17:02:53 #startmeeting ironic 17:02:54 Meeting started Mon May 29 17:02:53 2017 UTC and is due to finish in 60 minutes. The chair is dtantsur. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:55 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:02:57 The meeting name has been set to 'ironic' 17:03:01 hi all, sorry for the delay! 17:03:02 o/ 17:03:08 hello! :) 17:03:24 our agenda can be found, as usual, here: 17:03:27 #link https://wiki.openstack.org/wiki/Meetings/Ironic 17:03:32 o/ 17:03:41 #topic Announcements / Reminder 17:03:48 o/ 17:03:56 I have requested releases for stable branches of all projects 17:04:14 they're still on review, will hopefully happen this week 17:04:16 o/ 17:04:28 anything from anyone else? 17:04:42 dtantsur: do we want to try to get the backport for oneview? 17:04:54 dtantsur: for the upcoming releases 17:05:19 rloo: I'd prefer not to change anything already. if we want to, we can make another release even in a week 17:05:25 dtantsur: ok 17:05:28 but if it does merge really soon - I can reconsider this 17:05:43 anything else? 17:05:53 dtantsur: what are these releases? 17:06:17 ricardoas: I don't think I understand the question. we make patch releases from stable branches from time to time 17:06:35 dtantsur: that's it... :) 17:06:36 voices in my head told me that this is about the time to make another bunch of such releases 17:06:52 thanks! 17:06:52 they're not time-based, just my gut feeling (and some amount of spare time) 17:07:49 #topic Review subteam status reports (capped at ten minutes) 17:08:03 #link https://etherpad.openstack.org/p/IronicWhiteBoard starting with line 94 17:08:33 dtantsur: did you update the bug stats for today? 17:08:48 rloo: nope, sorry. will update for 2 weeks next time 17:08:59 dtantsur: ok, thx 17:09:04 * dtantsur is not feeling well recently, and confused the time of this meeting 17:09:30 hmm, I haven't seen any updates about rescue for some time 17:10:19 rloo: the deploy steps spec, does it need another update or are the comments there contentious? 17:10:42 dtantsur: needs an update. i actually haven't looked at it recently; will look this week. 17:11:09 ack 17:12:11 so, rescue is concerning me 17:12:24 anyone else has a feeling that it's not making Pike? 17:13:06 dtantsur: it is still possible, we should find out whether anyone is actually going to work on it soon 17:13:34 I think so, too 17:13:39 rloo, dtantsur as i know stendulker and aparnav are working onit 17:13:43 dtantsur: I guess Aparna is working on that 17:13:48 the patches seem to have some activity 17:13:53 but both of them are on leave 17:14:06 Nisha_Agarwal: could you light a fire under them -- are they updating the status so we know what's going on or not going on? 17:14:19 Nisha_Agarwal, vmud213, when they come back, please ask them to update the whiteboard. if they can't make it to the meeting, please ask them to do it in advance 17:14:33 rloo, sure. i just came back last week from vacation so dont know actual status 17:14:35 dtantsur:sure 17:14:44 thanks! 17:14:50 dtantsur, sure 17:14:52 :) 17:15:36 dtantsur: wrt redfish, that power thing is a bug, right? what's outstanding wrt the feature? 17:16:01 rloo: nothing. I just can't call it done, until we have power actions working the same way as in other drivers. 17:16:15 etags spec was approved; i'll update that status 17:16:26 dtantsur: ah, ok. so we have CI and docs done, just that 'bug'. 17:16:39 mm, I think I did a minute ago 17:16:43 yep 17:17:44 and YAY, etags spec merged ! 17:17:49 \o/ 17:18:12 does anyone know if the etag patches are ready to review? 