| opendevreview | Merged openstack/nova master: Extend the reproducer for 1953359 and 1952915 https://review.opendev.org/c/openstack/nova/+/820859 | 05:06 |
|---|---|---|
| opendevreview | mitya-eremeev-2 proposed openstack/nova master: Delete bogus attachments. https://review.opendev.org/c/openstack/nova/+/820935 | 06:42 |
| *** bhagyashris_ is now known as bhagyashris | 06:57 | |
| *** bhagyashris_ is now known as bhagyashris | 07:19 | |
| 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 | 09:56 |
| gibi | sean-k-mooney, slaweq: I looked at the gate bug https://bugs.launchpad.net/nova/+bug/1953478 and it seems to me that neutron sends the vif-plugged event at bind time instead of plug time in this case causing the timeout | 10:45 |
| gibi | so far I thought that with ml2/ovs neutron always sends that event at plug time | 10:45 |
| slaweq | gibi yes, afaik when You boot vm it will be sent when neutron will finish provisioning that port | 10:47 |
| gibi | slaweq: it is an unshelve after shelve_offload | 10:47 |
| slaweq | but during e.g. live migration it may be differently | 10:47 |
| slaweq | I don't know about shelve and unshelve | 10:47 |
| gibi | that is almost like a new boot | 10:47 |
| gibi | but from a previous snapshot | 10:48 |
| gibi | there is a new scheduling, a new port biding a new vif plug and spawn | 10:48 |
| slaweq | so it should be sent when plug is completed | 10:48 |
| slaweq | in such case | 10:49 |
| gibi | yes, that is how I would expect it (and how nova expects it) | 10:49 |
| slaweq | please move that bug to neutron then, I will check in our logs what happened there | 10:49 |
| opendevreview | mitya-eremeev-2 proposed openstack/nova master: Delete bogus attachments. https://review.opendev.org/c/openstack/nova/+/820935 | 10:49 |
| slaweq | and thx for checking that | 10:49 |
| gibi | slaweq: ok, I summarized this in the bug so you have the logs with timestamps from nova | 10:49 |
| slaweq | ++ | 10:49 |
| gibi | I haven't looked at the other case in the same bug the tempest.api.compute.servers.test_server_actions.ServerActionsTestJSON.test_resize_server_revert | 10:51 |
| gibi | but I can do that too | 10:51 |
| sean-k-mooney | slaweq: unshelve should be the same as first boot | 12:12 |
| lyarwood | does anyone know when/where during a spawn that we would expect the metadata service to have details on a given instance ready to serve up? | 12:13 |
| * lyarwood hasn't really touched this path before | 12:13 | |
| lyarwood | I'm looking at a CI failure downstream where cloud-init gets Connection refused everytime it tries to pull from the api but I can't see errors in the actual service logs | 12:14 |
| gibi | lyarwood: I have limited knowledge too, but the request is goes from the guest, goes to the neutron-metadata service via the guest's network and the neutron forwards the request to nova-metadata | 12:19 |
| lyarwood | thanks I think I was missing the neutron part | 12:23 |
| sean-k-mooney | ill get the code one sec | 12:24 |
| sean-k-mooney | lyarwood: it will always be ready before we call libvirt | 12:24 |
| sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/virt/netutils.py#L168 | 12:25 |
| sean-k-mooney | that is where we generate the network metadata | 12:25 |
| sean-k-mooney | which is called form the init of InstanceMetadata https://github.com/openstack/nova/blob/052cf963583ab7c6bbe4fcbf7bfe69f8f6733bdb/nova/api/metadata/base.py#L176 | 12:26 |
| lyarwood | ACK thanks | 12:27 |
| lyarwood | looks like an issue before that as the neutron-metadata-agent on the compute isn't even seeing the requests | 12:28 |
| lyarwood | I guess with plugging but AFAICT from the compute logs that worked | 12:28 |
| sean-k-mooney | the neutorn-metadata-agent is typeiclay not on the compute but on the contler | 12:29 |
