| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2025-02-21 | |||
| 21:21:35 | melwitt | mikal: it's continuing to run. I didn't think it used to work that way but this is nice | |
| 21:22:06 | mikal | I'll still have to drop the depends on for my WIP tempest change or it will get in the way. I'll do that now. | |
| 21:22:11 | opendevreview | Douglas Viroel proposed openstack/nova master: Add support for showing scheduler_hints in server details https://review.opendev.org/c/openstack/nova/+/938604 | |
| 21:22:52 | melwitt | sorry for duplicating. I wanted to try to help since the microversion race messed everything up | |
| 21:23:31 | mikal | Oh, I am very grateful for the help. I definitely don't want to be chasing this over the weekend, but I am also very worried about missing feature freeze at this point. | |
| 21:23:55 | mikal | fatal: Unpack error, check server log | |
| 21:23:55 | mikal | error: remote unpack failed: error Missing tree 4a45ed28aa51f7ae9791b3f90a1fe8fef3c42dab | |
| 21:24:00 | mikal | I have no idea what that means. | |
| 21:24:08 | melwitt | no, we will get it working | |
| 21:29:17 | melwitt | bleh. my DNM patch just failed. must be something in my tempest patch, I'll figure out what's wrong with it | |
| 21:30:39 | mikal | I am just arguing with gerrit to try and remove that depends-on I only just added, but for reasons too boring to explain git review takes like five minutes in my workflow. | |
| 21:30:48 | opendevreview | Michael Still proposed openstack/nova master: libvirt: direct SPICE console object changes https://review.opendev.org/c/openstack/nova/+/926876 | |
| 21:30:49 | opendevreview | Michael Still proposed openstack/nova master: libvirt: allow direct SPICE connections to qemu https://review.opendev.org/c/openstack/nova/+/924844 | |
| 21:30:49 | opendevreview | Michael Still proposed openstack/nova master: libvirt: direct SPICE console database changes https://review.opendev.org/c/openstack/nova/+/926877 | |
| 21:31:22 | mikal | Right, depends-on for my bad tempest change removed. melwitt, feel free to add a depends on to your change if that's easier. I am not offended by you uploading patches to that series. | |
| 21:33:36 | melwitt | sure. yeah adding depends-on would probably be better. tbh I didn't expect anyone would be doing anything with it today so I just stacked it on top :P | |
| 21:33:55 | melwitt | and will obviously abandon once this gets working | |
| 21:37:23 | mikal | Whatever works for you is fine with me. | |
| 21:37:49 | mikal | I need to go for a run before it gets any hotter outside, but I will check back in a couple of hours. | |
| 22:07:56 | opendevreview | Max proposed openstack/nova master: feat: nova-manage db instance events cleanup https://review.opendev.org/c/openstack/nova/+/916927 | |
| 23:28:49 | melwitt | sean-k-mooney, mikal: I think I got things working for 2.98 in tempest here https://review.opendev.org/c/openstack/tempest/+/942492. the spice series passes the spice tempest test with that cc gmann | |
| 23:39:11 | sean-k-mooney | so skimming that i think that looks corect to me | |
| 23:39:30 | sean-k-mooney | and it "shoudl" fix the spice test i belvie | |
| 23:39:38 | melwitt | I tested it here https://review.opendev.org/c/openstack/nova/+/942493 | |
| 23:39:54 | sean-k-mooney | oh awsome | |
| 23:40:49 | melwitt | it definitely works for now. but if we change the schema again (I saw the fix you are working on) then we'll need to correspondingly adjust tempest after I think | |
| 23:41:15 | sean-k-mooney | am we should not need to change it | |
| 23:42:12 | sean-k-mooney | if we change it from properties to image properties like ragesh suggested we woudl be ut that need a new microversion | |
| 23:42:19 | sean-k-mooney | so we shoudl avoid that | |
| 23:42:21 | melwitt | we definitely need to add 'null' as a type bc the api is currently returning None for some properties like hw_input_bus. I comment about it on your patch | |
