| opendevreview | Jorhson Deng proposed openstack/nova master: recheck the attachment_id after the reschedule successful https://review.opendev.org/c/openstack/nova/+/796209 | 01:49 |
|---|---|---|
| opendevreview | Ade Lee proposed openstack/nova master: Add check job for FIPS https://review.opendev.org/c/openstack/nova/+/790519 | 03:15 |
| opendevreview | Wenping Song proposed openstack/placement master: Dropping lower constraints testing https://review.opendev.org/c/openstack/placement/+/787863 | 07:57 |
| bbezak | Hi - Can we make a release os-vif for Victoria? 2.2.1? this bug is quite critical - https://bugs.launchpad.net/os-vif/+bug/1892132. And it is not part of 2.2.0 release | 08:45 |
| sean-k-mooney | its not actully a bug it was a kernel abi break | 09:02 |
| sean-k-mooney | but yes we can proably do a release we just need too propose it to the release repo | 09:02 |
| sean-k-mooney | bbezak: the release managmen it automated via a git repo | 09:03 |
| sean-k-mooney | so we just need to update this with the corerct sha https://github.com/openstack/releases/blob/master/deliverables/victoria/os-vif.yaml | 09:04 |
| sean-k-mooney | there is a script in the release repo for doing that or you can do it by hand | 09:04 |
| sean-k-mooney | https://github.com/openstack/os-vif/compare/2.2.0...stable/victoria | 09:05 |
| kashyap | sean-k-mooney: What's the kernel ABI break? | 09:06 |
| sean-k-mooney | looking at the delta it does not look like there is much else included but release are pretty cheap | 09:06 |
| bbezak | indeed - OFED 5.4 also introduced that change in renumeration at some level. | 09:06 |
| bbezak | https://bugzilla.redhat.com/show_bug.cgi?id=1918703 | 09:06 |
| kashyap | Ah, a commit is linked - https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/?id=123f0f53dd64b67e34142485fe866a8a581f12f1 | 09:06 |
| bbezak | kashyap: https://bugzilla.redhat.com/show_bug.cgi?id=1918703 | 09:06 |
| kashyap | Thanks | 09:06 |
| sean-k-mooney | kashyap: melonox moved where the reprentor netdevs were advertised in /sys | 09:07 |
| sean-k-mooney | we have fixed os-vif and libvirt to be able to handel both now | 09:07 |
| sean-k-mooney | but we need to actully release os-vif with the fix | 09:07 |
| kashyap | I see | 09:07 |
| kashyap | This seems to be a "forced" ABI break - because of an external vendor | 09:08 |
| kashyap | But yeah, a break nonetheless | 09:08 |
| sean-k-mooney | well yes they chose to do this to enable a new feature in the future | 09:08 |
| sean-k-mooney | we adapted on master a long time ago and the backports where held on review bandwith for some time | 09:09 |
| sean-k-mooney | now that they have actully landed we can do a realase for them | 09:09 |
| bbezak | yesterday ussuri was merged, but I think older ones are still waiting for +2s | 09:09 |
| bbezak | and/or rechecks | 09:10 |
| sean-k-mooney | i think train was reviewed. i should now have stable branch rights so ill take a look | 09:10 |
| sean-k-mooney | ah you are right | 09:10 |
| sean-k-mooney | they are still pending | 09:10 |
| * bauzas wonders whether we should comment in the prelude section which python versions we support | 09:12 | |
| sean-k-mooney | its in the governace repo | 09:12 |
| bauzas | ok, then it's not needed | 09:12 |
| sean-k-mooney | https://github.com/openstack/governance/blob/master/reference/runtimes/xena.rst | 09:13 |
| sean-k-mooney | technically i guess these are minimume but 3.6-3.8 is what is tested | 09:13 |
| sean-k-mooney | technially i think we can mostly run on 3.5 | 09:14 |
| sean-k-mooney | bbezak: i have rechecked the train version ill take a look at them again later today | 09:14 |
| sean-k-mooney | bbezak: are you going to porpos a release to the git repo or would you like me to do that | 09:15 |
| bbezak | sean-k-mooney: thx, I will try to propose git release for victoria/ussuri | 09:17 |
