| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-06-15 | |||
| 12:31:18 | cdent | bauzas, alex_xu: some helpful increased test coverage from rfolco, should be an easy merge: https://review.openstack.org/#/c/473479/ | |
| 12:33:00 | moshele | sfinucan: I just don't want to miss the pike time frame | |
| 12:33:53 | sfinucan | moshele: That's fair. I will, however, defer to sean-k-mooney's judgement, as he was the one that suggested waiting in the first place | |
| 12:34:26 | moshele | sfinucan: Ok I will try to ping him late today | |
| 12:38:50 | moshele | sfinucan: by the way do you know when OVS 2.8 should be release? | |
| 12:39:30 | sfinucan | moshele: iirc, OVS is tied to the OpenStack release cycle now so it should be some time shortly before Pike | |
| 12:39:49 | sfinucan | #openvswitch or the mailing lists might be the place to ask for sure though | |
| 12:42:27 | edleafe | alex_xu: I'm working on adding traits to the resource provider selection in placement. Are you or your team doing anything related? | |
| 12:44:36 | moshele | sean-k-mooney: hi | |
| 12:54:12 | openstackgerrit | Merged openstack/nova master: Centralize compute_node_search_by_hypervisor in os-hypervisors https://review.openstack.org/474381 | |
| 13:09:24 | alex_xu | edleafe: I only have this old poc https://review.openstack.org/429364, I think you can free to go that, and I'm happy if you need any help | |
| 13:10:00 | alex_xu | edleafe: are you talk about the resource provider selection based on this https://review.openstack.org/#/c/471927/? | |
| 13:10:24 | frickler | could someone pick up on https://launchpad.net/bugs/1697452 ? this is getting a FAQ in various channels | |
| 13:10:25 | openstack | Launchpad bug 1697452 in python-novaclient "openstack client dumps debug info on console even without specifying --debug option" [Undecided,Confirmed] | |
| 13:12:09 | mriedem | sdague: are you ok with this bug fix spec from alex? https://review.openstack.org/#/c/468619/ | |
| 13:12:16 | edleafe | alex_xu: yes, that's the spec. I think I've already got most of what was in your POC, but I may end up using some of yours, too | |
| 13:12:19 | sdague | mriedem: looking | |
| 13:13:02 | sdague | mriedem: yes | |
| 13:13:15 | alex_xu | edleafe: ok, cool, maybe add me as co-author, if you get something useful from my poc :) | |
| 13:13:41 | edleafe | alex_xu: sure! | |
| 13:13:50 | alex_xu | edleafe: cheers! | |
| 13:16:28 | mriedem | alex_xu: gmann: https://blueprints.launchpad.net/nova/+spec/fix-quota-classes-api is approved now | |
| 13:17:03 | alex_xu | cdent: +2 for that patch | |
| 13:17:21 | alex_xu | mriedem: sdague thanks! | |
| 13:17:27 | openstackgerrit | Merged openstack/nova-specs master: Fix to return server groups quota in the quota-classes API https://review.openstack.org/468619 | |
| 13:18:52 | cdent | thanks alex_xu | |
| 13:19:23 | alex_xu | cdent: np | |
| 13:20:28 | mriedem | sfinucan: do you still have a change with this one? https://review.openstack.org/#/c/472622/ | |
| 13:20:35 | mriedem | s/change/issue/ | |
| 13:21:13 | sfinucan | mriedem: I still don't like moving that, but it's a nit and I won't block on it | |
| 13:21:16 | sfinucan | switched to +2 | |
| 13:23:41 | mriedem | sfinucan: ok, i think the NETWORK_ATTACH_EXTERNAL one is wrong | |
| 13:24:06 | sfinucan | i.e. it should stay up top | |
| 13:24:07 | sfinucan | ? | |
| 13:24:20 | mriedem | no, the description, i'm commenting | |
| 13:25:17 | mriedem | https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L464 | |
| 13:26:48 | mriedem | commented | |
| 13:27:01 | mriedem | that rule doesn't have anything to do with os-server-external-events | |
| 13:27:11 | mriedem | it's a check to see that only admins can boot into an unshared external network | |
| 13:27:19 | mriedem | happens on the compute side, so it's really late and odd | |
| 13:28:45 | mriedem | http://lists.openstack.org/pipermail/openstack-dev/2016-June/096919.html | |
| 13:34:36 | gibi | mriedem: hi! I fixed your yesterday's comment on the snapshot_id notification patch https://review.openstack.org/#/c/453077/ | |
| 13:35:08 | mriedem | gibi: ok will look shortly | |
| 13:35:41 | gibi | mriedem: thanks | |
| 13:38:53 | openstackgerrit | Eric Berglund proposed openstack/nova master: Add PowerVM to nova support matrix https://review.openstack.org/470999 | |
| 13:45:20 | openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/os-vif master: Add 'datapath_type' field to OVS related objects https://review.openstack.org/474581 | |
| 13:47:33 | mriedem | gibi: +2 | |
| 13:51:33 | gibi | mriedem: thanks | |
| 13:53:00 | openstackgerrit | Merged openstack/nova master: [placement] Use util.extract_json in allocations handler https://review.openstack.org/471067 | |
