16:03:41 #startmeeting OpenStack-Ansible 16:03:42 Meeting started Thu Jun 9 16:03:41 2016 UTC and is due to finish in 60 minutes. The chair is mhayden. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:03:43 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:03:46 The meeting name has been set to 'openstack_ansible' 16:03:46 #topic Rollcall & agenda 16:03:53 o/ 16:03:56 * mhayden stumbles into the room 16:04:01 o/ 16:04:02 o/ 16:04:04 o/ 16:04:12 o/ 16:04:30 o/ 16:04:31 o/ 16:04:45 in another meeting so lurking 16:05:00 \o\ 16:05:52 * mhayden gives it a minute or two more since i started late 16:06:00 o/ 16:06:14 jmccrory: are you around? 16:06:15 o/ 16:07:35 okay, let's roll 16:07:39 #topic Review action items 16:07:56 #link Last week's meeting log -> http://eavesdrop.openstack.org/meetings/openstack_ansible/2016/openstack_ansible.2016-06-02-15.59.html 16:08:12 first two actions are about the mid-cycle -- we can get to those in a second 16:08:31 next was asettle and andymccr to send something to the ML about the testing repository 16:08:39 * mhayden remembers seeing that in his inbox 16:09:04 #link andymccr's repository -> https://github.com/andymcc/openstack-ansible-testing-core 16:09:11 asettle / andymccr: any updates there? 16:09:17 mhayden: mail was sent! 16:09:22 woah don't look at me aye. I edited. It got sent. Ta daaaa 16:09:22 :p 16:09:41 okay, so that one is still open -- has anyone had a chance to look over it yet? 16:09:44 it's still in my queue 16:10:31 #action Everyone review andymccr's email to see how we can help move that along 16:10:41 feel free to ping me if you have questions/feedback 16:10:56 im keen to gather, so far evrardjp has given some - which ive noted (in the email) 16:11:07 Next is reviewing phil_h's Astara etherpad 16:11:25 #link phil_h's Astara etherpad -> https://etherpad.openstack.org/p/astara-openstack-ansible 16:11:40 phil_h: looks like you got a good response in there! 16:11:46 I hope everyone has taken a look 16:11:56 speak now or forever hold your peace 16:12:02 We have tried to answer everyones questions 16:12:04 or at least hold your piece until the spec comes up 16:12:09 lol 16:12:10 *peace 16:12:12 :) 16:12:21 anything else you need from us, phil_h? 16:12:30 I don't think so 16:12:40 do you have enough to begin work on the spec? 16:12:44 I am looking at code and we will get a spec written 16:12:56 swee 16:12:58 t 16:13:03 spec is next 16:13:10 #action phil_h to start drafting the Astara spec soon 16:13:16 next action item is andymccr 16:13:18 errr 16:13:22 I will miss the next two meetings 16:13:29 next action item is antonym's Xen spec -- which actually merged today! 16:13:37 phil_h: we will miss you sir 16:13:45 antonym: nice going on the spec 16:14:02 mhayden: thanks :) 16:14:12 anything you need from us on the rest, antonym? 16:14:33 mainly just going to track how the multi os stuff is looking and try and replicate how that all looks for xen 16:14:47 i think the nova stuff was still in flight 16:15:02 quite a few of us have hacked on the multi-os stuff, so we will be glad to help point you in the right direction for things 16:15:07 just want to align it with how it looks for the other OS types 16:15:18 cool, i'm sure i'll be bugging some of you all 16:15:34 sounds good 16:15:50 #action antonym to get started on adding Xen support using the multi-os framework 16:16:11 next up, odyssey4me was working on Newton-1 tags and next tags for Liberty/Mitaka 16:16:46 jenkins appears to be unhappy about some things in the newton-1 tag -> https://review.openstack.org/#/c/325252/ 16:17:08 but the 12.0.14 and 13.1.2 releases did go out 16:17:21 odyssey4me: if you need a hand, let us know! 16:17:41 last item was evrardjp's review https://review.openstack.org/#/c/322249/ 16:17:53 which merged 16:17:54 hooray! 16:17:58 woo! 16:18:05 man this agenda is easy 16:18:17 its a commentary on how skilled you are mhayden 16:18:17 #topic Topics for Discussion 16:18:29 first up, the mid-cycle! 16:18:32 cue the fanfare 16:18:51 so automagically had a date for hosting at comcast, but he's out of the office at that time :( 16:19:04 i was able to get three dates for hosting at Rackspace's HQ in San Antonio, TX 16:19:10 * mhayden digs in email 16:19:20 Nice, nice 16:19:22 here are the options: 16:19:26 1) Aug 10-12 16:19:32 2) Aug 22-26 16:19:40 3) Aug 29 - Sep 2 16:19:51 (apologies to the UK folks who are confused by this date format) :) 16:19:58 mhayden: So we can have it in Philly but automagically wouldn't be there?:) 16:20:08 automagically said that Aug 29-Sep2 won't work for him 16:20:22 spotz: we could order some philly cheesesteak sandwiches to make it feel more like PA 16:20:44 does anyone have any strong feelings on these dates? 