[08:27:55] <wikibugs>	 10serviceops, 10Operations, 10media-storage, 10Patch-For-Review, 10User-jijiki: Swift object servers become briefly unresponsive on a regular basis - https://phabricator.wikimedia.org/T226373 (10fgiunchedi) Since the 1s timeout change didn't seem to have changed things, could we revert it please?
[09:58:27] <mark>	 hey
[09:58:34] <mark>	 so today US people are out, including daniel
[09:58:36] <mark>	 joe is out
[09:58:41] <mark>	 and effie is maybe out
[09:58:51] <mark>	 do we want to do something with the meeting, e.g. move it earlier?
[09:58:54] <mark>	 or cancel it
[09:59:03] <fsero>	 i think it makes more sense cancelling it
[09:59:05] <fsero>	 IMO
[09:59:08] <mark>	 there's goals to do for sure
[09:59:10] <mark>	 but they are draft for now
[10:02:42] <apergos>	 whatever folks want, I have nothing urgent for discussion
[10:03:20] <mark>	 how about I write some draft goal language for what we discussed last week and add that to the etherpad
[10:03:32] <mark>	 and we'll see how controversial that is ;)
[10:04:04] <mark>	 i'll start with php7, taking the existing goal and modifying it
[10:11:52] <mark>	 well i took the existing goal and did not modify it :P
[10:12:44] <mark>	 akosiaris: ping?
[10:28:34] <akosiaris>	 o/
[10:28:47] <mark>	 hey
[10:28:56] <mark>	 so we need to better define the deployment pipeline goal
[10:29:09] <mark>	 i was thinking of giving it a title like "Self-service deployment pipeline"
[10:29:12] <akosiaris>	 we have one already? I don't remember writing anything
[10:29:15] <mark>	 and then a bit on documention, to be better defined
[10:29:33] <mark>	 and then a bit on "support migration of X services... by service owners"
[10:29:33] <mark>	 brb
[10:30:39] <mark>	 we didn't, we just discussed one
[10:31:05] <mark>	 does this summary sound good to you?
[10:31:13] <mark>	 and fabian, who already agreed to write documentation as well :)
[10:31:27] <akosiaris>	 sounds good to me
[10:31:32] <mark>	 ok
[10:31:41] <mark>	 i think we should just tighten up what we're gonna do on documentation
[10:31:48] <mark>	 also the current goal has "Increase documentation quality" which is... meh
[10:32:16] <mark>	 "Document procedure for service owners to deploy a new service onto the pipeline" - something along those lines?
[10:32:42] <akosiaris>	 sounds great actually
[10:32:50] <mark>	 ok
[10:32:52] <mark>	 anything else?
[10:32:55] <mark>	 trainings?
[10:32:58] <mark>	 dunno what you were already planning
[10:33:25] <akosiaris>	 probably 1. Maybe 2, but that would be about it I think
[10:34:14] <mark>	 Conduct 2 trainings for service owners?
[10:35:32] <mark>	 did we decide to do any cluster work itself or none at all?
[10:35:36] <mark>	 i'm not entirely sure anymore :/
[10:36:11] <mark>	 not much, that I know hehe
[10:36:19] <akosiaris>	 decided ? no I don't think so. We had some ideas, but I am not all sure we will manage to do much
[10:37:14] <mark>	 ok
[10:37:20] <mark>	 then let's not add it, maybe as stretch at most
[10:37:26] <mark>	 but you're also on the bacula stuff as well
[10:37:30] <mark>	 and the general service catalog stuff etc
[10:37:39] <mark>	 which we should really discuss on monday
[10:37:49] <akosiaris>	 yup. so little time to work on it :-(
[10:37:51] <mark>	 and we should probably define which services we will support
[10:37:56] <mark>	 yeah
[10:38:07] <mark>	 ok, i added it to the etherpad
[10:38:08] <mark>	 take a look
[10:38:13] <mark>	 and do you agree canceling today's meeting?
[10:38:37] <akosiaris>	 sure, with all these people missing sounds ok
[10:38:46] <mark>	 alright
[10:39:09] <mark>	 holy crap, i have only a single meeting today
[10:39:13] <mark>	 i might actually get some work done ;)
[10:39:21] <akosiaris>	 lol
[11:04:57] <vgutierrez>	 that must be a gcalendar bug or something
[11:04:59] * vgutierrez hides
[11:12:18] <volans>	 jijiki: I've a new scap release to rollout, any suggested procedure? I see you did the last one
[11:26:39] <jijiki>	 volans: there is lovely doc which makes it super fas 
[11:26:47] <jijiki>	 I can build the package for you 
[11:26:54] <volans>	 I've already build the package
[11:26:56] <jijiki>	 and you can roll it out 
[11:26:57] <volans>	 and uploaded to APT
[11:27:08] <volans>	 it's just for the roll out, I've already the debdeploy file for it
[11:27:17] <volans>	 just to know how we deploy and test it
[11:27:27] <volans>	 given there are no deploys today so no real-time feedback
[11:27:34] <jijiki>	 oh right 
[11:27:42] <jijiki>	 then I am not sure :/
[11:27:52] <volans>	 I guess codfw canaries, then codfw, then eqiad canaries, than eqiad
[11:27:54] <jijiki>	 but you can ping a swat deployer
[11:27:56] <jijiki>	 like amir
[11:28:04] <jijiki>	 yeah last time I did canaries 
[11:28:04] <volans>	 I can see if the DBAs have anything to do in mw-config
[11:28:08] <jijiki>	 and then I did fleet
[11:28:11] <jijiki>	 the*
[12:37:22] <wikibugs>	 10serviceops, 10Operations, 10ops-eqiad: Upgrade firmware on ms-be1021  (Was: Degraded RAID on ms-be1021) - https://phabricator.wikimedia.org/T227076 (10jijiki)
[13:28:42] <wikibugs>	 10serviceops, 10Operations: conftool: upgrade fleet to use existing python3-conftool - https://phabricator.wikimedia.org/T226965 (10Volans) The blocker above has been fixed in scap, released and rolled out to the fleet. I'll proceed with the removal of python-conftool.
[13:38:40] <jijiki>	 volans: thank you!
[13:38:52] <volans>	 yw :)
[13:59:45] <wikibugs>	 10serviceops, 10Operations: conftool: upgrade fleet to use existing python3-conftool - https://phabricator.wikimedia.org/T226965 (10Volans) >>! In T226965#5295407, @Joe wrote: > What we need to do is: >  > [x] Upgrade python3-etcd to the latest version > [x] Upgrade python3-conftool to the latest version > [x]...
[14:01:51] <wikibugs>	 10serviceops, 10Operations: conftool: upgrade fleet to use existing python3-conftool - https://phabricator.wikimedia.org/T226965 (10Volans) 05Open→03Resolved