| opendevreview | Balazs Gibizer proposed openstack/nova master: DNM: check nova job results with placement transaction fix https://review.opendev.org/c/openstack/nova/+/807558 | 09:45 |
| bbezak | ussuri, victoria os-vif release has been proposed, as per docs - https://releases.openstack.org/reference/using.html#using-new-release-command. PTL needs to approve those, please take a look gibi: | 10:07 |
| bbezak | https://review.opendev.org/c/openstack/releases/+/807694 https://review.opendev.org/c/openstack/releases/+/807696 | 10:07 |
| sean-k-mooney | gibi: i have reviewed both for those ^ and they look good to me. we dont have any other pending backports that i see that we should wait for | 10:17 |
| sean-k-mooney | i have some patch for master that i plan to backport but we can do another release for those in a few weeks they are not urgent and release are pretty cheap so i have no issue with defering them for now | 10:18 |
| bauzas | gibi: permission to rewite your cycle highligts for the prelude ? | 10:18 |
| gibi | bauzas: grandted :) | 10:39 |
| gibi | sean-k-mooney: I will check them soon | 10:40 |
| gibi | bauzas: when you need some distraction there is two doc / reno patch up for review https://review.opendev.org/c/openstack/nova/+/807564 and https://review.opendev.org/c/openstack/nova/+/705667 | 11:10 |
| gibi | stephenfin, sean-k-mooney: I added a topic about tox.ini basepython pinning to the PTG etherpad https://etherpad.opendev.org/p/nova-yoga-ptg L81 | 11:25 |
| gibi | lyarwood: if you have time I could use your oppinion on https://bugs.launchpad.net/nova/+bug/1942766 | 11:42 |
| lyarwood | gibi: yeah live and then persistent could be an option but ultimately a hard reboot would recover the situation so it's a `low` bug at best IMHO | 11:54 |
| gibi | lyarwood: as the hard reboot will regenareate the xml from the db? | 11:55 |
| lyarwood | gibi: correct it destroys the live domain and undefined the persistent domain before recreating everything based on what we have in the db | 11:56 |
| lyarwood | undefines* | 11:56 |
| gibi | cool, thanks | 11:56 |
| opendevreview | Balazs Gibizer proposed openstack/nova master: Add more retries to TestMigrateFromDownHost tests https://review.opendev.org/c/openstack/nova/+/807714 | 12:45 |
| gibi | lyarwood: a further tuning on these tests ^^ | 12:52 |
| gibi | ohh I see you already checked | 12:52 |
| gibi | thanks | 12:53 |
| lyarwood | gibi: yup already reviewed, I did see a gate failure with this last week but didn't get time to look, thanks for sorting that again | 12:53 |
| * lyarwood needs to stop writing racey tests | 12:53 | |
| gibi | lyarwood: no problem, I like hunting down these :) | 12:53 |
| gibi | and I think there is no way to avoid races sometime, nova is complex | 12:53 |
| lyarwood | we all have our vices :D | 12:53 |
| gibi | :D | 12:54 |
| bauzas | man, I wish generated sphinx errors were simplier to debug with reno | 13:35 |
| kashyap | bauzas: What's the error that's giving you grief? | 13:40 |
| bauzas | /home/sbauza/git/openstack/nova/releasenotes/source/unreleased.rst:40: WARNING: Bullet list ends without a blank line; unexpected unindent. | 13:41 |
| kashyap | Ah, I've seen this enough no. of times that it is now written into my brain's ROM | 13:41 |
| bauzas | also, running the tox releasenotes target takes a while | 13:42 |
| bauzas | it's repopulating all the releases | 13:42 |
| opendevreview | Balazs Gibizer proposed openstack/nova master: Avoid unbound instance_uuid var during delete https://review.opendev.org/c/openstack/nova/+/805605 | 13:48 |
| opendevreview | Merged openstack/nova master: fup: Print message logging uncaught nova-manage exceptions https://review.opendev.org/c/openstack/nova/+/807358 | 14:28 |
| opendevreview | Merged openstack/nova master: console: Improve logging https://review.opendev.org/c/openstack/nova/+/778407 | 14:29 |
| gibi | cores: do we support providing the adminPassword to the guest via the metadata service? | 14:31 |
