| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2022-07-21 | |||
| 13:09:05 | bauzas | sean-k-mooney: nope | |
| 13:09:06 | gibi | nope | |
| 13:09:18 | bauzas | and I don't see a netsplit | |
| 13:09:34 | sean-k-mooney | looks like the vpn dropped again | |
| 13:09:42 | sean-k-mooney | ok ill fix that | |
| 13:54:26 | opendevreview | Rico Lin proposed openstack/nova master: Add locked_memory extra spec and image property https://review.opendev.org/c/openstack/nova/+/778347 | |
| 13:54:27 | opendevreview | Rico Lin proposed openstack/nova master: Add traits for viommu model https://review.opendev.org/c/openstack/nova/+/844507 | |
| 13:54:27 | opendevreview | Rico Lin proposed openstack/nova master: libvirt: Add vIOMMU device to guest https://review.opendev.org/c/openstack/nova/+/830646 | |
| 14:05:19 | bauzas | folks, I need to disappear for a dentist appointment | |
| 14:15:45 | EugenMayer | sean-k-mooney thank you for helping once again, sorry forgot the mention the obvious! | |
| 14:23:44 | sean-k-mooney | EugenMayer: no worries sorry i did not see how to make it work end to end but happy your prod it working again | |
| 15:28:59 | opendevreview | Amit Uniyal proposed openstack/nova master: For evacuation, ignore if task_state is not None https://review.opendev.org/c/openstack/nova/+/848886 | |
| 15:28:59 | opendevreview | Amit Uniyal proposed openstack/nova master: add regression test case for bug 1978983 https://review.opendev.org/c/openstack/nova/+/849104 | |
| 15:48:04 | colby_ | one instead of existing. Let me know if you need anything from me as manually deleting the mdevs is not ideal for production at this point. | |
| 15:48:04 | colby_ | sean-k-mooney: bauzas: Still trying to get our vGPU stuff working. Here is the bug I filed: https://bugs.launchpad.net/nova/+bug/1981631. Im happy to be a guinea pig for testing. I can also poke around and help gather any info you need to help identify the issue. I did some logging of python code variables to see what rp(pri device) its trying to use when they are all created already and its def trying to use a new | |
| 15:48:30 | bauzas | colby_: fwiw, I opened a bug against libvirt for it | |
| 15:48:47 | bauzas | colby_: https://bugzilla.redhat.com/show_bug.cgi?id=2109450 | |
| 15:51:30 | colby_ | ok thanks! Its good to know its on that end I guess. I guess that is the drawback of using centos8 stream which has the newer version of libvirt | |
| 15:52:59 | colby_ | So will using that new API be backported to older releases once you add that in the next release? | |
| 15:53:44 | bauzas | colby_: or libvirt will fix it | |
| 15:55:33 | colby_ | or are we stuck hoping redhat will add the old functionality back in? | |
| 15:55:37 | colby_ | ah ok | |
| 16:01:29 | bauzas | colby_: tbc, I'm working on a POC to use the new API | |
| 16:01:37 | bauzas | but ideally, the regression should be fixed | |
| 16:02:26 | colby_ | cool. thanks for pointing me to that bug. Ill keep an eye on it. | |
| 16:02:28 | opendevreview | Balazs Gibizer proposed openstack/nova master: Reproducer for bug 1982497 https://review.opendev.org/c/openstack/nova/+/850672 | |
| 16:23:01 | opendevreview | Sylvain Bauza proposed openstack/nova master: Reproducer for bug 1951656 https://review.opendev.org/c/openstack/nova/+/850673 | |
| 17:51:16 | opendevreview | ribaudr proposed openstack/nova master: Allow unshelve to a specific host (REST API part) https://review.opendev.org/c/openstack/nova/+/845897 | |
| 17:51:16 | opendevreview | ribaudr proposed openstack/nova master: Allow unshelve to a specific host (Compute API part) https://review.opendev.org/c/openstack/nova/+/831507 | |
