16:02:34 #startmeeting openstack_ansible_meeting 16:02:35 Meeting started Tue Aug 6 16:02:34 2019 UTC and is due to finish in 60 minutes. The chair is mnaser. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:02:36 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:02:36 #topic office hours 16:02:38 o\/ bonjour 16:02:38 The meeting name has been set to 'openstack_ansible_meeting' 16:02:49 o/ 16:02:59 bonjour 16:03:06 rholloway: I guess you could, but it might require some manual work. We are all doing ansible behind the scenes. We just have special modules and plugins. 16:03:13 o/ 16:03:18 but it's meeting time now, so le'ts talk about that later :) 16:03:20 o/ 16:03:30 yea it would require extra work, I shall be silent for a bit while you all smarter than me do important things 16:03:41 you're welcome to join :) 16:03:48 the more the merrier :) 16:04:02 i think we all have common interest in how we drive the deployment before the OSA part 16:04:07 i pushed this up to the mailing list btw -- https://etherpad.openstack.org/p/PVG-OSA-PTG 16:04:16 i could swear evrardjp made one before 16:04:35 but i lost it and couldnt google it anywhere (even using google-fu with site:eavesdrop.openstack.org) 16:04:35 I can recall it either... 16:05:41 I have maybe in bookmarks, but I said noonedeadpunk should bookmark if I forgot mine :) 16:05:42 hahah 16:06:12 * noonedeadpunk looks through bookmarks 16:06:54 https://etherpad.openstack.org/p/osa-shanghai-summit-planning 16:07:07 let's keep yours mnaser 16:07:20 we'll just add to the official one what we think we should add :) 16:07:23 ah i guess i can copy pasta 16:07:30 i emailed that one out 16:07:32 yeah, selective maybe 16:07:50 take the one you emailed as ref :) 16:07:56 mine was just a memo more than anything 16:08:39 ok voila 16:08:52 yup just added a few ones I can think of right now 16:08:56 i have someone figuring out logistics to see if i can get noonedeadpunk and guilhermesp to be there 16:09:04 that's awesome 16:09:10 that will be quite a trip for guilhermesp :) 16:09:17 everywhere is quite the trip for guilhermesp 16:09:23 that's true :) 16:09:24 :P 16:09:24 anywhere he goes its 30 hours .. i dunno how :) 16:09:50 btwf 16:09:52 or those interested 16:09:59 i would try to start the visa process early 16:10:11 it's a good 10 pages or so to fill ut 16:10:27 mnaser: I guess I'll be attending anyway (btw logistic is quite straighforward for me, since I've been in Shanghai several times) 16:10:42 TIL :P 16:11:06 so i anticipate will be there but i dunno about others 16:11:08 jrosser: ? 16:11:14 we struggled to get you out to denver :) 16:11:29 it's fairly well beyond my travelling appetite really 16:11:36 i figured as much 16:11:37 :p 16:11:49 jrosser: :) 16:11:54 logan- / spotz ? 16:11:55 hahhah evrardjp yeah, would be nice to go through pacific and not atlantic 16:12:06 hola 16:12:14 is shanghai seemingly a thing for you? :) 16:12:17 o/ 16:12:28 I am not planning on it no 16:12:37 ah okay fair enough 16:12:37 same 16:12:51 that's sad:( 16:13:19 My travel tends to be on the graciousness of others 16:13:34 we could open the video conferencing for those who don't ~join~ sleep 16:13:51 yeah it seems like we almost have no one going to be there at this point :( 16:14:02 * chandankumar is going to shanghai 16:14:18 well, no one as in like.. not much people :p 16:14:21 a much smaller audience 16:14:23 I know if I am coming I will be split into many small pieces before/during/and after the travel 16:14:40 yeah likewise.. so maybe we shouldnt ask for that much time for OSA 16:14:44 I think most groups will have a small prescense if any 16:15:25 Space is a premium so yeah if you only need a small chunk better to under commit then over 16:15:27 I guess one of the session proposed on the etherpad could have the presence of chandankumar so that's a plus :) 16:15:45 I mean NEED the presence of chandankumar 16:15:48 :D 16:16:15 i assume MAYBE a few ex-rackers that used to work on OSA might be there 16:16:16 evrardjp: mnaser we started devstack OSA thread on openstack-discuss does something happened with that? 16:16:18 we would be honored to have his presence all along but I suppose he is busy in other places too :) 16:16:27 but they'll be busy with their own projects i think 16:16:27 evrardjp: :-) 16:16:43 chandankumar: i got grilled in that thread and it seemed like people weren't too excited about it :P 16:16:55 felt like an overwhelming amount of work (or it was painted as such) 16:17:08 mnaser: I think you mean "a few others ex-rackers"? 16:17:19 yes :p 16:17:21 detail 16:17:24 there are still some hanging around here, don't know why 16:17:31 hahahah 16:17:51 mnaser: :) 16:17:54 evrardjp: do we want to poke them ? to know their status 16:18:06 cloudnull: odyssey4me you are coming to shanghai ptg? 16:18:41 btw, evrardjp I can remember that Rico promissed smth in Berlin, but don't remember what exactly... And returning to ex-rackers, I hope it wasn't Stella... 