| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-11-16 | |||
| 14:47:29 | elod | bauzas gibi : ^^^ | |
| 14:48:16 | elod | and if you are there, you could also have a look at this: https://review.opendev.org/#/c/762408/ o:) | |
| 14:50:05 | bauzas | elod: +1 | |
| 14:51:46 | elod | thx! | |
| 14:52:38 | bauzas | elod: fwiw, I just +w https://review.opendev.org/#/c/756404/1 for the stable/stein | |
| 14:52:57 | bauzas | I just looked at https://review.opendev.org/#/q/branch:stable/stein+project:openstack/nova and saw it | |
| 14:55:21 | elod | bauzas: does it worth another final Stein release? | |
| 14:55:23 | elod | :) | |
| 14:56:18 | bauzas | nope, I don't think | |
| 14:58:13 | elod | ack | |
| 15:10:49 | lyarwood | stephenfin: https://review.opendev.org/#/c/749315/5/nova/compute/rpcapi.py@1140 - You need to remove the rpcapi method right? | |
| 15:12:03 | stephenfin | Yup, I do | |
| 15:13:45 | lyarwood | cool I was worried I had missed yet more rpcapi weirdness | |
| 15:17:37 | lyarwood | stephenfin: stupid question, does osc not support setting extra_specs? | |
| 15:17:58 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Fix config option default value for sample config file https://review.opendev.org/762721 | |
| 15:17:59 | stephenfin | openstack flavor set --property foo=bar baz | |
| 15:18:07 | lyarwood | yeah sorry --property confused me | |
| 16:37:52 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: DNM: replace sleep with flags https://review.opendev.org/762872 | |
| 16:52:40 | gibi | elod: do you have a separate patch stein-em patch for placement and os-resource-classes? | |
| 16:53:42 | elod | gibi: yes, let me look it up | |
| 16:54:18 | elod | https://review.opendev.org/#/c/762412/ | |
| 16:54:21 | elod | gibi: ^^^ | |
| 16:57:12 | gibi | elod: was there any reason behind splitting the -em patches this way? | |
| 17:00:17 | gibi | elod: also there is a doc bug fix in placement stable/stein since the last release but I guess docs are generated from the branch not from the -em tag so the published docs will be up to date | |
| 17:02:01 | gibi | elod: +1d both -em patch, thanks for proposing them | |
| 17:07:43 | elod | gibi: well, the splitting is based on the 'team' tag in the yamls :] | |
| 17:10:04 | elod | https://opendev.org/openstack/releases/src/branch/master/deliverables/stein/placement.yaml#L5 | |
| 17:38:33 | openstackgerrit | Merged openstack/nova stable/stein: Sanity check instance mapping during scheduling https://review.opendev.org/756404 | |
| 17:49:24 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Add missing exception https://review.opendev.org/762898 | |
| 17:50:18 | openstackgerrit | melanie witt proposed openstack/nova stable/rocky: Sanity check instance mapping during scheduling https://review.opendev.org/762899 | |
| 17:53:02 | openstackgerrit | melanie witt proposed openstack/nova stable/queens: Sanity check instance mapping during scheduling https://review.opendev.org/762903 | |
| #openstack-nova - 2020-11-17 | |||
| 03:00:53 | xinranwang | gibi: Hi gibi, could you please review the smartnic spec when you got time? ;) | |
| 04:26:04 | openstackgerrit | Brin Zhang proposed openstack/nova master: Cyborg shelve/unshelve support https://review.opendev.org/729563 | |
| 07:58:18 | gibi | xinranwang: when I got time I will review :) | |
| 07:58:38 | gibi | I mean I'm trying to get to it | |
| 08:29:09 | xinranwang | gibi: thanks~ | |
| 09:08:19 | bauzas | good spec review day, Nova | |
| 10:00:18 | bauzas | gibi: easy peasy to review for a spec review day https://review.opendev.org/#/c/756242/ | |
| 10:43:19 | gibi | bauzas: done | |
| 10:54:18 | stephenfin | gibi, bauzas: I know it's spec review day, but could take a real quick look at https://review.opendev.org/#/c/762898/ while it's fresh in my mind? | |
| 10:54:25 | bauzas | gibi: cool, thanks | |
| 10:54:57 | bauzas | stephenfin: ack, I clicked for it, I'm just looking at a spec now | |
| 10:55:05 | stephenfin | all good, thanks :) | |
| 10:58:14 | openstackgerrit | Merged openstack/nova-specs master: Re-proposes Routed Networks https://review.opendev.org/756242 | |
| 11:09:59 | gibi | stephenfin: So the intention of this exception is to fail the startup of the compute service this why it is never caught? | |
| 11:10:33 | stephenfin | gibi: Yes, exactly. It's presence indicates a misconfiguration meaning we should hard fail | |
| 11:13:19 | stephenfin | tbc, we will only attempt to retrieve this information on the host if nova.conf tells us to | |
| 11:38:05 | gibi | stephenfin: thanks. Then my only question is about underscores :) https://review.opendev.org/#/c/762898/1/nova/tests/unit/virt/libvirt/test_driver.py@27505 | |
| 11:38:20 | sean-k-mooney | gibi: just reading the comment | |
| 11:38:28 | stephenfin | oh, there shouldn't be three /o\ | |
| 11:38:30 | sean-k-mooney | gibi: that is the style that stephenfin uses | |