| 17:57:23 | opendevreview | ribaudr proposed openstack/nova master: Allow unshelve to a specific host (REST API part) https://review.opendev.org/c/openstack/nova/+/845897 | |
| 18:24:31 | opendevreview | liuhuajie proposed openstack/nova master: The default value of the dictionary get method is None, so remove the default value of None in the get method https://review.opendev.org/c/openstack/nova/+/850450 | |
| #openstack-nova - 2022-07-22 | |||
| 07:11:33 | opendevreview | wangkuntian proposed openstack/nova stable/yoga: Modify the command of getting serial console https://review.opendev.org/c/openstack/nova/+/850723 | |
| 07:18:17 | bauzas | morning | |
| 07:24:05 | gibi | good morning | |
| 08:09:39 | Uggla | bauzas, hi, I pushed an update version of unshelve to host. If you can have a look. | |
| 08:10:02 | bauzas | Uggla: I'll try but I'm overloaded by the vgpu regression | |
| 08:23:08 | opendevreview | ribaudr proposed openstack/nova master: Allow unshelve to a specific host (Compute API part) https://review.opendev.org/c/openstack/nova/+/831507 | |
| 08:23:09 | opendevreview | wangkuntian proposed openstack/nova master: Modify the command of getting serial console https://review.opendev.org/c/openstack/nova/+/850729 | |
| 08:23:09 | opendevreview | ribaudr proposed openstack/nova master: Allow unshelve to a specific host (REST API part) https://review.opendev.org/c/openstack/nova/+/845897 | |
| 08:48:56 | opendevreview | wangkuntian proposed openstack/nova master: Modify the command of getting serial console https://review.opendev.org/c/openstack/nova/+/850734 | |
| 09:01:52 | opendevreview | wangkuntian proposed openstack/nova master: Modify the command of getting serial console https://review.opendev.org/c/openstack/nova/+/850729 | |
| 09:07:53 | gibi | bauzas: now I managed to get back to your keypair patch and left feedback https://review.opendev.org/c/openstack/nova/+/849133 | |
| 09:19:57 | bauzas | gibi: ack, will look | |
| 09:53:53 | bauzas | gibi: thanks for the review, will update it this afternoon | |
| 10:24:23 | auniyal | O/, please have a review this - https://review.opendev.org/c/openstack/nova/+/848886 | |
| 10:31:00 | sean-k-mooney | over all it looks ok there are proably some spelling nits that i cant see | |
| 10:31:29 | sean-k-mooney | we dont need the other section in the release note but i have leaft that feedback on the review | |
| 11:01:17 | gibi | I left feedback too | |
| 11:01:53 | sean-k-mooney | auniyal: it would be helpful if you could also mark some fo the coments as resolved | |
| 11:03:03 | sean-k-mooney | im currently in the middle of something downstream so cant really spend much time on the review currently but in general when you push a new revision ist good if you can look at the old commens in gerrit, confirm its adressed and mark them done. | |
| 11:08:06 | opendevreview | Balazs Gibizer proposed openstack/nova master: Prevent instance.save() under mutated migration context https://review.opendev.org/c/openstack/nova/+/850746 | |
| 11:09:56 | gibi | fyi: if somebody has an idea to handle the fallout of ^^ (and the parent bug) then I'm all ears as I can only solve this back hacking either instance.save() or the rollback_live_migration_at_destination driver call | |
| 11:10:12 | gibi | s/back/by/ | |
| 11:15:19 | sean-k-mooney | gibi: sorry cant load context now but remind me on monday and ill think about it | |
| 11:15:34 | gibi | sean-k-mooney: sure, I know you are knee deep in a DB dump | |
| 11:15:49 | sean-k-mooney | also didnto know that Instance.mutated_migration_context() was a thing | |
| 11:16:26 | gibi | it is a beautiful if you use it in a small scope, but applying to a big scope is dangerous like hell | |
| 11:16:41 | gibi | anyhow it is not super urgent | |
