| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-06-01 | |||
| 14:07:20 | ganso | sean-k-mooney: loved the term tribal knowledge! didnt know it! will now use it for many things lol! | |
| 14:07:30 | ganso | bauzas: yes, I plan to backport. Will look at the patch now | |
| 14:08:02 | bauzas | ganso: okay, then please provide a new PS and I'll directly +2/+W | |
| 14:08:13 | opendevreview | Balazs Gibizer proposed openstack/placement master: [doc] Redirect people to #openstack-nova https://review.opendev.org/c/openstack/placement/+/793832 | |
| 14:08:58 | ganso | bauzas: great! will work on that now! thank you! | |
| 14:09:12 | bauzas | ganso: ping me when you're done | |
| 14:13:38 | gibi | bauzas, sean-k-mooney, melwitt: when you have time please check back to the placement re-parenting spec. https://review.opendev.org/c/openstack/nova-specs/+/788243 | |
| 14:13:47 | gibi | thanks | |
| 14:13:50 | bauzas | gibi: sure thing | |
| 14:15:06 | gibi | :) | |
| 14:19:32 | sean-k-mooney | gibi: yes its on my list for today. just looking at the review priorty doc review now | |
| 14:19:43 | sean-k-mooney | gibi: do we have the team meeting here today by the way | |
| 14:19:58 | sean-k-mooney | or are we using #openstack-meeting-3 | |
| 14:20:02 | sean-k-mooney | if that exists here | |
| 14:23:21 | opendevreview | Stephen Finucane proposed openstack/nova master: Deprecate filters that have been replaced by placement filters https://review.opendev.org/c/openstack/nova/+/745605 | |
| 14:25:52 | gibi | sean-k-mooney: we will be on #openstack-meeting-3 | |
| 14:26:02 | sean-k-mooney | ack | |
| 14:26:05 | gibi | it do exists | |
| 14:26:17 | gibi | but I will raise the question if we want to move the meeting to here | |
| 14:31:24 | opendevreview | Rodrigo Barbieri proposed openstack/nova master: Error anti-affinity violation on migrations https://review.opendev.org/c/openstack/nova/+/784166 | |
| 14:31:46 | ganso | bauzas: ^ I improved the reno and commit message | |
| 15:06:10 | stephenfin | lyarwood: $ openstack server create ... --block-device source_type=volume,uuid=44d317a3-6183-4063-868b-aa0728576f5f,destination_type=volume,delete_on_termination=true --wait test-server | |
| 15:06:18 | stephenfin | Unexpected API Error. Please report this at http://bugs.launchpad.net/nova/ and attach the Nova API log if possible. | |
| 15:06:18 | stephenfin | <class 'nova.exception.VolumeNotFound'> (HTTP 500) (Request-ID: req-7fe03627-c4ce-4f4b-9d5c-3abd6b88d3e3) | |
| 15:09:12 | stephenfin | lyarwood: full traceback here http://paste.openstack.org/show/806238/ | |
| 15:15:46 | sean-k-mooney | stephenfin: your create a server with a prexising volumn | |
| 15:15:53 | stephenfin | yes | |
| 15:16:04 | sean-k-mooney | stephenfin: did you intent to create a server wiht a boot volume form an image? | |
| 15:16:08 | stephenfin | nope | |
| 15:16:14 | stephenfin | I'm updating delete_on_termination=true | |
| 15:16:15 | stephenfin | whoops | |
| 15:16:17 | opendevreview | Merged openstack/placement master: [doc] Redirect people to #openstack-nova https://review.opendev.org/c/openstack/placement/+/793832 | |
| 15:16:18 | stephenfin | doc/source/user/launch-instance-from-volume.rst | |
| 15:16:23 | stephenfin | to use OSC | |
| 15:16:38 | sean-k-mooney | well to use your new os command | |
| 15:16:43 | sean-k-mooney | not just osc | |
| 15:16:54 | sean-k-mooney | but that is weird that sould be correct | |
| 15:16:59 | stephenfin | both | |
| 15:17:11 | stephenfin | the command is correct except I did use the wrong UUID | |
