| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2022-11-24 | |||
| 11:19:46 | opendevreview | Amit Uniyal proposed openstack/nova stable/train: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/864055 | |
| 14:37:26 | opendevreview | sean mooney proposed openstack/nova stable/wallaby: Add compute restart capability for libvirt func tests https://review.opendev.org/c/openstack/nova/+/865571 | |
| 14:37:27 | opendevreview | sean mooney proposed openstack/nova stable/wallaby: enable blocked VDPA move operations https://review.opendev.org/c/openstack/nova/+/865572 | |
| 14:38:00 | sean-k-mooney | gibi: just an fyi i took a slightly diffent approch with xena->wallaby | |
| 14:38:19 | sean-k-mooney | for yoga and xena i included the double mocking patch | |
| 14:38:25 | sean-k-mooney | well patches | |
| 14:38:36 | gibi | ack | |
| 14:38:38 | sean-k-mooney | they were pretty ok to backport | |
| 14:38:55 | sean-k-mooney | but xena to wallaby it was getting a bit messy so i dropped them | |
| 14:39:12 | sean-k-mooney | i didnt really have many conflict when i did that so it shoudl be fine | |
| 14:40:01 | gibi | OK | |
| 14:40:03 | bauzas | don't get me into the ML thread about OSC/SDK and the project clients :D | |
| 14:40:08 | gibi | I added it to my review queue | |
| 14:40:30 | bauzas | IMHO, OSC : profiled for UX | |
| 14:40:44 | bauzas | SDK : profiled for powerusers | |
| 14:41:32 | bauzas | powerusers being in my terminology 'seasoned users of openstack', and not powerusers in terms of role | |
| 14:41:57 | bauzas | so, having the OSC defaulting to latest sounds to me right | |
| 14:41:58 | sean-k-mooney | so i attened some of the very early design summit sessions on osc and talked to dean troyer alot about it in the halway track | |
| 14:42:18 | bauzas | while we could tell that the SDK defaults to the bare min version | |
| 14:42:24 | sean-k-mooney | bauzas: we coudl default to latest but if we do it need to eb a new microverion | |
| 14:42:31 | sean-k-mooney | *new majory version | |
| 14:42:41 | bauzas | OSC or the SDK ? | |
| 14:42:50 | sean-k-mooney | technially both | |
| 14:43:00 | sean-k-mooney | sdk recently broke all of the ansible module integration | |
| 14:43:24 | sean-k-mooney | osc need to have a majory version if its going to enabel auto negociation | |
| 14:43:54 | sean-k-mooney | the sdk i think did but they broke ansible recently with a change in this area | |
| 15:00:52 | bauzas | sean-k-mooney oh when you say major version, you were talking about the semver of OSC releases, lol | |
| 15:02:51 | sean-k-mooney | yes | |
| 15:04:10 | sean-k-mooney | osc has a backward incompatible change dock and follows semver for the client releases | |
| 15:04:12 | sean-k-mooney | https://docs.openstack.org/python-openstackclient/pike/cli/backwards-incompatible.html | |
| 15:04:47 | sean-k-mooney | so if they want to turn on auto negociation there should be a major version of the client | |
| 15:05:37 | sean-k-mooney | https://github.com/openstack/ansible-collections-openstack#breaking-backward-compatibility-warning | |
| 15:05:58 | bauzas | I'm not opposed to signal it thru a .x | |
| 15:06:05 | bauzas | x. rather | |
| 15:06:46 | bauzas | this is not like we have an open bug for 1 year due to a libvirt behavioural breaking change in a .y bump | |
| 15:06:56 | sean-k-mooney | :) | |
| 15:07:16 | sean-k-mooney | i like to think that we have learn for other projects mistakes | |
| 15:25:01 | bauzas | that port_active_state failure gives me headaches https://zuul.opendev.org/t/openstack/build/c6dee3859e354e6d8ed69c7ac7826761 | |
| 15:31:34 | auniyal | bauzas, and Push the CA certificate to me https://zuul.opendev.org/t/openstack/build/e3bf378e2f8249feb1beec96fcde2802 | |
| 15:35:17 | sean-k-mooney | bauzas: that will hopefully get fixed soon althoguh we coudl skip those test for now until that happens | |
| 15:35:47 | sean-k-mooney | bauzas: oh that is a differnt one | |
| 15:36:13 | sean-k-mooney | oh no its the turnk one | |
| 15:36:27 | sean-k-mooney | so ya we can exclude it for now until the neutron patch lands | |
| 15:37:09 | sean-k-mooney | bauzas: ralonsoh has updated a patch form mlavalle to adress it | |
| 15:37:23 | ralonsoh | sean-k-mooney, not yet | |
| 15:37:30 | ralonsoh | I've pushed a patch to test the backends | |
| 15:37:41 | ralonsoh | but I'll remove the conditional branch now Nova patch is merged | |
| 15:37:43 | ralonsoh | and thanks! | |
| 15:37:54 | sean-k-mooney | ralonsoh: your still having issue with ovn right | |
| 15:38:03 | ralonsoh | yes but I found the issue | |
| 15:38:08 | sean-k-mooney | cool | |
| 15:38:29 | ralonsoh | (what a week with trunks a live migration, both OVS and OVN hehehe) | |
| 15:38:42 | sean-k-mooney | ralonsoh: would you object if we dropped the test until its fixed or show we just wait a few days | |
| 15:39:56 | ralonsoh | sean-k-mooney, maybe you can use "tempest_exclude_regex" for some days | |
