations, 10Patch-For-Review, 10User-Elukey: Analytics Kafka cluster causing timeouts to Varnishkafka since July 28th - https://phabricator.wikimedia.org/T172681#3515242 (10elukey) Another interesting thing is the fact that the error ***always*** happens once per minute at the first... [13:12:16] hiiii a-team :) [13:16:56] elukey, yt? [13:18:10] ottomata: o/ [13:18:17] hiiii [13:18:19] morning :) [13:22:18] 10Analytics-Kanban, 10User-Elukey: dbstore1002 /srv filling up - https://phabricator.wikimedia.org/T168303#3515322 (10Marostegui) >>! In T168303#3514253, @Marostegui wrote: > I will be removing around 120G from m3 instance from dbstore1002 (T156758#3514245) Done and around 100G are back https://grafana.wikime... [13:27:05] hiiii [13:27:09] elukey: https://gerrit.wikimedia.org/r/#/c/370865/ [13:27:10] :) [13:27:12] objections? [13:28:35] nope :) [13:28:54] gr8 [13:28:55] 10Analytics, 10Analytics-EventLogging, 10Performance-Team, 10Scap (Scap3-Adoption-Phase1): Use scap3 to deploy eventlogging/eventlogging - https://phabricator.wikimedia.org/T118772#3515344 (10Ottomata) > However, does this also mean we should deploy updates to eventlogging ourselves? Yup! > might also hel... [13:29:08] elukey: gimme a +1 ? [13:30:22] do you want it on paper to blame me if anything does wrong :P ? [13:31:02] ottomata: jokes aside, I think that we could also restrict to Hadoop only [13:32:08] elukey: we could ya [13:32:24] but i didn't really want to maintain a network constant for all hadoop workers? [13:32:28] . [13:32:56] I thought there was already one, but probably only for the masters :) [13:32:59] okok no problem [13:33:03] https://phabricator.wikimedia.org/T172681#3515242 is weeeird [13:33:10] if you have time today --^ [13:34:11] elukey: that is weird [13:34:18] that's in the broker log? [13:34:24] yep, server.log [13:34:30] wonder if it is happening on main eqiad [13:34:38] since changeprop uses it with the same kafka client [13:35:53] elukey: is this causing harm? this causes dropped webrequest messages ? [13:36:40] for the moment only one oozie data loss email, but I keep seeing varnishkafka queueing tons of requests and timing out in the logs.. not sure if the reqs eventually makes to the broker or not [13:36:52] I am worried that with more traffic this thing will exacerbate [13:36:59] this is why I am working on it :( [13:40:49] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515429 (10Ottomata) > As a reminder, there is still no real support for EventLogging data analysis using Hive/Hadoop. IT IS COMING! :) > ColumnStore Or something. There... [13:41:04] elukey: aye [13:41:20] elukey: let me see if i can test adding partitions to an eventstreams topic real quick and see what happens... [13:41:40] and elukey yes, i want +1 so it doesn't look like I just self merged it :) [13:41:47] ottomata: I checked net constants and only hadoop_masters is avaiable, sorry I was convinced that we had all :( [13:42:01] yaaa, i checked yesterday too :) [13:42:07] ottomata: mine was only a joke, hope that I didn't offend you.. I was about to +1 :) [13:42:36] haha didn't! :) [13:46:41] in the meantime I have the patches ready for migrating eventbus to deploy-service [13:46:52] worked with marko this morning [13:46:59] puppet should chown the files [13:47:31] so my plan is to disable puppet on kafka[12]00[123], then depool/run-puppet/restart-eventbus/check/repool each node [13:49:43] elukey: cool, thank you. [13:49:48] added comment on https://gerrit.wikimedia.org/r/#/c/371014/9 [13:49:51] but otherwise +1 [13:50:11] ottomata: I think it defaults to $title, so I skipped it.. but I can re-add [13:50:35] ah no that one is package name, checking [13:52:03] elukey: yeah, the service-name here needs to match what is installed for systemd in eventlogging::service::service { 'eventbus': [13:52:04] which is going to be [13:52:07] eventlogging-service-eventbus [13:52:51] aaaahhhh okok [13:53:06] brilliant, amending [13:59:09] 10Analytics-Kanban, 10User-Elukey: Calculate how much Popups events EL