Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-03-15
14:25:00 stephenfin 🎉
14:25:11 mriedem is that a lego brick? a touch tone phone?
14:25:27 dansmith party horn?
14:25:31 stephenfin https://emojipedia.org/party-popper/
14:25:46 mriedem oh, doesn't render for me
14:25:49 mriedem pidgin
14:25:57 stephenfin Windows?
14:26:01 mriedem yeah
14:26:47 stephenfin But yeah, assign me to that. Soon as I get my server back, I can take a look
14:26:54 stephenfin I'm on spec review duty all day today
14:33:52 mriedem test patch https://review.openstack.org/553412
14:33:56 mriedem mgoddard_: ^
14:35:33 mgoddard_ mriedem: nice, I'll keep an eye on it
14:40:15 openstackgerrit Andrey Volkov proposed openstack/osc-placement master: Get resource provider by uuid or name https://review.openstack.org/527791
14:41:43 openstackgerrit Stephen Finucane proposed openstack/nova-specs master: Few correction in the server filter/sort spec https://review.openstack.org/527019
14:44:29 Spaz-Home dansmith, I confirmed either Jianghua or Niachuan will be at the meeting later today, sir. You should be good to go on reviewing that spec with them.
14:44:50 Spaz-Home Time for me to head to bed.. have a good day folks.
14:45:16 dansmith Spaz-Home: niachuan asked us to do the meeting for him because he can't join
14:45:18 dansmith which I will
14:45:22 dansmith but yeah
14:45:25 Spaz-Home Ahh that's unfortunate
14:45:27 Spaz-Home It's hard on our hours
14:47:32 mriedem bhagyashri_s: i've abandoned your spec since it looks, well, abandoned https://review.openstack.org/#/c/511825/
14:47:51 mriedem needs to be moved to the rocky directory structure and incorporate the direction from the PTG
14:51:45 mriedem simondodsley: are you going to move this to rocky? https://review.openstack.org/#/c/340168/ it's proposed against pike
14:54:06 mriedem sean-k-mooney: does anyone care about this anymore? https://review.openstack.org/#/c/471815/
14:54:20 mriedem i don't know if baoli is around anymore re: cisco
14:54:40 mriedem mgagne: ^ seems like something that would be up your alley
14:56:03 mriedem or vdrok ^ i.e. vlan trunk port metadata bliss
14:56:44 openstackgerrit Nguyen Hai proposed openstack/python-novaclient master: Fix local test fails with pypy https://review.openstack.org/553426
14:57:24 openstackgerrit Nguyen Hai proposed openstack/python-novaclient master: Fix local test fails with pypy https://review.openstack.org/553426
15:00:00 openstackgerrit Surya Seetharaman proposed openstack/nova master: Add disabled field to CellMapping object https://review.openstack.org/550090
15:00:20 mgagne mriedem: thanks for the heads-up
15:00:22 dansmith mriedem: comments in that rebuild-bfv spec
15:00:27 mriedem mgagne: do you still plan on pushing this spec? https://review.openstack.org/#/c/312626/ it's targeted at pike and would need to be moved to rocky and address reviewer feedback
15:00:49 dansmith mriedem: I might be the only one, but doing a create, delete and re-pointing at a new volume seems like not a great experience to me
15:01:58 efried I assume (because I've not seen it done) that it's Wrong to point to specs from comments/docstrings.
15:03:26 efried Reason I ask: http://specs.openstack.org/openstack/nova-specs/specs/rocky/approved/update-provider-tree.html#computedriver-update-provider-tree contains a lot more information than I want to put into https://github.com/openstack/nova/blob/master/nova/virt/driver.py#L833
15:03:40 efried But that information sure would be useful for people implementing that method.
15:03:59 mriedem efried: just say "see the update_provider_tree spec for more details"
15:04:01 mriedem or something
15:04:21 efried Can I say "rocky" too?
15:04:33 efried That'd give them enough info to find it in one shot.
15:04:36 mriedem sure
15:04:46 mriedem or, move the dev details of that spec into devref
15:05:16 mriedem https://docs.openstack.org/nova/latest/reference/index.html
15:06:57 dansmith mriedem: tested by a real live person: https://review.openstack.org/#/c/552691
15:06:59 dansmith per your ask
15:08:04 openstackgerrit Surya Seetharaman proposed openstack/nova master: Add disabled field to CellMapping object https://review.openstack.org/550090
15:08:51 openstackgerrit Claudiu Belu proposed openstack/nova-specs master: Adds spec for instance live resize https://review.openstack.org/141219
15:08:51 efried mriedem: And then can I link to the devref, or do I still have to say, "See the [name] section of the devref for details" ?