| 13:53:54 | frickler | mriedem: did you see my comment about the novaclient bug earlier? would be good if someone could at least triage it, if not fix it. sadly no time to look deeper myself | |
| 13:54:12 | mriedem | frickler: i didn't, wasn't around | |
| 13:54:19 | mriedem | link? | |
| 13:54:50 | openstack | Launchpad bug 1697452 in python-novaclient "openstack client dumps debug info on console even without specifying --debug option" [Undecided,Confirmed] | |
| 13:54:50 | frickler | mriedem: https://launchpad.net/bugs/1697452 | |
| 13:55:43 | mriedem | frickler: when you say volume commands in there, you mean osc volume commands? | |
| 13:55:47 | mriedem | b/c those would be cinderclient | |
| 13:55:54 | mriedem | which would make me think it's a shared common library issue | |
| 13:56:03 | mriedem | nova meeting in 4 minutes | |
| 13:56:34 | frickler | mriedem: no, "volume show" does a "server show" internally in order to display attachments. so its really only novaclient=9.0.0 being broken | |
| 13:57:24 | mriedem | huh, well that seems totally wasteful and redundant since cinder's volume api shows attachments within it's own response | |
| 13:57:25 | mriedem | but ok | |
| 13:57:55 | mriedem | https://developer.openstack.org/api-ref/block-storage/v2/?expanded=show-volume-details-detail#show-volume-details | |
| 13:58:26 | bauzas | cdent: quick question, why https://review.openstack.org/#/c/473479/1/nova/tests/functional/api/openstack/placement/gabbits/inventory.yaml@654 is having generation by 8 ? | |
| 13:58:41 | bauzas | cdent: because we delete ? | |
| 13:58:44 | bauzas | oh, nvm | |
| 13:59:36 | frickler | mriedem: anyway, main issue is getting debug output on any nova related command like "server show" or "flavor list" | |
| 13:59:52 | mriedem | yeah, will have to do a diff between 8.0.0 and 9.0.0 | |
| 13:59:56 | mriedem | about to start a meeting though | |
| 14:04:21 | cdent | bauzas: you figured it out? | |
| 14:04:45 | bauzas | cdent: yup, because since we delete, we need to increment the generation | |
| 14:16:24 | openstackgerrit | Baodong (Robert) Li proposed openstack/nova-specs master: Expose vlan trunking in metadata/configdrive https://review.openstack.org/471815 | |
| 14:22:52 | frickler | mriedem: sdague: reverting https://review.openstack.org/463508 fixes 1697452 for me, but there might be a better solution than that | |
| 14:25:05 | sdague | frickler: right, so https://review.openstack.org/#/c/463508/5/novaclient/v2/client.py is the bit that's wrong? | |
| 14:25:29 | sdague | frickler: can you propose a revert of just that part of the patch? | |
| 14:26:04 | sdague | frickler: or would you like me to do it? | |
| 14:26:08 | sfinucan | mriedem: Went for tea when you were discussing https://review.openstack.org/#/c/472275/ | |
| 14:27:03 | frickler | sdague: I'm not sure whether the rest of the patch will break then. maybe rather do a complete revert and then work on a better fix. anyway, please do as you like, I'm off for today now | |
| 14:27:16 | sdague | frickler: ok, I'll play with it | |
| 14:27:41 | sfinucan | I've been talking to asettle about that. Happy to "own" it if there's work to be done moving stuff into tree | |
| 14:27:47 | mriedem | sdague: +1 to just https://review.openstack.org/#/c/463508/5/novaclient/v2/migrations.py | |
| 14:27:56 | mriedem | sdague: warnings.warn is what we've always done in the python api binding code | |
| 14:28:01 | sdague | mriedem: yeh | |
| 14:28:07 | sdague | it was just too aggressive | |
| 14:28:08 | mriedem | sfinucan: i'd love someone to be a point person on that docs transition | |
| 14:28:13 | sdague | I'm going to do a partial revert | |
| 14:28:24 | mriedem | sdague: ack, this is the bug https://bugs.launchpad.net/python-openstackclient/+bug/1697452 | |
| 14:28:24 | openstack | Launchpad bug 1697452 in python-novaclient "openstack client dumps debug info on console even without specifying --debug option" [Undecided,Confirmed] | |
| 14:29:10 | mriedem | frickler: thanks for bringing it up | |
| 14:29:33 | sfinucan | mriedem: Cool. I'm your guy, in that case. Assuming nothing is happening til Queens anyway | |
| 14:29:37 | mriedem | sfinucan: nicolasbock is also docs CPL so you can double team | |
| 14:29:46 | sfinucan | seeing as the spec need to be approved | |
| 14:29:53 | sfinucan | Sounds good | |
| 14:30:04 | mriedem | btw, does double-team translate in irish? | |
| 14:30:14 | mriedem | https://en.wikipedia.org/wiki/Professional_wrestling_double-team_maneuvers | |
| 14:30:22 | sfinucan | tag-team, I guess? | |
| 14:30:35 | sfinucan | or just "team up" | |
| 14:31:10 | openstackgerrit | Sean Dague proposed openstack/python-novaclient master: Revert "client.logger.warning wrongly used in migrations" https://review.openstack.org/474600 | |
| 14:31:34 | sdague | mriedem: ^^^ | |
| 14:31:46 | nicolasbock | Hi sfinucan | |
| 14:31:53 | sdague | oh, crap, I did the wrong revert direction | |