16:18:45 Also andymccr 16:19:04 I doubt mhayden 16:19:14 jamesdenton? 16:19:19 * mhayden wuts 16:19:21 hi hey 16:19:24 noonedeadpunk: :) 16:19:31 mhayden jamesdenton - Shanghai? 16:19:42 I will be there. 16:19:44 Going through the list of ex-rackers besides evrardjp and me:) 16:19:51 cloudnull: woot 16:19:55 not at this time 16:19:59 but things could change 16:20:05 congrats on core cloudnull btw :) 16:20:12 jamesdenton: :) 16:20:33 cloudnull: or should I say good luck ? :) 16:20:51 I don't know yet for andymccr and asettle 16:21:12 what about d34dh0r53 ? 16:21:54 so cloudnull might be there but i guess he'll be busy with tripleo-y things 16:22:01 we had far more than those but those were deeply involved in osa :) 16:22:04 but it would be good to find sometime to talk cross-project discussions 16:22:12 yeah 16:22:53 I know there's been a lot of cross project goodness with tripleo and OSA, any with kolla-ansible? 16:22:55 I think I could use that time for containers-sig, as there are related work that doesn't fully touch the containers but touch deployment tools 16:23:37 i think we tried to reach out but the project seems to be happy with their tooling and not interested 16:23:44 spotz: yeah mark said hello :) 16:23:54 but that's the gist indeed 16:24:21 evrardjp: Say hi back:) 16:25:19 mnaser: so what's the long story then? 16:25:41 do we keep something, only for cross team collab? 16:25:56 i think tripleo/osa is a good point of collab 16:26:08 so we'll have something there indeed 16:26:28 the rest can be done remotely I guess? 16:26:46 and doesn't need attendance _per se_? 16:26:51 mnaser: regarding container tooling what about checking with tripleo team if they want to collaborate 16:26:58 i guerss so 16:27:03 mnaser: ok 16:27:09 thanks for clarification 16:29:16 I have to go folks, ttyl everyone! 16:29:25 take carte 16:30:18 ty! (ok will play with them! ok, french speaking joke...) 16:30:30 lolll 16:34:49 btw manila still fails:( https://logs.opendev.org/33/660333/9/check/openstack-ansible-deploy-aio_metal-debian-stable/6a73e41/logs/openstack/aio1-utility/stestr_results.html 16:35:03 does anyone have any assumptions why this may happen? 16:38:43 Tons of reasons found https://logs.opendev.org/33/660333/9/check/openstack-ansible-deploy-aio_metal-debian-stable/6a73e41/logs/host/manila-share.service.journal-16-16-39.log.txt.gz#_Aug_06_16_15_19 16:38:48 evrardjp thanks! 16:39:57 I suspect I will be there busy with tripleo things, however, OSA is now part of some of those tripleo-y things, so I'm sure we'll have some time to talk shop and have a few Stellas 16:41:39 if i create /host_vars/node1.yml .and i want it to be lxd, is just adding 1 line nova_virt_type: lxd enough ? 16:42:40 Ok, found it! 16:43:13 to be exact: /etc/openstack_deploy/host_vars/hostname.yml 16:44:30 mnaser: is there something existing on the etherpad there which would cover tempest plugins being installed from pip rather than source 16:44:43 i.e how we could consume source for more things that break us 16:46:00 if i understand correctly thats what is stopping this from getting further https://review.opendev.org/#/c/671769/ 16:47:03 jrosser: can you look at this? - http://paste.openstack.org/show/755578/ 16:47:26 why "neutron-sriov-nic-agent.service" service running here? 16:47:48 jrosser: well i see this as two ways 16:47:56 we can either fix our stuff so we can help gate tempest plugins that we consume 16:48:04 and continue pulling in source 16:48:07 I have queens deployment which doesn't have that service running there, look like only stein has this issue 16:48:09 and i think thats like a a win-win-win 16:48:11 Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Add image for manila test https://review.opendev.org/674653 16:49:52 mnaser: there have been a couple of occasions recently where we have tried to increase test coverage, octavia and novnc, and both have failed becasue of bringing in only released versions of some $thing 16:50:01 so if we can improve that, it would be good 16:50:58 jrosser: hmm not sure i am following yet :( 16:51:37 the adding novnc test to os_nova fails becasue we don't use the master version of the tempest plugin for that, i think 16:52:39 maybe i understand this wrong? 16:52:51 jrosser: no that makes sense 16:52:58 this goes back to us pinning things inside master 16:53:03 idk if thats the best thing to do 16:53:10 tempest itself we pin the SHA for 16:53:12 but not the plugins 16:57:26 hmm 16:57:31 yeah ok that makes sense 16:57:41 imho we shouold just leverage zuul for everything in our gates but idk 16:58:43 Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add integrated coverage https://review.opendev.org/660333 17:08:26 mnaser endmeeting? 17:10:06 noonedeadpunk: welp 17:10:08 #endmeeting