| 11:18:17 | sean-k-mooney | im glad you were able to find the cause of that issue one way or another | |
| 11:20:01 | gibi | it was a tricky one as we have two RPC to the dest during migration rollback, the first is a cast the second is a call. The second one saves the proper instance data but the first one saves the wrong data. However as the first one is cast it can happen _after_ the second call. | |
| 11:20:29 | sean-k-mooney | ah | |
| 11:20:40 | sean-k-mooney | so this is also why its not determinitic | |
| 11:20:52 | sean-k-mooney | also cast_as_call-- | |
| 11:20:57 | gibi | it was deterministic for James but it wasn't in a func env | |
| 11:21:13 | sean-k-mooney | well likely because of the cast as call fixutre | |
| 11:21:22 | gibi | yepp | |
| 11:21:25 | sean-k-mooney | that will make it so the cast alwasy happens first | |
| 11:21:48 | gibi | that was the first thing to hide the issue, but then I still had to slow down the execution of the cast to let the call finish first | |
| 11:22:06 | gibi | in real env during the cast we do disk IO (delete disk) | |
| 11:22:24 | gibi | so that could take time and nova switch to the next RPC which is the call | |
| 11:22:29 | sean-k-mooney | well we can do what we did in other case | |
| 11:22:49 | sean-k-mooney | add a lock to the mocked op | |
| 11:23:21 | sean-k-mooney | so that it will block then to the call then unlock to let the cast continue | |
| 11:23:22 | gibi | while locking would force a better serialization of the events it will still create a time window where the db has the wrong instance data | |
| 11:23:39 | gibi | we would save the wrong data first | |
| 11:23:40 | sean-k-mooney | oh i just mean in the func test | |
| 11:23:47 | gibi | ahh | |
| 11:24:01 | gibi | in the func test I used threading.Condition to serialize the thing | |
| 11:24:14 | sean-k-mooney | ack ya same idea | |
| 11:24:16 | gibi | yepp | |
| 11:24:19 | gibi | that is basically a lock | |
| 11:24:44 | gibi | I actually needed two as I had to prevent instance.delete to happen before the cast too :D | |
| 11:24:58 | sean-k-mooney | :) fun | |
| 11:25:04 | gibi | it was indeed :) | |
| 11:25:44 | sean-k-mooney | i have stared and set rp +1 form me on both | |
| 11:25:50 | sean-k-mooney | so ill get back to them monday | |
| 11:26:07 | gibi | ack, thanks | |
| 11:37:03 | sean-k-mooney | gibi: if you respin care to pull https://review.opendev.org/c/openstack/nova/+/850672/1/nova/tests/functional/integrated_helpers.py into its own change first | |
| 11:37:22 | sean-k-mooney | just thinking that might be good to be able to backport seperatly | |
| 11:37:27 | sean-k-mooney | although not required | |
| 11:58:46 | gibi | sean-k-mooney: sure I can, but I guess we will backport the reproducer and the eventual fix anyhow | |
| 11:59:14 | sean-k-mooney | yep its not that important | |
| 11:59:37 | sean-k-mooney | just one of the reaosn i have used the freestanding approch in the past is to not include cast_as_call | |
| 12:00:03 | sean-k-mooney | i.e. if you just use the mixins that is not added unless you do it yourslef in teh func test | |
| 12:00:19 | gibi | sean-k-mooney: true | |
| 12:00:39 | gibi | I'm not sure how easy it would be to move the libvirt functional tests away from the _IntegratedBase | |
| 12:01:07 | sean-k-mooney | i would move it to https://github.com/openstack/nova/blob/master/nova/tests/functional/libvirt/integrated_helpers.py#L23= first | |
| 12:02:11 | sean-k-mooney | but as stephen notes https://github.com/openstack/nova/blob/master/nova/tests/functional/integrated_helpers.py#L1231-L1233= | |
| 12:02:24 | sean-k-mooney | but if we were to remove one or the ohter i would remove _IntegratedTestBase | |
| 12:03:07 | sean-k-mooney | crap meeting | |