| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-02-16 | |||
| 11:00:14 | gibi | stephenfin: I think the os-trait merge should not take long time | |
| 11:00:18 | bauzas | gibi: I added a comment about it | |
| 11:00:25 | bauzas | about the same company one | |
| 11:00:50 | bauzas | gibi: agreed about releasing it quickly | |
| 11:01:06 | gibi | bauzas: no worries about the same company rule | |
| 11:02:36 | gibi | stephenfin: is it OK for you to wait for this patch to merge and then release the os-traits? | |
| 11:02:45 | stephenfin | yup, wfm | |
| 11:02:46 | gibi | I think we can do that today | |
| 11:07:13 | lyarwood | so I've not a single gerrit review email this morning, odd. | |
| 11:09:13 | lyarwood | gah nvm it's offlineimap *again* | |
| 11:09:23 | lyarwood | I should really switch to gmail and give up with mutt etc | |
| 11:11:43 | openstackgerrit | Merged openstack/os-traits master: Add COMPUTE_EPHEMERAL_ENCRYPTION tratis https://review.opendev.org/c/openstack/os-traits/+/759878 | |
| 11:14:31 | gibi | that was faaaast ^^ | |
| 11:14:55 | stephenfin | yup, https://review.opendev.org/c/openstack/releases/+/775804 updated | |
| 11:15:01 | gibi | looking | |
| 11:15:53 | gibi | done | |
| 11:17:30 | stephenfin | thanks | |
| 11:52:03 | kashyap | lyarwood: OfflineIMAP isn't being ported to Python-3 :-( It's Python-2-only | |
| 11:52:16 | kashyap | lyarwood: I have been planning to migrate to `mbsync`, and keep delaying it | |
| 11:52:34 | stephenfin | lyarwood: Are you okay with me merging https://review.opendev.org/c/openstack/nova/+/775415 into https://review.opendev.org/c/openstack/nova/+/769920/5 ? | |
| 11:52:40 | kashyap | [I know, Python version isn't the problem in your case now] | |
| 11:52:41 | stephenfin | I need that work for the Secure Boot stuff also | |
| 11:52:52 | stephenfin | I can keep them separate too if you'd prefer | |
| 11:53:18 | lyarwood | stephenfin: yeah merge/nuke it given you've fixed things up in your change | |
| 11:53:40 | lyarwood | stephenfin: to be clear, close mine out. | |
| 11:55:31 | kashyap | stephenfin: Thanks for the review here; /me goes to read: https://review.opendev.org/c/openstack/nova/+/774240 | |
| 11:56:00 | kashyap | stephenfin: lyarwood: My capacity this week is significantly reduced due to some personal event :-( and next week I'm on PTO | |
| 11:56:13 | kashyap | So I'll be somewhat less in action; but I'd like to address the feedback on the CPU stuff | |
| 11:56:57 | lyarwood | ack np, I can help push things along while you're out this week, I'm also out Thursday/Friday next week FWIW. | |
| 11:57:54 | kashyap | Thank you; the CPU stuff is in shape, modulo Stephen's comments. (And just trying to close out some more pressing downstream stuff, too) | |
| 11:58:20 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Add 'nova.virt.libvirt.utils.get_arch' stub to fixture https://review.opendev.org/c/openstack/nova/+/774904 | |
| 11:58:20 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Remove unnecessary mock https://review.opendev.org/c/openstack/nova/+/774905 | |
| 11:58:21 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Poison os.uname https://review.opendev.org/c/openstack/nova/+/775415 | |
| 11:58:26 | kashyap | I'm still here this week, but functioning less than 100%. (Next week I'm off) | |
| 11:58:26 | stephenfin | b****** | |
| 11:58:43 | stephenfin | that shouldn't have rebased the whole series :-( | |
| 11:59:27 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Add 'nova.virt.libvirt.utils.get_arch' stub to fixture https://review.opendev.org/c/openstack/nova/+/774904 | |
| 11:59:28 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Remove unnecessary mock https://review.opendev.org/c/openstack/nova/+/774905 | |
| 11:59:28 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Poison os.uname https://review.opendev.org/c/openstack/nova/+/775415 | |
| 12:00:29 | lyarwood | -R ftw | |
| 12:01:12 | stephenfin | nah, I've a bad habit of using 'git rebase -i master' to do any history rewriting work | |
| 12:01:21 | stephenfin | vs. git rebase -i HEAD~2 or whatever I needed there | |
| 12:05:47 | lyarwood | huh does that edit the entire series from origin/master/HEAD? | |
| 12:06:24 | stephenfin | rebases and lets me do what I want to the entire series | |
| 12:07:05 | lyarwood | ah it rebases onto master, so if that changes the entire series does | |
| 12:07:14 | stephenfin | yup | |
| 12:07:20 | lyarwood | yeah I just use -i HEAD~$n | |
| 12:07:37 | stephenfin | a wiser approach | |
| 13:11:57 | kashyap | Randomly thinking out loud ... a big blocker in the way of secure boot is the handling of the NVRAM file (which stores boot-level variables) -- right now, on instance start/stop or any of the "move operations" (all kinds of migration), the NVRAM file gets reset | |
