Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2026-03-19
11:48:25 zigo I've set systemd's TimeOutStop to 185 as per release notes, but it doesn't seem to work as expected.
11:52:49 zigo Also, when spawning my first Gazpacho VM, I'm getting:
11:52:49 zigo nova.exception.ImageUnacceptable: Image 1f9640fd-75ff-4bee-87b0-e8f0c5c57ec2 is unacceptable: Image is not raw format
11:52:49 zigo even though the image really is a qcow2 and is uploaded as such without mistake.
11:54:00 zigo I tried:
11:54:00 zigo disable_deep_image_inspection=true
11:54:00 zigo but that's not helping.
11:55:43 zigo If I add 'qcow2' to the line:
11:55:43 zigo if image_meta.get('disk_format') not in ['raw', 'iso']:
11:55:43 zigo then it fails later on with: 'No image locations are accessible'
11:55:43 zigo when really, the image is in Glance, and can be downloaded.
11:56:50 zigo Ah no, checksum missmatch when doing "image save" ...
12:44:33 zigo Weird, whatever the image I upload, I always get the same os_hash_value ...
15:03:20 opendevreview ribaudr proposed openstack/nova master: Reproduce bug #2144660: NUMA cell sorting bug for non-PCI VMs https://review.opendev.org/c/openstack/nova/+/978885
15:03:21 opendevreview ribaudr proposed openstack/nova master: Fix NUMA cell sorting for non-PCI VMs with pack strategy https://review.opendev.org/c/openstack/nova/+/978954
15:25:53 dansmith Uggla: quick question on the second one
16:27:11 opendevreview Stephen Finucane proposed openstack/nova master: trivial: Clean manager.Manager, service.Service signatures https://review.opendev.org/c/openstack/nova/+/764806
17:05:57 dansmith sean-k-mooney: did you just comment on PS1 of Uggla's patch but -1 the current PS4?
17:07:36 sean-k-mooney dansmith: im review the patch again. i pushed the comment i had form last week
17:07:44 sean-k-mooney now im doing the full review
17:08:18 sean-k-mooney most of them have been adressed althoguht he commit message is slightly incorrect
17:08:28 sean-k-mooney the repoducer tests look mostly fine to me however
17:12:18 dansmith okay
17:13:46 sean-k-mooney i commented on ps1 but left a 0 cr on ps4 by the way i didnt actully leave a -1 since i needed to review the new reviesion . i left a +1 because the ai atributation is not quite corect
17:14:41 sean-k-mooney same issue on the fix patch but it looks mostly correct to me. im just reviewing the tests now
17:15:34 dansmith gerrit was showing your vote as -1 even on PS4
17:15:51 dansmith I was just thinking maybe you had an old link and didn't realize you were reviewing the old one
17:16:05 sean-k-mooney that not what i see in gerrit
17:16:31 sean-k-mooney but i did say `-1 until this is removed` i jsut didnt click the button
17:16:51 sean-k-mooney i was -1 on the fix patch on patchset 1
17:17:32 sean-k-mooney anyway Uggla it woudl be nice ot have a sort release note in teh fix patch. im still reviewing it but sofar it looks correct
17:17:39 dansmith not now, since you've +1d.. anyway, was just trying to point it out.. didn't know you were committing old review comments and were planning to circle back
17:18:15 sean-k-mooney Uggla: im double checkign the change you had to make to the other tests
17:20:06 Uggla sean-k-mooney, sure I can add a release note, no problem.
17:20:37 sean-k-mooney im currently thinking through thet logic of test_create_server_with_pci_dev_and_numa_fails
17:21:36 sean-k-mooney in your new code your bootign 2 vms the problem with that is the host shoudl now be rejected dueo to the cpu avialableity i htink
17:21:59 sean-k-mooney we may need to boot 2 vms then delete the first one
17:22:12 dansmith I had that comment and he adjusted the test
17:22:16 sean-k-mooney so that there are enouch cpus free btu the numa node with the pci device is full
17:22:48 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/978954/4/nova/tests/functional/libvirt/test_pci_sriov_servers.py#4654
17:22:59 sean-k-mooney its still not correct in v4
17:23:11 Uggla sean-k-mooney, it is what I did, boot 2 vm, delete 1, spawn a new one that must fail.
17:23:31 Uggla unless I miss something.
17:23:42 sean-k-mooney Uggla: i dont see the vm delete you ddidnt delete server 1
17:23:50 sean-k-mooney did you do that locally?
17:24:49 dansmith sean-k-mooney: L5870
17:25:09 dansmith and if you see my previous review feedback it'll point it out
17:25:09 sean-k-mooney thats a diffent test
17:26:12 sean-k-mooney test_create_server_with_pci_dev_and_numa_fails vs test_create_server_with_pci_dev_and_numa
17:26:23 sean-k-mooney the fails one is the one that is incorrect
17:27:01 sean-k-mooney the fails one now would fail even without the pci device becasue there are not enough dedicated cpus
17:28:41 dansmith in my previous feedback, I noted that that test was the same as the fails one above, and commented that we should cover the case where he deleted one to make sure it was failing for the right reason
17:28:56 dansmith seems like both cases are covered now, I'm not sure why it matters which one does it
17:30:20 sean-k-mooney there are only 8 pCPUs on the host
17:30:34 sean-k-mooney if you boot 2 vms as palce hoslder and dont delete 1 of them
17:30:48 sean-k-mooney (that both request 4 cpus)
17:31:01 sean-k-mooney the third vm will get a no valid host regardless fo the pci request
17:33:54 Uggla sean-k-mooney, please make a comment I'll try to fix it tomorrow morning.
