Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-07-20
20:59:14 melwitt nova meeting in one minute
20:59:28 mriedem is the end of voltron better then the end of seinfeld?
20:59:30 mriedem because that's not hard
20:59:41 jangutter mriedem: it's a 4 minute youtube video.
20:59:59 mriedem no time
21:00:07 mriedem that's 3:45 too long
21:02:09 openstackgerrit Matt Riedemann proposed openstack/nova master: Cleanup unnecessary logic in os-volume_attachments controller code https://review.openstack.org/485823
21:03:27 openstackgerrit melanie witt proposed openstack/nova master: Make Quotas object favor the API database https://review.openstack.org/410945
21:04:24 openstackgerrit melanie witt proposed openstack/nova master: Add online migration to move quotas to API database https://review.openstack.org/410946
21:22:28 cfriesen If anyone wants a short break from what they're doing, I'd welcome some additional reviewers for https://review.openstack.org/#/c/468203 and https://review.openstack.org/#/c/461456 . Sahid doesn't really like the proposed tweaks to cpu_realtime_mask but sfinucan is fine with it...would be useful to get a few more opinions.
21:45:20 melwitt dansmith: I removed the remotables from the Quotas object patch. those methods are only called in the quota classes API
21:45:30 melwitt thanks for pointing that out
21:45:39 dansmith melwitt: oh sweet
21:45:58 melwitt I must have had remotable fever when I first added them
21:46:12 dansmith a common affliction
21:46:34 melwitt haha, yup!!
21:47:03 mriedem cfriesen: so you mean, take a break from reviewing code to review some code?
21:47:40 cfriesen mriedem: but it's different code!
21:47:46 mriedem OH!
21:48:01 dansmith less critical for the release code?
21:48:03 dansmith sign me up.
21:48:13 cfriesen less stressful, more laid-back
21:48:14 mriedem cfriesen: did you miss this while you were on vacation?
21:48:21 mriedem this abuse
21:48:43 cfriesen all the time. had to go search out random strangers
21:49:16 mriedem random...canadian strangers?
21:49:22 cfriesen no, they're too polite
21:49:25 mriedem right
21:50:13 cfriesen so are multi-cells actually going to be completely usable in pike? anything that needs some reviews to make sure that happens?
21:50:57 mriedem the last big thing is moving quotas to the api
21:51:06 mriedem starts here https://review.openstack.org/#/c/410945/
21:52:01 mriedem cfriesen: and this doesn't have an owner https://bugs.launchpad.net/nova/+bug/1702935
21:52:02 openstack Launchpad bug 1702935 in OpenStack Compute (nova) "Unshelving an offloaded server does not restrict to the original cell" [High,Confirmed]
21:52:15 mriedem otherwise i
21:52:37 mriedem otherwise i'd like to see someone doing some performance testing, maybe with rally, comparing stable/ocata to current master
21:53:06 mriedem now that we have the counting quotas stuff in and some changes to the filter scheduler for using placement
21:53:21 mriedem really just try to break things
21:53:50 cfriesen our arch guys have decided to skip ocata and jump from newton right to pike (yeah, I know), so I won't be able to compare against ocata. We should never have made kilo->mitaka work.
21:54:56 melwitt at least you're upgrading? :P
21:55:08 cfriesen there is that
21:55:53 melwitt skipping releases the right way just means you pull the middle code and just run migrations, then move to the next I thought. I would think that shouldn't be too painful or is there more to it?
21:55:56 cfriesen has anyone looked at allowable latency between the API cell and the compute cells?
21:56:25 cfriesen melwitt: we're going to have computes running newton live migrating to computes running pike
21:56:42 melwitt oh, live, okay
21:56:53 cfriesen I expect, anyway. since I think we'll probably be doing an OS upgrade too
21:57:01 melwitt cfriesen: what do you mean by latency? there's no syncing of data back and forth so there's no latency in that sense
21:57:53 dansmith the latency requirements are exactly the same as the requirements today between your api services and the database server
21:58:06 cfriesen melwitt: I mean ping time basically. given the architecture I wouldn't expect it to be all that sensitive
21:58:18 dansmith i.e. "don't use this for geographically distributed stuff"
21:58:27 cfriesen dansmith: that's exactly what I'm wondering about. :)
22:02:15 openstackgerrit Ildiko Vancsa proposed openstack/nova master: Remove check_detach https://review.openstack.org/446671
22:02:16 openstackgerrit Ildiko Vancsa proposed openstack/nova master: Implement new attach Cinder flow https://review.openstack.org/330285
22:04:15 ildikov mriedem: VolumeUnattached removed ^^
22:23:56 openstackgerrit Alex Szarka proposed openstack/nova master: Transform instance.exists notification https://review.openstack.org/403660
22:27:14 mriedem ack, will review later
22:57:01 imacdonn think I found a little doc-bug ... https://docs.openstack.org/ocata/config-reference/compute/config-options.html talks about the reserved_host_cpus option, but https://review.openstack.org/#/c/460113/ is only in master (post ocata-release)
23:15:57 openstackgerrit Jay Pipes proposed openstack/nova master: claim resources in placement API during schedule() https://review.openstack.org/483566
23:16:13 jaypipes mriedem, dansmith, edleafe: k, fixed up comment from review ^
23:24:22 openstackgerrit Jay Pipes proposed openstack/nova master: claim resources in placement API during schedule() https://review.openstack.org/483566
23:24:23 jaypipes forgot to git add the releasenote...