| sean-k-mooney | it runs in either the router or dhcp agent network namespaces | 12:30 |
| sean-k-mooney | depening on your config | 12:30 |
| sean-k-mooney | ovn works slightly differntly and i think it does it on each compute | 12:30 |
| sean-k-mooney | but the agent there is jsut calling the nova metadta api | 12:30 |
| sean-k-mooney | the api i belive generates the instnace mentadta directly itself and caches it | 12:31 |
| sean-k-mooney | for libvirt we generate it here by the way as part of creating the config drive https://github.com/openstack/nova/blob/e537d90d6fc0977742f7126c3f8cfef6bf8b2a15/nova/virt/libvirt/driver.py#L4782 | 12:32 |
| sean-k-mooney | lyarwood: are we using memcache downstream | 12:36 |
| sean-k-mooney | we had to enable it upstream to make the ci stable | 12:36 |
| sean-k-mooney | without it differrnt request could go to differnt api workers | 12:36 |
| sean-k-mooney | cloud-init will only retry the first request | 12:36 |
| sean-k-mooney | and after that it assuem the data is avaiable and does not retry them | 12:37 |
| sean-k-mooney | if you hit a different worker and the node is under powered that can lead to timeouts | 12:37 |
| sean-k-mooney | or failure to retive the data as genertatd the metadata is actully quite expensive | 12:37 |
| lyarwood | sorry had to drop quickly for lunch | 12:56 |
| * lyarwood reads back | 12:56 | |
| lyarwood | this is part of a queens to train upgrade job downstream and it looks like the neutron metadata services are running on the computes | 12:57 |
| lyarwood | I think this is failing post upgrade to train but it's super confusing | 12:58 |
| lyarwood | jenkins-- | 12:58 |
| lyarwood | ah no it's pre upgrade on queens | 13:00 |
| gryf | asfdsdf | 13:09 |
| kashyap | gryf: Thanks; hope you don't change your password :D | 13:11 |
| gryf | kashyap, not really, just the ssh connection hiccup ;) | 13:13 |
| kashyap | Heh, was just trolling ya | 13:14 |
| gryf | I figured :) | 13:16 |
| pslestang | ccccccdlucfhelkhccngbvunedtddhhlecgrlcdverhc | 13:22 |
| pslestang | ouup's sorry | 13:22 |
| pslestang | hello by the way! | 13:22 |
| kashyap | That's a YubiKey :) | 13:22 |
| pslestang | ;-) | 13:22 |
| gryf | that's looks definitely like a sort of pass ;P | 13:23 |
| kashyap | pslestang: There's a setting to actually make YubiKey to _not_ send the key press automatically | 13:23 |
| kashyap | Lemme post my notes. I have a Nano | 13:23 |
| pslestang | kashyap: I do not have a press button, my yubikey is near the enter key and when I touch the edge of the enter key ccccccdlucfhekeerjhdnrejghllkrdvdnjnjnijknbt | 13:25 |
| pslestang | you see waht I mean | 13:25 |
| gryf | pslestang, just change usb port :) | 13:25 |
| pslestang | already done | 13:26 |
| gryf | :D | 13:26 |
| kashyap | pslestang: What I'm saying is that once you touch it, it automatically sends return keypress. That way you can still touch it while you're active on this channel, and it won't post it here | 13:27 |
| kashyap | See what I mean? | 13:27 |
| kashyap | pslestang: This one - https://kashyapc.fedorapeople.org/YubiKey-Nano-Config-for-Return-Key.html | 13:27 |
| pslestang | kashyap: thx for sharing this, it will definitely be part of my setup | 13:30 |
| kashyap | When I said "That way ..." above, I meant to say _once_ you use the above config. :) | 13:31 |
| pslestang | kashyap: just applied, works great, thx! | 13:33 |
| kashyap | Cool. | 13:34 |
| kashyap | pslestang: (I had broken formatting in the above page; fixed it now. I see that you were able to see through it) | 13:40 |
| pslestang | kashyap: yep I automatically 's/ /</br>/' whith my eyes in the corresponding lines :) | 13:43 |