| 13:18:51 | gibi | lyarwood: I read through the machine_type series, left commentas and questions but overall I'm OK with the direction | |
| 13:26:59 | lyarwood | gibi: ack thanks, in return I have your device detach series open and will go through it at some point today | |
| 13:32:13 | gibi | lyarwood: stephen had some comments on that so I have to go back and fix things up | |
| 13:32:18 | gibi | lyarwood: but review is appreciated | |
| 13:32:24 | lyarwood | gibi: ack understood | |
| 13:44:51 | elod | gibi: can you hit this one? (not that placement @ train has much to release, just this one: Add DEBUG logs to help troubleshoot no allocation candidates): https://review.opendev.org/c/openstack/releases/+/773093 | |
| 13:45:22 | openstackgerrit | Merged openstack/os-vif master: Drop use of deprecated collections classes https://review.opendev.org/c/openstack/os-vif/+/773360 | |
| 14:08:25 | openstackgerrit | Lajos Katona proposed openstack/nova master: Run nova-live-migration job with neutron-trunk enabled https://review.opendev.org/c/openstack/nova/+/775838 | |
| 14:49:36 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Add net & utils methods for routed nets & segments https://review.opendev.org/c/openstack/nova/+/773976 | |
| 14:49:39 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Add a routed networks scheduler pre-filter https://review.opendev.org/c/openstack/nova/+/749068 | |
| 14:49:52 | bauzas | gibi: stephenfin: sean-k-mooney: eventually done with routed networks ^ | |
| 14:50:35 | bauzas | I just splitted the functests into a whole separate module self-containing its own fixture, plus some nits from stephenfin | |
| 14:50:48 | bauzas | stylish nits | |
| 14:51:02 | sean-k-mooney | bauzas: since its fresh in your mind does the requested networks have the list of ports in it | |
| 14:51:05 | sean-k-mooney | it does right | |
| 14:51:13 | bauzas | sean-k-mooney: the fixture ? | |
| 14:51:16 | sean-k-mooney | well ports or networks | |
| 14:51:22 | sean-k-mooney | no in the request spec | |
| 14:51:22 | bauzas | ah the request ? | |
| 14:51:33 | sean-k-mooney | i need it for something else | |
| 14:51:34 | bauzas | well, for boot, we get the list of networks or ports | |
| 14:51:47 | bauzas | and for move, then we get the existing ports | |
| 14:52:09 | bauzas | the list of *asked* ports or networks | |
| 14:52:14 | sean-k-mooney | i might need to copy the same logic for the numa prot policy work | |
| 14:52:37 | bauzas | like, if I ask for -net <something> | |
| 14:52:47 | bauzas | then the RequestSpec will get the network ID | |
| 14:52:53 | sean-k-mooney | ya thats ok | |
| 14:53:10 | bauzas | (but that's only for booting) | |
| 14:53:10 | sean-k-mooney | for the port numa affinity policeis you will need to precreate the ports | |
| 14:53:18 | openstackgerrit | Belmiro Moreira proposed openstack/nova master: Change API unexpected exception message https://review.opendev.org/c/openstack/nova/+/628061 | |
| 14:53:22 | sean-k-mooney | so i can ignore networks | |
| 14:53:25 | bauzas | sean-k-mooney: if so, you'll get the ports in the spec | |
| 14:53:29 | bauzas | even for the boot | |
| 14:53:33 | sean-k-mooney | yep | |
| 14:53:37 | sean-k-mooney | just double checking | |
| 14:53:40 | bauzas | np | |
| 14:53:54 | sean-k-mooney | i basically have it implemented for sriov already | |
| 14:53:57 | bauzas | glad to help about network :p | |
| 14:54:10 | sean-k-mooney | i was checking how what else is needed for numa aware vswitchs | |
| 14:54:19 | sean-k-mooney | it look not that hard | |
| 14:54:38 | sean-k-mooney | but i need to pass the port info | |
| 14:54:51 | openstackgerrit | Merged openstack/nova master: libvirt: Don't drop CPU flags with policy='disable' from guest XML https://review.opendev.org/c/openstack/nova/+/775431 | |
| 14:55:55 | sean-k-mooney | ill need to do something similar to this https://review.opendev.org/c/openstack/nova/+/749068/16/nova/scheduler/request_filter.py#317 | |
| 14:59:25 | sean-k-mooney | i can either extend the Schduler limits object or i can look up the port policy form neutron in the hardware module | |
| 15:05:05 | gibi | elod: there is a newer commit in the stable/train branch of placement. see my comment in le checking | |
| 15:05:08 | gibi | 15:53 < bauzas> np | |
| 15:05:14 | gibi | ups | |
| 15:05:19 | gibi | wrong copy paster buffer | |
| 15:05:27 | gibi | elod: so see my comment in https://review.opendev.org/c/openstack/releases/+/773093 | |
| 15:05:29 | bauzas | hah lol | |
| 15:05:32 | bauzas | no worries | |