Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2021-12-09
10:30:27 pslestang yep
10:30:43 sean-k-mooney we could likely adapt that to use delete_at if its popluatedand fall abck to updated at if not
10:31:02 sean-k-mooney or just leave it as is
10:33:25 pslestang I prefer to adapt the code to make it consistent with the soft-delete of instance_actions
10:33:42 sean-k-mooney bauzas: im prety sure we dont look at the actions in nova they are write only
10:34:16 sean-k-mooney pslestang: ok would you prefer to write this all down in a short spec which we can try an appove in the review day on the 14th
10:34:29 sean-k-mooney or just proceed with this as a specless blueprint
10:34:57 pslestang I can write a short spec
10:34:58 sean-k-mooney i dont thinks we really need a spec but i think having one to document this would be good so im happy to review it if you write it
10:35:06 pslestang sure
10:35:20 sean-k-mooney cool feel free to add me and likely gibi/bauzas to it
10:35:22 pslestang is it better to change the one I wrote or to create a new one?
10:35:36 sean-k-mooney oh you already have one?
10:35:57 pslestang I was talking about the BP
10:36:26 sean-k-mooney oh reuse the blueprint you have now but file a spec with the same name
10:36:51 pslestang ok understand, I will dot it asap
10:37:03 sean-k-mooney cool
10:46:31 bauzas pslestang: thanks pslestang for writing up a quick spec
10:46:39 bauzas you can see the template and the already existing specs too
10:47:05 bauzas I'm not really asking about a paperwork, just want to be sure we don't miss any important issue
10:48:56 sean-k-mooney yep for me the spec is more just documentaiton/todo list in this case rather then a detailed design requirement because its complex
10:49:29 sean-k-mooney this is a relitively simple change we just dont want to miss any interop or upgrade impact
10:50:08 sean-k-mooney i think we are all supportive of doing it so you dont really need to convice use just detail what needs to be done
10:53:51 gibi bauzas: could you quickly mark this wontfix https://bugs.launchpad.net/nova/+bug/1953734 as it is talking about a deprecated API ? (I don't want to do it as I was asked downstream to open it :D)
10:54:45 bauzas gibi: hehe sure :p
10:54:54 bauzas do you want my bank account ?
10:55:04 bauzas :D
10:55:05 gibi I will pay in beer
10:55:05 gibi :D
10:55:14 gibi when we finally meet again
10:55:43 bauzas well, if I was paying for a vine in 2019 for the next meeting, then it would be a very nice one :)
10:56:05 bauzas after 3 years
10:56:42 sean-k-mooney gibi: what is the invalid input in this case by the way
10:57:00 bauzas hmmm, https://docs.openstack.org/api-ref/compute/?expanded=list-security-groups-by-server-detail#servers-security-groups-servers-os-security-groups
10:57:26 bauzas unfortunately, we don't document the fact that the POST action is now deprecated
10:57:37 gibi sean-k-mooney: security_group is expected to be a dict not a string
10:57:49 sean-k-mooney ah ok
10:57:55 gibi These APIs are proxy calls to the Network service. Nova has deprecated all the proxy APIs and users should use the native APIs instead. These will fail with a 404 starting from microversion 2.36. See: Relevant Network APIs.
10:58:03 bauzas gibi: yeah I know
10:58:15 gibi I think this is for the whole /os-security-groups resource
10:58:15 sean-k-mooney yep the post action i tough twas only vailid for nova networks
10:58:20 bauzas gibi: I'm just saying we don't document it in our API docs :p
10:58:56 bauzas gibi: well, then the GET /os-security-groups action should be told to by deprecated in our API docs :)
10:58:59 gibi I think the documentation is there in the header
10:59:13 bauzas oh, did I miss something in the docs ?
10:59:15 sean-k-mooney https://docs.openstack.org/api-ref/compute/?expanded=list-security-groups-by-server-detail#list-security-groups
10:59:30 sean-k-mooney bauzas: its in the os-security-groups section
10:59:38 sean-k-mooney These APIs are proxy calls to the Network service. Nova has deprecated all the proxy APIs and users should use the native APIs instead. These will fail with a 404 starting from microversion 2.36. See: Relevant Network APIs.
