| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2024-07-30 | |||
| 17:10:47 | sean-k-mooney | that sounds good | |
| 18:15:29 | opendevreview | Merged openstack/nova stable/2024.1: Refactor vf profile for PCI device https://review.opendev.org/c/openstack/nova/+/925181 | |
| 18:15:35 | opendevreview | Merged openstack/nova master: [tools] Ignore bot generated patches https://review.opendev.org/c/openstack/nova/+/924103 | |
| 18:15:42 | opendevreview | Merged openstack/nova master: [tools] Backport validator: handle unmaintained https://review.opendev.org/c/openstack/nova/+/924104 | |
| 20:32:46 | opendevreview | Masahito Muroi proposed openstack/nova master: Change instance_system_metadata value to Text type https://review.opendev.org/c/openstack/nova/+/925163 | |
| #openstack-nova - 2024-07-31 | |||
| 07:13:11 | opendevreview | Balazs Gibizer proposed openstack/nova master: [manila]Use client token when talking to manila https://review.opendev.org/c/openstack/nova/+/925277 | |
| 07:52:00 | opendevreview | Fabian Wiesel proposed openstack/nova master: Vmware: Remove uuid parameter from get_vmdk_info call https://review.opendev.org/c/openstack/nova/+/910627 | |
| 11:32:32 | opendevreview | Adam Oswick proposed openstack/nova master: Refactor _get_pci_passthrough_devices function https://review.opendev.org/c/openstack/nova/+/925382 | |
| 12:59:16 | jlejeune | sean-k-mooney: bauzas: hello, I found why the test_snapshot_running is now failing: https://bugs.launchpad.net/nova/+bug/2075178/comments/1 | |
| 13:00:02 | sean-k-mooney | jlejeune: yes we discussed that change in behavior with dansmith and it was noted in the commit message | |
| 13:00:37 | sean-k-mooney | jlejeune: https://github.com/openstack/nova/commit/d5a631ba7791b37e49213707e4ea650a56d2ed9e | |
| 13:01:08 | sean-k-mooney | """Note that this includes seemingly-unrelated test changes because we | |
| 13:01:10 | sean-k-mooney | were actually skipping the snapshot_running test for libvirt, which | |
| 13:01:12 | sean-k-mooney | has been a bug for years. In that test case, when we went to look | |
| 13:01:14 | sean-k-mooney | for image_meta.disk_format, that attribute was not set on the o.vo | |
| 13:01:16 | sean-k-mooney | object, which raised a NotImplementedError. That error is also checked | |
| 13:01:18 | sean-k-mooney | by the test to skip the test for drivers that do not support snapshot, | |
| 13:01:20 | sean-k-mooney | which meant that for libvirt, we haven't been running that case | |
| 13:01:22 | sean-k-mooney | beyond the point at which we create snapshot metadata and trip that | |
| 13:01:24 | sean-k-mooney | exception.""" | |
| 13:02:07 | sean-k-mooney | dan added some stubs https://github.com/openstack/nova/commit/d5a631ba7791b37e49213707e4ea650a56d2ed9e#diff-4104495b69c5ecf83423c1b9aafa2a8b55e2f8ebbd4cc36a5299206a004263d2R835 | |
| 13:02:16 | sean-k-mooney | buy i guess is missign one | |
| 13:02:27 | jlejeune | y es I saw them | |
| 13:04:55 | jlejeune | I will add new ones, but I have also to mock file_open() utils function, is it ok for you if I mock it at the test level ? | |
| 13:07:59 | sean-k-mooney | you shoudl not need to mock file_open | |
| 13:08:24 | sean-k-mooney | you idealy you woudl mock hihger then that as mocking file open may breakt things | |
| 13:09:16 | sean-k-mooney | we basically want to mock the file inspector methods i think | |
| 13:19:12 | sean-k-mooney | oh its nova.virt.libvirt.utils.file_open | |
| 13:19:30 | sean-k-mooney | not open() | |
| 13:19:42 | sean-k-mooney | ya we might want to add that to the global poion at some point | |
| 13:23:48 | jlejeune | :) | |
| 13:47:09 | elinux | can someone comment on this please https://bugs.launchpad.net/nova/+bug/2075150 | |
| 13:56:25 | elinux | anyone online yet ? | |
| 13:57:38 | opendevreview | Julien Le Jeune proposed openstack/nova master: Fix test_snapshot_running https://review.opendev.org/c/openstack/nova/+/925208 | |
| 14:07:08 | opendevreview | Julien Le Jeune proposed openstack/nova stable/2023.2: Refactor vf profile for PCI device https://review.opendev.org/c/openstack/nova/+/905135 | |
| 14:24:33 | jlejeune | elodilles: hello, I've updated the review by cherry-picking the right commit from 2024.1 branch if you want to have a look: https://review.opendev.org/c/openstack/nova/+/905135 | |
