15:04:18 #startmeeting openstack-helm 15:04:19 Meeting started Tue Sep 4 15:04:18 2018 UTC and is due to finish in 60 minutes. The chair is portdirect. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:04:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:04:22 The meeting name has been set to 'openstack_helm' 15:04:41 ok - think it's gonna be a quite one this week in the build up to the PTG 15:04:57 o/ 15:05:05 the agenda is here: https://etherpad.openstack.org/p/openstack-helm-meeting-2018-09-04 15:05:07 do we have a meeting schedule somewhere? 15:05:17 ptg meeting 15:05:26 jayahn: an important clarification :) 15:05:42 jayahn: we have three days - wed to friday 15:05:49 https://web14.openstack.org/assets/ptg/Denver-map.pdf 15:05:52 mattmceuen had an etherpad started with topics i think 15:05:55 https://www.openstack.org/assets/ptg/PTG-Dulin-Schedule.pdf 15:06:17 lol whould help if google cached the right one - thx mattmceuen 15:06:48 do we know how many will show up? 15:07:02 (guesstimate) 15:07:18 i am leaving on Friday, so I will be able to attend wed and thur 15:07:21 So far on the etherpad we have 15:07:21 Attendance: 15:07:21 Matt McEuen (mattmceuen) 15:07:21 Rich Wellum (rwellum) 15:07:21 Pete Birley (portdirect) 15:07:21 James Gu (jamesgu) 15:07:30 But I know that's not the full list :) 15:07:35 I think probably 5-10 people throughout the duration based on previous exp evrardjp 15:08:10 will you be able to spare some time from OSA? 15:08:11 I'll try to come, but I will be spread very thin, so if we have a schedule, that would help me :) 15:08:31 portdirect: OSA, Loci and others :) 15:08:40 #action portdirect to work out a schedule and send out by thurs this week 15:08:48 thanks portdirect 15:09:23 still on the PTG topic, what about having dinner together -- maybe hosted at the same location as OSA's team dinner? 15:09:40 the more the merrier -- war stories to share 15:09:43 evrardjp: that would be great, do you have a venue yet? 15:09:56 nope, previous one closed, but we have a short list. 15:10:42 Will probably do that on Wednesday, as there is an activity planned by the foundation on Thursday (game night) 15:10:45 iirc 15:11:01 that sounds great - I'll certainly try to make it regardless 15:11:03 what is OSA? 15:11:10 jayahn: openstack-ansible 15:11:15 ah.. 15:11:25 jayahn: :) 15:11:29 we stole their abbreviation ;) 15:11:51 certainly 15:12:05 We are not into fancy names, we are just caring about our stuff working and improving :p 15:12:17 ++ 15:12:18 so many names it went through 15:12:26 SamYaple: :) 15:12:40 SamYaple: final one is official! 15:12:50 anyway 15:12:54 agreed, its the best one. /derail 15:13:07 sorry for the interruption. I have said everything I wanted to say on this topic :) 15:13:07 mattmceuen: you want to talk about the fun topic of copyright? 15:13:30 yessir 15:13:32 ok 15:14:04 So as I mentioned last week I changed the copyright assigned to the "OpenStack-Helm Authors" to the OSF 15:14:27 We'd discussed and I had it on good authority that that was the right path 15:14:41 However the larger OSF found some issues with that approach 15:14:44 So we reverted 15:15:18 We are back to OpenStack-Helm Authors, and everyone please follow your corporation's or your personal copyright policy as applicable when editing files going forward 15:15:28 that is all 15:15:58 honestly - I wish we did not have a (c) header in everyfile, as they are virtually meaningless - but such is life 15:16:02 thx mattmceuen 15:16:07 portdirect: agreed 15:16:59 mattmceuen: ideally we should review with -1 new files that don't hold the proper ownership 15:17:30 evrardjp: in your understanding should all files have the licence, and a (c) applied? 15:17:41 maybe - but, I think the idea of "proper ownership" differs depending on the author 15:17:53 i.e. certain companies have rules and others don't care? 15:17:54 yes, with the individual (if individual work) or company name 15:18:07 mattmceuen: one or the author, we cannot validate that 15:18:13 or the other* 15:18:39 but something that looks like having signed either a ICLA, or a company's equivalent 15:18:50 well I am not a lawyer 15:19:01 :p 15:19:04 thankfully none of us are 15:19:12 though mattmceuen's getting close :P 15:19:22 you take that back.... 15:19:23 but that's my understanding :) Maybe we should point to the foundation wiki 15:19:34 portdirect: :) 15:19:39 we should prob have an effort to point to the wiki/docs in our own 15:19:50 you ok with getting that in the appropriate place mattmceuen ? 15:20:37 I will take a look and see if I can find it - if not I will seek evrardjp out since he is even closer to being a lawyer than me 15:21:09 awesome 15:21:26 ok to move on? 15:21:50 ok 15:22:04 #topic LMA - Hows it going? 