| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2024-06-13 | |||
| 13:17:40 | sean-k-mooney | i think "openstack server list --all-projects" will list all vms if you jsut dont have limit unless that changed with the sdk and its alwasy using pagination now | |
| 13:18:03 | ozzzo_work | In 5.5.0 the limit is 1000 | |
| 13:18:11 | ozzzo_work | let me test in 6.6.0 real quick | |
| 13:19:28 | sean-k-mooney | if you have more vms then the api limit | |
| 13:19:34 | sean-k-mooney | you need to use https://docs.openstack.org/python-openstackclient/latest/cli/command-objects/server.html#cmdoption-openstack-server-list-marker | |
| 13:19:44 | sean-k-mooney | passing the last server uuid as the marker | |
| 13:19:50 | sean-k-mooney | and make mulitple requests | |
| 13:20:17 | sean-k-mooney | but assuming you dont hit the api limit and if you dont pass --limit i would expelct it to list htem all | |
| 13:20:45 | ozzzo_work | we have >1000 VMs in some regions | |
| 13:20:50 | sean-k-mooney | i am not sure if osc handeles automatic pagination if you dont pass a limit | |
| 13:24:04 | ozzzo_work | Theoretically I could rewrite my scripts to use pagination, but "--limit -1" was nice because it was simple. Is there anything I can do to make it work in 6.6.0? I tried setting a different api version but that didn't seem to make a difference | |
| 13:26:13 | sean-k-mooney | no i dont think so. i belive this behaivor changed bcause osc is not using nova clinet anymore and the sdk does not supprot -1 for automatic client side pagination | |
| 13:26:32 | sean-k-mooney | you could try using 2.1 | |
| 13:26:45 | sean-k-mooney | but i think pagination for servers predates microverions | |
| 13:28:21 | ozzzo_work | 2.1 doesn't change the behavioor | |
| 13:30:50 | sean-k-mooney | then ya server list supported pagination before micorverions so there is no way to turn it off | |
| 13:30:57 | sean-k-mooney | which means it has to be done client side | |
| 13:36:13 | ozzzo_work | ok ty sean | |
| 13:36:27 | opendevreview | Sahid Orentino Ferdjaoui proposed openstack/nova master: scheduler: fix _get_sharing_providers to support unlimited aggr https://review.opendev.org/c/openstack/nova/+/921665 | |
| 13:39:46 | ozzzo_work | aha it looks like I can delay rewriting my script by using "nova list" with -1 | |
| 13:49:57 | sean-k-mooney | yes although nova client is deprecated. we proably wont delete it for another release or 3 becasue the python binding are still used in some other project | |
| 13:50:05 | sean-k-mooney | but eventually it will go away | |
| #openstack-nova - 2024-06-14 | |||
| 11:18:32 | stephenfin | bauzas: The OpenAPI work is kind of in no mans land atm: it's ready to go on my end but I'm waiting for reviews for a few weeks now. Is there someone I can ask to start reviewing that? I've a whole load of patches sitting locally to close it out but I don't want to push them until we're making progress | |
| 11:18:48 | stephenfin | (asking you as PTL, as much as a potential reviewer) | |
| 13:11:03 | bauzas | stephenfin: I have some problems this week due to my internet provider, but I want to review some specs | |
| 16:27:17 | gmann | stephenfin: sorry for delay, I had that in my list. I will start reviewing those soon. | |
| #openstack-nova - 2024-06-17 | |||
| 03:03:35 | opendevreview | Takashi Kajinami proposed openstack/nova master: WIP: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 03:10:28 | opendevreview | Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 03:32:32 | opendevreview | Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 06:44:52 | opendevreview | Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 06:45:49 | opendevreview | Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 07:11:42 | opendevreview | Takashi Kajinami proposed openstack/nova master: Remove unused default argument https://review.opendev.org/c/openstack/nova/+/922073 | |