| 14:25:45 | opendevreview | Adam Oswick proposed openstack/nova master: Refactor _get_pci_passthrough_devices function https://review.opendev.org/c/openstack/nova/+/925382 | |
| 14:30:34 | opendevreview | Julien Le Jeune proposed openstack/nova master: Fix test_snapshot_running https://review.opendev.org/c/openstack/nova/+/925208 | |
| 14:35:52 | elodilles | jlejeune: looks good, thanks! +2+W'd | |
| 16:49:12 | opendevreview | Merged openstack/nova stable/2024.1: [tools] Ignore bot generated patches https://review.opendev.org/c/openstack/nova/+/925278 | |
| 16:49:18 | opendevreview | Merged openstack/nova stable/2024.1: [tools] Backport validator: handle unmaintained https://review.opendev.org/c/openstack/nova/+/925279 | |
| 18:03:37 | opendevreview | Merged openstack/nova stable/2023.2: Refactor vf profile for PCI device https://review.opendev.org/c/openstack/nova/+/905135 | |
| 21:42:06 | opendevreview | Merged openstack/nova stable/2024.1: cpu: Only check governor type on online cores https://review.opendev.org/c/openstack/nova/+/924514 | |
| #openstack-nova - 2024-08-01 | |||
| 07:04:25 | opendevreview | Julien Le Jeune proposed openstack/nova stable/2023.1: Refactor vf profile for PCI device https://review.opendev.org/c/openstack/nova/+/905137 | |
| 08:41:11 | opendevreview | Merged openstack/nova stable/2023.2: [tools] Ignore bot generated patches https://review.opendev.org/c/openstack/nova/+/925280 | |
| 08:41:17 | opendevreview | Merged openstack/nova stable/2023.2: [tools] Backport validator: handle unmaintained https://review.opendev.org/c/openstack/nova/+/925281 | |
| 08:48:51 | Taketani_Ryo | Hi, I have a question about how to use GPU on OpenStack VM. Can I use Multi-Instance GPU(MIG) on OpenStack VM? I understand that we can use vGPUs on OpenStack. But, if possible, we use MIG instances on VM because the license is necessary for vGPU. | |
| 10:13:21 | bauzas | Taketani_Ryo : Nvidia MIG support relies eventually on VFIO mediated devices that are eventually created above SR-IOV VFs, which themselves are related to MIG GIs. | |
| 10:13:43 | bauzas | Taketani_Ryo: since the vGPUs are actually mdevs, Nova can use them | |
| 10:13:52 | bauzas | so yeah, it would work | |
| 10:32:36 | opendevreview | Merged openstack/nova stable/2023.1: [tools] Ignore bot generated patches https://review.opendev.org/c/openstack/nova/+/925282 | |
| 10:32:42 | opendevreview | Merged openstack/nova stable/2023.1: [tools] Backport validator: handle unmaintained https://review.opendev.org/c/openstack/nova/+/925283 | |
| 12:21:29 | opendevreview | Merged openstack/nova stable/2023.1: Refactor vf profile for PCI device https://review.opendev.org/c/openstack/nova/+/905137 | |
| 13:21:51 | opendevreview | Merged openstack/nova stable/2023.1: Fix exception catch when volume mount fails https://review.opendev.org/c/openstack/nova/+/898954 | |
| 14:12:43 | bauzas | elodilles: I guess we can now prepare a 2024.1 release | |
| 14:38:23 | elodilles | bauzas: ACK. i thought we want to release all stable versions at the same time, but you are right, we don't need to wait. let me generate the release patch | |
| 14:38:35 | bauzas | cool | |
| 14:38:50 | bauzas | elodilles: I'll disappear for a long PTO until Aug 25 btw. | |
| 14:39:01 | bauzas | like I said on the upstream meeting | |
| 14:41:34 | elodilles | bauzas: ACK | |
| 14:43:01 | elodilles | i'll ping the release liaisons for the next releases when you'll be on vacation ( <<< auniyal Uggla o:)) | |
| 14:43:41 | elodilles | bauzas: have a great & relaxing time off o/ | |
| 14:43:51 | bauzas | thanks | |
| 14:43:57 | bauzas | not yet ended for today tho :) | |
| 14:47:38 | elodilles | bauzas: in that case, here is the release patch :D https://review.opendev.org/c/openstack/releases/+/925493 | |
| 15:04:53 | elodilles | thanks o:) | |
| 17:48:59 | opendevreview | Ghanshyam proposed openstack/nova master: DNM: testing new RBAC default change in oslo policy https://review.opendev.org/c/openstack/nova/+/925515 | |
| 18:04:00 | opendevreview | Ghanshyam proposed openstack/nova master: Remove default override for RBAC config options https://review.opendev.org/c/openstack/nova/+/925519 | |
| 18:42:16 | opendevreview | Elod Illes proposed openstack/nova stable/2024.1: [CI] Replace deprecated regex https://review.opendev.org/c/openstack/nova/+/925522 | |
