16:01:50 #startmeeting openstack_ansible_meeting 16:01:51 Meeting started Tue Nov 28 16:01:50 2017 UTC and is due to finish in 60 minutes. The chair is evrardjp. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:52 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:54 The meeting name has been set to 'openstack_ansible_meeting' 16:02:20 #topic rollcall 16:03:16 o/ 16:03:48 o/ 16:03:48 * mhayden woots 16:04:13 that's 4 of us! 16:04:26 o/ 16:04:28 5! 16:04:31 woot! 16:04:44 o/ but leaving in 10-15 mins. 16:04:46 I was wondering if I should change the time of the meeting, but 5, it's byzance! 16:04:52 ok let's go then 16:05:05 #topic last meeting action items 16:05:09 \o/ 16:05:50 last meeting action items were to deprecate monasca, searchlight, and freezer (the roles, not the projects or the french dragon ball character) 16:06:19 I did it. We still have a flaky telemetry stack, but that isn't on the agenda. 16:06:23 next 16:06:29 evrardjp: Was someone interested in picking up one of them? I think I remember someone popping up after the meeting 16:06:37 no news. 16:06:47 I went ahead and deprecated what I could 16:07:01 #topic liaisons feedback 16:07:16 anything to report? 16:07:41 odyssey4me: anything to say about infra? mhayden: anything to say about CVE-2017-16239 ? 16:07:52 Nothing really from docs, we need Jimmy to make the CSS change to mark Newton EOL. I've got a patch ready to go about the repo change ones we move it though 16:08:00 evrardjp: not too much there -- i'm not aware of any workarounds other than bumping shas 16:08:07 ok 16:08:11 then again, you must be an authed user to exploit it 16:08:47 ok, so no real urgency 16:09:03 we'll do our regular cycle of updates and we'll get the sha update 16:09:04 good 16:09:26 Merged openstack/openstack-ansible stable/ocata: Update all SHAs for 15.1.13 https://review.openstack.org/522856 16:09:27 Merged openstack/openstack-ansible stable/newton: Update all SHAs for 14.2.13 https://review.openstack.org/522858 16:09:31 evrardjp only issue for infra at this stage is that we're hitting some pretty bad timeouts for master & pike, so I'll see how we can work around it or resolve them in the next week or two 16:09:32 on the bug sides, we haven't done much bug triages this month, due to summit, zuul and many things. So sorry for that. 16:09:52 I have some ideas to combat them and will try out some things. 16:10:06 odyssey4me: Yes. I will work with you on that. If anyone has more ideas, I guess I'd say "shoot!" 16:10:27 next topic then 16:10:39 #topic specs 16:10:48 anything to be pointed out there? 16:11:11 I guess we have a few pending ones that need some love 16:11:26 but I mean I don't need to talk about one particularily 16:11:29 anyone? 16:11:51 I've done some work on the python simplification and need another review for https://review.openstack.org/521754 to help move that along. It's a start, but there's still a long way to go. 16:13:02 odyssey4me: I voted only a +1 to let ppl the time to process it too, so ppl, get acquainted to this wonderful work :) 16:13:21 anyone else? 16:13:26 1x 16:13:28 2x 16:13:31 3x 16:13:33 4x 16:13:47 5x! done! let's move to a different topic! 16:13:57 switch* 16:14:10 #topic Helping infra reducing their inodes outage 16:14:23 is there anything that should still be done there? I've seen we've added things back 16:14:38 or at least there was a patch in, IIRC. 16:14:46 odyssey4me: you worked on it, right? 16:15:28 evrardjp yeah, I need to rework it which may just involve removing the old stuff but haven't had a chance yet, so I'll revisit that 16:15:50 I don't think there is an urgency anymore, should I remove that from the meeting topics? 16:16:02 or you want to keep that as a reminder? 16:16:08 I can do an action task 16:16:11 you wanted to have a detailed discussion about what is appropriate or not appropriate to store - I just wanted to tarball everything and go from there 16:16:21 but nah, there's a review up - that's reminder enough to me 16:16:35 ok. Let's go from there. 16:16:53 #topic open discussion 16:17:52 need another review for https://review.openstack.org/522731 which is a pretty bad bug that needs fixing with it 16:18:43 looking 16:19:15 oh yeah, backport of this bad boy 16:19:25 Mind if i say something about the documentation? 16:19:27 Done and could have sworn I'd done that one already unless it was a backport 16:19:37 RandomTech: go ahead! 16:20:29 ive noticed the documentation here: https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html isnt very clear about what is actually required to use ceph. it seems to just give an example of one way you can 16:20:58 my team has been stuck on it for a while because of that 16:21:02 what would you do to improve it ? 16:21:19 it's a fairly new doc IIRC 16:21:33 oh no it's not 16:21:47 it would be nice to see a list of things that have to be specified in the diffrent files. either in comments or as their own table 16:21:48 the example config appendices are meant to be examples to illustrate what configs need to be in place to get it to a working state 16:22:37 RandomTech: https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html#user-variables and https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html#environment-customizations seem quite explicit 16:22:41 those are the minimum required - any extra tweaks or details would be found in the individual role docs, or role code - for ceph we use ceph-ansible so you'd also have to look there 16:22:47 is there anything missing? 16:23:09 Merged openstack/openstack-ansible-tests stable/pike: run_tests_common.sh: Add more openSUSE packages https://review.openstack.org/523347 16:23:35 it may be better to try and discuss where you're getting stuck to figure out the root cause based on your actual config 16:23:49 but IIRC logan- actually tested with that config 16:23:56 okay we were slightly confused as it say we MAY need those variables set. 16:24:04 not that we need to overide all of those 16:24:07 odyssey4me: RandomTech is giving us a first hand, new deployer, experience, so it's great to hear it 16:24:19 RandomTech: ok. 16:24:33 ah ok - so it's that initial line in user_variables that's causing confusion 16:24:34 RandomTech: but what if you don't want to use ceph for x and y? 16:24:41 that really should not be there in an example 16:25:01 yeah. 16:25:10 that's easy to fix 16:25:31 yup 16:25:35 catching up 16:25:37 good feedback :) 16:25:50 RandomTech: thanks! 16:25:55 that said, they are *just* examples - it is not the *only* way 16:25:58 no problem, still may take me a bit to figure it out fully though. got to get the fsid figured out 16:26:27 the command given is not agreeing with my shell for some reason 16:26:58 RandomTech: Can you bug this? I should have time thi weekend to do some stuff 16:27:16 i think its something on my end. its not liking str 16:27:29 RandomTech you could generate a uuid using anything 16:27:31 And I meant the doc confusion:) 16:27:34 that's just a suggestion 16:27:58 wait cant ceph generate one on install? 16:28:18 you need it in user_vars because the uuid is fed into the integrations 16:28:27 so yes it can, but then it won't feed back into OSa 16:28:32 ahh okay. sounds good 16:28:45 thanks guys 16:28:50 so you need to generate it first, using anything - even https://www.uuidgenerator.net/ if you like - then put the value in there 16:29:26 according to https://serverfault.com/questions/103359/how-to-create-a-uuid-in-bash there is a bash tool too 16:29:46 any other topic? 16:29:54 Yeah it is, we used to have to do it randomly at one place I wirked 16:30:07 maybe the docs could provide a few examples of generating one, rather than just one way 16:30:25 anyway, time for me to depart - 'night all 16:31:45 good night odyssey4me ! 16:31:49 I will fix it spotz 16:32:00 evrardjp: okie 16:32:05 ok let's close this up! 16:32:09 #endmeeting