| gibi | I see that the metadata service trying to fetch the password from instance.system_metadata https://github.com/openstack/nova/blob/402fe188b4e7ff76109e8a5ea1f24a5e915eaa09/nova/api/metadata/password.py#L37 | 14:31 |
| gibi | but I don't see we ever store the adminPassword there | 14:31 |
| gibi | asking for due to https://bugs.launchpad.net/nova/+bug/1942709 | 14:32 |
| gibi | FYI: nova meeting starts in 9 minutes here in the channel | 15:50 |
| bauzas | gibi: I'll be a bit late, daughter's homework helping | 15:59 |
| gibi | ack | 15:59 |
| gibi | #startmeeting nova | 16:00 |
| opendevmeet | Meeting started Tue Sep 7 16:00:10 2021 UTC and is due to finish in 60 minutes. The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| opendevmeet | The meeting name has been set to 'nova' | 16:00 |
| gibi | o/ | 16:00 |
| dansmith | o/ | 16:01 |
| gibi | #topic Bugs (stuck/critical) | 16:03 |
| gibi | no critical | 16:03 |
| gibi | #link 13 new untriaged bugs (-4 since the last meeting): #link https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New | 16:03 |
| gibi | we have 1 bugs marked with xena-rc-potential tag #link https://bugs.launchpad.net/nova/+bugs?field.tag=xena-rc-potential | 16:03 |
| gibi | https://bugs.launchpad.net/nova/+bug/1942345 fix is going through the gate | 16:04 |
| gibi | and we have the placement transaction scope issue https://storyboard.openstack.org/#!/story/2009159 that is being worked on in https://review.opendev.org/c/openstack/placement/+/807014 | 16:04 |
| gibi | I guess we should mark that as RC critical too | 16:04 |
| gibi | I think melwitt's solution is OK but we cannot really make a reproduction test working in the functional env | 16:05 |
| gibi | so I started running nova jobs against the fix | 16:05 |
| gibi | in https://review.opendev.org/c/openstack/nova/+/807558 | 16:05 |
| gibi | so far the runs there are not producing the error | 16:06 |
| gibi | so probably we will land that fix without the reproduction test | 16:06 |
| gibi | is there any other bug we should consider as RC critical? | 16:07 |
| dansmith | you're saying that we can't repro the failure in functional, | 16:07 |
| dansmith | but that's kinda expected and normal for these kinds of load-based heisenbugs right? | 16:07 |
| * bauzas is now back and around | 16:07 | |
| gibi | dansmith: yes, it is a race that is hard to reproduce in a clean env (it needs mysql and it needs parallel transactions) | 16:08 |
| bauzas | dansmith: the problem is about mysql | 16:08 |
| dansmith | gibi: yeah | 16:08 |
| bauzas | hah, jinxed | 16:08 |
| bauzas | difficult to write a correct parallel test | 16:08 |
| dansmith | yeah, just sounded like gibi was expecting we wouldn't land until we had a repro test | 16:09 |
| dansmith | and I'm saying I'd have expected that to be impossible | 16:09 |
| gibi | we tried the repro test but we faild | 16:09 |
| gibi | so I'm OK to land this without a repro | 16:09 |
| bauzas | "I guess we should mark that as RC critical too" > yes, please | 16:10 |
| gibi | bauzas: OK, I will. It is already in the tracking etherpad https://etherpad.opendev.org/p/nova-xena-rc-potential | 16:10 |
| gibi | so if there is no other bug for the RC then I have a question about https://bugs.launchpad.net/nova/+bug/1942709 | 16:11 |
| gibi | do we support providing the adminPassword to the guest via the metadata service? | 16:11 |
| gibi | I see that the metadata service trying to fetch the password from instance.system_metadata | 16:11 |
| gibi | https://github.com/openstack/nova/blob/402fe188b4e7ff76109e8a5ea1f24a5e915eaa09/nova/api/metadata/password.py#L37 | 16:11 |
| gibi | but as far as I see we dont store the password in instance.system_metadata on master | 16:12 |
| gibi | and I was not able to track down changes around this in the git history | 16:12 |
| * bauzas dunoo | 16:12 | |
| dansmith | I don't remember if this works with libvirt | 16:13 |
| dansmith | I want to say no | 16:13 |
