20:00:05 <johnsom> #startmeeting Octavia 20:00:06 <openstack> Meeting started Wed Oct 11 20:00:05 2017 UTC and is due to finish in 60 minutes. The chair is johnsom. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:00:09 <openstack> The meeting name has been set to 'octavia' 20:00:19 <johnsom> Happy dance, meeting bot is working here.... 20:00:36 <johnsom> Hi folks 20:00:42 <rm_work> o/ 20:00:45 <johnsom> #topic Announcements 20:00:57 <jniesz> hi 20:00:58 <johnsom> A few reminders: 20:01:00 <johnsom> Next week is Queens milestone 1 20:01:23 <longstaf_> hi 20:01:28 <rm_work> wat 20:01:30 <rm_work> already 20:01:31 <johnsom> I will be doing MS1 releases next week 20:01:41 <johnsom> Yeah, seems like time has flown by.... 20:01:58 <johnsom> Also, Newton EOL coming 10/18 20:02:13 <rm_work> poor newton, we hardly knew ye 20:02:24 <johnsom> I have made final releases of the stable/newton projects. 20:02:51 <johnsom> I have also closed out any open patches against newton. 20:02:56 <rm_work> we might even upgrade our cloud to newton before Q1! :P 20:03:03 <johnsom> Ha 20:03:05 <johnsom> ummmm 20:03:07 <johnsom> Yeah, 20:03:32 <rm_work> errr sorry, i meant mitaka, nevermind :( 20:03:37 <jniesz> our cloud is on Newton, but we are on Ocata release of Octavia 20:03:57 <johnsom> Also, in case you missed it, zuul v3 has been postponed again. The Jenkins gates have been a bit broken so most folks are working on those at the moment. 20:04:07 <nmagnezi> o/ 20:04:09 <rm_work> yeah i think the official recommendation is to run Octavia as close to current as possible 20:04:37 <johnsom> Yeah, good stuff is usually found in the recent releases 20:04:38 <nmagnezi> johnsom, can we expand on CI in the open discussion part? (just to recap the current state of things) 20:04:54 <johnsom> nmagnezi Sure thing 20:05:01 <nmagnezi> johnsom, thanks :) 20:05:12 <johnsom> Another announcement, we have a new core reviewer: nmagnezi 20:05:32 <johnsom> (in case you don't read the OpenStack developer e-mail list) 20:05:40 <johnsom> Congratulations Nir! 20:05:43 <jniesz> congrats 20:05:48 <nmagnezi> thank you :) 20:06:31 <nmagnezi> I'm excited to join this great team! 20:06:36 <johnsom> We could still use at least one more, so please do quality reviews and you too can become core! 20:06:57 * rm_work glares at JudeC :P 20:07:07 <johnsom> Hahaha 20:07:14 <johnsom> Any other announcements today? 20:07:25 <JudeC> :P 20:08:07 <johnsom> #topic Brief progress reports / bugs needing review 20:08:17 <johnsom> Ok, on to progress updates / patches 20:09:12 <johnsom> I have been working on a bunch of different things this week. 20:09:23 <johnsom> One of which is a database diagram for our docs: 20:09:25 <johnsom> http://docs-draft.openstack.org/19/509319/5/check/gate-octavia-docs-ubuntu-xenial/b91b593//doc/build/html/contributor/devref/erd.html 20:09:33 <rm_work> Could use comments on https://review.openstack.org/#/c/509933/ as to whether anyone actually wants something like that, or if we really should just "let stuff failover" even on major maintenances. 20:09:43 <johnsom> That patch is up for review: https://review.openstack.org/509319 20:10:11 <johnsom> The layout is by graphviz, so not ideal, but it is generated when patches merge, so it's always up to date. 20:10:45 <rm_work> And regardless of whether we actually want to do internal maintenance API stuff, it would be helpful to land either of these two options for at least being able to fetch the AZ info: https://review.openstack.org/510225 or https://review.openstack.org/511045 20:10:56 <johnsom> This patch for the 404 issue with nics not coming up is still open: https://review.openstack.org/507986 20:11:23 <rm_work> So, please comment, don't need to be a core to give good feedback (in fact, I'd *specifically* like to hear from some of the non-cores that don't speak up very much) 20:11:38 <rm_work> Ah, that one was on our side? 20:11:41 <rm_work> I'll look at it today. 