[02:29:03] bd808: this was just merhged, please do take a look and see if it needs more additions: https://gerrit.wikimedia.org/r/#/c/analytics/refinery/source/+/538607/ [15:14:14] Hi! I want to replicate article histories of select 3-4 articles with their edit contents. I was planning to use quarry to get the data, but found out article text is not maintained there. Is there any other better way than wrangling with dumps? [15:31:01] codezee: you can use the rest or action api [20:05:04] Out of curiosity, do tools.wmflabs.org/ webpages get cached? [20:05:12] or is that up to the tool maintainer [20:21:21] !log git Upgrade all instances to linux-image-4.19.0-0.bpo.6-amd64 [20:21:24] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Git/SAL [20:53:26] Zppix: there is not a cache in front of tools.wmflabs.org currently. [20:54:56] Someone would need to do some research to figure out if adding one would be likely to be useful or not. [21:00:02] I'm seeing a lot of instability on the nginx web proxy - HTTP requests seem to be getting a random selection of the resource I'm expecting, a 500 Internal Server Error, or even ERR_CONNECTION_CLOSED from my browser. [21:00:25] (and this collection of different results is appearing for different resources on the same page load) [21:00:45] The specific endpoint I'm trying to reach is https://accounts.wmflabs.org/acc.php [21:00:59] andrewbogott: hi, I've not dealed recently with sentry project vms nor their configs on horizon [21:01:06] i was about to report the same thing stwalkerster mentioned [21:01:15] i get it from puppet-compiler.wmflabs.org [21:01:44] Framawiki: who should I ask instead then? Those VMs were created by you and are currently switched off. [21:02:22] stwalkerster, mutante I see that too at least some of the time [21:03:14] the last couple hours i used that without the issue but maybe an hour ago or so it started for me [21:03:57] andrewbogott: i'll check them later to see what they contains, if it's not urgent [21:04:15] Framawiki: not urgent, just would be good to cleanup unused VMs [21:05:35] Sure, 'll do, thanks for the ping [21:06:01] I've received a lot of down then up alerts for quarry.wmflabs.org today, is there some issues with wmcs infrastructure? [21:06:43] Perhaps same issue mentioned by mutante above? [21:07:44] Framawiki: same thing as above? nginx? [21:07:56] SSL error? [21:08:15] or The connection to puppet-compiler.wmflabs.org was interrupted while the page was loading.? [21:08:29] Same, all my VPS instances have had gone down and up over the past day or so [21:08:48] Perhaps front nginx shared by wmcs, I've not ssh access to debug [21:08:49] well.. VPS instance going down is pretty different [21:09:16] Framawiki: yea, it seems to be the domainproxy [21:09:40] Yeah could be the proxy, I'm going by automated reports that look for 200 OK [21:09:44] hm, http://www.koeri.boun.edu.tr is pulling zillions of map tiles [21:09:53] I think it's just more traffic than we can handle [21:11:54] university in Istanbul with an English page. but when you click email it's 404 [21:12:01] can we block them? [21:12:05] and then mail them? [21:12:31] I'm not sure I know how to block them but checking... [21:13:11] mutante: https://gerrit.wikimedia.org/r/#/c/operations/puppet/+/479041/ maybe [21:13:24] ah! [21:13:34] maybe he has a special user agent? [21:14:09] or instead of nginx config maybe we can just shut down his instance , heh :) [21:14:16] if you know where he come sfrom [21:35:21] mutante, stwalkerster, getting better behavior now? [21:36:21] much, thanks :) [21:36:37] andrewbogott: yes!:) [21:56:32] !log tools.wikibugs Updated channels.yaml to: 304a5e24d3b69bddb38cadc08a3c9a0a024e91de remove ##Zppix-Wikipedia from ZppixBot notifications [21:56:34] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.wikibugs/SAL