| *** takashin has joined #openstack-placement | 00:07 | |
| *** tetsuro has joined #openstack-placement | 00:20 | |
| openstackgerrit | Adam Spiers proposed openstack/os-traits master: Add HW_CPU_AMD_SEV trait https://review.openstack.org/635608 | 00:49 |
|---|---|---|
| *** mriedem_afk has quit IRC | 00:49 | |
| *** bhagyashris has joined #openstack-placement | 01:17 | |
| *** mriedem has joined #openstack-placement | 01:46 | |
| *** mriedem has quit IRC | 02:14 | |
| *** takashin has left #openstack-placement | 03:25 | |
| *** tetsuro has quit IRC | 04:07 | |
| *** tetsuro has joined #openstack-placement | 04:12 | |
| *** tetsuro has quit IRC | 06:07 | |
| *** e0ne has joined #openstack-placement | 06:50 | |
| *** tetsuro has joined #openstack-placement | 06:50 | |
| *** e0ne has quit IRC | 07:06 | |
| *** alex_xu has joined #openstack-placement | 08:05 | |
| *** rubasov has joined #openstack-placement | 08:08 | |
| gibi | efried: regarding mapping allocations to request groups here is the patch that adds the algorightm to that https://review.openstack.org/#/c/616239/22/nova/objects/request_spec.py@703 it is generic in nature but simplfied based on what neutron use in the request (only required traits and no aggregates) | 09:12 |
| *** e0ne has joined #openstack-placement | 09:20 | |
| *** tssurya has joined #openstack-placement | 09:25 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: Prepare in_tree allocation candidates https://review.openstack.org/635722 | 09:26 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: in_tree alloc_cands support with microversion 1.31 https://review.openstack.org/635723 | 09:26 |
| *** tetsuro has quit IRC | 09:27 | |
| *** mcgiggler has joined #openstack-placement | 09:30 | |
| *** bauzas is now known as bauwser | 09:39 | |
| *** bhagyashris has quit IRC | 09:47 | |
| *** david-lyle has joined #openstack-placement | 09:54 | |
| *** dklyle has quit IRC | 09:57 | |
| *** tssurya has quit IRC | 10:15 | |
| gibi | efried: and it is temporary in nova as I think this info should come from placement as per https://review.openstack.org/#/c/597601/ | 10:27 |
| *** stephenfin is now known as finucannot | 11:27 | |
| *** tssurya has joined #openstack-placement | 11:29 | |
| *** mcgiggler has quit IRC | 12:11 | |
| *** cdent has joined #openstack-placement | 12:11 | |
| openstackgerrit | Merged openstack/nova-specs master: AMD SEV: address final feedback received prior to merge https://review.openstack.org/628252 | 12:25 |
| *** e0ne has quit IRC | 12:29 | |
| openstackgerrit | Merged openstack/os-traits master: Add HW_CPU_AMD_SEV trait https://review.openstack.org/635608 | 12:56 |
| *** mcgiggler has joined #openstack-placement | 12:59 | |
| *** e0ne has joined #openstack-placement | 13:17 | |
| *** cdent has quit IRC | 13:23 | |
| *** mriedem has joined #openstack-placement | 13:31 | |
| *** cdent has joined #openstack-placement | 13:46 | |
| sean-k-mooney | cdent: o/ | 13:59 |
| cdent | ahoy | 14:00 |
| sean-k-mooney | cdent: so i found the issue | 14:00 |
| cdent | yeah? | 14:00 |
| sean-k-mooney | [Fri Feb 08 13:57:44.366386 2019] [proxy:error] [pid 8825:tid 140636590126848] (2)No such file or directory: AH02454: uwsgi: attempt to connect to Unix domain socket /var/run/uwsgi/nova-placement-api.socket (uwsgi-uds-nova-placement-api) failed | 14:00 |
| sean-k-mooney | the socket is called placement-api.socket now | 14:00 |
| sean-k-mooney | so apache is still configured to use the intree placement socket | 14:01 |
| sean-k-mooney | at least that is my assumtion as to why this has been renamed | 14:01 |
| cdent | oh, one sec | 14:01 |
| cdent | nm, was remembering something lese | 14:02 |
| cdent | that continues to sound like an issue with "old" apache configuration being reused rather than erased and replaced | 14:02 |
