| opendevreview | Sylvain Bauza proposed openstack/nova master: Add a Caracal prelude section https://review.opendev.org/c/openstack/nova/+/912973 | 10:01 |
|---|---|---|
| opendevreview | Sylvain Bauza proposed openstack/nova master: Update contributor guide for 2024.2 Dalmatian https://review.opendev.org/c/openstack/nova/+/913481 | 10:03 |
| bauzas | elodilles: the gate failed again for https://review.opendev.org/c/openstack/nova/+/899625 :( | 10:06 |
| bauzas | elodilles: I rechecked it so I hope it will work now | 10:06 |
| elodilles | ACK | 10:25 |
| opendevreview | Merged openstack/nova master: libvirt: Cap with max_instances GPU types https://review.opendev.org/c/openstack/nova/+/899625 | 12:31 |
| bauzas | sean-k-mooney: gibi: could you please look at https://review.opendev.org/c/openstack/nova/+/912973 ? | 13:49 |
| bauzas | thanks | 13:49 |
| artom | Uggla, I think after the weekend and sleeping on it, the least bad solution I could come up with, at least in the short term, is to add a separate helper that "filters out" that new cpu shared set field from dst_numa_info, and make my code use it | 14:32 |
| artom | But medium term we need to get rid of the conditional do_cleanup RPC call | 14:32 |
| artom | I think if we were to test an instance with mdevs or vpmem with Lee's functional test, it would fail as well | 14:33 |
| artom | For the exact same reason: the presence of either one of those causes the cleanup to get called | 14:33 |
| bauzas | artom: Uggla: because today is the last day for creating a RC1, I find we will punt the cpu_shared_set changes to Dalmatian :( | 14:34 |
| bauzas | I'm just awaiting for https://zuul.openstack.org/status#902084 + the prelude | 14:34 |
| artom | Err, CPU power management doesn't work properly without it... | 14:34 |
| artom | gibi's going to be very angry at you if we don't fix it entirely | 14:35 |
| artom | ;) | 14:35 |
| artom | I guess we can debate releasing it with a known issue (if you have different cpu_shared_set on source and dest and use libvirt CPU power management, not only does live migration doesn't update the pins, but also you might end up on CPUs that are powered down and the instance just doesn't start)... | 14:37 |
| bauzas | artom: Uggla: so, then tell me about https://review.opendev.org/c/openstack/nova/+/877773 | 14:37 |
| artom | IMO it's fixable with a relatively small and clean-ish (as much as you can be clean when the whole basic principle is a mad genius hack) adjustment | 14:39 |
| dansmith | bauzas: will you +W that patch yourself? | 14:40 |
| dansmith | ... or do you need another review? looks like its 2x+2 already | 14:42 |
| artom | Uggla seems to be AFK at the moment, so I can update it myself since we're in a time squeeze | 14:42 |
| gibi | artom: bauzas: my powermgmt needs are targeting antelope. So if the cpu_share_set fix only lands in D then that will not prevent the backport to antelopej | 14:42 |
| bauzas | dansmith: which change you mean ? | 14:42 |
| dansmith | I'm talking about 902084 | 14:42 |
| bauzas | dansmith: it's already accepted, but it's running on the gate | 14:42 |
| dansmith | 902084 has no +W | 14:43 |
| bauzas | https://zuul.openstack.org/status#902084 | 14:43 |
| bauzas | ?* | 14:43 |
| dansmith | it's in check now | 14:43 |
| dansmith | "WorkflowNo votes" | 14:43 |
| bauzas | oh right | 14:43 |
| artom | gibi, fair... when does master switch to D? Tomorrow after today's RC1? | 14:43 |
| bauzas | shit | 14:43 |
| bauzas | dansmith: no worries, I'll +W it | 14:43 |
| dansmith | bauzas: cool | 14:43 |
| bauzas | thanks btw. | 14:44 |
| bauzas | artom: gibi: so you would want to have https://review.opendev.org/c/openstack/nova/+/877773 to be merged in C ? | 14:44 |
| bauzas | fwiw, we're still missing a second core here https://review.opendev.org/c/openstack/nova/+/903706/2 | 14:45 |
| artom | bauzas, I think gibi's saying that as long as it gets backported to Antelope, he's happy, regardless if it's merged on master in C or D | 14:45 |
| bauzas | my concern is that we couldn't backport https://review.opendev.org/c/openstack/nova/+/903706/2/nova/objects/migrate_data.py#246 | 14:45 |
| gibi | artom: yes, after RC1 we will have master open for D, but I guess bauzas wants to limit the changes on master until final RC to make sure any last minute change to C is easy to backport from masterj | 14:46 |
| artom | bauzas, right, that's why there's a backportable hack below it in https://review.opendev.org/c/openstack/nova/+/877773/9 | 14:46 |
| bauzas | I just +2d both of the changes | 14:46 |
