| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-11-23 | |||
| 15:49:30 | gibi | yeah, i have no problem ignoring this qemu issue if it happens once in two years | |
| 15:50:01 | sean-k-mooney | yep same | |
| 15:50:15 | sean-k-mooney | we just know that it can fail this way | |
| 15:50:18 | kashyap | sean-k-mooney: Also, what's happening in the guest? | |
| 15:50:32 | kashyap | Is it just a garden-variety CirrOS instance just idling? | |
| 15:50:35 | sean-k-mooney | its just a migtrion of a cirror guest | |
| 15:50:41 | sean-k-mooney | yep | |
| 15:51:20 | sean-k-mooney | so beyound the ocational dhcp request or other idel work it should be doing nothing most of the time | |
| 15:54:55 | kashyap | sean-k-mooney: Dave says if it's a migrate-during-a-reboot, it could be a potential problem | |
| 15:54:58 | kashyap | That said ... | |
| 15:55:38 | kashyap | sean-k-mooney: gibi: I learn from Dave that we get that "guest index inconsistent" error when the migrated RAM is inconsistent with the migrated 'virtio' device state. | |
| 15:55:44 | kashyap | And a common case is where a 'virtio' device does an operation after the vCPU is stopped and after RAM has been transmitted. | |
| 15:55:52 | kashyap | (I'll add a comment in the bug from 2018 with notes) | |
| 15:58:22 | gibi | kashyap: thanks | |
| 16:10:36 | stephenfin | bauzas: Think you could take a look at this this week? https://review.opendev.org/c/openstack/nova/+/756549 | |
| 16:10:49 | bauzas | stephenfin: sure, will look | |
| 16:10:53 | stephenfin | ta | |
| 16:14:13 | openstackgerrit | Merged openstack/nova-specs master: Re-proposes 'Proposal for a safer remote console with password authentication https://review.opendev.org/c/openstack/nova-specs/+/759828 | |
| 16:28:53 | openstackgerrit | Merged openstack/nova master: Ignore InstanceNotFound exception when host rejoin cluster https://review.opendev.org/c/openstack/nova/+/757053 | |
| 16:39:14 | stephenfin | dansmith: I know it's your final day before the new year, but if you've a chance to provide guidance at [1] I'd appreciate it. Would like to close that out this year :) https://review.opendev.org/c/openstack/nova/+/759114/3/nova/db/sqlalchemy/models.py#1212 | |
| 16:40:19 | dansmith | stephenfin: ack | |
| 16:50:11 | kashyap | gibi: sean-k-mooney: I've outlined some notes here after my chat with QEMU migration maintainer, FWIW: https://bugs.launchpad.net/nova/+bug/1737625/comments/4 | |
| 16:50:11 | openstack | Launchpad bug 1737625 in OpenStack Compute (nova) "live migration failed with "Failed to load virtio-blk:virtio" (libvirt/qemu)" [Undecided,Confirmed] | |
| 16:51:03 | sean-k-mooney | kashyap: we might want to use the other bug | |
| 16:51:13 | sean-k-mooney | and close that as a duplicate | |
| 16:51:35 | kashyap | sean-k-mooney: Isn't this more accurate? It is talking about 'virtio-blk' and the Gate issue also is virtio-blk | |
| 16:51:44 | sean-k-mooney | https://bugs.launchpad.net/nova/+bug/1761798 | |
| 16:51:44 | openstack | Launchpad bug 1761798 in OpenStack Compute (nova) "live migration intermittently fails in CI with "VQ 0 size 0x80 Guest index 0x12c inconsistent with Host index 0x134: delta 0xfff8"" [Medium,Confirmed] | |
| 16:51:56 | sean-k-mooney | kashyap: no i dont think it is i think its the same underlying issue | |
| 16:52:56 | kashyap | In any case, I'm copying my comment into this one, too: https://bugs.launchpad.net/nova/+bug/1761798 | |
| 16:52:56 | openstack | Launchpad bug 1761798 in OpenStack Compute (nova) "live migration intermittently fails in CI with "VQ 0 size 0x80 Guest index 0x12c inconsistent with Host index 0x134: delta 0xfff8"" [Medium,Confirmed] | |
| 18:14:26 | openstackgerrit | Merged openstack/nova master: virt: Remove 'change_instance_metadata' API https://review.opendev.org/c/openstack/nova/+/749316 | |
| 18:42:34 | openstackgerrit | Stephen Finucane proposed openstack/nova master: virt: Remove 'reset_network' API https://review.opendev.org/c/openstack/nova/+/749315 | |
| 18:42:34 | openstackgerrit | Stephen Finucane proposed openstack/nova master: virt: Remove 'get_all_bw_counters' API https://review.opendev.org/c/openstack/nova/+/749312 | |
| 18:42:35 | openstackgerrit | Stephen Finucane proposed openstack/nova master: objects: Remove 'BandwidthUsage', 'BandwidthUsageList' https://review.opendev.org/c/openstack/nova/+/759114 | |
