[00:01:23] !log wm-bot (logging) Restarted from #MacFan4000 where there was a functioning instance [00:01:26] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Wm-bot/SAL [14:11:03] dcaro: hi! I'm auditing icinga contacts that are using the legacy paging system, AFAICS you are getting paged via victorops so I'm going to remove the SMSes sent directly by icinga, seems reasonable ? [14:14:05] godog:yep, sounds great, thanks! [14:14:55] dcaro: np! [15:53:40] * arturo wasn't in this channel today for whatever reason [16:08:12] * qedk sighs at irccloud [16:21:24] arturo: irccloud had a hiccup, probably why you werent in here [16:22:00] yeah [16:37:11] Reposting for help: I have been running some queries in toolforge (grid) and it seems to fail for enwiki with `mysql connection got lost`error. Surprisingly the same query works perfectly fine in PAWS and locally (through ssh). Code and ways to replicate are here (https://phabricator.wikimedia.org/T272822). Can someone help troubleshoot why this is, I thought running in grid should be faster and better? [16:39:09] tanny411: 'Lost connection to MySQL server during query' generally means that the "query killer" running on the wiki replica servers themselves are closing your connection for taking too long. [16:40:14] A "LIMIT n" is not going to help with a query like this that contains a "group by" clause. All the limit does in that case is truncate the returned results, but the full result set must be computed by the database server first. [16:41:40] tanny411: one thing to check is which wiki replica server class you are querying. The "analytics" hostnames are the ones you need to use for long queries. [16:42:32] IRCCloud is really not having a good day, it seems [16:46:20] or a good week [16:46:22] AntiComposite: yes, they are -- https://twitter.com/IRCCloud/status/1354472963169538050 [16:46:56] * bd808 assumes all of AWS's processing power is being eaten by stonks trading [16:49:57] given their hatred for IPv6, it seems they may be using digitalocean's cutting edge VPSes 💀💀 [17:02:48] @bd808 @tanny411 fwiw it seems to run just fine on quarry: https://quarry.wmflabs.org/query/51954 [17:10:49] qedk: *nod* I'm pretty sure that indicates that the query is too long for the "web" replicas, but fits into the limits on the "analytics" replicas. [17:28:08] bstorm: hey, these two are the last ones for hiera() in wmcs: https://gerrit.wikimedia.org/r/c/operations/puppet/+/659336 and https://gerrit.wikimedia.org/r/c/operations/puppet/+/659141 [17:28:47] I'll take a look in a bit :) [17:38:37] Thanks [19:14:20] !log tools.lexeme-forms deployed cb0855af48 (simplify current_url) [19:14:22] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.lexeme-forms/SAL [19:44:08] Could anyone give me a pointer on how to extract commons structured data on a larger scale (e.g. getting the about 16 million camera locations). the `wb_property_info` table on `commonswiki.analytics.db.svc.eqiad.wmflabs` `commonswiki_p` is empty. Where is that data? I'd like to avoid running SPARQL queries via HTTP, unless the performance is [19:44:09] comparable to SQL queries. [19:44:35] !log admin shutting down cloudcontrol2001-dev because it's in a partially upgraded state; will revive when it's time for Train [19:44:38] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Admin/SAL [19:45:05] !halp :-) [19:48:03] !help [19:48:03] If you don't get a response in 15-30 minutes, please create a phabricator task -- https://phabricator.wikimedia.org/maniphest/task/edit/form/1/?projects=wmcs-kanban [19:51:12] [telegram] dschwen: it’s a good question, but I have no idea how structured data on commons works at the db level. I have a hunch that the data is not in the wiki replica databases and instead in the content ‘slots’ like wikitext is. [19:51:34] Ohh [19:52:09] I'm currently extracting location data through the external links table, but that's a bit of a hack [19:52:32] any image that has the structured data, but no location template (with a geohack link) won't get picked up [19:52:57] and I want to go with the times here since structured data is obviously the way forward [19:54:02] [telegram] You might try asking at https://commons.wikimedia.org/wiki/Commons_talk:Structured_data [19:57:14] Thx bd808, I've done that now. I was hoping a WMF cloud DB guru would know the answer... [20:03:53] !log tools.lexeme-forms deployed 868bccbbe7 (fall back to en) [20:03:55] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.lexeme-forms/SAL [21:34:50] bstorm: Thanks! It's just I can't merge it 😓 [21:34:58] if you have time and if you want to [21:49:01] sorry, this is the last wmcs batch, there's nothing left [21:55:59] Ah crud, I forgot that :) [21:56:02] Sorry Amir1 [21:56:10] I'll ge tit [21:56:12] *get it [22:43:57] anyone with admin access to PAWS about? someone found a file with a password in it [22:45:37] What sort of password? [22:46:06] SUL [22:46:47] bleugh [22:47:05] Just trying to remember what we normally do in these sorts of cases. Scramble the password and force them to reset IIRC? [22:47:16] a steward locked the account [22:47:47] so they get to deal with T&S now (https://meta.wikimedia.org/wiki/Help:Compromised_accounts) [22:48:07] but I'd like to see if someone can remove it from the file anyway [22:50:14] Hmm. no chico [22:51:15] AntiComposite: PM me the file location, I should be able to remove it in theory [22:53:20] PM'd [23:08:32] sorted [23:10:13] thank you