| 23:43:32 | mikal | Thanks for chasing this down guys, I really appreciate it. | |
| 23:43:40 | melwitt | the previous failure of the spice test was because 'null' wasn't allowed by the schema, so I added it to the schema on the tempest side | |
| 23:44:00 | mikal | Is there anything I can do here to be helpful, or do you think its largely sorted out now? | |
| 23:45:33 | melwitt | afaict I think it's sorted. just need to merge this tempest patch and then we need to re-apply +2s and +W the bottom 3 patches. I assume we can diff against PS40 to see that no code changed | |
| 23:45:50 | melwitt | *3 spice patches | |
| 23:47:23 | mikal | The first two still have their +2s in gerrit, the third one was clobbered. | |
| 23:50:40 | sean-k-mooney | i guess we have 2 chocies | |
| 23:50:55 | sean-k-mooney | we coudl add null or we could jsut not print that key in that case | |
| 23:51:16 | sean-k-mooney | is None a valid value or is it just printign that because its unset? | |
| 23:52:06 | sean-k-mooney | i had to step away to let freya (my frieds dog im minding for the week) out | |
| 23:52:16 | sean-k-mooney | but i started lookign at the third spice patch | |
| 23:52:45 | sean-k-mooney | melwitt: we can add a depends on on my image property patch to make sure its tested with your tempest patch to make sure we confrom to it | |
| 23:54:06 | melwitt | you know better than I do whether None is a valid/meaningful value. I would think it's not but I don't know for sure | |
| 23:54:48 | sean-k-mooney | we would have had to require it to be set to null in glance i think | |
| 23:55:10 | sean-k-mooney | i suspect not but we can adress that in my bugfix patch | |
| 23:55:25 | melwitt | ++ | |
| 23:55:59 | melwitt | mikal: ah ok, thanks | |
| 23:56:23 | sean-k-mooney | i readded +2 on hte third | |
| 23:58:05 | mikal | sean-k-mooney: I assume the door has now closed on that os-traits change landing this cycle, yeah? | |
| 23:58:05 | sean-k-mooney | so reading your schema patch comment | |
| 23:58:21 | sean-k-mooney | for the bools ectra i thinks we can make sure we alwasy convert the vlues to stings | |
| 23:58:50 | sean-k-mooney | mikal: am technically the release should have haappend today | |
| 23:59:10 | sean-k-mooney | but in realisty i dont know if it has so we could try and squeeze it in | |
| 23:59:13 | sean-k-mooney | let me check | |
| #openstack-nova - 2025-02-22 | |||
| 00:01:28 | sean-k-mooney | technially the final reales patch for os-traits has not been done yet... that may have been an oversite. | |
| 00:02:09 | mikal | If its too late its too late, but it does mean there's no chance of getting the USB stuff in and we should let shadow.tech know that. | |
| 00:03:46 | melwitt | does the os-traits have usefulness without the remaining spice patches? IMHO I'm not sure it's realistic to land those by the end of next week | |
| 00:04:21 | sean-k-mooney | not really | |
| 00:04:34 | sean-k-mooney | the vid series has some utility without the speice patches | |
| 00:04:35 | mikal | No, its only helpful if the usb and sound series lands as well. That said, that functionality has been trying to get in for two releases now from multiple proposers. | |
| 00:04:51 | sean-k-mooney | but the os-triats one doesnt have unitlyt with out the usb/sound changes | |
| 00:06:02 | sean-k-mooney | in either case i think there is a stong argument to try and land the usb/sound change after rc1 is cut early in 2025.2 | |
| 00:06:16 | sean-k-mooney | i.e. before the ptg | |
| 00:07:39 | mikal | I am not currently passionate about writing the spec for that bit for a third time. | |
| 00:08:04 | sean-k-mooney | i was going to suggest a specless bluepirnt for the remaing part | |
| 00:08:30 | sean-k-mooney | but i dont know how others would feel about that | |
| 00:22:58 | opendevreview | sean mooney proposed openstack/nova master: only show standard image properties in server show. https://review.opendev.org/c/openstack/nova/+/942413 | |