| 18:42:35 | openstackgerrit | Stephen Finucane proposed openstack/nova master: db: Drop models for removed services, features https://review.opendev.org/c/openstack/nova/+/698706 | |
| 19:12:20 | sean-k-mooney | stephenfin: ah those are related to xenapi removal | |
| 19:16:50 | sean-k-mooney | i tought it was related to the compute_monitors config intially | |
| 19:16:57 | sean-k-mooney | https://github.com/openstack/nova/blob/120d866df7ab81b666c0e43b97705a90483533bc/nova/conf/compute.py#L107-L126 | |
| 19:24:02 | sean-k-mooney | stephenfin: by the way we talked about the metrics weigher and weighing on things that were not compute weights in the past | |
| 19:24:06 | sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/compute/monitors/__init__.py#L40-L48 | |
| 19:24:17 | sean-k-mooney | the monitors are a stevaldor entry point | |
| 19:24:30 | sean-k-mooney | and the numa memory bandwtih ones were actully plugins not in tree | |
| 19:24:57 | sean-k-mooney | https://github.com/openstack/nova/blob/c250aca00ba89b6fad86bfaebb3083e9f5c55f3b/nova/compute/monitors/__init__.py#L44 the doc used to reference them years ago | |
| 19:28:43 | sean-k-mooney | well it was propsoed upstream https://review.opendev.org/c/openstack/nova/+/270344 | |
| 19:28:53 | sean-k-mooney | but it was not merged | |
| 19:31:20 | sean-k-mooney | https://review.opendev.org/c/openstack/nova-specs/+/180983 was the spec | |
| 21:21:27 | sean-k-mooney | gibi: stephenfin i just realised that https://review.opendev.org/c/openstack/nova/+/653506 is still not merged but its been so long that i have lost most of the context. its still proably should be fixed but im not sure when ill get to it. hopefully soon but i tought this was merged years ago... | |
| #openstack-nova - 2020-11-24 | |||
| 00:10:54 | gmann | gibi: stephenfin: can you check this sample config fix- https://review.opendev.org/c/openstack/nova/+/762721 | |
| 00:49:36 | brinzhang | gibi, dansmith: do you have time to review the cyborg shelve/unshelve support patch, it has addressed all your patch https://review.opendev.org/c/openstack/nova/+/729563 | |
| 01:38:09 | openstackgerrit | DatLQ proposed openstack/nova stable/rocky: Fix unplugging VIF when migrate/resize VM https://review.opendev.org/c/openstack/nova/+/763898 | |
| 01:58:13 | openstackgerrit | DatLQ proposed openstack/nova stable/rocky: Fix unplugging VIF when migrate/resize VM https://review.opendev.org/c/openstack/nova/+/763898 | |
| 05:51:22 | openstackgerrit | Kevin Zhao proposed openstack/nova master: Support Cpu Compararion on Aarch64 Platform https://review.opendev.org/c/openstack/nova/+/763928 | |
| 07:20:24 | openstackgerrit | Jorhson Deng proposed openstack/nova-specs master: Add instance_state to ensure volume attachment successful https://review.opendev.org/c/openstack/nova-specs/+/758920 | |
| 08:05:31 | openstackgerrit | MIKI Nobuhiro proposed openstack/nova-specs master: Add IP address to libvirt guest metadata https://review.opendev.org/c/openstack/nova-specs/+/760750 | |
| 08:26:25 | openstackgerrit | Nobuhiro MIKI proposed openstack/nova-specs master: Add IP address to libvirt guest metadata https://review.opendev.org/c/openstack/nova-specs/+/760750 | |
| 08:35:59 | openstackgerrit | Jorhson Deng proposed openstack/nova-specs master: Add instance_state to ensure volume attachment successful https://review.opendev.org/c/openstack/nova-specs/+/758920 | |
| 10:46:29 | openstackgerrit | Xinran WANG proposed openstack/nova-specs master: SRIOV Enabled Nic Support Specification https://review.opendev.org/c/openstack/nova-specs/+/742785 | |
| 11:09:34 | lyarwood | gibi: morning, could I pester you for a review of https://review.opendev.org/c/openstack/nova/+/756434 - RH owner and stephenfin has already ACK'd so I didn't want to +W it without further review. | |
| 11:09:55 | gibi | lyarwood: ack, looking | |
| 11:10:06 | lyarwood | gibi: thanks | |
| 11:11:41 | openstackgerrit | Stephen Finucane proposed openstack/nova master: objects: Remove 'BandwidthUsage', 'BandwidthUsageList' https://review.opendev.org/c/openstack/nova/+/759114 | |
| 11:11:42 | openstackgerrit | Stephen Finucane proposed openstack/nova master: db: Drop models for removed services, features https://review.opendev.org/c/openstack/nova/+/698706 | |
| 11:14:21 | gibi | lyarwood: +a | |
| 11:16:11 | lyarwood | gibi: cheers | |