| bauzas | we need to look at the code honestly | 16:13 |
| gibi | bauzas: please, I got stuck tracking done if it ever worked | 16:13 |
| gibi | OK moving on then | 16:14 |
| gibi | any other bug we need to discuss? | 16:14 |
| gibi | #topic Gate status | 16:15 |
| gibi | Nova gate bugs #link https://bugs.launchpad.net/nova/+bugs?field.tag=gate-failure | 16:15 |
| gibi | The nova-next job should be passing now after the new osc-placement release on Friday | 16:15 |
| gibi | Allocation deletion conflict still can happen, and it is being worked on in https://review.opendev.org/c/openstack/placement/+/807014 | 16:15 |
| bauzas | gibi: wish me good luck for finding about the crap in sysmetadata | 16:15 |
| gibi | bauzas: good luck! :) | 16:16 |
| gibi | I'm not tracking other high frequency bugs in the gate I think we are able to merge code to master | 16:16 |
| gibi | Placement periodic job status #link https://zuul.openstack.org/builds?project=openstack%2Fplacement&pipeline=periodic-weekly | 16:16 |
| gibi | placement periodics are green | 16:16 |
| gibi | any other gate issue we need to talk about? | 16:17 |
| gibi | #topic Release Planning | 16:18 |
| gibi | Release tracking etherpad #link https://etherpad.opendev.org/p/nova-xena-rc-potential | 16:18 |
| gibi | We are in Feature Freeze. | 16:18 |
| gibi | I'm not tracking any FFEs | 16:18 |
| gibi | and not tracking any feature patches approved but not landed yet | 16:18 |
| gibi | We need to produce RC1 latest on 17th of September which is in less than 2 weeks. | 16:19 |
| * bauzas raised fist at reno btw. | 16:19 | |
| gibi | yeah, bauzas writing the reno prelude as you see | 16:19 |
| gibi | :) | 16:19 |
| bauzas | I can't upload the prelude change because of a weird issue with my rst file | 16:19 |
| bauzas | (actually, yaml one) | 16:19 |
| gibi | bauzas: push it up and tomorrow I can look at it | 16:19 |
| bauzas | hope to fix it sooner than later | 16:19 |
| bauzas | gibi: yeah, if I can't find it, I'll throw it to the gate | 16:20 |
| gibi | the above etherpad has the links to RC bugs and other TODOs | 16:20 |
| bauzas | probably something obvious enough that I missed. | 16:20 |
| gibi | any question about the coming release? | 16:20 |
| gibi | #topic PTG Planning | 16:21 |
| gibi | every info is in the PTG etherpad #link https://etherpad.opendev.org/p/nova-yoga-ptg | 16:21 |
| gibi | we don't have much topics yet, but there is still plenty of time | 16:22 |
| gibi | If you see a need for a specific cross project section then please let me know | 16:22 |
| gibi | any question about the PTG? | 16:22 |
| mloza | is there way to set extra_specs to an existing instance without modifying the db? | 16:22 |
| gibi | I guess you can turn to bauzas about those as well as he is the PTL elect :)P | 16:22 |
| bauzas | could we have a yoga mat for the PTG ? :D | 16:23 |
| gibi | mloza: let get back to you at the Open Discussion section | 16:23 |
| gibi | bauzas: sure you can, please buy one in decatlon | 16:23 |
| gibi | :) | 16:23 |
| * bauzas misses swag | 16:23 | |
| gibi | #topic Stable Branches | 16:24 |
| gibi | elodilles_pto is off | 16:24 |
| gibi | so we don't have status update on the wiki | 16:24 |
| gibi | but I guess stable is OK :) | 16:24 |
| gibi | any stable issue? | 16:24 |
| bauzas | either way, we're entering a pretty quiet period about stable | 16:24 |
| gibi | I have one news we pushed stable os-vif releases today | 16:25 |
| gibi | https://review.opendev.org/c/openstack/releases/+/807694 | 16:25 |
| gibi | https://review.opendev.org/c/openstack/releases/+/807696 | 16:25 |
| gibi | victoria and ussuri | 16:25 |
| gibi | anyhow moving on | 16:25 |
| gibi | #topic Sub/related team Highlights | 16:26 |
| gibi | Libvirt (bauzas) | 16:26 |
| gibi | ? ;) | 16:26 |
| bauzas | nothing to say | 16:26 |
| gibi | #topic Open discussion | 16:26 |
| gibi | nothing on the agenda | 16:26 |
| gibi | mloza: your turn | 16:26 |
| gibi | mloza: do you mean flavor extra_specs? | 16:26 |