| gsantos | Hello, folks. I'm the owner of https://review.opendev.org/c/openstack/nova/+/815373 , which was just merged, and I was wondering if a backport of this fix to the previous branches would be feasible. I'm looking at Ussuri, specifically. | 14:29 |
| lyarwood | gsantos: I would think so but it's conditional on the version of libvirt used right? | 14:31 |
| gsantos | Yes, that is a concern. It needs at least libvirt v4.3.0, and the minimum libvirt version for Ussuri seems to be 4.0.0 | 14:37 |
| lyarwood | gsantos: kk that's easy enough to handle in the backport however | 14:40 |
| gsantos | Great! So, in order to do that, do I cherry pick (and resolve conflicts for) this commit to the xena, wallaby, victoria and ussuri branches? | 14:46 |
| lyarwood | gsantos: Correct, https://docs.openstack.org/project-team-guide/stable-branches.html#processes | 14:52 |
| lyarwood | gsantos: and add in the libvirt version check once our MIN_LIBVIRT_VERSION dips below 3.2.0 | 14:52 |
| lyarwood | sorry 4.3.0 | 14:52 |
| opendevreview | Alexey Stupnikov proposed openstack/nova master: Test aborting queued live migration https://review.opendev.org/c/openstack/nova/+/776250 | 14:57 |
| opendevreview | mitya-eremeev-2 proposed openstack/nova master: Delete bogus attachments. https://review.opendev.org/c/openstack/nova/+/820935 | 15:06 |
| *** artom__ is now known as artom | 15:08 | |
| opendevreview | mitya-eremeev-2 proposed openstack/nova master: Delete bogus attachments. https://review.opendev.org/c/openstack/nova/+/820935 | 15:12 |
| EugenMayer | when getting `--os-compute-api-version 2.26 or greater is required to support the --not-tag option` while using xena, is that expected? | 15:15 |
| *** efried1 is now known as efried | 15:19 | |
| EugenMayer | interestingly, that is only the case for the openstack cli, using nova list --not-tags works without issues | 15:20 |
| opendevreview | Merged openstack/nova-specs master: Allow project admin to list hypervisors https://review.opendev.org/c/openstack/nova-specs/+/793011 | 15:21 |
| gsantos | lyarwood: I will try to make these cherry-picks today. Thank you! | 15:33 |
| sean-k-mooney | lyarwood: if you have time can you look at https://review.opendev.org/c/openstack/nova/+/820531 i think the patch looks ok but it would be good to have your input. its pretty small | 15:39 |
| lyarwood | sean-k-mooney: yeah that doesn't look correct | 15:59 |
| lyarwood | I'll write up some notes after some downstream calls | 15:59 |
| sean-k-mooney | ack | 16:00 |
| opendevreview | Merged openstack/nova master: Reattach mdevs to guest on resume https://review.opendev.org/c/openstack/nova/+/815373 | 16:08 |
| EugenMayer | Trying to us `--not-tags` with the openstack cli - tells me that my API version '--os-compute-api-version 2.26 or greater is required to support the --not-tag option' - is this expected with xena? seems like https://docs.openstack.org/releasenotes/nova/en_GB/xena.html tells me, that xena has v2.90. - any hints? | 16:54 |
| melwitt | EugenMayer: it is expected, you have to pass --os-compute-api-version 2.26 with the command else by default OSC uses the oldest available microversion 2.1. it does not default to latest microversion | 16:56 |
| EugenMayer | ah ! thank you sir! | 17:03 |
| eandersson | Anyone know if the bug with orphaned neutron ports when deleting a VM on a offline compute fixed in newer versions of OpenStack? Also not super sure if it is a Neutron or Nova bug :p | 19:54 |
| *** tobias-urdin3 is now known as tobias-urdin | 20:10 | |
| *** EugenMayer3 is now known as EugenMayer | 20:10 | |
| opendevreview | Gustavo Santos proposed openstack/nova stable/xena: Reattach mdevs to guest on resume https://review.opendev.org/c/openstack/nova/+/821126 | 20:55 |
| *** artom_ is now known as artom | 23:01 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!