15:00:15 #startmeeting third-party 15:00:16 Meeting started Mon May 11 15:00:15 2015 UTC and is due to finish in 60 minutes. The chair is anteaya. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:20 The meeting name has been set to 'third_party' 15:00:24 hello 15:00:44 last monday meeting prior to summit 15:00:51 if anyone in additon to myself is here 15:01:10 hi anteaya 15:01:16 hello wznoinsk 15:01:16 * ctlaugh_ is here 15:01:22 welcome ctlaugh_ 15:01:31 how are you today? 15:01:35 hi 15:01:39 great, thank you 15:01:41 o/ 15:02:01 glad to hear it 15:02:09 do we have anything to discuss today? 15:02:42 I just wanted to make sure I know when all the third-party -related sessions are 15:02:54 good idea 15:03:03 not sure if asselin is about yet 15:03:06 hi 15:03:34 ctlaugh_, I have a partial list here -> https://etherpad.openstack.org/p/liberty-third-party-ci-working-group 15:04:03 hey asselin_ 15:04:08 hey 15:04:11 have you a link to the schedule? 15:04:26 ctlaugh_, there's a few infra sessions for the common ci solution: https://etherpad.openstack.org/p/infra-liberty-summit-planning 15:04:33 hey lennyb I need to discuss your account after we address the summit questions 15:04:49 asselin_: did that get posted to the main schdule yet, do you know? 15:04:54 anteaya: np 15:05:22 anteaya, not sure 15:05:32 not as of this morning 15:07:26 krtaylor: The session where these (https://etherpad.openstack.org/p/liberty-third-party-ci-working-group) were going to be discussed is what I wanted to make sure I had on my list. Is there a time / location for it? 15:07:50 #link http://libertydesignsummit.sched.org/overview/type/design+summit/Infrastructure#.VVDFqeQnFz0 15:08:05 ctlaugh_, it didnt get scheduled 15:08:32 the 3:10pm and 4:10pm sessions are thursday 15:09:12 the main focus of these two sessions is furthering asselin_'s work 15:09:32 ctlaugh_, it didn't make the cut, but we have the BoF to attend for cross-project topics 15:09:38 to refactor openstack's infrastructure modules so they are consumable by everyone 15:09:39 #link http://specs.openstack.org/openstack-infra/infra-specs/specs/openstackci.html 15:09:54 including openstack's infrastructure team 15:10:00 lot's going on now 15:10:06 this is really important 15:10:10 Sorry... BoF? 15:10:15 and we need to focus on this 15:10:22 and this 15:10:26 #link https://review.openstack.org/#/c/178887/ 15:10:29 asselin_: thanks 15:10:59 asselin_: I believe openstack's log server is now driving your module? 15:11:17 hi o/ 15:11:24 yes...fully merged and live! 15:12:06 yay 15:12:11 congratulations 15:12:14 hi fbo 15:12:22 anteaya, thanks :) 15:12:26 so this is a huge step forward for all of us 15:12:38 and is just the beginning of the work that needs to happen 15:13:00 so the more focus we can have on asselin_'s work, the better for everyone 15:13:07 #link https://storyboard.openstack.org/#!/story/2000101 15:13:08 remember these are working sessions 15:13:29 show up prepared to work, reviewing, testing, writing 15:13:35 stories are being tracked here^^ 15:14:45 nodepool & apache are looking for owners 15:14:49 oh hi, so what's the "Refactor apache" story about ? 15:14:53 I have to apologize to haven't started mine yet 15:15:09 I am committed to do something this week 15:15:16 devvesa: no apologies, and yay 15:15:39 devvesa: stay in communication and tell us how we can help you, or ask if you hit a block or mind blank 15:15:56 sure I'll do 15:15:59 tristanC, actually...that one may be a noop...but there for completeness 15:16:06 devvesa: thanks 15:16:30 asselin_: did you want to expand with some notes to that effect? just so we don't have a trip hazard 15:16:48 anteaya, ?? trip hazard?? 15:17:13 asselin_: trip hazard, something that is observable but only serves to show people that it is usless 15:17:23 time waster, noop that appears to be an actual task 15:17:29 I have more 15:17:44 asselin_: oh well, system-config has a dependency on puppetlabs-apache 0.0.4 which is pretty old.... I was wondering if it would be possible to work toward a more recent version. 15:18:14 anteaya, well...the task is really to investigate and see what needs to be done, if anything. 