| artom | bauzas, err, want to jump on a call for me to explain stuff? | 14:47 |
| bauzas | guys, that's like 4 days i told about rc1 | 14:48 |
| bauzas | I wonder what you guys want | 14:48 |
| gibi | I want to enable powermgmt in antelope :) | 14:49 |
| gibi | backporting the fix from C or D does not matter for mej | 14:51 |
| dansmith | bauzas: I'm trying to decide if that cpu shared set thing is okay or not | 15:13 |
| dansmith | I don't love deleting a non-versioned property and replacing it with a versioned one like it's doing | 15:14 |
| dansmith | I'm trying to decide if it's actually a problem, or just _very_ confusing (the latter is definitely true, IMHO | 15:14 |
| dansmith | ah, basically what sean-k-mooney said on PS1 | 15:16 |
| bauzas | dansmith: gibi: artom: Uggla: okay, so now I understand more | 15:18 |
| artom | dansmith, heh, bauzas and I were just on a call where I explained the nuances | 15:18 |
| artom | I can do the same for you if you wish :) | 15:18 |
| dansmith | bummer | 15:18 |
| bauzas | so, basically, we have an already bug with live-migration for cpu_shared_set | 15:18 |
| dansmith | yeah I get that | 15:19 |
| dansmith | how far back does it need to be backported? | 15:19 |
| bauzas | now with power management, you can live-migrate an instance that will use a wrong cpu which is a dedicated one, like before, but which will be off | 15:19 |
| bauzas | my point is that I don't think it's a RC1 needed bugfix | 15:20 |
| bauzas | what we can rather do is to merge https://review.opendev.org/c/openstack/nova/+/877773 in Dalmatian and then backport it down to Antelope if we want | 15:20 |
| bauzas | https://review.opendev.org/c/openstack/nova/+/903706/ won't be backported | 15:21 |
| bauzas | (as it's a new object version) | 15:21 |
| bauzas | so I'm just asking Uggla to modify https://review.opendev.org/c/openstack/nova/+/877773 to make sure it fixes the wrong functional test which is https://github.com/openstack/nova/blob/master/nova/tests/functional/regressions/test_bug_1899835.py | 15:21 |
| bauzas | (the last assert is wrong in the regression bug) | 15:22 |
| * bauzas goes off for 20 mins | 15:22 | |
| bauzas | (school) | 15:22 |
| gibi | elodilles: could you look at this b | 15:28 |
| gibi | ... this backport series https://review.opendev.org/c/openstack/nova/+/913198 | 15:28 |
| elodilles | gibi: ACK, i'll try to look into it when i get there | 16:03 |
| bauzas | gibi: dansmith: we need approval for the prelude patch https://review.opendev.org/c/openstack/nova/+/912973 | 16:03 |
| dansmith | ack | 16:04 |
| bauzas | thanks | 16:04 |
| gibi | elodilles: thanks | 16:07 |
| dansmith | bauzas: done | 16:22 |
| bauzas | cool thanks ! | 16:22 |
| * bauzas will rebase now | 16:23 | |
| opendevreview | Sylvain Bauza proposed openstack/nova master: Add a Caracal prelude section https://review.opendev.org/c/openstack/nova/+/912973 | 16:26 |
| bauzas | dansmith: ^ (when you have time) | 16:26 |
| opendevreview | Merged openstack/nova master: vgpu: Allow device_addresses to not be set https://review.opendev.org/c/openstack/nova/+/902084 | 16:58 |
| zigo | bauzas: Typ3 on the last 3 lines: | 17:18 |
| zigo | and yet another one another one (twice "another one") | 17:18 |
| zigo | Live-mig of vGPU is a nice feature, thanks! | 17:19 |
| zigo | Also, good thing for SEV. We were looking at implementing it, now I'll make sure we do that over Caracal! :) | 17:20 |
| opendevreview | Merged openstack/nova stable/2023.2: Add cpuset_reserved helper to instance NUMA topology https://review.opendev.org/c/openstack/nova/+/913193 | 18:35 |
| opendevreview | Merged openstack/nova stable/2023.2: Reproducer for not powering on isolated emulator threads cores https://review.opendev.org/c/openstack/nova/+/913194 | 18:35 |
| opendevreview | Sylvain Bauza proposed openstack/nova master: Add a Caracal prelude section https://review.opendev.org/c/openstack/nova/+/912973 | 18:59 |
| bauzas | dansmith: melwitt: could you please quickky approve now the prelude above $ so I could provide a new RC1 release patch version tonight? | 19:00 |
| dansmith | bauzas: done | 19:10 |
| bauzas | dansmith: melwitt: ta folks, as soon as it merges, I'll update the RC1 patch | 19:17 |
| opendevreview | Merged openstack/nova master: Add a Caracal prelude section https://review.opendev.org/c/openstack/nova/+/912973 | 19:46 |
| opendevreview | Merged openstack/nova stable/zed: Reproducer for bug 2003991 unshelving offloaded instance https://review.opendev.org/c/openstack/nova/+/888093 | 22:15 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!