17:18:23 * dtantsur does not 17:18:46 I'll take a look and update the whiteboard 17:18:57 rloo: https://review.openstack.org/#/q/topic:bug/1605728 17:19:02 and needs removing your -2 17:19:50 I'm done reviewing the status 17:19:58 dtantsur: done, thx for reminder. but i guess nothing is ready to be reviewed yet. 17:20:08 +1 done with status 17:20:10 does not look like that 17:20:14 everyone done? 17:20:19 dtantsur, for py3.5 i wanted to discuss 17:20:40 Nisha_Agarwal: feel free to bring it on the open discussion 17:20:48 dtantsur, sure 17:20:57 #topic Deciding on priorities for the coming week 17:21:08 so, we've finished one item 17:21:42 dtantsur: we have to remove rolling upgrades, nothing to review yet. 17:22:07 this is sad, but yeah. I hope we can re-introduce it next week 17:22:15 dtantsur: me hopes too! 17:22:46 the next item we could take is https://review.openstack.org/#/c/439907/ 17:22:59 "OSC commands for ironic driver-related commands" 17:23:14 which is quite important for feature parity between CLI tools we have 17:23:24 dtantsur: that's on my plate; i have to write/update 17:23:31 dtantsur: hoping to do that this week too 17:23:51 rloo: do you think we could try finishing the spec this week? 17:24:12 dtantsur: would like to. 17:24:28 rloo: ok. I can help you with it, if you'll be running out of time. 17:24:45 dtantsur: thx. i need to sit down and think about sambett's comments 17:25:11 are there any network-related things that we should be looking at? (not that i have time...) 17:25:25 vdrok: ^^^? 17:25:33 * dtantsur does not see vsaienk0 around 17:25:54 is vsaienk0 back this week or next? 17:26:10 I think I saw him this morning 17:26:14 rloo: yeah he's back 17:26:21 just the things that are here https://review.openstack.org/#/q/project:openstack/networking-baremetal 17:26:29 but they were reviewed already 17:26:32 good :-) 17:26:34 good, yeah 17:26:42 I don't think I understand this code well, tbh 17:26:46 we should get an update from vsaienk0; he's probably catching up on stuff. 17:27:03 ok, we have 3 items: from BFV, from driver composition and OSC work 17:27:20 does this list look good? do you want to add and/or remove something? 17:27:26 btw dtantsur, wrt redfish. 'we need to fix it' -- do you know who will fix it or do you need someone to volunteer to fix it? 17:27:48 rloo: I will. but we decided with sambetts to do some refactoring first 17:27:59 dtantsur: ok, i'll update to indicate you :-) 17:28:04 we want to stop copy-pasting the "wait for power state" loop to all drivers 17:28:13 so he has a WIP patch to move it to the conductor 17:28:31 dtantsur: ah, yes, i recall those discussions.ok, so it'll be fixed soonish. 17:28:58 #link https://review.openstack.org/#/c/449454/ 17:29:45 ok, any changes to the list? 17:29:51 i'm good with the priorities 17:30:19 #topic Open discussion 17:30:33 Nisha_Agarwal: you wanted to discuss something? 17:30:49 dtantsur, yes 17:31:27 go ahead Nisha_Agarwal... 17:31:33 i raised couple of patches to add experimental gates for ironic governed projects 17:32:27 but got the comment that this lot of duplicating in the project-config project, so instead we should add python3 in existing gates 17:33:04 Nisha_Agarwal: link handy? 17:33:37 but i am afraid that if i address that comment then gates will start failing as swift is not ready fr py3.5 17:33:41 yes 17:33:49 i updated in whiteboard 17:34:43 #link https://review.openstack.org/462706 17:34:59 #link https://review.openstack.org/462701 17:35:26 #link https://review.openstack.org/462487, https://review.openstack.org/462695 17:35:28 Nisha_Agarwal: I don't read Ian's comments the same way. I think what he asks for is to use the same builder, not the same jobs 17:35:47 dtantsur, we use the same builder 17:36:00 right. this does not mean that existing voting jobs will start using python 3 17:36:05 no 17:36:20 it just means adding (yet another) option to our main builder 17:36:25 dtantsur, i added one experimental gate per project 17:37:03 whch is kind of duplication of already running gates, just added python3 builder 17:37:28 yeah, so the ironic change seems good 17:37:44 ironic was already in when i took over 17:37:46 I guess the ironicclient one is contentious, right? 