| 07:35:38 | opendevreview | Takashi Kajinami proposed openstack/nova master: Report availability of stateless firmware support https://review.opendev.org/c/openstack/nova/+/908888 | |
| 07:35:39 | opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Launch instances with stateless firmware https://review.opendev.org/c/openstack/nova/+/908890 | |
| 09:22:16 | opendevreview | Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814 | |
| 11:57:23 | opendevreview | Konrad Gube proposed openstack/nova master: Use Cinder's os-extend_volume_completion volume action. https://review.opendev.org/c/openstack/nova/+/873560 | |
| 12:14:18 | opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Launch instances with stateless firmware https://review.opendev.org/c/openstack/nova/+/908890 | |
| 14:22:24 | dansmith | sean-k-mooney: are you going to reply to this? https://review.opendev.org/c/openstack/nova/+/904424/9 | |
| 14:31:47 | sean-k-mooney | oh i spoke to stephen about that in person | |
| 14:32:14 | sean-k-mooney | tldr the missed that the monkey patch was done as a sideefect of the __init__.py | |
| 14:33:05 | sean-k-mooney | that was however goign to be one fo the things i changed based on other feedback | |
| 14:33:48 | dansmith | ack, maybe you could update it so it doesn't look like it's hanging? | |
| 14:33:55 | sean-k-mooney | sure | |
| 14:34:10 | dansmith | and I assume that means we're waiting on a revision? | |
| 14:34:15 | sean-k-mooney | in mels version fo the scater gather change you ask for this to be more explict liek this https://review.opendev.org/c/openstack/nova/+/650172/28/nova/cmd/__init__.py | |
| 14:34:39 | sean-k-mooney | dansmith: yes i was ment to work on it last week but got pulled into other things | |
| 14:34:52 | sean-k-mooney | the core would be the same but one thing i noticed in neutron is | |
| 14:35:07 | sean-k-mooney | instead of using /bin at the top level they use /cmd and /cmd/eventlet | |
| 14:35:22 | sean-k-mooney | where only the eventlet subdir is monkney patched | |
| 14:35:35 | dansmith | ack | |
| 14:35:36 | sean-k-mooney | i was wondering if that would be better for consitency | |
| 14:36:05 | dansmith | yeah I mean, if there is prior art that makes sense to me, and also makes it easier to see where major eventlet deps are | |
| 14:36:44 | sean-k-mooney | yep that was the idea behind /bin but stephen said having a bin directory at the top level could be confusing | |
| 14:36:52 | sean-k-mooney | since technially they files are not executable | |
| 14:36:59 | sean-k-mooney | they are used by pbr to create teh executables | |
| 14:37:06 | dansmith | meh | |
| 14:37:45 | sean-k-mooney | the pathces are reviewable in general but i do plan to respine them this week | |
| 14:38:24 | dansmith | ack, I just figure I'll wait until they're respun, but I think we should make sure we don't lose track of this | |
| 14:39:57 | sean-k-mooney | thanks for looking in anycase. ill let you know when its up | |
| 15:39:11 | zigo | How may I check why the nova-scheduler is filtering my destination host when doing a live-migration ? | |
| 15:41:12 | zigo | I'm trying to do: | |
| 15:41:12 | zigo | openstack --os-compute-api-version 2.96 server migrate --live-migration --host compute-1.infomaniak.ch <vm-id> | |
| 15:41:12 | zigo | and then in the scheduler logs, I'm getting: | |
| 15:41:12 | zigo | Host filter only checking host compute-1 and node compute-1 | |
| 15:41:14 | zigo | Host filter ignoring hosts: compute-1 | |
| 15:41:16 | zigo | Filtered [] _get_sorted_hosts /usr/lib/python3/dist-packages/nova/scheduler/manager.py:708 | |
