[05:36:17] <_joe_>	 fsero: well it all went pretty well, after all?
[07:45:52] <elukey>	 morning! I'd like to deploy https://gerrit.wikimedia.org/r/#/c/operations/puppet/+/514656/ today if nobody opposes
[07:46:30] <elukey>	 it seems working fine on canaries
[07:47:37] <elukey>	 to avoid any service disruption I could go with one mw host at the time, depool/run-puppet/pool
[07:47:51] <elukey>	 or even small batches
[07:47:56] <moritzm>	 +1
[07:48:39] <elukey>	 thanks :)
[11:08:52] <elukey>	 ok so mcrouter restarts almost done
[11:09:05] <elukey>	 the codfw proxies are still to be completed
[11:09:27] <elukey>	 I did only one and the impact (from the mcrouter's metrics) was minimal
[11:09:38] <elukey>	 same thing reported from Manuel from the db side
[11:09:51] <elukey>	 will do the rest after the SWAT
[12:18:22] <elukey>	 as FYI, two codfw proxies are still to restart. Need to run errand for a bit, will complete the work when back :)
[12:18:32] <elukey>	 (puppet will be disabled for this time frame)
[12:20:04] <_joe_>	 thanks :)
[12:20:14] <_joe_>	 and good to know
[12:20:24] <_joe_>	 lmk when puppet is enabled again
[15:03:11] <moritzm>	 ottomata: we currently don't have the eventschemas servers covered at https://wikitech.wikimedia.org/wiki/Service_restarts, is there more to restarting/rebooting them other than a standard depool/repool?
[15:26:26] <ottomata>	 Nope!
[15:31:15] <moritzm>	 ok, I'll amend the service restarts page, then
[15:31:22] <ottomata>	 i'm updating now moritzm 
[15:31:23] <ottomata>	 !
[15:31:28] <moritzm>	 ok!
[15:32:02] <ottomata>	 i also updated the Druid stuff, that was out of date
[15:32:09] <moritzm>	 ack
[16:35:51] <_joe_>	 jijiki: are you attached to my tmux on cumin1001?
[16:36:23] <_joe_>	 it just closed...
[16:40:41] <jijiki>	 yesh, my keyboard fucked up 
[16:41:01] <jijiki>	 because I hadnt replaced the batteries
[16:41:39] <jynus>	 oh, it happened to me once, very annoying
[16:42:11] <jijiki>	 _joe_: I am looking at which servers are remaining 
[16:42:22] <jijiki>	 and I will continue the restarts
[16:42:55] <jijiki>	 I am using the laptop keyboard so this should be fin 
[16:43:00] <jijiki>	 e
[16:43:03] <_joe_>	 jijiki: I have an idea about that
[16:43:11] <_joe_>	 let's deploy my nrpe check
[16:43:16] <_joe_>	 and see what warns/alerts
[16:43:26] <_joe_>	 then you can restart on those
[16:46:27] <jijiki>	 they  are 11 servers
[16:46:29] <jijiki>	 sure
[16:46:34] <jijiki>	 ok let me look at the patch 
[17:00:59] <_joe_>	 the last version works better :D
[17:02:49] <jijiki>	 lol I thought you'd go with uptime 
[17:02:56] <jijiki>	 :p
[17:03:07] <_joe_>	 no hits is more accurate
[17:03:20] <jijiki>	 ok should I merge it ?
[17:03:34] <_joe_>	 if you wish
[17:04:05] <jijiki>	 let me do a verification
[17:05:16] <wikibugs>	 10serviceops, 10Citoid, 10Release Pipeline, 10Core Platform Team Backlog (Watching / External), 10Services (watching): Figure out how to test Citoid with Zotero in the pipeline - https://phabricator.wikimedia.org/T225236 (10mobrovac)
[17:07:51] <jijiki>	 lol _joe_ I was running the compiler catalogue :p
[17:08:08] <_joe_>	 there is an error, not caught by the compiler
[17:08:21] <_joe_>	 fixing it
[17:40:51] <_joe_>	 I'm ashamed of myself, I did a gross miscalculation. We have servers in warning that shouldn't be (the opcache hit ratio)