| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2023-01-19 | |||
| 13:55:42 | sean-k-mooney | so it might be when i get back | |
| 13:55:45 | kashyap | Sure. | |
| 13:56:17 | sean-k-mooney | yep i want to merge both although i also think we should simplfy it so we only do one check | |
| 13:56:22 | gibi | kashyap: that nit from me does not worth a respin | |
| 13:56:50 | sean-k-mooney | e.g. only check with model+extra_flags | |
| 13:57:08 | kashyap | Yep; ie. remove the workaround-code | |
| 13:57:17 | kashyap | gibi: Nod; thx | |
| 13:57:34 | kashyap | (Remove later, i.e. not now :)) | |
| 13:57:59 | sean-k-mooney | anyway ill be back hopefully in an hour | |
| 13:58:10 | gibi | kashyap: so https://review.opendev.org/c/openstack/nova/+/869950 will be top of https://review.opendev.org/c/openstack/nova/+/870794 ? | |
| 13:58:21 | kashyap | gibi: Right | |
| 13:58:27 | kashyap | Makes sense to you? | |
| 13:58:35 | sean-k-mooney | well the workaround right now is disabling both check right | |
| 13:58:44 | gibi | kashyap: yeah that make sens | |
| 13:58:58 | sean-k-mooney | i have not looked at that patch in a few revisions | |
| 13:58:59 | kashyap | sean-k-mooney: Yes, only if you enable the workaround, obv. | |
| 13:59:27 | kashyap | sean-k-mooney: It's dead-simple - if workaround -> just skip the first _compare_cpu() call. | |
| 13:59:53 | kashyap | sean-k-mooney: And thanks, I see you've also +2ed the API-replacement patch, but I need to rebase it on top of the workaroud patch | |
| 14:00:09 | sean-k-mooney | and we still do the second call in init wiht the flags | |
| 14:00:49 | sean-k-mooney | ok am ill leave this for now and come back to it when i have time to review both again | |
| 14:01:39 | kashyap | Yeah, the second call correctly succeeds as it also takes into account the flags | |
| 14:02:22 | sean-k-mooney | yep so as long as that is alwasy there i was ok with removing the first call | |
| 14:02:38 | sean-k-mooney | i was not ok with removeing both of them and that why i wanted the workaround | |
| 14:02:51 | sean-k-mooney | anyway got to go be back later | |
| 14:05:58 | opendevreview | Sofia Enriquez proposed openstack/nova master: Implement is_luks_inside_qcow2 funtion https://review.opendev.org/c/openstack/nova/+/854030 | |
| 14:22:26 | opendevreview | Kashyap Chamarthy proposed openstack/nova master: libvirt: Replace usage of compareCPU() with compareHypervisorCPU() https://review.opendev.org/c/openstack/nova/+/869950 | |
| 14:23:07 | kashyap | gibi: --^ (No hurry) Done, rebased it on top of the workaround. | |
| 14:33:17 | gibi | kashyap: done :) | |
| 14:34:10 | kashyap | (シ_ _)シ | |
| 14:34:16 | kashyap | (It's a Japanese bow) | |
| 14:47:56 | gibi | ahh, I see :) | |
| 15:00:36 | dansmith | gibi: nice | |
| 15:00:48 | dansmith | gibi: I'm working on the real fix, but it's giving me a hard time.. lots of layers to cut through | |
| 15:06:54 | opendevreview | Alexey Stupnikov proposed openstack/nova stable/ussuri: fixtures: Handle binding of first port https://review.opendev.org/c/openstack/nova/+/871048 | |
| 15:08:22 | opendevreview | Alexey Stupnikov proposed openstack/nova stable/ussuri: Neutron fixture: don't clobber profile and vif_details if empty https://review.opendev.org/c/openstack/nova/+/871049 | |
| 15:08:25 | dansmith | ...aaand I got it working. sleep helps. | |
| 15:08:38 | dansmith | now to clean up the giant mess I've made | |
| 15:10:18 | gibi | dansmith: sounds good | |
| 15:28:02 | opendevreview | Alexey Stupnikov proposed openstack/nova stable/victoria: fixtures: Handle binding of first port https://review.opendev.org/c/openstack/nova/+/796905 | |
| 15:28:42 | opendevreview | Alexey Stupnikov proposed openstack/nova stable/victoria: Neutron fixture: don't clobber profile and vif_details if empty https://review.opendev.org/c/openstack/nova/+/796906 | |
| 17:45:07 | bauzas | gibi: all good to know | |
| 17:45:49 | bauzas | hopefully https://review.opendev.org/c/openstack/tempest/+/870974 will be merged tonight | |
| 20:50:03 | opendevreview | Sofia Enriquez proposed openstack/nova master: WIP: Implement encryption on backingStore https://review.opendev.org/c/openstack/nova/+/870012 | |
| 21:11:49 | opendevreview | Sofia Enriquez proposed openstack/nova master: WIP: Implement encryption on backingStore https://review.opendev.org/c/openstack/nova/+/870012 | |
| #openstack-nova - 2023-01-20 | |||
| 09:10:15 | opendevreview | Hiroki Narukawa proposed openstack/nova master: libvirt: retry libvirt connection on live_migration_monitor https://review.opendev.org/c/openstack/nova/+/867077 | |
| 09:33:48 | opendevreview | Nobuhiro MIKI proposed openstack/os-traits master: Add new 'COMPUTE_ADDRESS_SPACE_*' traits https://review.opendev.org/c/openstack/os-traits/+/871226 | |