23:24:24 jaypipes done.
23:57:52 mriedem imacdonn: hmm, weird
23:58:10 mriedem yeah, i'm not really sure how those config docs are generated
23:58:14 mriedem stephenfin: ^?
23:58:31 imacdonn heh, yeah, I was just going to say the same thing (about the process for generating docs)
23:59:00 mriedem i'm asking in #openstack-doc
23:59:07 imacdonn ok
23:59:09 mriedem those are all getting moved into nova now rather than openstack-manuals
23:59:15 mriedem but stephenfin has been the one working on it
#openstack-nova - 2017-07-21
00:09:54 openstackgerrit Monty Taylor proposed openstack/python-novaclient master: WIP Use keystoneauth for version discovery https://review.openstack.org/485868
00:13:15 openstackgerrit Merged openstack/nova master: Fix and optimize external_events for multiple cells https://review.openstack.org/445142
00:17:33 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/485634
00:20:16 openstackgerrit Felipe Monteiro proposed openstack/nova master: Update policy description for 'instance_actions' https://review.openstack.org/484609
00:25:44 trinaths1 sean-k-mooney: ping
00:26:45 trinaths1 sean-k-mooney: http://paste.openstack.org/show/616099/ it went well. but there was a iptables issues.
00:26:52 openstackgerrit Steve Noyes proposed openstack/nova master: update live migration to use v3 cinder api https://review.openstack.org/463987
00:39:18 openstackgerrit melanie witt proposed openstack/nova master: Request zero root disk for boot-from-volume instances https://review.openstack.org/428481
00:39:19 openstackgerrit melanie witt proposed openstack/nova master: Claim and report zero root disk for boot-from-volume instances https://review.openstack.org/428505
01:26:46 openstackgerrit Matt Riedemann proposed openstack/nova master: Cleanup unnecessary logic in os-volume_attachments controller code https://review.openstack.org/485823
01:29:59 mriedem need a +W on the change to remove check_detach from the code https://review.openstack.org/#/c/446671/
01:39:03 openstackgerrit Tony Breeds proposed openstack/nova master: Move the last_bytes util method to the libvirt driver. https://review.openstack.org/472228
01:39:03 openstackgerrit Tony Breeds proposed openstack/nova master: Avoid chowning console logs in libvirt https://review.openstack.org/472229
01:39:04 openstackgerrit Tony Breeds proposed openstack/nova master: Move libvirt usages of chown to privsep. https://review.openstack.org/471972
01:39:04 openstackgerrit Tony Breeds proposed openstack/nova master: First attempt at adding a privsep user to nova itself. https://review.openstack.org/459166
01:39:05 openstackgerrit Tony Breeds proposed openstack/nova master: Decpouple the os-brick and nova privsep modules. https://review.openstack.org/485908
01:39:05 openstackgerrit Tony Breeds proposed openstack/nova master: Do not import nova.conf into nova/exception.py https://review.openstack.org/485907
01:39:57 mriedem holy f it's a wild tonyb
01:40:10 tonyb :D
01:40:21 tonyb mriedem: You can blame mikal
01:40:41 mriedem "no one is reviewing my privsep thing so you do it and maybe they'll help"
01:40:42 mriedem like that?
01:41:23 tonyb More like "why does my privsep stuff fail in the gate, can you help me replicate it"
01:41:33 tonyb mriedem: so I did that thing and then started playing
01:41:56 mriedem oh i see
01:42:41 mriedem maybe document that somewhere sometime
01:42:44 mriedem since no one else knows how
01:43:04 tonyb mriedem: If I could work out what to document sure.
01:44:29 trinaths1 sean-k-mooney: ping
01:46:19 mriedem tonyb: FAQ?
01:46:28 mriedem when you see x, what does it mean - this is what i found out!

Earlier   Later