15:00:30 #startmeeting ironic 15:00:31 o/ 15:00:31 Meeting started Mon Dec 2 15:00:30 2019 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:32 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:34 The meeting name has been set to 'ironic' 15:00:36 o/ 15:00:38 Good morning everyone! 15:00:41 o/ 15:00:42 o/ 15:00:46 o/ 15:00:49 o/ 15:00:52 o/ 15:00:56 o/ 15:01:12 o/ 15:01:22 * TheJulia makes one last moment change to agenda that likely nobody will notice because she has CDO 15:01:34 Our agenda can be found on the wiki 15:01:36 #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting 15:01:58 #topic Announcements / Reminders 15:02:03 \o 15:02:05 o/ 15:02:17 o/ 15:02:27 #info Cycle priorities have merged. 15:02:39 #info Whiteboard has been updated to reflect cycle priorities. 15:03:02 #info Older/still-in-flight items have been moved into the lower section of the etherpad accordingly. 15:04:31 As a reminder, the Operators mid-cycle is scheduled to take place in London January 6th and 7th. If anyone from ironic is in attendance, it would be good for a community representation and relationship building context. If you have any questions, please feel free to reach out to me as I have attended a few of their mid-cycles over the years. 15:04:43 Does anyone have any other Announcements or Reminders for us this morning? 15:05:11 I will probably be in London for the ops meetup 15:05:36 btw should be January 7-8, not 6-7 15:05:37 rpittau: Awesome! 15:05:39 doh! 15:05:43 :) 15:05:45 Thanks! 15:05:54 Anything else or shall we proceed onward? 15:06:51 * TheJulia takes silence as that we can proceed 15:06:56 HP 3rd Party CI is with POST_FAILURE since past week =) 15:07:24 I've sent an e-mail to the main contacts from the wiki Page =) 15:07:33 iurygregory: Awesome! 15:07:50 #topic Action Items from the prior meeting 15:08:30 We had only one action item last week which was mine to reach out to the regarding use cases regarding the baremetal whitepaper. I didn't get a chance to do that because I simply ran out of time last week. 15:08:49 #action TheJulia to reach out to the foundation regarding the baremetal program use cases. 15:09:05 Moving on! 15:09:08 #topic Review subteam status reports 15:09:17 #link https://etherpad.openstack.org/p/IronicWhiteBoard 15:09:25 Starting at line 275 15:10:17 arne_wiebalck: I've not had a chance to circle back to the ml thread from Nova regarding nova-compute performance, but we still had try to make scale/performance improvements/testing, so I've added at least one possible item to the theme on the whiteboard. 15:11:08 TheJulia: ok, thanks 15:11:18 TheJulia: will try to have a look at this asap 15:11:35 TheJulia: with our local nova experts 15:11:35 arne_wiebalck: thanks 15:12:03 TheJulia: just a bookkeeping thing; L277. Are you still updating this based on our priorities, or has it been updated already. 15:12:04 etingof: It would be good if you could update line 310's section at some point in the next week 15:12:45 rloo: Deleted line 277, good catch, I already indicated earlier on that I had updated the etherpad from the priority list 15:12:56 TheJulia: thanks :) 15:13:00 struckout items on the etherpad I'll delete after this meeting 15:13:56 I think the list mostly looks good. 15:14:17 iurygregory: I saw the client library py27 removals were hitting issues? Is there anythign requiring assistance there? 15:14:40 TheJulia, according to the community goals we can only drop on Phase 2 15:14:46 dtantsur|afk: or jerrywang, for replacing WSME, would you please clean that up, show the current status? (Or did people want history?) 15:14:53 L315ish 15:14:55 thats why I've mark Workflow -1 15:15:00 iurygregory: ack 15:15:00 ok. 15:15:48 A little history is always useful for context, but I concur, I Was looking at that item earlier and thinking that it wasn't clear what the current status really was at a glance. 