17:05:39 #startmeeting self-healing 17:05:40 Meeting started Wed Jan 2 17:05:39 2019 UTC and is due to finish in 60 minutes. The chair is aspiers. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:05:41 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:05:43 The meeting name has been set to 'self_healing' 17:05:46 #topic backlog review 17:06:00 BTW I noticed that storyboard has nice permalinks now 17:06:14 #link https://storyboard.openstack.org/#!/project/openstack/self-healing-sig for SIG backlog 17:06:40 #link http://eavesdrop.openstack.org/meetings/self_healing/2018/self_healing.2018-12-19-09.02.html last meeting 17:07:05 we didn't have a meeting this morning, but I had a quick chat with ifat_afek 17:07:38 in Dec she already delivered on the action from last meeting about submitting a spec 17:07:59 #link https://review.openstack.org/#/c/627180/ 17:08:23 I think that's all you missed recently 17:08:58 oh got it that’s great 17:09:32 very helpful work to document how to identify with monasca alarms based on dimensions. 17:09:40 yeah 17:09:44 will be useful even beyond vitrage. 17:09:52 I need to get my head around that, it's new stuff for me 17:10:34 just looking at what else is ongoing 17:10:51 damn, I still didn't send out a summary of Berlin 17:10:59 aspiers: I see you’ve signed up for quite a few SIG admin things. I can off load some if you’d like just let me know. 17:11:06 that would be great :) 17:11:21 I feel like we have to keep hammering away at https://storyboard.openstack.org/#!/story/2004537 17:11:36 It's gonna take regular publicity from multiple angles 17:12:06 maybe you could take 28285 or 28309? 17:13:04 great. 17:13:11 I can take both. 17:13:36 changed on board. 17:15:12 awesome thanks! 17:15:38 I guess it should be easy to finish https://storyboard.openstack.org/#!/story/2004536 17:15:52 I can do that now 17:16:30 of back to f2f events: if you already have events in mind to look into pls let me know so I can make sure I don’t overlook them in my search. 17:16:41 I don't have any in mind 17:16:55 not sure where to get visibility of that 17:17:03 maybe we can reach out to the UC? 17:17:18 I guess they will know about all the OpenStack Days and Meetups 17:17:33 ok great I’ll dig around and update. 17:17:36 I'm gonna ask on #openstack-uc now 17:17:42 ok 17:18:17 OK done 17:18:32 #action clarify on wiki page where to find docs / storyboard 17:18:56 hrm forgot to include action owner 17:20:10 #action aspiers to ask the UC about how to find out upcoming events 17:20:26 regarding https://storyboard.openstack.org/#!/story/2004535 I emailed the guy who proposed it but never heard back 17:20:36 so I guess it's up to us to submit this use case (task 28279) 17:21:02 I see3 17:21:18 I can try to find time for that, unless it's particularly appealing to you 17:22:14 I am interested. But I’m quite unfamiliar with that territory. 17:22:25 No less familiar than me I guess ;-) 17:22:55 It's just a matter of taking the template and replacing the boilerplate with the info about a memory leak scenario 17:23:08 cool I’ll take a first crack at it. 17:23:13 cool thanks! 17:23:26 the example he gave was a leak in ovs-vswitchd 17:23:32 https://etherpad.openstack.org/p/berlin-self-healing-sig-brainstorm line 53 17:23:38 but it could be any service really 17:24:39 I’ll need to read up on current practices on how to even detect memory leaks over time. 17:24:50 yeah 17:25:13 the use case doc doesn't necessarily have to go into details, at least not at first 17:25:19 I think it's OK if it's initially quite vague 17:25:24 ok got it. 17:25:35 we can flesh it out later, or add details in a separate spec if coding is required 17:26:00 maybe it can all be done simply using Vitrage + Mistral + $monitoring_component 17:26:31 it would still be nice just to have the idea documented, to get people's creative juices flowing :) 17:26:41 perfect. 17:27:17 I think we can abandon https://storyboard.openstack.org/#!/story/2003423 17:27:36 based on discussion in Berlin, or maybe it was Denver - I can't remember :) 17:27:41 yes that makes sense. 17:27:45 denver. 17:27:49 fungi suggested using survey.openstack.org instead 17:27:55 and we already have a task for that 17:28:14 OK cool, I'll add a comment and close it 17:29:13 #action find the Denver etherpad which covered the User Survey discussion, and close the corresponding story 17:29:22 doh, forgot to assign myself again 17:30:24 I guess we don't need to discuss https://storyboard.openstack.org/#!/story/2002684 17:30:33 since that's already in Ifat's very capable hands :) 17:30:58 next one is https://storyboard.openstack.org/#!/story/2002129 17:31:06 ugh, I have been the bottleneck on that forever :-( 17:31:25 need to resurrect that doodle poll 17:33:05 do you think my idea of a kickoff video conference still makes sense? I feel like the key to success here would be to get SUSE + Red Hat + one or two other companies to all donate ~0.5 engineers to the cause 17:33:28 to form a virtual team 17:33:48 without a bit of dedicated resource it seems unlikely it will happen any time soon, since it's not trivial 17:33:59 sampath would definitely join the effort 17:34:15 since he's probably the only one working on it right now 17:35:02 hmm. that makes sense. 17:35:18 maybe rather than spend the next 25 mins discussing the rest of the backlog, we could wrap up this meeting early and I could actually see how many things I can close off in 25 mins ;-) 17:35:29 for example https://storyboard.openstack.org/#!/story/2001628 should be low-hanging fruit too 17:35:31 not sure how best to take the initial approach tho. 17:36:06 well, we can at least try to get enough interested people in the same conf call 17:36:12 great. 17:36:16 if it works, great - if not, we tried :) 17:36:27 right. 17:36:42 alright 17:36:52 anything else you wanna raise right now? 17:37:01 nothing on my side at the moment =) 17:37:05 OK cool 17:37:18 I'll be around on this channel anyway 17:37:30 thanks! 17:37:36 #endmeeting