| 10:38:33 | kashyap | bauzas: sean-k-mooney: Can anyone please put this through? -- https://review.opendev.org/c/openstack/nova/+/870794/6 -- libvirt: At start-up allow skiping compareCPU() with a workaround | |
| 10:38:55 | bauzas | kashyap: sure, adding on my today's list | |
| 10:41:09 | kashyap | bauzas: FWIW, it is already tested in a real env by a downstream colleague | |
| 10:41:17 | bauzas | kashyap: yup, I knew | |
| 10:41:27 | kashyap | Ah, thx | |
| 10:41:46 | sean-k-mooney | kashyap: so that is not really what i was asking for by th eway with the workaround | |
| 10:41:58 | kashyap | sean-k-mooney: Sigh, what then? | |
| 10:42:01 | sean-k-mooney | your orginal patch did this https://review.opendev.org/c/openstack/nova/+/869587/4/nova/virt/libvirt/driver.py | |
| 10:42:08 | sean-k-mooney | removing both checks | |
| 10:42:14 | sean-k-mooney | unconditionally | |
| 10:42:24 | kashyap | sean-k-mooney: Right; but please see the discussion in the patch itself | |
| 10:42:25 | sean-k-mooney | i wanted you to add the workaroudn if you were removing both checks | |
| 10:42:43 | sean-k-mooney | the new one only removes the first check condtionally | |
| 10:42:45 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/870794/6/nova/virt/libvirt/driver.py | |
| 10:42:54 | kashyap | sean-k-mooney: Yes, the new one only removes the first check | |
| 10:43:08 | kashyap | sean-k-mooney: That's because, the second check evaluates both model _and_ flags. So that is useful. | |
| 10:43:09 | sean-k-mooney | if you wer only removing the first check but keeping the second i was ok with that being done unconditionally | |
| 10:43:15 | kashyap | sean-k-mooney: Aaah | |
| 10:43:16 | kashyap | I see | |
| 10:43:26 | sean-k-mooney | right | |
| 10:43:35 | sean-k-mooney | i wanted to keep the second correct check | |
| 10:43:37 | kashyap | sean-k-mooney: So okay ... that is also a possibility | |
| 10:43:53 | sean-k-mooney | so im ok with approving your currnt patch or you can just delete the first check | |
| 10:43:58 | sean-k-mooney | its up to you | |
| 10:43:59 | kashyap | sean-k-mooney: I can remove the first one unconditionally and keep the second under the workaround | |
| 10:44:11 | sean-k-mooney | i dont think you need the workaround | |
| 10:44:22 | sean-k-mooney | if you just delete the second one | |
| 10:44:25 | sean-k-mooney | *first one | |
| 10:44:45 | kashyap | Right. | |
| 10:44:49 | sean-k-mooney | but if you want to future proof then sure you can put the remain check under a workaroudn | |
| 10:44:57 | sean-k-mooney | anyway let me knwo which you would prefer | |
| 10:45:06 | sean-k-mooney | the curernt patch https://review.opendev.org/c/openstack/nova/+/870794/6/nova/virt/libvirt/driver.py | |
| 10:45:16 | sean-k-mooney | or just delete the first check | |
| 10:45:17 | kashyap | sean-k-mooney: Yes, I want to future-proof it a bit by keeping the 2nd part under the workaround | |
| 10:45:28 | kashyap | We can't test all possible CPU models on all hardware | |
| 10:45:34 | kashyap | This check can prove to be useful | |
| 10:45:36 | kashyap | sean-k-mooney: How about this: | |
| 10:45:42 | sean-k-mooney | ok gibi ^ does that work for you | |
| 10:45:46 | kashyap | remove the first check and keep the second one under the workaround. | |
| 10:46:05 | sean-k-mooney | sure the second check is not under the workaround in the current patch by the way | |
| 10:46:14 | sean-k-mooney | but im ok with what you are suggesting | |
| 10:46:16 | kashyap | sean-k-mooney: Yes, yes. I know :) | |
| 10:46:45 | bauzas | kashyap: shall I hold reviewing ? | |
| 10:47:27 | kashyap | bauzas: Yes, pls | |
| 10:47:48 | bauzas | k | |
| 10:47:54 | bauzas | anyway, I need to do gym | |
| 11:24:53 | gibi | sean-k-mooney, kashyap: removing the first check unconditionally and making the second check conditionally (and by default executed) works for me | |
| 11:25:09 | kashyap | gibi: Nod; reworking ... the unit test | |
| 11:42:54 | artom | bauzas, hey, so after gym, would https://review.opendev.org/c/openstack/nova/+/869812 be good to go? | |
| 11:49:22 | opendevreview | Kashyap Chamarthy proposed openstack/nova master: libvirt: At start-up rework compareCPU() usage with a workaround https://review.opendev.org/c/openstack/nova/+/870794 | |
| 11:49:22 | opendevreview | Kashyap Chamarthy proposed openstack/nova master: libvirt: Replace usage of compareCPU() with compareHypervisorCPU() https://review.opendev.org/c/openstack/nova/+/869950 | |
| 11:49:28 | kashyap | gibi: sean-k-mooney: --^ | |
| 11:52:10 | kashyap | gibi: (I've also readjusted the wording where appropriate from the old patch) | |
| 11:53:09 | kashyap | (Lost the +W on the 2nd patch) | |