| sean-k-mooney | i uninstalled apache adn rebooted | 14:03 |
| sean-k-mooney | i did not nuke the apache etc dir however | 14:03 |
| sean-k-mooney | i can try that | 14:03 |
| cdent | i think it's the config handling in devestack is just not as good as it should be | 14:04 |
| cdent | but we don't encounter that in the gates because it is all from scratch | 14:05 |
| sean-k-mooney | if this fixes it i will look at the devstack code and see if i can guard against this | 14:07 |
| sean-k-mooney | when did the devstack change to use extracted placement go in? the devstack i was useing previouslwas form earlay january or late december | 14:08 |
| sean-k-mooney | so its likely im only hitting this sice i did a git pull of my devstack code | 14:09 |
| * cdent checks | 14:09 | |
| cdent | 2018-11-27 | 14:10 |
| sean-k-mooney | ok i dont think it was that old but that siad i dont update devstack unless i hit issues stacking so maybe | 14:10 |
| sean-k-mooney | anyway ill let you know if this fixes it | 14:11 |
| cdent | \o/ | 14:11 |
| sean-k-mooney | i said if :) but so far so good ill know in about 5 mins | 14:13 |
| cdent | I was merely expressing my happiness that someone instead of me was looking into it :) | 14:13 |
| sean-k-mooney | ah ok well untill i fix it my sriov dev env is broken and i dont really want to do are reinstall of the os. i coudl but i dont really want to so debuging devstack it is | 14:15 |
| sean-k-mooney | yep nuking the directories fixed it now to go do that in devstack or otherwise force the config to be updated. | 14:25 |
| sean-k-mooney | cdent: want me to add you as a review to the patch when i push it? | 14:25 |
| cdent | yes please, thanks very much | 14:26 |
| efried | gibi: Thanks. I hope we can get that sucker done in Train. | 14:34 |
| *** efried is now known as fried_rice | 14:35 | |
| sean-k-mooney | efried: bandwith schuleing? or something else | 14:35 |
| fried_rice | sean-k-mooney: having alloc cand response provide information associating request group with allocation RP | 14:35 |
| fried_rice | so we don't need to repeat placement logic on the client side. | 14:36 |
| fried_rice | sean-k-mooney: https://review.openstack.org/#/c/597601/ | 14:36 |
| sean-k-mooney | ah ok that will be needed for both cyborg and the bandwith stuff | 14:36 |
| fried_rice | correctamundo | 14:36 |
| sean-k-mooney | gibi currenly is implementing it on the nova side right so ye are going to modify the api respoce to just provide it im assuming. | 14:37 |
| sean-k-mooney | ill add the spec to my list | 14:37 |
| sean-k-mooney | ah yes https://review.openstack.org/#/c/597601/1/specs/stein/approved/placement-resource-provider-request-group-mapping-in-allocation-candidates.rst@426 is more or less what i was expecting | 14:39 |
| *** cdent has quit IRC | 15:31 | |
| *** tssurya has quit IRC | 15:52 | |
| *** cdent has joined #openstack-placement | 16:04 | |
| *** e0ne has quit IRC | 16:17 | |
| *** rubasov has quit IRC | 16:24 | |
| mnaser | https://review.openstack.org/#/c/618820/ | 16:30 |
| *** mcgiggler has quit IRC | 17:06 | |
| *** jaypipes is now known as leakypipes | 17:43 | |
| mnaser | cdent: the placement role landed today as well so i hope to try to find sometime to integrate it soon | 18:00 |
| cdent | woot! | 18:01 |
| *** cdent has quit IRC | 19:14 | |
| *** mriedem has quit IRC | 19:27 | |
| *** mriedem has joined #openstack-placement | 19:27 | |
| *** e0ne has joined #openstack-placement | 20:29 | |
| *** e0ne has quit IRC | 21:06 | |
| *** e0ne has joined #openstack-placement | 21:11 | |
| *** e0ne has quit IRC | 21:24 | |
| *** mriedem has quit IRC | 23:29 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!