| 11:38:58 | gibi | sean-k-mooney: I know he likes underscores, but I think there is one extra now | |
| 11:39:02 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Add missing exception https://review.opendev.org/762898 | |
| 11:39:04 | sean-k-mooney | yep | |
| 11:39:12 | stephenfin | done; my mistake /o\ | |
| 11:39:20 | sean-k-mooney | test___ instead of test__ | |
| 11:40:02 | sean-k-mooney | im still fine with this | |
| 11:40:25 | gibi | stephenfin: now you have one underscore, which is fine by me, but I guess your style would cause two underscores | |
| 11:40:52 | stephenfin | it would, but I know not everyone likes that so I'm okay to stick with one | |
| 11:41:29 | gibi | ok | |
| 11:41:47 | bauzas | sean-k-mooney: FWIW, I think I drafted a good nova-cyborg relationship in https://review.opendev.org/#/c/750116/9/specs/wallaby/approved/support-vGPU-nova-cyborg-interaction.rst@183 | |
| 11:42:01 | bauzas | based on the open issues we have atm | |
| 11:42:09 | bauzas | with the mdev persistence and the likes | |
| 11:42:16 | bauzas | this would hugely benefit to Cyboth | |
| 11:42:19 | bauzas | cyborg* | |
| 11:42:34 | bauzas | sean-k-mooney: anyway, lunch | |
| 11:42:41 | gibi | bauzas: does it have an impact on https://review.opendev.org/#/c/742785/6..9/specs/wallaby/approved/support-sriov-smartnic.rst too? | |
| 11:42:42 | openstackgerrit | Stephen Finucane proposed openstack/nova master: virt: Remove 'change_instance_metadata' API https://review.opendev.org/749316 | |
| 11:42:42 | openstackgerrit | Stephen Finucane proposed openstack/nova master: api-ref: Move 'os-agents' API to obsolete section https://review.opendev.org/755729 | |
| 11:42:43 | openstackgerrit | Stephen Finucane proposed openstack/nova master: virt: Remove 'get_all_bw_counters' API https://review.opendev.org/749312 | |
| 11:42:43 | openstackgerrit | Stephen Finucane proposed openstack/nova master: virt: Remove 'reset_network' API https://review.opendev.org/749315 | |
| 11:42:44 | openstackgerrit | Stephen Finucane proposed openstack/nova master: objects: Remove 'BandwidthUsage', 'BandwidthUsageList' https://review.opendev.org/759114 | |
| 11:42:55 | bauzas | gibi: I haven't reviewed this spec yet | |
| 11:43:05 | gibi | bauzas: nvm, I will read your comment in the vGPU spec | |
| 11:43:17 | bauzas | gibi: but tl;dr: I propose to leave cyborg manage the mdevs and just ask nova to bind them to the guest definition | |
| 11:43:32 | bauzas | which is sometimes we already have | |
| 11:43:42 | bauzas | s/sometimes/something | |
| 11:43:55 | sean-k-mooney | bauzas: ill go read it shortly | |
| 11:44:08 | bauzas | gibi: sean-k-mooney: the only difference would be the inventory reporting | |
| 11:44:41 | bauzas | but this way, this would allow cyborg to keep a persisted state of mdevs, which is something I don't wanna managed in nova and which creates problems for us | |
| 11:45:18 | bauzas | we haven't discussed this at the PTG, I reckon, but I feel we would all benefit of this solution | |
| 11:45:48 | bauzas | like, my customers would have the choice to either manage the mdev fleet by themselves and just use nova, or play with cyborg | |
| 11:46:09 | sean-k-mooney | bauzas im not sure they __need__ to precreate the mdev but i agree they could and it would ok to do so | |
| 11:46:30 | bauzas | sean-k-mooney: precreating is way better for many reasons | |
| 11:46:51 | bauzas | I mean, precreating after a config modification, that's it | |
| 11:46:53 | sean-k-mooney | i just read you last comment i need to read the rest of the doc and your comments for context | |
| 11:47:07 | bauzas | because once you're done with the config, your inventory won't change | |
| 11:47:09 | sean-k-mooney | i think it was on v2 the last time i looked | |
| 11:47:22 | bauzas | either way, I need to go lunching | |
| 11:47:32 | bauzas | my kids need to go to school | |
| 11:47:49 | sean-k-mooney | ya cyborg might want to be more dynmaic but i agree it has advantages | |
| 11:47:56 | sean-k-mooney | simplcity being one of them | |
| 12:08:10 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.binary_type/integer_types/string_types https://review.opendev.org/728094 | |
| 12:08:35 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.text_type (1/2) https://review.opendev.org/728109 | |
| 12:20:31 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.text_type (2/2) https://review.opendev.org/728117 | |
| 12:28:17 | openstackgerrit | Stephen Finucane proposed openstack/nova-specs master: Add modernize-os-hypervisors-api spec https://review.opendev.org/755109 | |
| 12:29:43 | Yumeng | sean-k-mooney, @gi | |
| 12:31:08 | Yumeng | good morning nova ^ | |
| 12:31:12 | sean-k-mooney | o/ | |
| 12:33:01 | Yumeng | hi sean, I wanna discuss some issues in the vGPU specs with you. https://review.opendev.org/#/c/750116/ | |