19:01:26 #startmeeting Ironic 19:01:26 \o 19:01:26 #chair devananda 19:01:27 Welcome everyone to the Ironic meeting. 19:01:27 Meeting started Mon Aug 25 19:01:26 2014 UTC and is due to finish in 60 minutes. The chair is NobodyCam. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:01:27 o/ 19:01:28 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:01:30 The meeting name has been set to 'ironic' 19:01:30 o/ 19:01:30 o/ 19:01:31 Current chairs: NobodyCam devananda 19:01:32 :) 19:01:43 Of course the agenda can be found at: 19:01:44 #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting 19:01:50 #topic Greetings, roll-call and announcements 19:01:51 Roll-call: Who's here for the Ironic Meeting? 19:01:57 * jroll is 19:01:57 o/ 19:02:00 \o/ 19:02:00 me 19:02:02 o/ 19:02:02 * Shrews too 19:02:05 hi 19:02:09 me 19:02:13 \o/ 19:02:21 welcome all :) 19:02:30 announcements: 19:02:32 Devananda is still off burning thing in the desert. 19:02:39 as you may know our gate was again broken last week :( 19:02:48 :( 19:03:04 rloo: are we borken again with the doc thing or is that a red harring? 19:03:14 :/ it's fixed now, but there's some inconsistencies with docs and py26 afaict 19:03:26 :/ 19:03:28 :( 19:03:29 I don't know how broken we are, but the doc thing seems to be broken a bit. 19:03:43 uggh 19:03:44 rloo: i'll try to help you track down the doc thing after the meeting 19:03:57 we'll need to check on htat after the meeting 19:03:58 the py26 is out of our control it appears. I updated our etherpad. you can recheck if py26 fails 19:04:03 Ty Shrews :_ 19:04:06 Shrews: thx. 19:04:14 are there bugs for either issues? 19:04:22 o/ 19:04:29 https://bugs.launchpad.net/ironic/+bug/1361295 19:04:30 Launchpad bug 1361295 in ironic "gate-ironic-docs failed due to TypeError from sphinx" [High,New] 19:04:30 yes, there are bugs. see etherpad. 19:04:42 thanks 19:04:43 :) 19:04:45 the doc thing would be good to check with ceilometer as well, they use the same docs as we do 19:04:55 sphinx with the wsme extension 19:05:05 oh good poing lucasagomes :) 19:05:16 point even 19:05:18 * adam_g wonders if PYTHONHASHSEED strikes again 19:05:31 no that landed this weekend 19:05:35 I saw it :-p 19:05:36 in tempest 19:05:39 oh 19:05:43 i mean, in ironic for the docs 19:06:02 adam_g: can you maybe help rloo and Shrews with that? 19:06:20 tox -evenv -- python setup.py build_sphinx 19:06:20 sure 19:06:25 :) 19:06:29 ill throw up a patch now just to see if that eliminates the issue after a number of rechecks 19:06:43 awesome TY rloo Shrews ^^^^ 19:06:56 any other announcements? 19:06:58 thx adam_g; I am crossing my fingers. 19:06:59 lol :) 19:07:05 hah 19:07:23 if not we'll jump right in to it 19:07:53 #topic Release cycle progress report 19:07:56 Lucas that you 19:08:00 ? 19:08:00 alright 19:08:03 #link https://launchpad.net/ironic/+milestone/juno-3 19:08:09 yeah I think it's me :) 19:08:15 stats: 7 Implemented, 5 Needs Code Review, 2 Started (no code upstream yet) 19:08:26 we have ~2 weeks until our feature freeze (Sep 4) 19:08:31 #link https://wiki.openstack.org/wiki/Juno_Release_Schedule 19:08:37 compared with last week we have 4 more blueprints implemeted this week :) 19:08:42 which is great! 19:08:53 woo hoo 19:08:59 two approved specs still have no code upstream yet :( 19:09:10 one is the Drac Managment Interface, and I'm helping with this one 19:09:14 lucasagomes: anyone we can poke on those? 19:09:14 I'm planning to put the code upstream tomorrow! 19:09:21 NobodyCam, yup, will update about it 19:09:32 :) 19:09:40 the other one is the Power Driver for SNMP 19:09:52 I sent an email to the author last thursday, and the tl;dr is 19:10:09 the driver is written and in use on their site but they need more time to refine and improve unit testing, also, he have not been through the company procedures for external code release yet 19:10:15 so I hope they can sort out the tests and the company thing, but there's some good chances that it's not going to make it :( 19:10:25 agree, I'd be shocked if that got in 19:10:40 all other bps seems to be in good shape, and have the code upstream 19:10:49 I think we are doing