17:38:04 let's start with limiting the scope of the problem 17:38:27 let's get only a job on ironic first. it will already cover ironic-lib, python-ironicclient and python-ironic-inspector-client implicitly 17:38:49 let's make it passing, overcome the swift-related issue, then consider moving on 17:38:53 yes ironicclient, ironic-inspector, ironic-lib anf ironi-inspector-client 17:39:08 dtantsur, ok 17:39:30 so will wait for ironic related patch to get merged 17:39:34 the second one will be ironic-inspector, as it's another service 17:39:52 dtantsur, yes but it also has the similar comment 17:40:03 that will get us complete coverage already. as soon as the jobs are voting, we can declare the goal fulfilled 17:40:03 ipa patch is merged 17:40:49 Nisha_Agarwal: I did not have a chance to look at the IPA change, and I wonder if it *actually* tests IPA with python 3. but this is the 3rd priority, after ironic and ironic-inspector 17:41:55 dtantsur: Just wanted to check 17:42:02 if it is possible to take a look at https://review.openstack.org/#/c/422572/ this week. 17:42:45 it's not impossible, but I cannot make any promises 17:43:43 As per the earlier discussion in Virtual meetup, this is one of the pending patch not getting much attention for a while .. 17:44:27 i'll try to take a look 17:44:33 I'd not say that not having reviews in 4 days is "for a while", though I'd love to improve on it 17:45:01 I'd like to ask a question to someone. I'm in trouble with Ironic tempest test. I can't do scenario test. Is my environment something wrong or bug? 17:45:23 vdrok: THanks 17:45:35 dtantsur: we fixed the version of python-oneviewclient for newton here https://review.openstack.org/#/c/468925/ should I ping you and rloo when the CI evaluate it? 17:45:48 izumi777: hard to tell without going into details, but chances are high that it's related to your environment, as we run the tests in our CI 17:46:21 ricardoas: quick review: please expand your release note with the detailed explanation of the situation 17:46:48 dtantsur: sure! 17:46:56 ricardoas: the release team would like to know why we bump this version. it's critical to mention in both the commit message and the release note that the service has never worked with the version in question 17:47:08 * Nisha_Agarwal got disconnected in between.... 17:47:13 ricardoas: I'd also appreciate a +2 from someone from the global stable team, not only rloo and me 17:47:18 datnsur: hmm, I see. 17:47:28 ricardoas: please find them on #openstack-stable 17:47:54 dtantsur: ok, thanks! 17:48:11 Nisha_Agarwal: back to your initial question: I think what they ask is to reuse the builder for python 2 and 3 by parametrizing it 17:48:27 this does not mean changing the existing job, only making the builder more flexible 17:48:32 ricardoas: you're welcome 17:48:53 dtantsur, ok. let me see 17:49:16 dtantsur, thanks. will check how we can do this in infra 17:49:21 yes please 17:49:38 :) 17:49:43 Does anyone encounter the phenoma: tempest.lib.exceptions.TimeoutException: Request timed out Details: (BaremetalBasicOps:test_baremetal_server_ops) Server 3497db75-4a12-447c-b45f-8814ff9bc35c failed to reach ACTIVE status and task state "None" within the required time (196 s). Current status: BUILD. Current task state: spawning. ? 17:50:02 any hint? 17:50:18 izumi777: this is a very generic issue. it only means the server timed out to deploy. it may be problems with your PXE environment or DHCP 17:50:48 izumi777: your next step, I guess it to figure out what *actually* happened to the node. maybe by looking at its console. 17:51:44 I could deploy maucally successfully, but failed in tempest case. 17:52:08 izumi777, may be some of the tempest variables are not getting populated 17:52:15 this may mean something is wrong with networking configuration. I guess double-checking tempest.conf is a good idea 17:53:23 dtantsur, Nisha_Agarwal: Thanks for giving me a hint! I'll check tempest.conf. 17:53:28 any other questions for the meeting? I guess the debugging session can be moved to the channel. 17:53:46 * dtantsur does not mind wrapping up early 17:54:37 crickets 17:54:39 aaaaaaannnnndd 17:54:45 thank you everyone! 17:54:50 :) 17:54:55 #endmeeting ironic