15:16:26 I Update latest WSME status 15:16:39 TheJulia: similar thing with Python 3 (L368). I'm not sure what the status is... 15:17:17 and Python 3 is an 'other'. did we finish up what we wanted to do? 15:17:41 rloo: I think that should be moved under community goals 15:18:01 rpittau: done 15:18:10 I'm striking out py3 related stuff that looks out of date 15:18:13 thanks :) 15:18:23 oh ok I removed py3.5 already 15:18:45 that is fine 15:18:52 is six not python3-compatible? 15:18:53 xXraphXx reworked & rebased the UEFI s/w RAID patches if anyone wants to have a look, I'll also try to review test this week 15:19:03 okay, I think that covers things from my pov 15:19:04 *review and test 15:19:09 i just looked at our priorities. 'drop python 2.7 support' is there. iurygregory and rpittau are the contacts. 15:19:33 rpittau, iurygregory: you ok if I put your names next to Python 3? (L366) 15:19:34 kaifeng: six is just a py2/py3 compatability library 15:19:34 kaifeng: it is, but it's not required anymore since we're dropping py2.7 15:19:42 rloo, sure 15:19:47 arne_wiebalck: thanks! 15:19:53 I'll try to look this week. 15:19:54 rloo: syre 15:19:59 s/syre/sure 15:20:08 not sure if we want to specific python3 version through it 15:20:13 thank YOU rpittau & iurygregory :) 15:20:32 rloo, yw 15:20:55 kaifeng: can you elaborate please ? 15:21:52 wrt zuul v3 migration, it looks like we're blocked on grenade not migrated? is that right? (L390) 15:22:12 rloo: afaik, yes, that seems to be the issue for everyone afaik 15:22:28 any idea when/if grenade will be migrated? 15:23:03 Looks like gmann is working on it 15:23:19 and some others it looks like 15:24:42 I suspect it is just going to be a waiting game 15:25:13 TheJulia: ok, i noted that we're blocked on that. we can wait. patience is a virtue etc :) 15:25:15 rpittau: oh i need a new brain, i mean stuff like sys.version_info 15:25:24 we can only move after they have the grenade migrated =) 15:26:07 I'm keeping an eye on ML to see when it's done so we can give a try to move our jobs 15:26:09 kaifeng: I suspect lots of that can go away, but we only had minimal amounts of such checks/logic paths 15:26:46 thx iurygregory! I'll put your name down next to that. 15:27:06 rloo, ty 15:27:16 TheJulia: that's true 15:27:56 Are we good to move on? 15:28:23 ++ 15:28:36 let's 15:28:59 Excellent 15:29:21 #topic Priorities for the coming week 15:29:40 #link https://etherpad.openstack.org/p/IronicWhiteBoard 15:29:46 Starting at line 167 15:30:29 arne_wiebalck: is the patch on line 224 still needed? 15:30:52 mmm sorry if I go back in time but have you seen the note on L154 about the HP CI ? 15:30:52 TheJulia: no 15:30:59 looks like https://review.opendev.org/#/c/686585/ is what is needed in the IPA list 15:30:59 patch 686585 - ironic-python-agent - Software RAID: Use UUID to find root fs - 5 patch sets 15:31:19 TheJulia: I should clean this up as well (I did at another place on the etherpad) 15:31:30 rpittau: That is about two months old 15:31:40 arne_wiebalck: okay 15:31:57 oh ok, thought they had another one! 15:32:35 TheJulia: it's cleaned up around line 330 (Ussuri priorities, probably the wrong place) 15:32:55 TheJulia: we have kind of a double bookkeeping for these 15:33:03 It seems we're going to need to discuss virtual media based L3 deploy, but dtantsur|afk is not here so I guess that item can stay in review fo rnow, otherwise the list lgtm minus raid cleanups :) 15:33:17 arne_wiebalck: kind of yeah, more so "what is out there, and what to focus on this week" 15:34:18 arne_wiebalck: are those two patches needed in any particular order? 15:35:05 TheJulia: the rootfs and the ueif one? 15:35:09 uefi 15:35:16 the one for ironic and the one for ipa 15:35:26 in the list around line 330 15:36:09 w/o the ironic one, the IPA one will break things (as it would expect a proper UUID) 15:36:26 so the ironic one should go first 15:36:32 ok 15:37:11 okay, I see the dependency. Uhhh... that makes me wonder if newer IPA were to break on a train ironic then? 15:37:40 yes, it would 15:37:52 that is a little bit of a problem :\ 15:38:31 how can we solve this? I guess the IPA cannot really check if the passed UUID is valid. 