| 15:17:19 | stephenfin | that's my image UUID, not the volume | |
| 15:17:30 | stephenfin | but the server should be handling this and spitting out a better error | |
| 15:17:37 | sean-k-mooney | ya that would make sense | |
| 15:17:43 | sean-k-mooney | it should not be a 500 | |
| 15:17:47 | sean-k-mooney | it should be a 400 | |
| 15:17:47 | stephenfin | exactly | |
| 15:18:25 | sean-k-mooney | ya so i guess we just need to wrap the volumen not found expciton | |
| 15:18:27 | sean-k-mooney | although | |
| 15:18:30 | stephenfin | I can fix but I'm notifying lyarwood since this seems like something that wouldn't go unnoticed for long, so I suspect it's only broken recently (like, last year) | |
| 15:18:34 | sean-k-mooney | we really should be validating this in the api | |
| 15:19:06 | sean-k-mooney | im not sure about that | |
| 15:20:22 | sean-k-mooney | stephenfin: i would expect most peopple to use "openstack server create --flavor 2 --volume VOLUME_ID myInstanceFromVolume" | |
| 15:20:38 | stephenfin | I'd say that will fail with the same error message | |
| 15:20:45 | stephenfin | since it's doing the same thing under the hood | |
| 15:21:00 | sean-k-mooney | it will but i thik they would be less likely to make the mistake of mixing the uuids | |
| 15:22:02 | sean-k-mooney | am i right that you had just previouly tested "-block-device source_type=image,uuid=44d317a3-6183-4063-868b-aa0728576f5f,destination_type=volume,delete_on_termination=true" | |
| 15:22:24 | stephenfin | nope, I just made a mistake | |
| 15:22:32 | stephenfin | the previous command had been 'openstack image list' | |
| 15:22:33 | sean-k-mooney | and then updated the source_type to volumn with out actully updatign the uuid | |
| 15:22:35 | sean-k-mooney | oh ok | |
| 15:22:52 | stephenfin | so my lazy brain grabbed the first UUID it saw on screen :) | |
| 15:23:52 | sean-k-mooney | stephenfin: well the current error is happing on the compute so this should defintly be caught in the api before we create a server record | |
| 15:27:24 | sean-k-mooney | maybe here https://github.com/openstack/nova/blob/da57eebc9e1ab7e48d4c4ef6ec1eeba80d867d81/nova/api/openstack/compute/servers.py#L470 | |
| 15:30:09 | sean-k-mooney | oh in _validate_and_build_base_options https://github.com/openstack/nova/blob/c0c2888acaaccfa4266874523f62039cb9e143e0/nova/compute/api.py#L968 | |
| 15:45:51 | kashyap | stephenfin: [Off-topic] I have an inkling you'll like this: https://www.merriam-webster.com/words-at-play/can-a-ship-flounder | |
| 15:46:40 | kashyap | stephenfin: Also did you know both noun and verb usage of 'flounder and 'founder'? (Before reading the article :) | |
| 15:46:59 | sean-k-mooney | they are two differnt things | |
| 15:47:41 | sean-k-mooney | a fish can flounder a ship cannot as far as im aware | |
| 15:48:20 | stephenfin | and vice versa, iirc | |
| 15:49:06 | sean-k-mooney | a wale that is beached on land can flounder, when it tries to get back into the water | |
| 15:50:28 | gibi | nova meeting starts in 10 minutes in #openstack-meeting-3 on OFTC ;) | |
| 15:52:55 | sean-k-mooney | hum i see how it could be used and it makes sense but it a bit arcic to use flounder for an in animiat object even by irish standard even if its technicaly allowable | |
| 15:53:37 | kashyap | sean-k-mooney: Yeah; they are two different things; but using "founder" as a verb for ship (but not "flounder") was a new thing for me | |
| 15:54:28 | sean-k-mooney | both words honestly have little usage day to day | |
| 15:55:15 | kashyap | Yeah; flounder is more often seen in writng, though | |