16:20:57 i'm good with #1 + #2, but #3 is not preferred 16:21:02 Uhhhh no strong feelings? 16:21:04 I'd say throw out #3 16:21:22 does anyone else wanna strike #3? 16:21:24 And I guess comcast is out as an option too 16:21:42 +1, #3 is not great for me 16:21:52 okay, so we could probably toss #3 16:22:06 Aug 10-12 is Wed-Fri 16:22:07 yeah if 2 cores can't make it it's no bueno 16:22:12 Aug 22-26 is Mon-Fri 16:22:36 Aug 22-26 conflicts with LinuxCon (Toronto) 16:22:47 I might be out 8/12 but you don't need me there 16:23:51 You don't need me either but I'll gate crash for fun 16:23:55 any opinions from the other core folks? d34dh0r53 jmccrory andymccr (forgetting a few here) 16:24:14 cloudnull - but he's still on vaca 16:24:17 August in san antonio is a little toasty, so be prepared 16:24:28 1 or 2 16:24:36 we can take this to the ML to decide, but i need to get back with my room booking person fairly soon :) 16:25:10 #action mhayden to send an email to the ML about choosing between Aug 10-12 or something during Aug 22-26 16:25:24 * mhayden is still working on an executive sponsor on this end for budget (food and such) 16:25:48 and we need funds for d34dh0r53's security detail 16:25:50 :) 16:25:54 8/23 and 8/24 openstack days east 16:26:03 ah, good catch mrhillsman 16:26:16 So looks like option 1 by default? 16:26:19 so it sounds like Aug 10-12 may be the preferred date -- i'll get something out to the ML to verify 16:26:43 we can work on agendas and stuff soon, i'm sure, but i'll put that ball in odyssey4me's court ;) 16:26:50 Party at mhayden's house! 16:26:51 are we good on mid-cycles for the moment? 16:26:57 spotz: that's possible 16:27:25 okay, next topics for discussion 16:27:38 we already got updates from andymccr, phil_h, and andymccr 16:27:56 we need an update on pip_lockdown from automagically, but he's not here 16:28:05 it looked to be moving well in reviews 16:28:25 I can randomly talk about pips and lockdowns if that suits? :p 16:28:25 it looked like there were maybe 3-4 reviews left that jmccrory was cleaning up 16:28:35 i guess we can table that for now 16:28:44 oh silly me, i haven't been changing the topic 16:29:01 #topic ppc64le mirror issue for mariadb (thorst and adreznec) 16:29:08 thorst / adreznec: you have the mic! :) 16:29:40 So we're testing out on ppc64le and the current repos point to x86 16:29:45 Yeah, so the issue here is that some of the package repos used don't support ppc64le for things like mariadb and percona 16:30:17 are we talking about the upstream mariadb/percona repos or something else? 16:30:23 Those packages do exist in the upstream repo for Xenial from Canonical/Ubuntu directly, but that's not where we're point today 16:30:35 Yeah, like https://mirror.rackspace.com/mariadb/repo/10.0/ubuntu/ 16:30:46 Containers amd64/i386 only 16:30:49 *Contains 16:30:56 oh, i see what you're saying 16:31:03 so they're missing from the rackspace repository? 16:31:07 but they're in other repositories? 16:31:21 Yep basically 16:31:34 so it sounds like we need to chat with the rackspace mirror maintainers to get that added 16:31:37 So we either need to figure out how to build them there or be able to point elsewhere, I guess? 16:31:41 i imagine they filter it out to save bandwidth/disk space 16:31:59 Yeah I thought that might be the case 16:32:04 adreznec: i'd imagine we can temporarily check for ppc and use a different mirror (short term) 16:32:06 the point elsewhere should be possible today 16:32:08 and get the rackspace mirrors fixed 16:32:28 can i get a racker to volunteer to reach out to the mirror team? :) 16:32:49 mhayden: Yeah, that's what we're doing for now in our tests. I know wangqwu is pointing to another mirror 16:33:02 Yeah that seems like a reasonable approach. No idea what the turnaround is on that from the repo update side 16:33:20 the turnaround on the rackspace side shouldn't be big, unless we're short on disk space 16:34:06 So for things that are pulled from the rackspace repo that should be no problem then 16:34:16 antonym: would you be able to sync up with the mirror folks about ppc on mirror.rackspace.com? 16:34:21 i know you have some ppc work going as well ;) 16:35:02 mhayden: yeah, we can ping them on that 16:35:07 antonym: thanks sir 16:35:10 There are some other things that are pulled from non-Rackspace repos like Percona that points to https://www.percona.com/downloads for the deb 16:35:29 #action antonym to talk to rackspace's mirror team about getting PPC packages added 16:35:43 okay, was there anything else on that topic, adreznec / thorst? 