| mloza | yup | 16:26 |
| gibi | you have to resize the vm | 16:27 |
| bauzas | embedded flavors can't be changed | 16:27 |
| bauzas | unless you resize, indeed | 16:28 |
| mloza | right now live migrating is failing because the instance doesn't have extra_specs that the flavor has. I cannot afford to the resize since it incurs a downtime | 16:29 |
| dansmith | I don't think that's the reason for a live migration failure, right? | 16:30 |
| mloza | i have host aggregates with metadata | 16:31 |
| mloza | the metadata is associated to the flavors | 16:31 |
| dansmith | I see, so it's the aggregate not the flavor | 16:32 |
| dansmith | force migrate should skip that, no? bauzas ? | 16:32 |
| bauzas | wait, was distracted, sorry | 16:33 |
| gibi | force still runs the scheduler isn't it? | 16:33 |
| bauzas | gibi: that depends on the microversion you ask | 16:33 |
| gibi | true | 16:33 |
| gibi | with old enough microversion you can skip the scheduler | 16:33 |
| bauzas | mloza: which exact command do you run ? | 16:33 |
| dansmith | I was going to say, I thought there was a way.. I mean that's how you break your AZ right? :) | 16:33 |
| gibi | host=<new-host> force=True microversion <= 2.67 | 16:34 |
| mloza | openstack server migrate --os-compute-api-version 2.30 --live-migration $id | 16:34 |
| bauzas | with no target ? | 16:34 |
| mloza | yep with no taret | 16:34 |
| mloza | target* | 16:35 |
| gibi | you need < 2.30 and you nee d target | 16:35 |
| bauzas | if so, this goes to the scheduler | 16:35 |
| bauzas | which checks this indeed | 16:35 |
| gibi | https://docs.openstack.org/api-ref/compute/?expanded=live-migrate-server-os-migratelive-action-detail#live-migrate-server-os-migratelive-action | 16:35 |
| dansmith | the scheduler is doing what you asked, in that case.. not sending the host to the aggregate that requires specific specs | 16:35 |
| sean-k-mooney | mloza: if you dont pass a target then it will use the az form the request spec | 16:35 |
| bauzas | gibi: or you can use 2.30 with the explicit force flag | 16:35 |
| gibi | bauzas: yes, or <=2.67 and force=True | 16:35 |
| bauzas | (which was deprecated later) | 16:35 |
| bauzas | that, yes | 16:35 |
| bauzas | I'd say, the contract is granted. | 16:36 |
| bauzas | you're asking to migrate something you shouldn't migrate | 16:36 |
| bauzas | I know this is weird | 16:37 |
| sean-k-mooney | if you for the migration will that bypas the aggreate extra spec affintiy filter | 16:37 |
| sean-k-mooney | i assume that is the one that is blocking the host | 16:37 |
| sean-k-mooney | id did not think it would | 16:37 |
| bauzas | but if all your hosts are within aggregates that set metadata and you have a filter that verifies this, then you explicitly write a contract | 16:37 |
| sean-k-mooney | i guess with force we will just check the host exist and skip everything else with the old microverion? | 16:38 |
| sean-k-mooney | that would leave the vm in a state the future move operation will try to move ti out of the aggreate you moved it into | 16:39 |
| mloza | got it to live migrate after passing a target host | 16:39 |
| gibi | OK, I guess that is solve then | 16:40 |
| gibi | :0 | 16:40 |
| gibi | :) | 16:40 |
| gibi | any other topic for today? | 16:40 |
| gibi | then lets closed this | 16:42 |
| gibi | thanks for joining | 16:42 |
| gibi | #endmeeting | 16:42 |
| opendevmeet | Meeting ended Tue Sep 7 16:42:23 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:42 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/nova/2021/nova.2021-09-07-16.00.html | 16:42 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/nova/2021/nova.2021-09-07-16.00.txt | 16:42 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/nova/2021/nova.2021-09-07-16.00.log.html | 16:42 |
| * bauzas just figured https://docs.openstack.org/reno/2.4.1/usage.html#debugging | 16:42 | |
| bauzas | holy shit, I was already knowing about it | 16:42 |