10:59:47 bauzas I'm blind
10:59:58 sean-k-mooney you were looking at the instance action
11:00:01 sean-k-mooney that does not have the hearder
11:00:28 sean-k-mooney but the os-security-groups endpoint section does
11:00:35 bauzas what the f***
11:00:40 sean-k-mooney bauzas: https://docs.openstack.org/api-ref/compute/?expanded=list-security-groups-by-server-detail#delete-deallocate-floating-ip-address
11:00:45 sean-k-mooney the big red box ^
11:00:49 bauzas yeah I get it now
11:00:56 sean-k-mooney :)
11:01:12 bauzas but... can't get why I wasn't able to see it in my own link
11:01:24 sean-k-mooney bauzas: because you were not looking in the right section
11:01:42 sean-k-mooney you were looking at /servers/uuid/os-secuirty-groups
11:01:49 sean-k-mooney that was not deprecated
11:01:59 bauzas Ctrl+F
11:02:12 sean-k-mooney but /os-security-groups and /os-security-group-rules were
11:02:13 bauzas not the right endpoint
11:02:28 sean-k-mooney yep easy mistake to make
11:02:30 sean-k-mooney brb
11:07:37 bauzas gibi: done, you owe me one :p
11:07:49 gibi thanks, and noted :)
13:40:33 plibeau lyarwood: To explain what append in my case: context, the image used during the boot of the instance is now private and the base image on the source compute don't exsit.
13:40:36 plibeau during the resize on the instance in the code we go in "_try_fetch_image_cache" (nova/virt/libvirt/driver.py +10359) and go in except because image.cache can't found image (because now the image is private). In the except image.cache is call again without image in parameter so create_image in call also without image_id. generating is True because image_id not found in kwargs
13:40:37 plibeau and the self.exists because it's a resize so the disk already exist.
13:40:39 plibeau I don't change in image.cache method the line 280 (if os.path.exists(base) and size > self.get_disk_size(base):) because it's the same code for other type of type (qow2,etc) and it's not good idea IMPOV to change that but mayY
13:40:50 plibeau https://review.opendev.org/c/openstack/nova/+/820531
13:46:11 plibeau *but maybe I'm wrong
14:02:24 opendevreview sean mooney proposed openstack/nova master: This change adds generative envs https://review.opendev.org/c/openstack/nova/+/804292
14:04:08 opendevreview sean mooney proposed openstack/nova master: This change adds generative envs https://review.opendev.org/c/openstack/nova/+/804292
14:41:08 opendevreview sean mooney proposed openstack/nova master: [WIP] allow cpus to be externally managed https://review.opendev.org/c/openstack/nova/+/821228
14:55:36 lyarwood plibeau: Apologies was on calls downstream
14:59:02 lyarwood I think I see what you mean
15:00:53 lyarwood and apologies I was mixing resize and rebuild
15:13:19 opendevreview mitya-eremeev-2 proposed openstack/nova master: Delete bogus attachments. https://review.opendev.org/c/openstack/nova/+/820935
15:24:04 plibeau1 okay I can copy my detail in the change if you want? And you prefer another change for the refacto?
15:31:51 lyarwood plibeau1: same change, new patchset would be fine I think?
15:33:14 plibeau1 probably yes I will check to apply your proposal on unit test
15:45:14 lyarwood yeah just amend the commit and git review again
16:31:15 opendevreview Pierre Libeau proposed openstack/nova master: Nova resize don't extend disk in one specific case https://review.opendev.org/c/openstack/nova/+/820531
16:38:43 gmann gibi: sean-k-mooney bauzas pslestang : I am fine not to change API as no one asked it. but +1 on spec to get clear view/documentation
17:27:59 sean-k-mooney gmann: ack
17:47:18 opendevreview Balazs Gibizer proposed openstack/nova master: Log which instance event was timed out https://review.opendev.org/c/openstack/nova/+/819817
19:24:53 mloza /j #openstack-keystone
19:28:56 opendevreview sean mooney proposed openstack/nova-specs master: add per process healthcheck spec https://review.opendev.org/c/openstack/nova-specs/+/821279
21:17:37 opendevreview melanie witt proposed openstack/nova master: Enable unified limits in the nova-next job https://review.opendev.org/c/openstack/nova/+/789963
#openstack-nova - 2021-12-10
01:19:34 opendevreview melanie witt proposed openstack/nova stable/xena: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821348
01:21:32 opendevreview melanie witt proposed openstack/nova stable/wallaby: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821349
01:26:52 opendevreview melanie witt proposed openstack/nova stable/victoria: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821350
01:28:40 opendevreview melanie witt proposed openstack/nova stable/ussuri: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821351
01:30:57 opendevreview melanie witt proposed openstack/nova stable/train: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821352
03:26:15 opendevreview Wenping Song proposed openstack/nova master: Fill the AcceleratorRequestBindingFailed exception msg info https://review.opendev.org/c/openstack/nova/+/817326
05:45:45 opendevreview Wenping Song proposed openstack/nova master: Generate cyborgclient when there is device_prfoile in instance https://review.opendev.org/c/openstack/nova/+/821368
08:00:20 opendevreview Wenping Song proposed openstack/nova master: Cleanup guest process when vm evacuate failed and then deleted https://review.opendev.org/c/openstack/nova/+/814746

Earlier   Later