databases can host - https://phabricator.wikimedia.org/T172322#3515527 (10Ottomata) > @Ottomata, would you be able to demonstrate in detail how to rewrite this query in Hive (and possibly update these instructions in the process)? Yes, but... [13:59:41] man, things get stale so fast [13:59:41] i can't run eventstreams in mw vagrant anymore [13:59:42] unnGhhHh [13:59:44] whyyy [13:59:54] ottomata: I was thinking that another way to find exactly the client that is causing the exception in the kafka brokers could be to hack log4j on say kafka1012, restart and check kafka-request.log (that is atm empty) [13:59:57] noderdkafka, libsasl2, service-runner [13:59:57] ahhh [14:00:12] elukey: not a bad idea [14:00:32] I tried in labs and with TRACE it logs all the connections [14:00:45] it is so weird that we currently don't have a sort of access log for kafka [14:01:09] i think it would just be too noisy [14:01:13] but ya elukey that could be good [14:14:26] a-team: it occurs to me we're doing wikimania wrong [14:14:42] we can actually attend remotely, and instead we're sucking the people from wikimania back into real work [14:15:11] +1 [14:15:11] hm, I'm gonna check out this remote attend thing, people spent a lot of energy making it better this time around [14:15:26] several of our colleagues even donated their own money [14:18:02] I didn't get the point sorry :) [14:19:46] haha [14:20:14] milimetric: the hackathon seemed a little not for me yesterday. i kinda signed up to mentor, but of course no one was that interested in the stuff i had to mentor :p [14:20:25] so, i'm gonna skip/go late today [14:20:30] will go to real wikimania tomorrow [14:20:56] yeah, real wikimania seems to be start tomorrow [14:21:23] elukey: I'm just saying we can attend Wikimania without being there [14:22:00] ah okok :) [14:22:25] and I was referencing the collection of money to sponsor the recording and streaming that Victor and Melody sent out a while ago [14:22:47] and at the same time realizing that instead of talking about this we're all just working as usual [14:23:04] so I brought it up :) [14:27:26] elukey: i am currently unsure why, but eventstreams (and kafkasse) does not automatically pick up an added partition [14:27:35] if you force a consumer to reconnect [14:27:37] it will [14:27:42] but an existing connection will not pick it up [14:27:48] so we might need to restart the service when we add a partition [14:27:57] i'm going to look at code and try to figure out why its not doing that properly [14:29:10] ottomata: thanks! [14:58:27] ergh, elukey i'm pretty sure it won't reassing partitions because i'm using the low level consumer in order to avoid saving state on kafka cluster [14:58:31] so the rebalance event is not fired for it [14:58:35] since it never 'subscribes' to a topic [14:58:44] the consumer client is just assigned specific partitions when it starts up [14:58:45] hm. [14:58:49] adding partitions should rarerly happen [14:59:04] so, i think for our case, we should just add partitions and restart eventstreams [14:59:05] haha [14:59:06] but also [14:59:11] now we need to check what happens to mirrormaker [14:59:17] i'll make a ticket to fix this for eventstreams for now [15:01:57] super ignorant about consumer types [15:02:53] 10Analytics, 10Wikimedia-Stream: EventStreams (via KafkaSSE) does not consume from newly added partitions in topic - https://phabricator.wikimedia.org/T173006#3515824 (10Ottomata) [15:03:02] 10Analytics, 10Wikimedia-Stream: EventStreams (via KafkaSSE) does not consume from newly added partitions in topic - https://phabricator.wikimedia.org/T173006#3515837 (10Ottomata) p:05Triage>03Low [15:03:22] so elukey, we wanted eventstreams to save as little state on the brokers as possible [15:03:26] since it is open to the public internet [15:03:35] usually, subscribe will register a consumer in a consumer group with the brokers [15:03:46] however, assign(), can be used to manually assign the consumer's topic-partitions [15:04:09] in that case, the broker will (mostly) be unaware about consumer client to topic assignments, as it is managed on the consumer side [15:04:40] but, it also means that the broker doesn't have a way of triggering the consumer to reassign if the partition count changes [15:04:49] so yaaas [15:04:50] hm [15:05:00] mirror maker shoudl be fine for this [15:05:20] elukey: we are probably just considering increasing partitions for these topics on the analytics/jumbo cluster, ya? [15:05:22] probably not main? [15:05:53] AH standup [15:05:54] ! [15:05:56] i'm coming today forgot! [15:06:45] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515845 (10Halfak) As I see it, the outcome is to phase out the use of the multisource database commonly referred to as analytics-store. This involved determining what the m... [15:09:15] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515849 (10Halfak) @jcrespo, I thought it was clear from your comments in IRC that you'd like to begin phasing out the singular big multisource database host. I created this... [15:12:48] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515874 (10jcrespo) > Phase out and replace analytics-store / As I see it, the outcome is to phase out the use of the multisource database (many databases on a single host)... [15:15:24] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515883 (10Ottomata) > , there are other tickets about eventlogging in hadoop so let's please keep that conversation on those. +1. Just to be clear T162610 is about EventLog... [15:16:28] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515888 (10jcrespo) Also, please stop T172410#3507261 re-adding me to this ticket- when I explicitly asked to be left aside and have nothing to do with whatever you want to do. [15:18:14] 10Analytics-Data-Quality, 10Analytics-Kanban, 10Readers-Web-Backlog (Tracking): Pageview drop in ro.wikipedia hu.wikipedia and fr.wikipedia - https://phabricator.wikimedia.org/T170845#3444534 (10Nuria) a:03Nuria [15:18:33] ottomata: oh yes only analytics/jumbo, kafka main seems ok [15:19:28] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3515898 (10Ottomata) @jcrespo, earlier in this ticket, you said: > we are going to deprecate multi-source replication, so instead of one big fat server that is unmaintained,... [15:19:59] ottomata: interestingly enough https://grafana.wikimedia.org/dashboard/db/kafka?refresh=5m&panelId=34&fullscreen&orgId=1&from=now-30d&to=now&var-cluster=main-eqiad&var-kafka_brokers=All&var-kafka_servers=All looks similar to what we are seeing in kafka-analytics [15:27:28] elukey: that makes sense for the same reasons, single partition, single leader [15:27:35] so all consumers for a topic-partitions use same broker [15:38:10] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3515978 (10Nuria) mmm.. what is going on, i swear i saw those changes yesterday but they are no longer there? https://wikimedia.org/api/rest_v1/#!/Pageviews_data/get_metrics_pageviews [15:38:22] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3515983 (10Nuria) 05Resolved>03Open [15:38:28] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3510280 (10Nuria) ping @Pchelolo [15:44:26] 10Analytics-Kanban, 10Operations, 10Patch-For-Review, 10User-Elukey: Analytics Kafka cluster causing timeouts to Varnishkafka since July 28th - https://phabricator.wikimedia.org/T172681#3516024 (10elukey) Example of more verbose logging from kafka1012: ``` [2017-08-10 14:16:01,259] DEBUG Processor 0 liste... [15:48:22] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3516036 (10Nuria) 05Resolved>03Open [15:51:24] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3516046 (10Pchelolo) Pong @Nuria I've been in a hurry reverting #restbase to sorting stable because I've broke VE so I might have reverted it too far away. It will all get back today [16:07:32] 10Analytics, 10Operations, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3516106 (10Nuria) How did this added jaime again? Sorry. [16:07:43] 10Analytics, 10Research: Phase out and replace analytics-store (multisource) - https://phabricator.wikimedia.org/T172410#3516110 (10Nuria) [16:25:00] 10Analytics: add a more friendly message to ladp authentication box for pivot - https://phabricator.wikimedia.org/T163797#3210388 (10mforns) The message should be something like: Use your Wikitech LDAP login and add link to Wikitech that explains how to get access. [16:25:24] elukey: we have the new kafka nodes, ya? [16:39:39] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3516289 (10Nuria) Ok! let's close ticket when things are back to where they should be, thank you! [16:46:09] nuria_: ping [16:46:24] Pchelolo: hola [16:46:31] hola! [16:46:56] So I indeed reverted a bit more then I should have, I was in a bit of a panic [16:47:16] Pchelolo: do not worry, i have done that before more than once [16:47:18] Are you ok if we get it fixed on Monday? [16:47:32] Pchelolo: yes, it is NP, just leave ticket open so it is a remainder [16:47:41] that this is still wip cc mforns [16:48:10] Pchelolo: do concentrate on functional issues, these other can certainly wait [16:48:35] :] [16:48:37] 10Analytics-Kanban, 10Operations, 10Patch-For-Review, 10User-Elukey: Analytics Kafka cluster causing timeouts to Varnishkafka since July 28th - https://phabricator.wikimedia.org/T172681#3516339 (10elukey) >>! In T172681#3511453, @elukey wrote: > > {F8994394} > Interesting finding: https://gerrit.wikimedi... [16:48:44] oke, great, gracias [16:49:39] Pchelolo: o/ - if you have time can you check https://phabricator.wikimedia.org/T172681 and let me know if any idea pops up in your mind about what could be causing it? [16:49:47] not urgent, whenever you have time [16:51:40] elukey: I've seen these exceptions on eventbus kafka as well I think [16:53:23] Pchelolo: ah really? [16:54:51] the api version idea will certainly do nothing as node-rdkafka uses librdkafka 0.9.something and the api.version got implemented only in 0.11.something [16:56:46] Pchelolo: yeah there is a bug about that for 0.9.x (mentioned in the task) but should be solved specifing the version [16:57:01] weird thing is that the exception happens every minute at first second [16:57:10] like it was metadata related or something [16:57:37] the error started exactly on the 28th at 18:00 UTC [16:58:11] So is the issue really pressing? We can either way for the blizzard people to update to librdkafka 0.11 or I can make a PR for them to speed it up [16:59:02] s/way/wait [16:59:04] Pchelolo: not really pressing but it is affecting other producers like varnishkafka, so might be good to fix it asap.. https://gerrit.wikimedia.org/r/#/c/367740/1 was also deployed around that time (late EU evening on the 26th) [16:59:46] Pchelolo: main problem is that I am not sure where it comes from.. Eventstreams? Other? [17:00:09] That should not have affected anything, it basically rebuilt the same c++ code.. [17:01:10] I don't think I can provide any useful info without looking closer. Lemme poke around and I'll get back to you [17:01:27] ottomata might be a better sourse here though [17:02:27] we have already discussed this, not much came out.. Whenever and if you'll have time it would be really great [17:03:21] oke I'll find some time later today [17:04:26] thanks :) [17:10:14] * elukey going offline! [17:10:18] ttl [17:31:48] milimetric, nuria_, I looked into the wikimetrics problem and indeed seems a bug [17:32:16] don't remember the details of the code now, but I checked a couple users and they should be recognized by wikimetrics [17:32:19] but they are not [17:32:33] is community tech maintaining wikimetrics? [17:35:04] mforns: i do not think so maybe milimetric knows? [17:39:13] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3516502 (10Nuria) Closing ticket, eman to open restabse one [17:39:22] 10Analytics-Kanban, 10Patch-For-Review: Add Licensing info to analytics.wikimedia.org - https://phabricator.wikimedia.org/T172814#3516503 (10Nuria) 05Open>03Resolved [17:40:45] 10Analytics-Kanban, 10RESTBase-API, 10WMF-Legal, 10Patch-For-Review, 10Services (done): License