great job, the review jam last week was pretty good and I would like 19:10:49 lucasagomes: are they requesting a ffe yet? 19:10:58 to do it again this week, if people agree with it (we can talk about time/day at the open-discussion) 19:11:01 NobodyCam, not yet 19:11:11 lucasagomes: +1 from me on jam 19:11:23 NobodyCam, he said he will try to put the code up asap, but you know how things are inside companies 19:11:26 it may take a while 19:11:32 and I'll try and make the whole thing this time too 19:11:34 NobodyCam: I think that's not worth it for an ffe? 19:11:40 +1 on a jam tomorrow 19:11:47 yeah that's a low priority bp 19:11:47 (or whenever) 19:12:07 jroll: I agree was just checking to see if they had asked for one 19:12:08 so we well see when time comes, if there's no code until the last day, we may just block it until K 19:12:18 and that's pretty much it about the updates 19:12:19 +1 19:12:23 let's keep with the reviews :) 19:12:34 great update lucasagomes thank you 19:12:56 any questions / comments for lucasagomes ? 19:13:25 ok then moving on :) 19:13:28 #topic SubTeam: integration & testing work 19:13:29 adam_g 19:13:32 thats you? 19:13:46 ilo drivers are having doc issue. Will the revew jam reivew code that have doc issue? 19:13:56 heya 19:14:03 wanyen: yes, that's a known bug 19:14:04 so we're in the final stretch for our CI/QA goals for this cycle 19:14:12 wanyen: we're looking into it, no worries 19:14:21 wanyen: lets tough on that in open discussion 19:14:23 jroll. ty 19:14:39 we should hopefully be wrapping up our tempest and grenade work this week 19:14:40 adam_g: w00t :) 19:14:46 yay 19:14:48 niiiice 19:14:50 and have a new grenade job popping up in our pipeline hopefully before FF 19:15:13 #link https://review.openstack.org/#/q/topic:ironic_tempest,n,z 19:15:13 sweet 19:15:16 (2 patches left to merge there) 19:15:23 great! 19:15:30 #link https://review.openstack.org/#/q/topic:ironic_grenade,n,z 19:15:33 (a few more there) 19:15:53 adam_g: do you need ny help pingging folks on getting the two patches reviewed, or are you doing good there? 19:15:56 confident we should be wrapping this up this week and getting the infra changes put in place to turn the job on 19:16:08 NobodyCam, i think we are okay now that everyone from qa is back from conference travel 19:16:13 adam_g: are you pushing to get this in this week as well? https://review.openstack.org/#/c/112134 19:16:34 jroll, yeah, thanks for reminding. looks like it needs a rebase 19:16:38 (I'm relying on that for getting IPA jobs in) 19:16:40 yeah 19:16:51 adam_g: if you're fine with it, I can help babysit that one 19:16:51 jroll, is the IPA stuff ready and blocked? 19:17:15 adam_g: devstack-gate patch has 1 +2, then I can land infra/config changes that depend on 112134 19:17:56 :) 19:17:56 jroll, okay, ill rebase it now and check with shrews after meeting about some minor concerns about coordinating the merge /w other things. i think we're okay, but want to check (dont want to risk gate breakage again) 19:17:58 I have some other work up after that; specifically IPA repo gate jobs, that stuff needs some more love before I can land it though 19:18:15 adam_g: cool, just making sure it's on your radar :) 19:18:34 thanks 19:18:56 any other q/c? 19:19:26 thats all from me 19:19:34 awesome Ty adam_g :) 19:19:47 moving on downhte road 19:19:49 #topic SubTeam: Bug team 19:19:53 dtantsur thats you? 19:19:55 o/ 19:20:08 \o/ 19:20:16 people like numbers, so here are some numbers: Open: 136 (+3). 8 new (0), 41 in progress (+6), 0 critical, 16 high (+1) and 5 incomplete (-3) 19:20:56 amount of bugs is not decreasing. also, I forgot to ping people who promised me to help with triaging smth nova-related 19:21:02 dtantsur: we should add a catagory for broke our gate 19:21:11 heh 19:21:32 if we create a tag in launchpad 19:21:35 :) 19:21:38 hehehe 19:21:40 and one more thing from me: 19:22:00 #link https://etherpad.openstack.org/p/IIe0cnnJ5L possible candidates for Juno 3 19:22:12 ah nice 19:22:14 these are my ideas, what could also be prioritized 19:22:22 feel free to add/delete/edit 19:22:40 cool 19:22:50 #link http://ironic-bugs.divius.net/ shows which bugs still have 'New' status 19:22:51 please cross out do not just delete :) 19:23:00 (some are cached, so don't be surprised) 19:23:29 that's all I wanted to tell you 19:23:40 great Thank you dtantsur :) 19:23:58 questions / comments for dtantsur 19:24:00 ? 