15:12:23 mriedem efried: idk, choose your own adventure
15:12:48 mriedem linking to a devref is probably ok, the spec URL is long as hell
15:16:15 mriedem dansmith: agree the create/delete volume thing sucks
15:16:27 mriedem don't know how easy it would be for cinder to add an api to re-image a volume
15:16:30 mriedem smcginnis: ^?
15:17:23 smcginnis So detach, re-lay down an image on the vol, reattach?
15:17:27 openstackgerrit Merged openstack/python-novaclient master: Updated from global requirements https://review.openstack.org/553255
15:17:46 mriedem not sure if we'd need the detach/re-attach part
15:17:50 dansmith smcginnis: yeah
15:17:51 mriedem but that's what happens during rebuild anyway yeah
15:17:54 dansmith mriedem: I assume we would for some backends
15:17:55 mdbooth mriedem: Would you mind taking a look at https://review.openstack.org/#/c/551302/ ? I think it's your bailiwick.
15:18:34 mriedem is a bailiwick related to a throat sweet?
15:18:51 smcginnis mriedem: I think we would need the detach reattach.
15:19:03 smcginnis We need to attach it to a cinder node to reimage it.
15:19:04 mriedem smcginnis: that's fine probably b/c we do it anyway for rebuild
15:19:14 smcginnis Unless nova wants to take care of that part and leave cinder out of it.
15:19:19 hrw morning
15:19:22 smcginnis Basically just duplicate what we do in cinder.
15:19:41 smcginnis Pull the image to the compute node, write it over the volume, then start up the instance again.
15:19:43 mriedem it seems like nova doing things with the internals of the volume is a bad idea
15:19:49 mdbooth mriedem: Not especially :) I think there's still a literal bailiwick in the channel islands.
15:20:04 dansmith mriedem: I dunno, it's no different than a regular disk image,
15:20:04 hrw can someone help me with understanding how nova compares cpus on live migration?
15:20:08 openstackgerrit Mathieu Gagné proposed openstack/nova-specs master: Multiple Fixed-IPs support in network information https://review.openstack.org/312626
15:20:13 smcginnis mriedem: I agree, but it would be an "optimization" to not have to tear things down and start over.
15:20:19 dansmith mriedem: the thing is I would expect cinder to maybe be able to do it more efficiently in some cases (i.e. ceph)
15:20:29 smcginnis Especially as I'm not sure how that would work with a boot volume. We can't detach those normally, right.
15:20:42 dansmith smcginnis: we're talking about bfv here
15:20:50 dansmith smcginnis: we're doing a rebuild, so we'd detach
15:20:52 smcginnis dansmith: Yeah, so that would be an extra complication.
15:20:59 smcginnis dansmith: Oh, right!
15:21:00 openstackgerrit Mathieu Gagné proposed openstack/nova-specs master: Multiple Fixed-IPs support in network information https://review.openstack.org/312626
15:21:29 mriedem the volume would remain in-use during all of this from a state pov
15:21:30 smcginnis So delete instance, detach volume, cinder rewrites it, reattach, create new "rebuilt" instance using it.
15:21:56 mdbooth dansmith: Could also be a multi-attach volume, btw.
15:22:01 openstackgerrit Surya Seetharaman proposed openstack/nova master: [WIP] Add CellMappingList.get_all_enabled() query method https://review.openstack.org/550188
15:22:18 mriedem this is the code that detaches the volume today for rebuild https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L3072
15:22:20 dansmith mdbooth: if people are multi-attaching bfv then I give up
15:22:33 mriedem we create a 2nd attachment to keep the volume 'in-use' and then delete the first attachment
15:22:40 dansmith well, unless it's read only I guess
15:22:45 mdbooth dansmith: It's a thing with real hardware.
15:23:03 dansmith but yeah, if we can't undo all the attachments, we'd have to fail I think
15:23:38 mriedem before we spawn the guest again, we 'attach' again https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L3103
15:23:47 hrw https://bugs.launchpad.net/nova/+bug/1756118 reported
15:23:48 openstack Launchpad bug 1756118 in OpenStack Compute (nova) "No live migration on aarch64" [Undecided,New]
15:23:55 mdbooth ... unless the volume was attached RO
15:24:00 openstackgerrit Jianghua Wang proposed openstack/os-traits master: GPU: define traits for maximum display heads https://review.openstack.org/553277
15:24:44 mriedem hrw: "nova doesn't support operation x for an architecture that no one started caring about nova supporting until recently" isn't a bug :)

Earlier   Later