| 15:40:02 | ralonsoh | not to block the CI | |
| 15:40:10 | ralonsoh | same as we did in https://review.opendev.org/c/openstack/neutron/+/865295/1/zuul.d/tempest-multinode.yaml#171 | |
| 15:42:44 | bauzas | sean-k-mooney: yup, this is a networking issue, I'm happy to hear a fix is coming up | |
| 15:43:00 | bauzas | in order to hold my rechecks, any gerrit link I should take care of ? | |
| 15:46:09 | sean-k-mooney | ralonsoh: ya that is what i was going to propose | |
| 15:46:20 | sean-k-mooney | bauzas: we dont have a patch currently but i can create one or you can | |
| 15:46:38 | sean-k-mooney | we only run the trunk live migration tests in a couple of jobs | |
| 15:46:55 | sean-k-mooney | so we can juse exculde them for a few days and revert it when the fix lands | |
| 15:46:55 | ralonsoh | but a patch for tempest or nova? I would suggest for nova only, as in | |
| 15:46:56 | ralonsoh | https://review.opendev.org/c/openstack/neutron/+/865295/1/zuul.d/tempest-multinode.yaml#171 | |
| 15:47:03 | bauzas | https://opensearch.logs.openstack.org/_dashboards/app/discover?security_tenant=global#/?_g=(filters:!(),refreshInterval:(pause:!t,value:0),time:(from:now-1d%2Fd,to:now))&_a=(columns:!(_source),filters:!(),index:'94869730-aea8-11ec-9e6a-83741af3fdcd',interval:auto,query:(language:kuery,query:'message:%22Call%20_is_port_status_active%20returns%20false%22'),sort:!()) | |
| 15:47:06 | sean-k-mooney | ralonsoh: for nova | |
| 15:47:09 | ralonsoh | thanks | |
| 15:47:14 | sean-k-mooney | in our zuul.yaml | |
| 15:47:20 | bauzas | I'm not able to catch the failures on the opensearch site ^ | |
| 15:47:54 | sean-k-mooney | bauzas: we only do live migration in the multi node jobs so and this only happens in ml2/ovs | |
| 15:48:07 | sean-k-mooney | so we only use ml2/ovs in one or two most are ovn now | |
| 15:49:14 | sean-k-mooney | nova-ovs-hybrid-plug and nova-next infact | |
| 15:52:25 | bauzas | did I messed up my query ? | |
| 15:52:29 | bauzas | lemme check | |
| 15:53:52 | sean-k-mooney | bauzas we dont use trunk bridge with ovn so the status issue cant happen there | |
| 15:54:09 | sean-k-mooney | and unless we overrided the devstack default all our jobs use ovn | |
| 15:54:16 | sean-k-mooney | we only override for those too | |
| 15:54:26 | sean-k-mooney | so that why i know it wont cause other jobs to fail | |
| 15:55:32 | bauzas | sean-k-mooney: sorry I was talking of the logstash query against opensearch | |
| 15:55:51 | bauzas | all the jobs are logged there, right? (ie. the ones including multinode) | |
| 15:55:53 | sean-k-mooney | oh i now im not sure why that is not finding things | |
| 15:55:57 | sean-k-mooney | yes | |
| 15:56:03 | sean-k-mooney | with a little delay buyt yes | |
| 16:01:13 | sean-k-mooney | i think there might be an issue with the data | |
| 16:02:20 | sean-k-mooney | there are no results for nova in the last 24 hours | |
| 16:06:18 | bauzas | only the grenade job | |
| 16:06:59 | bauzas | a8-11ec-9e6a-83741af3fdcd',interval:auto,query:(language:kuery,query:''),sort:!()) | |
| 16:06:59 | bauzas | https://opensearch.logs.openstack.org/_dashboards/app/discover?security_tenant=global#/?_g=(filters:!(),refreshInterval:(pause:!t,value:0),time:(from:now-1h,to:now))&_a=(columns:!(_source),filters:!(('$state':(store:appState),meta:(alias:!n,disabled:!f,index:'94869730-aea8-11ec-9e6a-83741af3fdcd',key:project,negate:!f,params:(query:openstack%2Fnova),type:phrase),query:(match_phrase:(project:openstack%2Fnova)))),index:'94869730-ae | |
| 16:07:07 | bauzas | I can see some | |
| 16:13:04 | sean-k-mooney | do you have an example fo the failure | |
| 16:13:18 | sean-k-mooney | i have got some results back but wonderign what im looking for | |
| 16:17:55 | sean-k-mooney | bauzas i only get two hits | |
| 16:18:05 | sean-k-mooney | https://opensearch.logs.openstack.org/_dashboards/app/discover?security_tenant=global#/?_g=(filters:!(),refreshInterval:(pause:!t,value:0),time:(from:now-6w%2Fw,to:now))&_a=(columns:!(build_status,project,message,filename,build_change,build_name),filters:!(('$state':(store:appState),meta:(alias:!n,disabled:!f,index:'94869730-aea8-11ec-9e6a-83741af3fdcd',key:build_status,negate | |
| 16:18:07 | sean-k-mooney | :!t,params:(query:SUCCESS),type:phrase),query:(match_phrase:(build_status:SUCCESS)))),index:'94869730-aea8-11ec-9e6a-83741af3fdcd',interval:auto,query:(language:kuery,query:'project:%20%22openstack%2Fnova%22%20and%20message:%20%22%20_is_port_status_active%20returns%20false%22'),sort:!()) | |
| 16:18:14 | sean-k-mooney | let me shorten that | |
| 16:18:31 | sean-k-mooney | https://tinyurl.com/4rz4hnxe | |
| 16:38:53 | gibi | 14 | |
| 18:24:46 | sean-k-mooney | gibi: thanks for the backport reviews. i think i see how to simplfy things | |
| 18:24:59 | gibi | cool | |
| 18:25:24 | gibi | I had not big issues with it. just needed some time to get thought the MAC change as that is big | |