| mloza | Thanks. I think the instance should have extra_specs and prefer live migrating without a target host. If I have to add to extra_specs in the instance, is it instance_extra table that I have to modify ? | 16:44 |
| mloza | instance_uuid='a65b26f5-a278-4200-a573-84e01c685699'; which has "extra_specs": {"aggregate_instance_extra_specs:cephcomputestorage": "true"} | 16:45 |
| mloza | https://paste.opendev.org/raw/bMf35OK3I9aYbCNCv9jm/ | 16:45 |
| gibi | mloza: yepp if you want to hack the db then that is the place | 16:45 |
| sean-k-mooney | well there are 2 palces | 16:47 |
| sean-k-mooney | the instance extra table is what woud be used for generating the vm xml and on the compute node | 16:47 |
| opendevreview | melanie witt proposed openstack/placement master: Narrow scope of set allocations database transaction https://review.opendev.org/c/openstack/placement/+/807014 | 16:47 |
| sean-k-mooney | i woudl hope that is the same falvor we use for move operation inthat are not resize but we do also have flavor info in the request spec which is what is passed to the schduler filters | 16:48 |
| sean-k-mooney | the schduler never uses the info in the instance_extra tabel driectly | 16:49 |
| mloza | where does the scheduler pull info if not from instance_extra table ? | 16:51 |
| sean-k-mooney | i belive we invoke the schduler with a populated request spec object when we do the select destioantion call form the conductor | 16:52 |
| sean-k-mooney | but technially there is a request_sepcs table in the api db | 16:52 |
| sean-k-mooney | with a serialised copy of the requst spec | 16:52 |
| bauzas | sean-k-mooney: this is correct | 16:53 |
| bauzas | but when you resize (eg.) you update the request spec | 16:53 |
| sean-k-mooney | yes | 16:53 |
| bauzas | for the flavor, I mean | 16:53 |
| sean-k-mooney | but what about live migrate and cold migrate | 16:53 |
| sean-k-mooney | do they just use the api db copy | 16:53 |
| sean-k-mooney | or reconstruct it form the embeded flavor in the cell db | 16:54 |
| bauzas | cold and live migrate don't modify the resources, right? | 16:54 |
| bauzas | resize and rebuild do | 16:54 |
| bauzas | so, yes, we just call out the request spec from the api table | 16:55 |
| bauzas | we never lookup the cell db | 16:55 |
| sean-k-mooney | right which has an enbeded copy of the flavor extra specs | 16:55 |
| bauzas | the instance table, you mean? | 16:55 |
| sean-k-mooney | mloza: so you would have to update both the instnace_extra table for the compute node to use and request_specs table for the scuduler to use | 16:56 |
| sean-k-mooney | bauzas: no the request_spec has a full embeded copy of the flavor as a json blob | 16:56 |
| sean-k-mooney | well seriasised nova object | 16:56 |
| bauzas | ah this | 16:56 |
| bauzas | yes | 16:56 |
| bauzas | we have the embedded flavor stored in two places | 16:57 |
| sean-k-mooney | so if mloza is changing it to make live migrateion work they need to update the request_sepc version | 16:57 |
| sean-k-mooney | bauzas: yes | 16:57 |
| bauzas | one in the api db within the request_specs table | 16:57 |
| bauzas | one in the cell db within the instances table | 16:57 |
| bauzas | the instances table is never used for scheduling decisions | 16:57 |
| sean-k-mooney | instance_extra but yes | 16:57 |
| bauzas | my bad, yes indeed | 16:57 |
| mloza | i looked at request_specs in nova api db and indeed, there is extra_specs | 16:58 |
| sean-k-mooney | so mloza you would have to keep both in sync if you do decide to hack the db | 16:58 |
| mloza | noted. thanks for the info | 16:59 |
| mloza | nbeed to figure out the query to modify the json blob in the table extra_specs and request_specs | 17:01 |
| bauzas | use the python objects directly | 17:01 |
| bauzas | simple and quick | 17:01 |
| bauzas | but you need to run the python script somewhere where you have some nova code | 17:02 |
| * sean-k-mooney really whish we had that recreate api. | 17:02 | |