| 00:23:44 | sean-k-mooney | melwitt: mikal i assume we wong get the tempest fix merged over the weekend. | |
| 00:24:17 | sean-k-mooney | in the unlikely event that happens ill check back sunday but we can revisity this on monday any hopefuly get everythin lined up | |
| 00:24:43 | mikal | Yeah cool. Fair enough. Once again I find myself thanking you for pulling a late night. | |
| 00:25:39 | sean-k-mooney | well i was actully just turning off my laptop after taking a nap and rememeber in left it streamign a game :) | |
| 00:25:50 | sean-k-mooney | so it was good? timing | |
| 00:26:14 | sean-k-mooney | i had litrally jsut sad down when melwitt push the patch | |
| 00:27:43 | mikal | Well if OpenStack still did physical events, and if I ever went to one, both of which seem unlikely, I'd definitely owe you a drink. | |
| 00:27:58 | mikal | (Also I think that sentence had too many commas) | |
| 00:28:16 | sean-k-mooney | if the foundation becomes aprt of the linux foundation you never knwo they could come back | |
| 00:28:56 | sean-k-mooney | its ok you can use my allocation of punctuation :) i rarely do | |
| 00:30:15 | sean-k-mooney | mikal: by the way are you based out of austrialia or is my bring entrily miss rememebrign that | |
| 00:32:21 | sean-k-mooney | the reason i ask is i find it funny how often you melwitt and i tend to be online at similar tiems even those we are goegraphically spread | |
| 00:35:18 | sean-k-mooney | althogh i think as a programmer and irish person im geneitclly alergic to daylight and being awake before 11 am, that and the fact i fundemtly object to having to 7:30's in a day but am quite happy with two 4 o'clocks :) | |
| 00:36:11 | sean-k-mooney | with that said freya will be up early so i better go to bed because she will wake me regarless of if i do or not o/ | |
| 00:38:18 | mikal | Yeah, I'm in Canberra Australia (three hours drive south of Sydney). When I am talking to you in your early evening that's normally about 6am my time. That's why I disappear after a couple of hours to go to work. | |
| 00:38:45 | mikal | I've always been a morning person which helps with overlapping with US people. | |
| 00:39:12 | melwitt | haha, I tend to be up really late in general. gnight sean-k-mooney o/ | |
| 00:39:52 | melwitt | I'm lucky our dog is also not a morning person so don't really have to get up early for him | |
| 00:40:43 | mikal | Yeah, it helps that you're on PST too -- overlapping with the US east coast is hard, especially in winter here. I might be a morning person, but 4am meetings are a bridge too far. | |
| 00:42:56 | gmann | melwitt: ack, will check once gate is finished | |
| 00:43:36 | melwitt | thanks gmann | |
| 00:45:46 | sean-k-mooney | im waiting for my local fucntional test to finish before i actuly leave but as an asseie the eventlet deprecaate messages add a lot of spam to the test output | |
| 00:45:59 | sean-k-mooney | we proably need to supress that message | |
| 00:46:45 | melwitt | yeah. if I knew how to do it I would propose it 😛 | |
| 00:48:52 | sean-k-mooney | we need to install a python log filter like this https://github.com/openstack/nova/blob/master/nova/config.py#L56C5-L89 in the tests | |
| 00:50:00 | melwitt | ah | |
| 00:50:06 | opendevreview | sean mooney proposed openstack/nova master: only show standard image properties in server show. https://review.opendev.org/c/openstack/nova/+/942413 | |
| 00:50:21 | sean-k-mooney | i did actuly have a bug but im not going to wait for ^ to reun again o/ | |
| 00:51:21 | sean-k-mooney | i often forget the for key, val in {} | |
| 00:51:40 | sean-k-mooney | doesnt actully work because iter on dict gives you the keys | |
| 00:52:48 | sean-k-mooney | im a little upset that the unit test passed ant that only showed up in functional test | |
| 00:52:49 | melwitt | ah yeah | |
| 00:53:04 | sean-k-mooney | that tells me i need more unit test coverage in my patch | |