15:18:14 but it may be to much of a work for now 15:18:24 I don't consider that a trip hazard 15:18:44 asselin_: ah okay, you had said noop 15:18:58 tristanC newer versions lack features 15:19:26 tristanC, I remember reading that infra did not want to upgrade b/c the newer versions "do too much magic" 15:19:33 we have the openstack-infra "fork" of 0.0.4 under a dofferent name to avoid conflicts with newer apache module though 15:19:41 asselin_ and lack features 15:20:57 anteaya, may be noop.... 15:21:27 basic issue is 0.0.4 allows you to do arbitrary things to vhosts. newer module versions drop that support and push more thibgs ibto puppet resources which is sad times if you need large sets of mod rewrite rules for example 15:21:56 can we get some notes in storyboard to reflect this information? 15:22:23 and perhaps we can earmark this work for some work during the work times at summit? 15:22:43 clarkb: oh I see, thanks for the details. 15:23:32 anteaya, not sure storyboard has a place to add notes. The spec explains what needs to be done for each module. 15:23:41 (generically) 15:23:44 yes 15:23:55 this one seems to be a bit of an outlier though, yes? 15:24:59 or did I mis-interpret this conversation 15:25:07 anteaya, well...I haven't done a thorough investigation, so I should take back my noop statement until that is done. 15:25:39 okay can we identify this task as something taht should be discussed at summit? 15:26:47 or do folks disagree? 15:27:07 anteaya, I don't think this one needs particular attention until an investigation shows there is a need. 15:27:18 okay fair enough 15:27:23 let's move on then 15:27:39 well to be honest, functional testing of puppet module seems more important atm 15:27:58 tristanC: okay that has scheduled time as well 15:28:11 I find it hard to review other downstream-puppet changes without tests 15:28:24 thursday 11:50am 15:28:32 tests are good 15:28:56 clarkb: asselin_ we haven't started migrating things to use httpd yet, but we should 15:29:26 does that have an etherpad, plan or gerrit topic yet? 15:29:29 it sounds like testing will get agreed to in some form at a fishbowl session at the summit so after that we can get testing in place 15:29:44 well it will be discussed 15:29:48 anteaya: no, its much less important (imho) than the other puppety infra efforts 15:29:51 hopefully there will be agreement 15:30:16 okay, I find it hard to follow up on things without a reference of some sort 15:30:52 so if you feel migrating things needs to happen, and I support that, some form of reference to help people see status and know what to do next is helpful 15:31:07 anteaya, testing effort is being currently https://review.openstack.org/#/c/178887/discussed in the spec proposal: 15:31:40 #link https://review.openstack.org/#/c/178887 15:31:45 great thanks asselin_ 15:31:56 anteaya, I believe the fishbowl session is intended to get agreement on that spec 15:32:07 yes, that is my understanding as well 15:32:24 tristanC, in the meantime, I (and others) have been using the simple bash script approach to testing. 15:32:48 tristanC, better than nothing, but doesn't seem to be the desired long-term approach 15:33:04 e.g. https://review.openstack.org/#/c/169117/ 15:33:20 asselin_: oh well, I used those script to make sure manifest still applying after edits, so they are usefull but feels very temporary 15:33:38 #link https://review.openstack.org/#/c/169117/ 15:33:39 asselin_, do you think we can go with the simple shell script to test or wait for a proper selected solution ? 15:33:53 we are one week out from summit 15:34:06 and possible agreement on a solution 15:34:06 tristanC, fbo imho we need the simple shell script as a separate dependent patch to test 15:34:27 and we can abandon once the real approach is decided and implemented 15:34:43 asselin_, ok well I'll do that for the zuul manifest 15:34:46 can I encourage folks to review clarkb's latest spec? 15:34:52 for the log server...those tests found issues that we able to be resolved before merge. 15:35:06 and they're really not that much effort to write 15:35:36 so even if temporary, they're helpful 15:35:40 and please review clarkb's latest patchset for the testing spec 15:36:05 and definitely better than nothing 15:36:23 glad you are championing testing 15:36:26 thank you 15:36:48 do we have more to say in this topic? 15:37:21 does anyone have anything else they want to discuss about summit? 15:37:52 keep in mind the infra work sessions are ment for work, not for a huge crowd and general discussion 15:37:59 meant 15:38:12 are we done talking about summit? 