19:24:25 just one thing 19:24:28 if not we'll move on too: 19:24:31 #topic SubTeam: Ironic Python Agent 19:24:33 jroll go 19:24:37 "-p 19:24:39 I'd like to go over those bugs in the review jam 19:24:46 and target as needed 19:24:48 :) 19:24:56 jroll: the one's on hte etherpad? 19:24:56 so, IPA 19:24:56 jroll, right, I posted today for everyone to prepare :) 19:25:00 NobodyCam: indeed 19:25:11 on the IPA front... not much new 19:25:17 still working on CI 19:25:27 #link https://etherpad.openstack.org/p/ipa-todos 19:25:30 jroll: should-we-release questions? should they be brought up here or wait? 19:25:38 ^ reviews are mentioned there 19:25:55 JayF: yeah, we should, I'll let you talk about that :) 19:26:01 JayF, I think it's related so +1 19:26:20 just generally wondering what folks think about how IPA should be released 19:26:22 jroll, NobodyCam, i have one point to bring 19:26:24 yea ipa questions go _here_ 19:26:31 initially the idea was releases would be cut like with clietns 19:26:37 rameshg87: for IPA? 19:26:39 but there's rarely a good natural place to cut an agent release 19:26:43 jroll, NobodyCam, the ilo-virtualmedia-ipa driver requires a change in ipa : https://review.openstack.org/#/c/115275/ 19:26:48 aside from release time with ironic 19:26:55 I didn't think of it 19:26:58 so I wonder if we should tag a stable branch for a juno agent, and do a 'release' 19:27:03 rameshg87: it's on my radar, last time I looked it was WIP :) 19:27:03 sorry, posted earlier 19:27:14 *I* would never use that, but wanted to check with folks who have other usaeg patterns 19:27:21 jroll, thanks, its fully functional code which you can review now 19:27:47 I wanted to say (and I barely manage my keyboard already) that distro folks may need some "know good version for J/K/..." 19:28:07 dtantsur: well, I highly disagree there should be deb/rpm/etc 19:28:13 but imbw 19:28:18 jroll: would it hurt to do that? ie, if it is useful for folks and only a minor inconvenience for you to do it, why not? 19:28:22 jroll, what do you propose for them? 19:28:27 Well, if the distro did something like build a deploy ramdisk and make that rpm/deb installable, that would be OK 19:28:51 but I don't think I would personally ever condone being able to apt-get install ironic-python-agent and have ipa installed, as a python module, in the distro directories 19:28:55 yeah, I guess it could be used in a ramdisk build 19:28:59 but agree with JayF 19:29:07 JayF: vs packaging up the scripts to build the image? 19:29:12 if someone is doing a ramdisk build, having a release cut and in pypi would be useful 19:29:25 NobodyCam: What other distro packages fit that pattern? 19:29:34 NobodyCam: install a binary package to build an image for something? 19:29:46 I just can't think of any, so it leads me to think folks may just do that one their own... I know I would 19:29:46 any src package 19:30:27 ironic-ipa-source-rpm 19:30:39 s/-rpm/.rpm/ 19:30:41 I think it should not be RPM/Deb but we still need some "know good version for J" somewhere, I guess 19:30:57 agree 19:31:01 we should tag a release 19:31:03 put on pypi 19:31:20 jroll want an action item for that? 19:31:21 jroll, ++ 19:31:24 sounds good 19:31:33 we should probably hit the mailing list too, to see what others think 19:31:35 NobodyCam: sure 19:31:46 NobodyCam: although the actual tagging is a long way out 19:32:21 #action IPA looking into how to cut a know good version for J 19:32:34 that work ^^ 19:32:40 sure 19:33:10 great update. any q/c for the IPA team 19:33:48 okay moving on to: 19:33:50 #topic SubTeam: Oslo 19:33:51 Ghe 19:33:54 is that you? 19:33:58 are you here? 19:34:00 * jroll will be back in 5 or so 19:34:07 ack TY jroll 19:34:12 here but nothing to report 19:34:32 :) woo hoo.. .no news is good news 19:34:33 :) 19:34:38 :) 19:34:39 Thank you GheRivero 19:34:59 questions / comments for GheRivero ? 