20:12:16 <nmagnezi> johnsom, I asked there (https://review.openstack.org/#/c/507986/ ) how to test this one, I can review it next week (when I'm back from PTO) 20:12:39 <johnsom> Yes, reviews from non-cores are super helpful! 20:12:47 <rm_work> I feel like it's hard to verify it actually fixes the issue? but just... as long as it doesn't BREAK anything... i'm happy 20:12:53 <johnsom> Also, please have a look at the provider driver spec: https://review.openstack.org/#/c/509957/ 20:12:54 <rm_work> (just in this case) 20:13:40 <nmagnezi> maybe we should just that #link thing so all patches that needs attention will go into the meeting minutes 20:13:47 <johnsom> nmagnezi I see that. It is super hard to reproduce that issue. I have only been able to do it once locally. 20:14:02 <johnsom> Sigh, yes, should be using line 20:14:04 <johnsom> link 20:14:10 <johnsom> #link https://review.openstack.org/#/c/509957/ 20:14:21 <rm_work> #link https://review.openstack.org/#/c/509933/ 20:14:24 <rm_work> #link https://review.openstack.org/510225 20:14:26 <johnsom> #link https://review.openstack.org/507986 20:14:30 <rm_work> #link https://review.openstack.org/511045 20:14:34 <rm_work> I wonder if it lets me link 20:14:35 <nmagnezi> johnsom, fair enough. in that case I'll give it a shot and if it won't reproduce I'll just check that the code itself looks okay 20:15:16 <johnsom> You might be able to artificially trigger the code, but getting the 404 error to occur is just cycles and cycles of LB creates 20:15:31 <johnsom> probably not worth the time to try to reproduce it. 20:15:45 <johnsom> rm_work Everyone can link 20:15:50 <rm_work> k 20:15:50 <nmagnezi> okay. agreed. 20:16:10 <johnsom> Other than that, I am trying to get back to working on the active/active patches. 20:16:26 <johnsom> Any other progress updates? 20:16:55 <johnsom> #topic Open Discussion 20:17:18 <johnsom> nmagnezi Thanks for reminding me to use the link tag... It's been a bit of a day already... 20:17:29 <nmagnezi> johnsom, np :) 20:17:44 <johnsom> nmagnezi I think you had a CI question? 20:17:49 <nmagnezi> johnsom, yes 20:18:34 <nmagnezi> johnsom, I was (and still am) out this week so just wanted to know what is the latest with CI. I'm gonna review stuff next week and want to know if I should even try to recheck stuff 20:18:49 <nmagnezi> so sorry if I'm asking something and everyone else are already up to date 20:19:08 <nmagnezi> I understand we reverted back from the Zuul upgrade? 20:19:16 <johnsom> Yeah, so right now we are still running on zuul v2, but they want to switch over to zuul v3 soon (again) 20:19:40 <nmagnezi> and currently.. CI works okay? 20:19:41 <johnsom> I have not seen firm dates of when this is all going to happen. 20:20:07 <johnsom> So, my take is go ahead and try a recheck unless you see a channel topic or banner asking you not to. 20:20:23 <johnsom> I have seen some CI issues off and on the last few days. 20:20:53 <johnsom> Currently, until the zuul v3 switch, you will see results for zuul, but they are not actually voting until they make the switch. 20:21:20 <johnsom> I am watching the OpenStack developer mailing list to try to keep track of the state of things. 