| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2022-08-17 | |||
| 18:29:43 | sean-k-mooney | then in the current one for the migration update it it | |
| 18:29:48 | JayF | got it | |
| 18:29:55 | JayF | yaml anchors are awesome and terrible :D | |
| 18:30:33 | sean-k-mooney | ok im going to go figure out dinner ill check back in a while but feel free to add me to the review and/or ping me here | |
| 18:30:58 | sean-k-mooney | in the mean time youcan likelly just recheck once the gate change fails | |
| 18:31:30 | sean-k-mooney | with a message plaease :) "recheck ironic does not support live migration" or similar | |
| 18:32:11 | sean-k-mooney | we have an issue with volume detach intermitently failing and i think that is what you are hitting | |
| 18:32:59 | opendevreview | Jay Faulkner proposed openstack/nova master: nova-live-migration tests not needed for Ironic https://review.opendev.org/c/openstack/nova/+/853529 | |
| 18:33:12 | JayF | sean-k-mooney: ^ | |
| 18:33:23 | JayF | Thanks sean :) | |
| 18:34:25 | sean-k-mooney | ok that looks ok i think | |
| 18:34:43 | sean-k-mooney | test only changes and releasenote only change are already skipped | |
| 18:34:50 | sean-k-mooney | so just the ironic virt driver | |
| 18:34:52 | JayF | That's why I didn't add the tests line, yep | |
| 18:35:13 | JayF | I know how the stanzas for zuul config work, I just can't ever find the #@%$# configs anymore lol | |
| 18:36:10 | sean-k-mooney | you can technially go form teh zuul ui to the job definition mostly at least to the repo where there defiend | |
| 18:36:26 | sean-k-mooney | i normally fall back on https://codesearch.opendev.org/ if i cant find it quickly | |
| 18:36:45 | sean-k-mooney | anyway chat to you tomorrow o/ | |
| 20:27:19 | opendevreview | Merged openstack/nova stable/victoria: [ironic] Minimize window for a resource provider to be lost https://review.opendev.org/c/openstack/nova/+/800873 | |
| 20:27:42 | JayF | \o/ | |
| 20:30:32 | opendevreview | Jay Faulkner proposed openstack/nova stable/ussuri: [ironic] Minimize window for a resource provider to be lost https://review.opendev.org/c/openstack/nova/+/853540 | |
| 21:19:03 | opendevreview | Rico Lin proposed openstack/nova master: Add locked_memory extra spec and image property https://review.opendev.org/c/openstack/nova/+/778347 | |
| 21:19:04 | opendevreview | Rico Lin proposed openstack/nova master: Add traits for viommu model https://review.opendev.org/c/openstack/nova/+/844507 | |
| 21:19:04 | opendevreview | Rico Lin proposed openstack/nova master: libvirt: Add vIOMMU device to guest https://review.opendev.org/c/openstack/nova/+/830646 | |
| 21:20:39 | ricolin | gibi: was asking about the necessary of raise exception for locked memory, but I guess the right behavior should be raise it. just update the patch accordingly. thanks for your review, please help to review again. ^^^ | |
| 22:04:15 | opendevreview | Jay Faulkner proposed openstack/nova stable/train: [ironic] Minimize window for a resource provider to be lost https://review.opendev.org/c/openstack/nova/+/853546 | |
| 22:05:56 | JayF | Train is as far back as you all go, right? | |
| 22:23:36 | sean-k-mooney[m] | i think we still might have older branches but i think we have gate issues beyond train | |
| 22:24:18 | sean-k-mooney[m] | we still have stable queens branches | |
| 22:24:43 | sean-k-mooney[m] | but i dont think the gate will pass. you can try | |
| 22:25:09 | sean-k-mooney[m] | we have been discussing disabling some broken jobs on really old branches | |
| 22:35:21 | JayF | train is probably the most reasonable to go back to anyway | |
| 22:57:26 | melwitt | JayF: iiuc, the eventual fix in the ironic driver was this https://review.opendev.org/q/Iba87cef50238c5b02ab313f2311b826081d5b4ab but it hasn't been reviewed on older branches yet. fyi | |
| #openstack-nova - 2022-08-18 | |||
| 03:10:56 | opendevreview | Merged openstack/nova master: block_device_info: Add swap to inline https://review.opendev.org/c/openstack/nova/+/826523 | |
| 05:19:53 | opendevreview | Amit Uniyal proposed openstack/nova master: Adds check for VM snapshot fail while quiesce https://review.opendev.org/c/openstack/nova/+/852171 | |
| 06:52:52 | opendevreview | Merged openstack/nova master: libvirt: Improve creating images INFO log https://review.opendev.org/c/openstack/nova/+/826524 | |
| 06:57:46 | opendevreview | Merged openstack/nova master: libvirt: Remove defunct comment https://review.opendev.org/c/openstack/nova/+/826525 | |
| 09:38:42 | opendevreview | Balazs Gibizer proposed openstack/nova master: Trigger reschedule if PCI consumption fail on compute https://review.opendev.org/c/openstack/nova/+/853611 | |