17:34:12 sean-k-mooney yep i have it written im jsut reviewing the rest so i can leave them all in one go
17:35:45 Uggla sean-k-mooney, also if you have not seen 946945: Add reproducer test for bug 2105896 | https://review.opendev.org/c/openstack/nova/+/946945 I rebase and stack the fix on top. Coz I would like to close this bug too.
17:36:57 sean-k-mooney Uggla: thanks for rebasing it. no i didnt see that you did that
17:37:33 sean-k-mooney and yes it would be nice to fix that im fine with stacking them or if you want to take that over feel free
17:37:51 Uggla I had to signoff the patch, so that's the reason why I set me as co author.
17:38:01 sean-k-mooney ya that ok
17:38:17 sean-k-mooney i just created the repoducer to help the person who was developign the bug fix
17:38:18 Uggla in any case keeping you as the original author.
17:39:22 sean-k-mooney well not quite i guess i created trhe repoducer ealier but never got around to fixing it
17:39:40 Uggla you can have a look at the fix too. Basically I have just added the reproducer part in the fix that was missing (so setting me as co author on the fix too).
17:40:46 Uggla So if you can confirm the direction is ok, then I'll look for cores to merge them.
17:49:35 sean-k-mooney sure proably not today but i can take a look tomorrow
17:49:51 sean-k-mooney ill see if i can review it today when i finish with your chagne
18:00:34 sean-k-mooney Uggla: comments pushed https://review.opendev.org/c/openstack/nova/+/978954/4/nova/tests/functional/libvirt/test_pci_sriov_servers.py there are a few minor testign regressions but the over all patch looks good
18:01:51 Uggla sean-k-mooney, thx, I'll try to fix that tomorrow morning with a fredh mind.
18:02:18 sean-k-mooney no worries. the changes are subtle so best not to do it while tired
19:33:12 opendevreview Lajos Katona proposed openstack/nova master: WIP: Use SDK for Neutron security-groups https://review.opendev.org/c/openstack/nova/+/981141
19:34:08 sean-k-mooney Uggla: oh i see did orgianly take over https://review.opendev.org/c/openstack/nova/+/946079/6 and update it for the first person tha ttried to fix it but they abandoned the patch and https://review.opendev.org/c/openstack/nova/+/970337 is a new attempt to fix it
19:35:47 sean-k-mooney so it not usign the orginal fix i was working at least not in the same way
20:51:24 sean-k-mooney Uggla: if you want to chat more about https://review.opendev.org/c/openstack/nova/+/970337/comment/4abdf064_10cf13ce/ tommorow let me know i think https://review.opendev.org/c/openstack/nova/+/946079 was more compelte fix and it shoudl not have been abandoned
21:52:57 opendevreview Dmitriy Chubinidze proposed openstack/nova master: docs: mention q35 machine type for UEFI guests https://review.opendev.org/c/openstack/nova/+/981413
21:59:53 opendevreview Dmitriy Chubinidze proposed openstack/nova master: docs: mention q35 machine type for UEFI guests https://review.opendev.org/c/openstack/nova/+/981413
#openstack-nova - 2026-03-20
10:25:35 opendevreview ribaudr proposed openstack/nova master: Reproduce bug #2144660: NUMA cell sorting bug for non-PCI VMs https://review.opendev.org/c/openstack/nova/+/978885
10:25:35 opendevreview ribaudr proposed openstack/nova master: Fix NUMA cell sorting for non-PCI VMs with pack strategy https://review.opendev.org/c/openstack/nova/+/978954
11:05:05 Uggla sean-k-mooney, dansmith, I have answered comments ^ so if you can have a new review run. It will be great.
11:28:01 opendevreview Dmitriy Chubinidze proposed openstack/nova master: docs: mention q35 machine type for UEFI guests https://review.opendev.org/c/openstack/nova/+/981413
12:28:07 sean-k-mooney Uggla: im +2 on both of those now. if you need to respine a release not would be nice but its not required
13:20:40 Uggla sean-k-mooney, yes I will write a release as discussed yesterday.
13:21:20 Uggla sean-k-mooney I'll keep you posted when it is done.
13:21:49 opendevreview ribaudr proposed openstack/nova master: Fix error when multiple security group in a project with same name https://review.opendev.org/c/openstack/nova/+/970337
13:22:45 Uggla sean-k-mooney I have just pushed an update of https://review.opendev.org/c/openstack/nova/+/970337. Tell me if I understand well what you wanted to achieve.
13:31:24 sean-k-mooney ack ill take a look after im off a call
13:31:54 opendevreview ribaudr proposed openstack/nova master: Fix NUMA cell sorting for non-PCI VMs with pack strategy https://review.opendev.org/c/openstack/nova/+/978954
13:36:51 Uggla sean-k-mooney release notes added in ^
13:53:52 sean-k-mooney +2 on the sorting series ill re review https://review.opendev.org/c/openstack/nova/+/970337/4 in a little more detail
13:54:31 sean-k-mooney i think you made the changes i asked for but it woudl also be nice to have a very short release note for the security group bug too
13:55:42 sean-k-mooney nova.tests.unit.api.openstack.compute.test_schemas.SchemaTest.test_schemas
13:56:05 sean-k-mooney seems to be hitting the per test timeout on slower nodes in the coverage job
13:56:20 sean-k-mooney i wonder if we shoudl tweak that in the tox.ini
13:59:28 Uggla sean-k-mooney ok for the release note. "seems to be hitting the per test timeout on slower nodes in the coverage job" how do you see that ?
14:13:44 sean-k-mooney i looked at the coverage job on the repoducer
14:14:05 sean-k-mooney the schema validation test timed out but i have seen that on one or two other patches i think

Earlier   Later