19:35:30 nope... moving on to: 19:35:31 #topic SubTeam: Nova Driver status 19:35:31 mrda thats you 19:35:38 hey 19:35:45 So a number of changes to the driver were required in the last week , both internal and external, which have all been applied. I've also been fixing up various 19:35:47 Nova check failures. 19:35:55 Hopefully there will be less churn this coming week. 19:36:06 But I have struck a git issue yesterday. During fixing a rebase conflict I accidentilly squashed a change by 'committing' when I should have 'continued'. So now I need to replay the sequence of patches on top of the pre-conflict patch to fix the dependencies. 19:36:26 mrda: i have a patch up that I was going to bring up in Open Discussion but prob should bring up here 19:36:31 sure 19:36:45 https://review.openstack.org/#/c/116316 19:36:55 #link https://review.openstack.org/#/c/116316 19:37:02 * mrda looks 19:37:21 NobodyCam, we need it as part of graduation? 19:37:22 I am not sure that is the correct place for that review? 19:37:30 lucasagomes: I think so 19:37:38 NobodyCam, I think we could just put that on nova directly no? 19:37:44 like at the end of the chain or something 19:37:57 It probably could be added to the end of the chain 19:38:06 lucasagomes: thats why I pointed it out to mrda 19:38:06 why at the end of the chain? our chain doesn't touch that file? 19:38:10 because it would be dependent on everything else being in 19:38:21 mrda: and nova bm being out 19:38:27 so it can be a separate patch that depends on the chain, but not part of the chain? 19:38:36 rloo: I think so 19:38:42 ie, I don't want that patch to cause the rest of the chain not to be approved/merged. 19:38:44 rloo, I don't think the driver has anything to do with that change, that's more about proxying nova bm commands to ironic 19:39:06 ya I would not include it with the driver patches 19:39:20 correct, but we don't want the nova proxy to land until we have a ironic driver in nova, right? 19:39:22 nova bm should be removed before it lands 19:39:30 lucasagomes: right, so it shouldn't be put 'at the end of the chain'. it isn't part of that chain, but it needs that chain as a dependency. 19:39:55 NobodyCam: fwiw, there are other patches that are -2 19:40:01 rloo, right... well it should merge after the driver is merged, but in a different topic 19:40:02 d cuz they are nova changes. 19:40:03 NobodyCam: I'm not sure nova bm is going to be removed for a cycle 19:40:04 afaict 19:40:40 ie nova bm will exist for a while in tree even afer ironic lands 19:40:43 mrda: we will need to figure out how and when that needs to land 19:40:56 NobodyCam: we should discuss that proxy patch. if you make the change to baremetal_nodes, and BM is removed at some point, will we still need to proxy those nova bm commands? 19:40:57 of whether it is switchable? 19:41:13 I wouldnn't want to remove the nova bm commands untill nova bm is actually removed 19:41:13 s/of/or/ 19:41:37 NobodyCam: right, which might be anotgher cycle 19:41:53 we could add a switch and let operators set it 19:42:04 +1 19:42:10 but I think we have time on that patch to work it out 19:42:30 I agree 19:42:50 So hopefully no more actual changes to the driver :) 19:42:54 mrda: do you need any help poking folks for reviews on hte nova patches? 19:43:06 NobodyCam: I'm curious if that's going to get blocked by "feature freeze" that has already happened in nova 19:43:12 mrda, +1 /me fingers crossed! 19:43:22 So I hope to resolve the rebase issue today, then it'll be back to bugging Nova. That should be fine. I'll holler if I need help. 19:43:32 jroll: nova team requested it... we have it, so I'm not sure 19:43:40 NobodyCam: hm, ok :) 19:43:45 The hardest part is fixing my squashed patch :-S 19:43:49 lucasagomes: are you in nova meeting this week? 19:44:06 NobodyCam, oh, I forgot that, when is it? yeah I can go this week 19:44:10 I can attend Nova this week. 6:30am is a fine timeslot for me 19:44:19 lol :( 19:44:22 "fine" 19:44:24 :P 19:44:34 mrda, heh not that fine, if needed I can join there for u 19:44:38 might be worth askin about how we want ot handle that patch 19:44:40 It's better than 2:30am or whatever 19:44:45 it was last week 19:45:14 :/ 19:45:22 lucasagomes: I'll be there, but you are also welcome to join, of course :) 19:45:39 ack... I will put on my calendar 19:45:41 awesome Thank you lucasagomes and mrda :) 19:46:16 any other questions / comments on nova driver stuff? 