| 10:04:07 | gibi | sean-k-mooney: for the yesterday pci filter bug discussion. The PciDevicePool object is differently set up depending on where it is used. in the compute it has a devices dict with PciDevice object, but that is not present from the scheduler perspective. | |
| 10:04:43 | gibi | so we cannot simply replace support_requests with consume_requests | |
| 10:04:56 | gibi | as consume_requests depends on the pool['devices'] to be present | |
| 10:06:43 | gibi | we can call apply_request and that will decrease the pool count so it will detect the double booking. the only downside is that it does not support dependent devices | |
| 10:07:07 | gibi | as that would again require pool['devices'] to be present in the stats | |
| 10:07:28 | sean-k-mooney | ack | |
| 10:07:36 | gibi | btw support_requests is called from both the PciPassthroughFilter and the NumaTopologyFilter | |
| 10:07:37 | sean-k-mooney | ya the schduler just works on the count | |
| 10:07:48 | sean-k-mooney | yes | |
| 10:08:00 | sean-k-mooney | the numa toplogy filter needs to validate teh device numa toplogy | |
| 10:08:05 | gibi | yepp | |
| 10:08:12 | gibi | so I will replace support with apply | |
| 10:08:25 | gibi | that will enhance the scheduling logic | |
| 10:08:39 | sean-k-mooney | you proably want to call both no | |
| 10:08:49 | sean-k-mooney | support then apply | |
| 10:09:13 | gibi | apply does what support do but also decrease counts | |
| 10:09:16 | sean-k-mooney | supporst is just a simple dict comprehention so its cheap | |
| 10:09:32 | sean-k-mooney | ack but is apply much more complex or about the same | |
| 10:09:35 | sean-k-mooney | have not looked in a while | |
| 10:09:35 | gibi | I can call both but I think apply is a superset of support | |
| 10:10:05 | gibi | support is basically _filter_pools, apply is _filter_pools + _decrease_pool_count | |
| 10:10:10 | sean-k-mooney | ya im just wondering about cost but i guess apply is already called for multi create | |
| 10:10:21 | gibi | apply is called at the end yes | |
| 10:10:25 | gibi | anyhow | |
| 10:10:37 | sean-k-mooney | ack so i guess that is fine | |
| 10:11:02 | sean-k-mooney | will apply decresase the pools if they all dont fit | |
| 10:11:20 | sean-k-mooney | i.e. can it handel partial cases | |
| 10:11:45 | sean-k-mooney | im just wonderign do we need to do an atomic swap | |
| 10:12:10 | gibi | I can call apply on a local copy of stats | |
| 10:12:14 | gibi | then drop the copy | |
| 10:12:23 | sean-k-mooney | ack | |
| 10:12:24 | gibi | so no interference between parallel requests | |
| 10:12:42 | sean-k-mooney | well i was thinking swap the copy with the orginal if it succeeeds | |
| 10:12:51 | sean-k-mooney | so that multi create works | |
| 10:13:16 | sean-k-mooney | althogh | |
| 10:13:18 | sean-k-mooney | maybe not | |
| 10:13:29 | sean-k-mooney | droping it might be better to not break the numa toplogy filters | |
| 10:13:33 | gibi | we have two phases for multicreate 1) running filters (this will apply on a copy) 2) consume the selected host (this already apply on a shared stats) | |
| 10:13:57 | sean-k-mooney | right we will need to do apply 3 times | |
| 10:14:08 | sean-k-mooney | pci filter, numa toplogy filter and host manager | |
| 10:14:16 | gibi | yes | |
| 10:14:20 | sean-k-mooney | the first two should be copies to not break each other | |
| 10:14:23 | gibi | yes | |
| 10:14:34 | gibi | so I will keep the support call to do the apply on a copy | |
| 10:14:49 | gibi | and keep the apply work on the shared stats | |
| 10:14:56 | sean-k-mooney | you could make supports call apply internally on a copy | |
| 10:15:23 | gibi | but then I need to signal to apply when to copy and when not to copy | |
| 10:15:47 | sean-k-mooney | i have not looked at the signiture but do we pass in the pools to apply | |
| 10:15:55 | sean-k-mooney | or does it get them itself | |
| 10:16:03 | sean-k-mooney | i was assumign we passed them in | |
| 10:16:11 | sean-k-mooney | so supports could do the copy for it | |
| 10:17:02 | sean-k-mooney | _apply takes the pools https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L622-L653 | |
| 10:17:48 | sean-k-mooney | its up to you which you think is cleaner | |
| 10:17:50 | gibi | yes, but apply_requests doesnt | |
| 10:17:58 | gibi | anyhow I will push the code soon | |
| 10:17:59 | sean-k-mooney | let me know when its ready to review | |
| 10:18:02 | gibi | and we can look at that | |
| 10:18:03 | sean-k-mooney | cool | |
| 10:18:09 | gibi | thanks for the discussion | |
| 10:24:40 | sean-k-mooney | im going to try an power through and finish the vdpa patches today just an fyi | |
| 10:24:48 | sean-k-mooney | its mainly just tests and docs at this point | |
| 10:24:50 | opendevreview | efineshi proposed openstack/python-novaclient master: Fix nova host-evacuate won't work with hostname is like a.b.c https://review.opendev.org/c/openstack/python-novaclient/+/853465 | |