| 18:42:53 | opendevreview | Elod Illes proposed openstack/nova stable/2023.2: [CI] Replace deprecated regex https://review.opendev.org/c/openstack/nova/+/925523 | |
| 18:44:58 | opendevreview | Elod Illes proposed openstack/nova stable/2023.1: [CI] Replace deprecated regex https://review.opendev.org/c/openstack/nova/+/925524 | |
| #openstack-nova - 2024-08-02 | |||
| 04:55:28 | auniyal | elodilles, sure :) | |
| 13:05:19 | opendevreview | Julien Le Jeune proposed openstack/nova master: nova-condutor put instance in error state https://review.opendev.org/c/openstack/nova/+/901655 | |
| 13:24:55 | opendevreview | Julien Le Jeune proposed openstack/nova master: Fix request_specs not updated after failed rebuild https://review.opendev.org/c/openstack/nova/+/899379 | |
| 13:28:15 | jlejeune | bauzas: elodilles: what do you think about that https://review.opendev.org/c/openstack/nova/+/899379 ? | |
| 13:30:20 | opendevreview | Merged openstack/nova stable/2023.2: cpu: Only check governor type on online cores https://review.opendev.org/c/openstack/nova/+/924517 | |
| 15:37:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: [manila]Lock/Unlock share by the share_mapping https://review.opendev.org/c/openstack/nova/+/925606 | |
| 15:37:30 | opendevreview | Balazs Gibizer proposed openstack/nova master: [manila]Lock/Unlock share by the share_mapping https://review.opendev.org/c/openstack/nova/+/925606 | |
| 15:40:34 | elinux | hi folks | |
| 15:40:40 | elinux | anyone online yet ? | |
| 16:06:10 | elinux | anyone online ? | |
| 16:08:45 | clarkb | elinux: people generally lurk. If you have a question its often best to just ask and if people are able they can answer | |
| 16:11:09 | elinux | I am trying to a add a new filter to nova scheduler. during that process, I want to execute a db command on nova db along these lines "select uuid from instances where project_id=<blah blah> and vm_state is not 'error'" | |
| 16:12:03 | elinux | i want to execute that command in that filter testing file to make a decission for custom filtering. How do I do it using any predefined nova or db objects ? | |
| 16:14:40 | elinux | my main target is to get the number of rows returned by that query | |
| 16:21:12 | gibi | elinux: the filters are filtering HostState objects and HostState already has an instances field https://github.com/openstack/nova/blob/bb2d7f9cad577f3a32cb9523e2b1d9a6d6db3407/nova/scheduler/host_manager.py#L144 that holds a dict of instance_uuid: Instance ovo pairs. So your custom filter just needs to look at the instances dict in each host state and count the instances is bad state. | |
| 16:22:30 | elinux | gibi: but can I filter out those instances per compute host per tenant or project ? | |
| 16:23:12 | gibi | the hoststate object is per compute 1:1. The Instance object has project_id field | |
| 16:24:04 | gibi | https://github.com/openstack/nova/blob/bb2d7f9cad577f3a32cb9523e2b1d9a6d6db3407/nova/objects/instance.py#L126 | |
| 16:41:51 | elinux | gibi: can you show me how to call this instance object for a host under host_pass function in a filter , for a specific project id ? | |
| 16:43:19 | elinux | gibi: unfortunately I am not able to trace the code using pdb for this. it is failing . so I am trying to use eventlet backdoor to attach to nova api wgi process but not able to get any source code loaded at pdb prompt | |
| 17:03:43 | gibi | https://paste.opendev.org/show/bcAOtsWiVKbbiTe94HBh/ something like that but I don't have an env right now where I can hack on this. | |
| 17:06:18 | elinux | gibi: how to trace the nova api service like we do with pdb in python ? but pdb in this case is not working since it is eventlet | |
| 17:06:44 | gibi | honestly I tend to debug by adding extra LOG statement | |
| 17:06:56 | dansmith | yeah, debug with LOG/print :) | |
| 17:07:14 | elinux | dansmith: no other way ? | |
| 17:09:01 | dansmith | elinux: I dunno, a debugger doesn't really lend itself to use in something like an eventlet environment.. we're telling you want nova core developers do, so ... | |
| 17:09:37 | dansmith | best thing is to write a test where you can control the environment and reproduce quickly | |
| 17:09:53 | gibi | ^^ this | |
| 17:09:54 | dansmith | you can't use a debugger with the tests normally either, but if you run them manually then you probably can | |