15:38:49 okay, let's move on 15:38:54 lennyb: glad you are here 15:39:05 lennyb: can you look at your wikipage please? https://wiki.openstack.org/wiki/ThirdPartySystems/Mellanox_CI 15:39:06 who here's going to the summit? 15:39:29 I will be at the summit 15:39:30 anteaya: yes 15:39:37 asselin_, Unfortunately I won't be there 15:39:43 lennyb: do you see it no loger renders as a table 15:39:46 fbo: no 15:39:53 fbo: I thought for sure you would be there 15:39:57 fbo: :( 15:40:01 tristanC: will you be there? 15:40:09 anteaya: yes! 15:40:15 lennyb: no longer renders as a table 15:40:19 tristanC: great 15:40:21 whew 15:40:22 anteaya, yes. I will fix it. 15:40:27 anteaya, I apologize but be sure I'll continue to work on that effort :) 15:40:28 ctlaugh_, ? 15:40:29 lennyb: great 15:40:36 yes - I will be there 15:40:44 fbo: oh I sense you would be there if you could 15:40:45 question regarding latest gerrit update 15:40:48 devvesa, ? 15:40:54 fbo: I just really wanted to see you again 15:41:00 lennyb: go ahead 15:41:05 We have started to see the following error: ERROR zuul.GerritEventConnector: Received unrecognized event type 'ref-replicated' from Gerrit. any ideas ? 15:41:59 anteaya, Yes but I cannot for this time 15:42:02 I think that means you need to upgrade zuul? 15:42:15 fbo: understood 15:42:29 the gerrit version we just upgraded to has replication events on the stream events notifier. Your zuul doesn't understand those events but latest should 15:42:49 asselin_: yes. I'm going to the summit 15:43:03 I have the latest zuul from the github ( I think ) 2.0.0.post386 15:43:37 clarkb: what is the latest version of zuul? 15:44:05 oh if it is latest then maybe zuul doesn't understand, in that case it should be fine its would just discard the event if it doesn't understand it 15:44:49 lennyb: what is happening on the error, do you just see it in logs, or do things cease to happen? 15:45:32 anteaya: I believe just logs. 15:46:02 okay perhaps it is clarkb says then and zuul will just discard the event if it doesn't understand it 15:46:41 lennyb: is that fair? 15:47:26 anteaya: yes, I just want to minimize a chance of our false alarms :) 15:47:32 good call 15:47:46 and yes perfectly fine and encouraged to ask about new behaviour 15:47:51 well done 15:48:11 so after you fix your wikipage, link me once you have, please talk with nova developers 15:48:37 as I don't know what moshele had agreed when (s)he talked with jogo to get your account re-enabled 15:49:05 and I'm not supportive of status updates sent to mailing lists, they go on the wikipage 15:49:22 so I will be denying the post you sent to -announce 15:49:33 it is best if you talk to developers in the projects to test 15:49:43 so they know and agree with your decisions 15:49:47 does that make sense? 15:49:55 do you know anyone in neutron yet? 15:50:01 btw, the issue that we had with 'merge failure' was very strange. We already talked with nova developers including jogo, our account is enabled, but I am still running silent to see that there are no 'merge fail' comments that wi poet 15:50:17 great 15:50:36 so rather than communicating that to a mailing list, use your wikipage and irc channels 15:50:41 and thank you 15:50:47 ok 15:50:49 do you know anyone in neutron yet? 15:51:28 right now you aren't in either #openstack-nova nor #openstack-neutron 15:51:45 please join those channels and be available for the developers to talk to you 15:52:06 after this meeting I can introduce you to some neutron folks in #openstack-neutron 15:52:14 it would be good for you to meet them 15:52:19 okay let's move on 15:52:20 I know. it's technical irc client issue 15:52:32 lennyb: ah okay 15:52:36 anteaya: thanks 15:52:56 lennyb: do ask in #openstack-dev if you need a hand to get irc client issues sorted 15:53:05 it is an important tool to have working 15:53:11 moving on 15:53:23 does anyone have anything else they would like to discuss today? 15:53:45 reminder that next week's meeting in this time slot is cancelled due to summit 15:54:01 the next meeting in this time slot will be May 25th 15:54:18 nothing more to discuss today? 15:54:41 thanks everyone for your participation 15:55:03 if you can't be at summit, follow along on etherpads so you can stay involved 15:55:25 and since we have nothing more to discuss, let's end here 15:55:38 thanks everyone, enjoy the rest of your day/evening 15:55:43 see you in 2 weeks 15:55:46 #endmeeting