| 11:54:32 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Make PCI claim NUMA aware during live migration https://review.opendev.org/c/openstack/nova/+/748453 | |
| 12:22:21 | openstackgerrit | Balazs Gibizer proposed openstack/nova stable/victoria: Warn when starting services with older than N-1 computes https://review.opendev.org/c/openstack/nova/+/761923 | |
| 12:22:26 | openstackgerrit | Balazs Gibizer proposed openstack/nova stable/victoria: Add upgrade check about old computes https://review.opendev.org/c/openstack/nova/+/761924 | |
| 12:28:45 | gibi | elod: ^^ did the squash | |
| 12:43:33 | sean-k-mooney | gibi: stephenfin since the autor of https://review.opendev.org/c/openstack/nova-specs/+/760750 ignored my attempt to colabarte and have just one spec i want to get your input on "do i really need a spec" | |
| 12:44:11 | sean-k-mooney | basically i was plannign to update the libvirt metadata with teh flavor extra specs and image proeprties via a specless blueprint | |
| 12:44:44 | sean-k-mooney | this is an internal api used only for debugging so i kind of would prefer to just propose a patch and debate the implemeantion there | |
| 12:45:19 | sean-k-mooney | i can write a spec but its not an api change so it feels kind fo heavy weight to me | |
| 12:46:42 | stephenfin | I think gibi's the only person that can really answer that | |
| 12:47:05 | gibi | do you remember why we asked for a spec for the first place? Now it would be strange to treate the two features differently | |
| 12:47:33 | stephenfin | I think because there were a few design questions and implications for things like attach/detach | |
| 12:48:26 | sean-k-mooney | i can write a spec but the only time the flavor and image will change are resize and rebuild where im regenreating the xml anyway | |
| 12:48:44 | gibi | and resize while changes the flavor it does regen... ^^ | |
| 12:49:09 | gibi | then I'm OK with a specless bp | |
| 12:49:24 | gibi | this makes the flavor part easier | |
| 12:50:03 | sean-k-mooney | yep basically im sick of having to figure out what the flavor and image propertise are form customer logs | |
| 12:50:25 | stephenfin | you'd be using the embedded flavor, presumably? | |
| 12:50:25 | sean-k-mooney | haveing them in the xml output would mean i dont have to ask them | |
| 12:50:28 | sean-k-mooney | stephenfin: yep | |
| 12:51:20 | sean-k-mooney | which is also important as i can say well this vm was booted before you updated the flavor as we can see form the xml | |
| 12:53:25 | sean-k-mooney | gibi: ill add it to the adgenda for the team meeting and if other want a spec ill add a small one. | |
| 12:53:54 | gibi | sean-k-mooney: cool | |
| 13:37:35 | Qianbiao | Hello, Nova. | |
| 13:37:56 | Qianbiao | Hello, i build a customer vendor_data metadata service. | |
| 13:38:18 | Qianbiao | And i can get vendor through "curl http://169.254.169.254/openstack/2018-08-27/vendor_data2.json" | |
| 13:38:43 | Qianbiao | But cloud-init does not try to load this file, it seems it only loads http://169.254.169.254/openstack/2018-08-27/vendor_data.json | |
| 13:38:48 | Qianbiao | what config should i update? | |
| 13:59:22 | Qianbiao | and i do not see how it works, from the source code: https://github.com/canonical/cloud-init/blob/master/cloudinit/sources/helpers/openstack.py#L230-L255 | |
| 13:59:35 | Qianbiao | it seems vendor_data2.json is never loaded. | |
| 14:40:11 | bauzas | dansmith: do you have time for a quick question about the RPC API bump ? | |
| 14:41:13 | bauzas | dansmith: could I just drop https://github.com/openstack/nova/blob/master/nova/compute/rpcapi.py#L786 now (and all the nova-network methods) or should I wait until X ? | |
| 14:42:17 | bauzas | dansmith: afaicu, if the operator would like to pin the RPC API to 5.x, then we would still need to support the method in the manager, but not in the rpcapi since nobody would use this method, right ? | |
| 14:42:47 | bauzas | as only computes would be in Victoria but none of the conductors | |
| 15:10:50 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: WIP: Bump the Compute RPC API to version 6.0 https://review.opendev.org/c/openstack/nova/+/761452 | |
| 15:22:36 | openstackgerrit | Stephen Finucane proposed openstack/nova master: db: Drop models for removed services, features https://review.opendev.org/c/openstack/nova/+/698706 | |