| 15:56:26 | opendevreview | Elod Illes proposed openstack/nova stable/wallaby: DNM: test ceph checkout with branch https://review.opendev.org/c/openstack/nova/+/794000 | |
| 15:56:51 | gibi | lyarwood, melwitt: FYI ^^ with elod we might caught a ceph testing issue | |
| 16:41:08 | opendevreview | Stephen Finucane proposed openstack/nova master: api: Handle invalid volume UUIDs during spawn https://review.opendev.org/c/openstack/nova/+/794006 | |
| 16:41:08 | opendevreview | Stephen Finucane proposed openstack/nova master: docs: Replace 'nova boot' with 'openstack server create' https://review.opendev.org/c/openstack/nova/+/794007 | |
| 16:43:05 | stephenfin | lyarwood: ^ easy fix. I didn't add unit tests because they seemed silly (see reasoning inline) but I can if people insist | |
| 17:14:15 | masterpe | Hi, I try to extend a in-use volume on Openstack Train. In the logs I get the following "error: Forbidden: Policy doesn't allow os_compute_api:os-server-external-events:create to be performed" | |
| 17:14:30 | sean-k-mooney | gibi: so we did tweak the realtime handeling to allow alll cores to be used but we required the emulator policy to be used | |
| 17:14:32 | sean-k-mooney | https://github.com/openstack/nova/blob/50fdbc752a9ca9c31488140ef2997ed59d861a41/releasenotes/notes/bug-1884231-16acf297d88b122e.yaml | |
| 17:15:16 | masterpe | Is there an article where it explains how to change the policy so that it is allowed? I searched but I was not able to find it. | |
| 17:15:18 | sean-k-mooney | gibi: so what we allow today is you can use a specific emulator policy which wont have the downstream issue or you can use a realtime mask | |
| 17:15:32 | gibi | seem good | |
| 17:15:34 | masterpe[m] | Hi, I try to extend a in-use volume on Openstack Train. In the logs I get the following "error: Forbidden: Policy doesn't allow os_compute_api:os-server-external-events:create to be performed" | |
| 17:15:34 | masterpe[m] | Is there an article where it explains how to change the policy so that it is allowed? I searched but I was not able to find it. | |
| 17:16:18 | sean-k-mooney | masterpe[m]: i think this is a know cinder bug | |
| 17:16:49 | sean-k-mooney | masterpe[m]: basicaly when cinder calls back the nova api its incorrectly using the user token to call the admin only external events api | |
| 17:21:55 | masterpe[m] | sean-k-mooney do you know if there is a patch for this cinder bug? | |
| 17:31:25 | sean-k-mooney | masterpe[m]: if you try this this as an admin i think it willl work | |
| 17:31:31 | sean-k-mooney | masterpe[m]: but im not sure | |
| 17:31:39 | sean-k-mooney | lyarwood: do you remember ^ | |
| 17:33:17 | sean-k-mooney | masterpe[m]: what cinder backend are you using | |
| 17:33:32 | masterpe[m] | End ceph | |
| 17:33:58 | masterpe[m] | I wanted to write rbd but auto correct | |
| 17:34:28 | sean-k-mooney | there was this previous bug for nfs https://bugzilla.redhat.com/show_bug.cgi?id=1640443 https://bugs.launchpad.net/tripleo/+bug/1802347 | |
| 17:34:28 | opendevmeet | bugzilla.redhat.com bug 1640443 in openstack-tripleo-heat-templates "volume_extend while attached - Policy doesn't allow os_compute_api:os-server-external-events:create" [High,Closed: errata] - Assigned to abishop | |
| 17:34:29 | opendevmeet | Launchpad bug 1802347 in tripleo "Need ability to configure cinder's settings for nova" [High,Fix released] - Assigned to Alan Bishop (alan-bishop) | |
| 17:35:16 | dansmith | masterpe: this channel has moved to oftc, so you'll need to re-ask there | |