15:38:33 the intent is for IPA to support ironic n-1 15:38:56 I'd have to look at the code to really be able to discuss it 15:39:11 we could try with the UUID and fall back to the current behaviour 15:39:20 but, yeah, we can take this offline 15:39:39 * TheJulia goes and makes notes 15:40:47 I'm good with the list, are we good to proceed? 15:41:14 Merged openstack/ironic-python-agent master: Remove the configuration sample file https://review.opendev.org/696862 15:41:47 * TheJulia hears crickets 15:41:59 everyone good with the list? 15:42:03 yep 15:42:05 Merged openstack/ironic-python-agent master: Remove the unused dependency on iso8601 https://review.opendev.org/696379 15:42:06 yes 15:42:11 ++ 15:42:57 Okay, onward! 15:43:10 #topic Discussion 15:43:36 We still have the discussion item about the midcycle, I guess there is not much to really discuss since some people have indicated they are seeking budget approvals 15:43:50 arne_wiebalck: you mentioned something about double checking hotels? 15:44:29 TheJulia: No, the rooms (as we may shift to Feb 26/27) 15:44:42 Oh, okay 15:44:42 TheJulia: We can put up a list of hotels, though. 15:44:51 26/27 would be more diffcult 15:45:06 we'll need to decide at some point 15:45:16 if 25/26 is ok, I suggest we stick with it 15:45:28 unless there are major objections? 15:46:25 I don't have any objections, 15:46:27 * arne_wiebalck hears crickets 15:46:33 no objections 15:46:33 * arne_wiebalck is too slow 15:46:57 then it'd be great if people registered on https://indico.cern.ch/event/863986/ 15:47:17 #link https://indico.cern.ch/event/863986/ 15:47:38 so I get the number of participants, can prepare the site access (and of course tours :) 15:47:49 okay 15:47:56 it's really just to get an idea 15:48:07 arne_wiebalck: even if we're not 100% sure ? 15:48:24 rpittau: that is the impresison I'm getting :) 15:48:25 rpittau: I thought you are 120% sure? 15:48:39 * TheJulia wonders if we're creating new number systems of counting 15:48:48 rpittau: yes, you can of course also cancel if things do not work out 15:48:57 Can we do percentages in hex? 15:49:03 rpittau: but esp for the tours I need to be early 15:49:18 TheJulia: :-D 15:49:27 * iurygregory registered even without 100% sure =) 15:49:34 \o/ 15:49:35 arne_wiebalck: I understand, and I'm optimist, so goingto register anyway :D 15:49:40 Okay, well, I guess we're good to carry on 15:49:44 iurygregory: nice 15:50:03 We have nearly 10 minutes left, I'm not sure there is anything to really discuss with the Baremetal SIG 15:50:13 at least right now 15:50:26 I don't think so ... we should ping potential authors now. 15:50:32 And there are no RFE's for review, so I think the next step is Open Discussion? 15:50:36 arne_wiebalck: ++ 15:50:49 * arne_wiebalck is getting mails about mid-cycle registrations 15:50:53 Merged openstack/ironic master: Add indicator management to redfish hw type https://review.opendev.org/652740 15:51:03 arne_wiebalck: \o/ 15:51:18 Moving to Open Discussion 15:51:25 #topic Open Discussion 15:51:28 Anyone have anything? 15:52:27 there is an interesting requirement to add a no-deploy interface, want to know if it has value to the upstream? 15:53:42 the idea is to do no real deployment, but utilize the network switching for vlan trunk configuration. 15:54:13 kaifeng: to what end? just like environment management? 15:54:30 there is a preexisting deploy server to do the rest. 15:54:54 kaifeng, I had a similar use-case in the past 15:55:08 yeah, something like that, we are about to do so for our bare metals 15:55:20 kaifeng: I could see that making a lot of sense for some of the mixed lab enironments out there... the idea is really growing on me 15:56:59 Anything else for discussion today? 15:57:49 * TheJulia hears cricket 15:58:16 TheJulia: I'll recap this after all this is done, still in verification stage, but things seems working fine. 15:58:33 kaifeng: awesome 15:58:42 Well, thanks everyone! 15:58:51 TheJulia: Thanks! 15:58:55 thanks! 15:58:59 TheJulia: thanks! 15:59:08 thejualia, thanks! 15:59:25 TheJulia: Thank you! 15:59:57 #endmeeting