| sean-k-mooney | this come up often | 17:02 |
| bauzas | in the past, we said those instances can be rebuilt | 17:02 |
| sean-k-mooney | not for flavor | 17:02 |
| sean-k-mooney | but ya resized | 17:03 |
| bauzas | or resized | 17:03 |
| sean-k-mooney | but this comes up alot | 17:03 |
| bauzas | and we said we are cloud | 17:03 |
| sean-k-mooney | it is a significant downstream and upstream | 17:03 |
| bauzas | for some reason, people thought we were vmware | 17:03 |
| sean-k-mooney | right and there is a cloud native way to do it | 17:03 |
| sean-k-mooney | well k8 swould suport this too alto0ugh it would delete it and recreate it | 17:04 |
| bauzas | resize --live ? | 17:04 |
| bauzas | you could even use bfv | 17:04 |
| sean-k-mooney | no resize to same flavor updates embeded extra specs | 17:04 |
| sean-k-mooney | we agreeed i should write a spec for that | 17:04 |
| sean-k-mooney | just didnt do it last cycle | 17:04 |
| sean-k-mooney | still a move op with downtime | 17:04 |
| sean-k-mooney | like normal resie just we don tallow resize to same flavor today | 17:05 |
| sean-k-mooney | so you have to duplicate the flavor rather then update it and then update all your exsiting vms | 17:05 |
| dansmith | resize to same flavor might involve a reschedule though right? | 17:09 |
| dansmith | so you'd need to have a "move me if necessary" or "don't move me and abort if I would have to move" option | 17:09 |
| sean-k-mooney | dansmith: yes its a normal resize just removing the requirement to have a different flavor | 17:10 |
| dansmith | right but people wanting this will likely want to violate those assumptions | 17:10 |
| dansmith | the person asking for this wanted to resize to the same flavor, but without moving or even rebooting, so they could then live migrate | 17:11 |
| sean-k-mooney | i think the assumtion would be its alwasy going to change host | 17:11 |
| sean-k-mooney | it might now but proably will | 17:11 |
| dansmith | that wouldn't do what the person wanted though | 17:11 |
| sean-k-mooney | *not | 17:11 |
| dansmith | so I'm just saying, it's sticky :) | 17:11 |
| dansmith | or slippery, or whatever | 17:11 |
| sean-k-mooney | true its the only safe way to do it if we allow you to update anything | 17:11 |
| sean-k-mooney | e.g. numa or cpu pinning | 17:12 |
| sean-k-mooney | by the way we dont correctly update extra spec on resize https://review.opendev.org/c/openstack/nova/+/805882/1/nova/tests/functional/libvirt/test_pci_sriov_servers.py | 17:12 |
| sean-k-mooney | artom is working on that | 17:13 |
| artom | Once I surface from these esclations, but yes | 17:13 |
| sean-k-mooney | https://launchpad.net/bugs/1941005 | 17:13 |
| sean-k-mooney | im pretty sure we regressed this at some point | 17:13 |
| sean-k-mooney | but ya that is also a thing right now | 17:13 |
| sean-k-mooney | although im slight confusted why this set of tempest test is passign https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/806239/6/whitebox_tempest_plugin/api/compute/test_vpmu.py | 17:16 |
| artom | sean-k-mooney, maybe we update the request spec when we confirm the resize? | 17:19 |
| sean-k-mooney | oh ya that could be it | 17:19 |
| sean-k-mooney | so we need that refactoring i suggested where we can assert thing during resize verify | 17:19 |
| artom | Aye | 17:20 |
| sean-k-mooney | that would also explain why we have not notcied this if it update on confirm | 17:20 |
| sean-k-mooney | the window to observe it would be only during the migration | 17:20 |
| artom | And if it causes problems during scheduling... | 17:21 |
| opendevreview | Sylvain Bauza proposed openstack/nova master: Add the Xena prelude section https://review.opendev.org/c/openstack/nova/+/807786 | 17:55 |
| opendevreview | Merged openstack/nova master: Parse alias from domain hostdev https://review.opendev.org/c/openstack/nova/+/806943 | 18:05 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!