16:35:44 We'll have to work through those and point at alternate repos where needed 16:36:12 Nope, I think that's it 16:36:29 adreznec: lets sync with kylek3h on the Percona one 16:36:31 awesome -- thanks for your work on ppc support in OSA! :) 16:36:42 adreznec: ok 16:36:44 #topic Release planning and decisions 16:36:49 thorst: Sounds good 16:36:54 so we talked about 13.1.2 and 12.0.14 going out already 16:37:04 the last kilo tag (11.2.17) is still in progress 16:37:12 does anyone have any input on that ag? 16:37:15 s/ag/tag/ 16:37:50 i'm not sure if the appropriate kilo-eol tags have gone out on other projects 16:37:56 we'll leave that for odyssey4me 16:38:12 #topic Open discussion 16:38:21 asettle: anything on the docs side you want to update us on ? 16:38:30 hello! 16:38:31 (and pardon me for ending with a preposition) 16:38:53 Thanks to everyone contributing to the spec with their thoughts and comments - it has been merged 16:39:11 As of today also, darrenc opened up a PR to begin work on the draft folder for the install guide 16:39:16 Give me two seconds and I'll grab a link 16:39:20 I'm having some issues getting nova to pass tempest testing on 16.04. It's complaining about a non existent AZ, but no AZ is specified when the tempest test calls the api 16:39:25 This has also been merged 16:39:47 local tempest testing has shown I don't know how to run tempest testing locally, lol. 16:39:54 https://review.openstack.org/#/c/327467/ 16:39:55 See here 16:40:07 If you have any questions regarding this, please message myself or Darren 16:40:10 but, if anyone can look at the log output and see if there is anything jumping out at you, I'd appreciate it. 16:40:32 This is visible for a build, but it is not currently available publicly. It is an exact reflect of the current install guide and will just be the platform for all incoming changes 16:40:58 darrenc will be working on updating this folder to reflect the new ToC for the install guide 16:41:02 Which is super exciting. 16:41:28 I'll be away form the 13th to the 23rd. Once I return, Darren and I will begin working on some of the smaller intricacies along with mhayden and Travis 16:41:35 Questions, concerns, comments? 16:41:37 woot 16:41:42 :) 16:42:14 Hopefully I will also be at the OSA midcycle meetup, so I'll be working on the docs there too! 16:42:20 The deadline for this is Newton :) 16:43:04 michaelgugino: do you have a review up? and/or tempest output from jenkins? 16:43:11 asettle: thanks for leading the charge there 16:43:21 mhayden: not a problem. Thanks for being a key reviewer! 16:43:31 https://review.openstack.org/#/c/324045/ 16:43:33 * mhayden woots 16:43:34 And thanks to everyone that took the time to review and keep up to date with our changes 16:44:49 michaelgugino: looking now -- we can take this over to #openstack-ansible 16:44:52 anything else? 16:45:00 * mhayden starts the timer... 16:45:05 Just wanted to drop out here that a thorst, ashana, and I have started picking through the issues with multi-architecture/multi-repo 16:45:09 mhayden: nothing from me, thanks guys 16:45:20 I've started documenting the work items/issues out at https://etherpad.openstack.org/p/osa-multi-arch-repo 16:45:28 we need to work out the lxc container create process a little more on xenial 16:45:46 lxcbr0 is not coming up after reboot due to conflicts with lxc-net service 16:45:47 adreznec: if you want any help with documentation, feel free to ping me :) 16:46:01 asettle: Thanks, will do! 16:47:07 #link adreznec put together an etherpad for multiple cpu architectures in OSA -> https://etherpad.openstack.org/p/osa-multi-arch-repo 16:47:11 anything else? 16:47:27 yes 16:47:44 adreznec: can you propose this spec in gerrit so we can iterate on it? 16:47:52 I was thinking we should but a role requirements file in each of the repos, because the meta lists dependencies 16:48:03 mhayden: Sure, I'll throw it up there for review 16:48:04 this would help people that want to use some of the plays outside of OSA 16:49:50 michaelgugino: want to summarize the idea there and get it out to the mL ? 16:50:15 mailing list? 16:50:42 yeah 16:51:10 sure, I can do that 16:52:05 okay folks, i'll close it up 16:52:09 thanks for coming! 16:52:12 #endmeeting