15:22:11 srwilkers: think the floors yours 15:22:16 oh hey 15:22:23 LMA's coming along -- have a few things in flight 15:22:50 now that ceph is living on openstack-helm-infra, we can properly exercise the work ive been doing with getting elasticsearch to use the ceph radosgw s3 api for snapshots 15:22:58 patchset can be seen here: https://review.openstack.org/#/c/559417/ 15:23:23 notably: http://logs.openstack.org/17/559417/46/check/openstack-helm-infra-dev-deploy-ceph/6405246/primary/pod-logs/osh-infra/elasticsearch-register-snapshot-repository-rphtk/register-snapshot-repository.txt 15:23:28 so that's exciting 15:23:47 also 15:24:35 we've still got the changes in flight for pruning the metrics we gather with prometheus by default -- would like to get these shored up properly next week in denver, along with getting some in-person feedback on the docs ive been writing up 15:24:47 along with the list requested wrt the exporters and metrics we've got across the platform 15:25:00 jayahn: i know you mentioned you all would look at prometheus 15:25:11 jayahn: any luck there yet? 15:25:15 yes, we do 15:25:46 that's awesome srwilkers 15:25:56 we reviewed list of metrics from most of exporter, and now reviewing some of alert rules.. 15:26:16 any feedback yet? or still in the process of reviewing it? 15:26:51 I have documents on wiki, can share if I can find a place to copy "wiki docs" to 15:27:14 jayahn: awesome 15:28:02 im looking forward to talking more about it at the PTG next week 15:28:10 that's most of what i've got for the LMA stuff 15:28:10 sure thing 15:28:20 would having a dedicated LMA session at ptg help here? 15:28:30 say a half day? 15:29:04 portdirect: it would 15:29:23 portdirect: ill work on getting topics added to the etherpad as a starter if that helps 15:29:37 Setting aside a half day on our schedule might also help bring in anyone outside the team who wants to talk LMA 15:30:19 #action srwilkers: Set aside 0.5 days for LMA review and planning at ptg (also produce outline for topics/structure) 15:30:59 i know there's some talk on the ML wrt fluentd and triple-o 15:31:12 but can deal with that later 15:31:39 lets find out what time they are free, always good to get some extra input/help 15:31:44 we done here? 15:31:50 portdirect: yep 15:32:20 #topic push for 1.0 15:32:32 (i added this cheekily) 15:32:52 keep on pushing! 15:33:05 there is a lot of low-hanging, but disruptive items on out 1.0 list of tasks 15:33:12 eg moving charts to the correct repos 15:33:25 I started this, but it would be great if someone could give me a hand 15:33:37 i think its pretty much essential that we have this done by monday 15:34:05 any volunteers to help? 15:34:52 portdirect: i'll help 15:35:18 portdirect: ceph moved pretty quick. if we can get ceph moved quickly, i can't imagine anything else being super painful 15:35:44 great - thanks srwilkers 15:35:59 lets get that done tonight/tomorrow 15:36:04 portdirect: works for me 15:36:19 so - and moving full circle 15:36:31 #topic Reminder: PTG Etherpad 15:36:34 ;) 15:36:42 https://etherpad.openstack.org/p/openstack-helm-ptg-stein 15:37:14 please can people add anything they want discussed here by wed 15:37:34 and I'll try and get a slot (at least penciled) in 15:38:36 yep. i'll flesh out the particulars for the LMA stuff. i've got some basics up there for placeholders 15:40:46 and with that, im fresh out of topics 15:41:10 as always we have a few reviews that could do with some eyes: 15:41:25 https://www.irccloud.com/pastebin/7YRs6Fku/ 15:41:39 please go forth and click 15:42:03 any more people want to add? 15:42:10 missing a line? 15:42:31 sorry my bad 15:43:01 ok 15:43:08 #topic Roundtable 15:43:37 I've got one 15:44:16 I'm currently looking at security in OSH - and comparing this to the openstack security guide 15:45:03 a lot of this is not quite applicable in a k8s world 15:45:37 evrardjp: (or anyone else) how have you handled deltas from the 'canonical' source in practice? 15:46:08 I had thought of making some patches over there - or would they best be kept 'in tree' as amendments to the guide? 15:46:42 portdirect: could you further explain what you mean? 15:47:28 yeah - for example there is a standard mask requested for config files 15:47:45 to prevent modifiction and being read by non-service users 15:47:49 even if the technical aspects might be different, some ideas of the security guide still applies 15:47:52 ^ this is hard to do in k8s 15:48:07 but we can (and do) mount the config files read-only 15:48:33 and also enforce only the service user to be able to run things in the container (this is in flight atm) 15:49:01 so if we had an audit that followed the 'letter' or the security guide, we would fail 15:49:12 I think the project can come up with a series of internal procedures for reviewing and code submitting -- and that's up to the project. We can then align with the evolution of the technology or expertise? 15:49:12 though we would meet the intent i think. 15:49:55 portdirect: oh many projects would probably fail,but failure is not a problem, the idea is to go forward 15:50:43 I am not sure if we should debate this in the meeting, but I am pretty sure some deltas can be documented in the project, and we iterate? 15:50:57 sounds great evrardjp 15:51:11 I'll try and formulate my thoughts better, and bring them to the ptg ;) 15:51:15 for the technicalities, I am still new, and I will be focusing on that in the future, I just don't know when yet. 15:51:38 any one else have anything they want to bring up? 15:52:06 im godo 15:52:08 *good 15:52:19 nice 15:52:27 * portdirect goes to find endmeeting command 15:52:52 #endmeeting