| 15:41:18 | zigo | There are 0 hosts available but 1 instances requested to build. _ensure_sufficient_hosts | |
| 15:41:20 | zigo | How do I debug this? | |
| 15:41:25 | zigo | I've spent the whole day on it ... :/ | |
| 15:41:48 | zigo | FYI, this is a new Caracal deployment... | |
| 15:42:38 | zigo | I by the way couldn't get nova-compute to start with cpu_models= set to EPYC-<something> (tried them all, even tried backports kernel and backported libvirt to bookworm). | |
| 15:42:57 | zigo | Each time, libvirt replies with 0 (ie: incompatible CPU) | |
| 15:43:27 | zigo | So I'm in host-passthrough mode to avoid this, and now I'm getting the above form the scheduler ! :/ | |
| 16:01:20 | bauzas | dansmith: hah, I now see https://opendev.org/openstack/project-config/src/branch/master/nodepool/nodepool.yaml#L352-L363 | |
| 16:01:33 | dansmith | bauzas: there was an email to the list last week too | |
| 16:01:59 | dansmith | (the service-announce list I mean) | |
| 16:02:24 | bauzas | oh, that's why I missed it | |
| 16:02:32 | bauzas | I know I should track that ML too | |
| 16:02:47 | dansmith | yeah it would probably have been helpful if that went to openstack-discuss as well, but alas | |
| 16:03:04 | sean-k-mooney | zigo: that message implies its currenly on compute-1 | |
| 16:03:47 | sean-k-mooney | Host filter ignoring hosts: compute-1 i think is only added if its in the ignored host list which should only be the current hsot | |
| 16:03:48 | bauzas | https://lists.opendev.org/archives/list/service-announce@lists.opendev.org/thread/EFRZGKYS7AO44Q4NFVXBNGAGQVBU3WUX/ | |
| 16:03:53 | bauzas | got it ^ | |
| 16:05:20 | bauzas | dansmith: would you be okay if I create some experimental job using nodeset:ubuntu-noble that would enable mtty ? | |
| 16:05:39 | bauzas | I need to configure Tempest to create the according flavors of course | |
| 16:12:43 | dansmith | bauzas: sure, at least for starters | |
| 16:12:55 | bauzas | okay | |
| 16:16:16 | opendevreview | Konrad Gube proposed openstack/nova master: Use Cinder's os-extend_volume_completion volume action. https://review.opendev.org/c/openstack/nova/+/873560 | |
| 16:18:37 | opendevreview | Stephen Finucane proposed openstack/nova master: api: Add request body schemas for SG APIs https://review.opendev.org/c/openstack/nova/+/915733 | |
| 16:18:53 | opendevreview | Stephen Finucane proposed openstack/nova master: api: Add remaining missing request body schemas https://review.opendev.org/c/openstack/nova/+/915734 | |
| 16:18:58 | opendevreview | Stephen Finucane proposed openstack/nova master: tests: Ensure all APIs have a request body schema https://review.opendev.org/c/openstack/nova/+/915735 | |
| 16:19:04 | opendevreview | Stephen Finucane proposed openstack/nova master: api: Fix bugs in schemas https://review.opendev.org/c/openstack/nova/+/915843 | |
| 16:19:12 | opendevreview | Stephen Finucane proposed openstack/nova master: tests: Ensure API schemas are valid https://review.opendev.org/c/openstack/nova/+/915844 | |
| 16:19:19 | opendevreview | Stephen Finucane proposed openstack/nova master: api: Migrate to JSON Schema Draft 2020-12 https://review.opendev.org/c/openstack/nova/+/915845 | |
| 16:19:43 | bauzas | dansmith: do you know a way to create a specific flavor in .zuul.yaml except by creating a specific playbook ? | |
| 16:20:36 | bauzas | (I can configure the nodes easily in the job definition, but I can't find how to define flavors except by creating them thru a playbook script) | |
| 16:21:23 | bauzas | maybe the best way is to create a specific Tempest environment in Tempest | |
| 16:35:52 | dansmith | bauzas: nope, that's the only way I know | |
| 16:35:59 | bauzas | ok | |
| 16:36:13 | bauzas | for a first starting point, I'll all do this in nova's zuul.yaml | |