20:21:41 <nmagnezi> okay. So in the meantime I should expect this to somehow not block us from getting result 20:21:45 <nmagnezi> s 20:21:57 <nmagnezi> ack. will watch for that one as well 20:22:00 <johnsom> We can hope.... 20:22:14 <johnsom> The last zuul v3 switch basically blocked us for a few days. 20:22:52 <johnsom> I keep telling myself, it will be nice when we are switched over.... 20:23:28 <rm_work> yeah... i really am looking forward to it 20:23:35 <tongl> http://lists.openstack.org/pipermail/openstack-dev/2017-October/123337.html 20:23:36 <rm_work> and yeah, if "zuul" -1's something, ignore it 20:23:37 <nmagnezi> mm i guess we can hope here as well 20:23:42 <rm_work> only "Jenkins" counts right now 20:24:03 <rm_work> even if it says scary things like "Patch in Merge Conflict" like it did for me, Jenkins will override it 20:24:14 <johnsom> Yep 20:24:32 <tongl> It seems it will start rollback at 11:00 UTC today 20:24:37 <nmagnezi> #link http://lists.openstack.org/pipermail/openstack-dev/2017-October/123337.html 20:25:00 <johnsom> tongl That time has come and gone, they posted in the channel that they could not start. 20:25:24 <johnsom> So... who knows. 20:25:36 <johnsom> Other open discussion items today? 20:25:53 <nmagnezi> a minor question about the client 20:26:09 <nmagnezi> I recall we had some gap about setting quotas 20:26:23 <nmagnezi> did we manage to get in fixed by now? 20:26:34 <johnsom> Yeah, OSC is missing: status tree, stats, and quotas if I remember correctly 20:26:38 <rm_work> Oh, I have a client bug open that we also need to get in... 20:26:40 <rm_work> #link https://review.openstack.org/505851 20:26:54 <nmagnezi> rm_work, shameless plug :P 20:26:56 <kong> morning guys 20:27:01 <johnsom> Yeah, I need to look at that 20:27:09 <johnsom> hi kong 20:27:11 <nmagnezi> johnsom, where are those being tracked? I may try to find some cycles to help with those 20:27:17 <rm_work> :P we were talking about the client ;) 20:27:18 <rm_work> morning kong 20:27:33 <nmagnezi> can't promise just yet, but I want to try to get this fixed in queens 20:27:39 <kong> for q-1, i'd like to know if there is a chance for the tempest patch to merge 20:27:43 <johnsom> nmagnezi https://storyboard.openstack.org/#!/project/911 20:27:50 <nmagnezi> johnsom, thanks! 20:28:05 <johnsom> Looks like I dont' have a story for quota, so I will add that after the meeting 20:28:16 <nmagnezi> #link https://storyboard.openstack.org/#!/project/911 20:28:39 <rm_work> Ah, I have also had a couple of things waiting for a while, one which is a quality-of-life change to batch members... Another is the addition of support for PKCS12 certfiles, which I'd like to get in asap so people can get used to it 20:28:43 <rm_work> #link https://review.openstack.org/477034 20:28:48 <rm_work> #link https://review.openstack.org/504175 20:28:56 <rm_work> kong: we're working on it again today 20:29:12 <rm_work> kong: actually if you want to help speed that up, possibly jump in a video-conf with us in a bit? 20:29:32 <kong> not sure it's convenient for me in the office 20:29:43 <johnsom> Oh, one other thing, there are a few patches with "WIP" in the title but do not have -1 in the workflow. Could you please check that your WIP patches have -1 in workflow? 20:30:22 <johnsom> It helps with my patch review filter... 20:30:33 <nmagnezi> noted. 20:30:37 <johnsom> xgerman_ has this one: https://review.openstack.org/499807 20:30:54 <rm_work> kong: k np, actually i think we have a couple things to sort out first anyway 20:30:57 <johnsom> I will ping swami on his 20:31:04 <kong> rm_work: we could take about that specific issue after the meeting 20:31:21 <johnsom> Other topics today? 20:32:05 <johnsom> Ok, thanks folks! 20:32:07 <johnsom> #endmeeting