19:46:47 Thank you for the great report mrda :) 19:46:53 thanks :) 19:47:03 ok we added this one last week: 19:47:05 #topic Outstanding items for graduation 19:47:27 I had the proxy patch listed here but we talked about it already 19:47:30 so 19:47:40 #link https://docs.google.com/spreadsheets/d/1Hxyfy60hN_Fit0b-plsPzK6yW3ePQC5IfwuzJwltlbo/edit?usp=sharing 19:48:01 any one have ANY concerns or comments? 19:48:07 questions too :) 19:48:09 how's the DB migration stuff? 19:48:33 and the horizon panel, is that going well? 19:48:45 do we need the onductor take-over for graduation? 19:48:53 * jroll checks on horizon 19:49:10 take over I don't believe so 19:49:23 has romcheg been on line? 19:49:24 for horizon, aweeks has been hammering away on it. i don't think there's anything submitted yet, but i'll give him a hand this week and have some patches by next meeting 19:49:28 the horizon is not part of graduation is it? 19:49:36 lucasagomes: it is 19:49:39 lucasagomes: no 19:49:43 :O 19:49:47 wait 19:49:50 hah... well I don't think it is 19:49:56 oh, right, we have to have some things proposed 19:49:58 but not landed 19:50:00 chicken/egg thing 19:50:01 ^ 19:50:11 So horizon won't take patches from non graduated projects 19:50:12 I think it's more of a "good faith effort toward it" 19:50:14 correct 19:50:26 so it can merge soon afterwards 19:50:36 so we'll show some good faith work this week 19:50:36 maybe someone could update that and add a comment about chicken/egg, proposed but ok if not landed? 19:50:47 sure 19:50:49 * jroll does it 19:50:53 rloo: On whiteboard? 19:51:01 Ty jroll :) 19:51:03 mrda: on that google doc link 19:51:11 mrda: https://docs.google.com/spreadsheets/d/1Hxyfy60hN_Fit0b-plsPzK6yW3ePQC5IfwuzJwltlbo/edit#gid=1357128775 19:51:11 who is working on ironic horizon? UEFI boot mode might need changes to horizon work 19:51:29 wanyen: aweeks and i 19:51:44 rloo: thnx 19:51:58 does horizon allow a user to specify which boot mode to deploy? 19:52:15 anyone need help with ANYTHING for graduation readyness? 19:52:17 wanyen: it won't land in juno, so don't worry about it yet 19:52:34 wanyen: (anything with horizon at all) 19:52:48 jroll, when is horizon work targeted toland? 19:53:03 wanyen: kilo, assuming ironic graduates in juno 19:53:15 jroll, I see. tx 19:53:34 jroll: horizon will not accept our patches until we graduate no? 19:53:41 jroll: I assume you're referring to the horizon patches (such as they are)? 19:53:45 at least thats what I thought 19:53:54 That's my understanding 19:54:25 any thing else on graduation? 19:54:31 NobodyCam: right, so horizon work will land the cycle after we graduate, yes? 19:54:40 (which hopefully will mean kilo) 19:54:42 thats what I thought 19:55:12 the nova bm to ironic job upgrade is in 19:55:22 ok with our last 5 minutes 19:55:26 so you can start running it as experimental 19:55:27 awesome TY clarkb 19:55:41 nice, thanks clarkb 19:55:56 clarkb: 'check experimental' to run it 19:56:18 ok with our last four minutes: 19:56:19 #topic Food for Thought / Open Discussion 19:56:22 yup against grenade devstack and devstack gate iirc 19:56:33 maybe ironic too 19:56:35 awesome TY clarkb 19:56:40 nicee 19:56:43 thanks clarkb! 19:57:09 #action test nova bm -> ironic upgrade with 'check experimental' 19:57:39 any Food for Thought? 19:57:49 I thought I had something... lost it 19:58:15 just to toss it out there: https://bugs.launchpad.net/ironic/+bug/1325942 19:58:19 Launchpad bug 1325942 in ironic "Ironic should support IPv6" [Wishlist,Confirmed] 19:58:36 two minutes 19:58:37 NobodyCam: :( 19:58:38 hah 19:58:47 NobodyCam: kilo 19:58:50 we should start puting together things we want to do in paris, soon 19:58:51 yep 19:58:58 jroll: ++ 19:59:14 yup +1 19:59:17 FWIW; the decom talk from myself and JoshNang was accepted for the Summit 19:59:21