[06:02:18] Why does "webservice stop" keep telling me "ValueError: get() more than one object; use filter"? [06:17:24] enterprisey: could be https://phabricator.wikimedia.org/T156626. could you paste the full traceback there? we are currently unable to reliably reproduce the issue [06:19:41] zhuyifei1999_: oh, I'm reliably getting it :) I'll post there [13:07:28] I am trying to setup a redirect on https://livingstyleguide.wmflabs.org/ and failing to do so. [13:08:09] Should I be setting 'Redirect permanent / www.example.com' on /etc/apache2/site-confs/devwiki/00-default.conf or /etc/apache2/site-confs/php5-devwiki/00-default.conf [13:08:15] Or is that the wrong way to do it completely? [13:31:09] prtksxna hi, see https://www.digitalocean.com/community/tutorials/how-to-create-temporary-and-permanent-redirects-with-apache-and-nginx which should help [13:31:32] Redirect / http://www.domain2.com [13:32:05] RedirectMatch ^/(.*)$ https:///$1 [13:39:13] andrewbogott: everything went well? [13:43:36] Yep, no changes as far as I can see [13:43:53] I haven't found anything awry yet [13:47:33] great, thanks guys :) [20:52:04] i'm constantly getting "OperationalError: (2013, 'Lost connection to MySQL server during query')" for a longer-running query on PAWS http://paws-public.wmflabs.org/paws-public/User:HaeB/new%20pages%20scratchpad.ipynb [20:52:24] (with constantly = 5 times in a row and longer-running = more than 1h) [20:52:30] anyone familiar with this? [20:55:26] iirc yuvipanda works on this... they dont seem to be on irc atm though [20:57:14] Zppix: yeah, yuvi is the expert regarding PAWS... but it may actually not be a PAWS-specific issue but an issue with database access on labs in general [20:58:45] HaeB: are you able to query the same thing but maybe split it so it doesnt take as long? [21:12:35] that's going to be difficult unfortunately (this is actually just a smaller test version of the actual query that's needed ;) [21:12:42] madhuvishy: any thoughts? ^ [21:36:29] HaeB: that's the long running query killer. I'm actually surprised you are getting to 60m before it cuts you off [21:37:06] ok thanks, i didn't know there was a limit like this (on the analytics servers they give us some more leeway ;) [21:37:35] *nod* the analytics cluster has far fewer users too :) [21:38:57] bd808: but labs has indices that are not available there :/ [21:39:47] it would be neat to have a hadoop cluster or something for Cloud Services that could handle really huge queries, but that's not on the roadmap right now. There is something related to "data lake" that the analytics team is looking to bring into our cloud next quarter but I don't know what that looks like yet [21:46:07] bd808: yep, we are already using that right now on the private servers https://wikitech.wikimedia.org/wiki/Analytics/Data_Lake/Edits and it's pretty cool for various purposes ... unfortunately it doesn't contain some of the tables needed in this case [22:10:33] bd808, madhuvishy: PS, I tried to document this here https://wikitech.wikimedia.org/w/index.php?title=PAWS&diff=1761125&oldid=1753303 [22:10:48] BTW is that for PAWS only or labs in general? [22:21:28] HaeB: its the Labs replicas generally that have the long query killers running