for pageview data - https://phabricator.wikimedia.org/T170602#3516507 (10Nuria) Now opening correct ticket, this deploy was reverted (due to unrelated issues) . @Pchelolo to close when licensing code gets depl... [17:40:54] 10Analytics-Kanban, 10RESTBase-API, 10WMF-Legal, 10Patch-For-Review, 10Services (done): License for pageview data - https://phabricator.wikimedia.org/T170602#3516509 (10Nuria) 05Resolved>03Open [17:58:08] (03Draft2) 10MarcoAurelio: Add hi.wikivoyage to stats [analytics/refinery] - 10https://gerrit.wikimedia.org/r/371100 (https://phabricator.wikimedia.org/T173013) [17:58:14] (03Draft1) 10MarcoAurelio: Add hi.wikivoyage to stats [analytics/refinery] - 10https://gerrit.wikimedia.org/r/371100 (https://phabricator.wikimedia.org/T173013) [17:58:50] (03CR) 10MarcoAurelio: Add hi.wikivoyage to stats (031 comment) [analytics/refinery] - 10https://gerrit.wikimedia.org/r/371100 (https://phabricator.wikimedia.org/T173013) (owner: 10MarcoAurelio) [18:08:00] nuria_ + mforns: they haven't taken over yet, I think their plans were to re-write some parts of it, like maybe the UI? [18:08:15] but we still own it [18:08:25] I seriously doubt it's a bug though :) I can take a look later mforns [18:08:43] or... I'll take a look now, why not [18:09:04] (03CR) 10Nuria: [V: 032 C: 032] Add hi.wikivoyage to stats [analytics/refinery] - 10https://gerrit.wikimedia.org/r/371100 (https://phabricator.wikimedia.org/T173013) (owner: 10MarcoAurelio) [18:11:31] fdans: any issues with the piwik deploy? [18:12:27] milimetric, are you lookin? can I pair? [18:12:33] nuria_: no, it's deployed and working in the piwik dashboard :) [18:12:50] mforns: to the cave [18:23:21] 10Analytics, 10Analytics-EventLogging, 10Contributors-Analysis, 10EventBus, 10Community-Tech-Sprint: Add index to mediawiki_page_create_1 table - https://phabricator.wikimedia.org/T170990#3516713 (10Marostegui) [18:27:17] * fdans late lunch! 🇨🇦 [19:08:53] elukey: around? I couldn't find anything even remotely valuable regarding the kafka issue, but it seems the node-rdkafka is gettinng really close to using librdkafka 0.11 https://github.com/Blizzard/node-rdkafka/issues/232 [19:09:16] So I duess I won't spend any more time on this and we can just update and see what happens [19:09:33] yeah, i kinda agree [19:18:18] fdans: it has a php beacon in case js is not enabled but in teh case of this site you can just remove that one as w/o js the wikistats site does not work [19:18:26] fdans: let me know if this makes sense [19:27:32] 10Analytics-Kanban, 10Analytics-Wikistats: "see our contributing guide" link points to wrong place - https://phabricator.wikimedia.org/T173043#3516967 (10Nuria) [20:06:52] !log stopped Wikimetrics web and queue on wikimetrics-01.eqiad.wmflabs because the queue ran into errors connecting to the database (max 10 connections limit reached) [20:06:53] Logged the message at https://www.mediawiki.org/wiki/Analytics/Server_Admin_Log [20:19:55] 10Analytics, 10Analytics-Cluster, 10Patch-For-Review: Productionize Tranquility (or shut it off) - https://phabricator.wikimedia.org/T168550#3517248 (10Ottomata) Yargh. I keep getting stuck on Jackson databind dependency issues. This is the same issue that druid has when running Hadoop indexing task, that... [21:31:56] sorry nuria_ hadn't seen your msg... ok I'll remove the beacon and push it with the next deployment [21:32:09] fdans: thank you [22:01:34] (03PS36) 10Ottomata: JsonRefine: refine arbitrary JSON datasets into Parquet backed hive tables [analytics/refinery/source] - 10https://gerrit.wikimedia.org/r/346291 (https://phabricator.wikimedia.org/T161924) (owner: 10Joal) [22:05:29] (03CR) 10jerkins-bot: [V: 04-1] JsonRefine: refine arbitrary JSON datasets into Parquet backed hive tables [analytics/refinery/source] - 10https://gerrit.wikimedia.org/r/346291 (https://phabricator.wikimedia.org/T161924) (owner: 10Joal) [23:03:38] 10Analytics-Kanban, 10Analytics-Wikistats: Fix wikistats 2.0 footer links - https://phabricator.wikimedia.org/T173043#3517534 (10Nuria)