| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-01-31 | |||
| 17:53:59 | openstackgerrit | Dan Smith proposed openstack/nova master: Make host_manager use scatter-gather and ignore down cells https://review.openstack.org/539617 | |
| 17:54:01 | dansmith | tssurya: they haven't started running tests yet, so ^ | |
| 17:54:03 | dansmith | tssurya: thanks! | |
| 18:01:41 | openstackgerrit | Elod Illes proposed openstack/nova master: Functional test: cold migrate to compute down https://review.openstack.org/496280 | |
| 18:27:09 | mriedem | esberglu: can you post a patch that adds powervm to the hpc feature support matrix page? https://docs.openstack.org/nova/latest/user/feature-classification.html#matrix-hpc | |
| 18:29:18 | mdbooth | dansmith: jaypipes Thanks. NP holding off for queens either. | |
| 18:30:10 | mriedem | kashyap: re http://logs.openstack.org/66/539266/2/check/build-openstack-sphinx-docs/f7a1712/html/admin/virtual-gpu.html#how-to-know-a-gpu-type where it says, "Note that the current libvirt API doesn’t provide a way to access that information." | |
| 18:30:16 | mriedem | is that true at least for libvirt 4.0.0? | |
| 18:30:18 | mriedem | which is the latest release? | |
| 18:30:29 | mdbooth | dansmith: I responded to your comments and I'm expecting to make some changes. Given that it's not going to land in Queens it's hardly urgent, but just in case you wanted to respond while you still have context... I'm looking for further feedback on which changes are really required. | |
| 18:34:17 | mriedem | efried_rollin: the in-tree powervm driver doesn't support gpu passthrough right? | |
| 18:49:22 | mriedem | esberglu: nvm i've just about got this | |
| 18:52:45 | openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: update the GPU passthrough HPC feature entry https://review.openstack.org/539638 | |
| 19:02:31 | mriedem | artom: we don't throw exceptions in python! | |
| 19:02:33 | mriedem | :P | |
| 19:02:44 | artom | mriedem, eh? | |
| 19:02:54 | mriedem | https://review.openstack.org/#/c/539266/2/doc/source/admin/virtual-gpu.rst@116 | |
| 19:03:05 | artom | I said something stupid, didn't I? | |
| 19:03:23 | artom | "Libvirt will throw an exception" | |
| 19:03:26 | artom | Yep :( | |
| 19:09:00 | artom | Oh wait, you mean we raise, not throw. | |
| 19:09:05 | artom | Yeah, that too. | |
| 19:09:13 | artom | It's not my fault they taught us Java :( | |
| 19:13:10 | melwitt | well, libvirt itself throws because it's C right :P | |
| 19:13:37 | dansmith | C doesn't throw | |
| 19:14:07 | dansmith | it returns -1 and sets ERRNO | |
| 19:14:25 | melwitt | oh -_- guess I was thinking of C++ | |
| 19:14:42 | dansmith | *gasp* | |
| 19:15:09 | openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: update the GPU passthrough HPC feature entry https://review.openstack.org/539638 | |
| 19:15:10 | mriedem | artom: i've tried to address your comments ^ but there were a lot of them... | |
| 19:15:10 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Provide support matrix and doc for VGPU https://review.openstack.org/539266 | |
| 19:15:22 | mriedem | and i'm just helping a brotha out | |
| 19:17:06 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Provide support matrix and doc for VGPU https://review.openstack.org/539266 | |
| 19:17:06 | openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: update the GPU passthrough HPC feature entry https://review.openstack.org/539638 | |
| 19:17:32 | artom | mriedem, aha, cheers! | |
| 19:19:25 | mriedem | and because you overloaded me with comments, i screwed something up | |
| 19:20:52 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Provide support matrix and doc for VGPU https://review.openstack.org/539266 | |
| 19:21:02 | artom | Yeah, I think rescue and resize got merged/switched/borked | |
| 19:22:36 | mriedem | fixed in the latest revision | |
| 19:22:48 | mriedem | i think it's ready for canadian-style scrutiny | |
| 19:23:02 | artom | Quebec, *ahem* | |
| 19:23:08 | mriedem | that's not canada? | |
| 19:23:13 | mriedem | french canada | |
| 19:23:14 | mriedem | but still | |
| 19:23:15 | artom | Not if we can help it ;) | |
| 19:23:21 | mriedem | you're like the texas of canada? | |
| 19:23:32 | mriedem | le yeeha | |
| 19:23:35 | mriedem | *yeehaw | |
| 19:23:46 | artom | L'osti de takarnak | |
| 19:23:53 | artom | *tabarnak | |
| 19:23:55 | artom | Dammit :( | |
| 19:24:21 | artom | mriedem, can I circle back to that doc later tonight? I promised daughter we'd assemble her new bed while she's in daycare :P | |
| 19:25:03 | artom | I wasn't expecting Sylvain to react to the review until tomorrow morning :/ | |
| 19:25:59 | mriedem | just throw some blankets and pillows on the ground | |
| 19:26:10 | mriedem | damn kids and their demands for beds | |
| 19:26:25 | mriedem | wa wa feed me bathe me love me | |
| 19:31:04 | dims | mriedem : wait for a few years... it turns into "leave me alone!" | |
| 19:31:08 | dims | :) | |
| 19:34:47 | mriedem | i almost thought i was going to see tears yesterday morning about the pants that laura laid out | |
| 19:34:55 | mriedem | so yeah i know what it's going to be like already | |
| 19:35:20 | mriedem | only *i* get to cry about the pants that laura lays out for me to wear! | |
| 19:35:37 | edleafe | mriedem: you wear pants? | |
| 19:35:45 | mriedem | still do yup | |
| 19:35:48 | mriedem | and shave and get haircuts | |
| 19:35:54 | mriedem | hasn't been a year of working at home yet | |
| 19:35:59 | dansmith | mriedem: I was going to say.. my pants still get laid out for me for important pants-wearing occasions | |
| 19:36:26 | edleafe | that's why I only have one pair. Prevents all the pants-related arguments | |
| 19:37:12 | dansmith | business in the front, party in the back? | |
| 19:38:27 | dansmith | not the kind that require specific pant selection, that's for sure | |
| 20:02:01 | openstackgerrit | Ameed Ashour proposed openstack/nova master: add "--until-complete" option for nova-manage db archive_deleted_rows man page https://review.openstack.org/539654 | |
| 20:05:14 | openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: placement upgrade notes for queens https://review.openstack.org/539658 | |
| 20:09:36 | mriedem1 | need a final +2 on the reno for versioned notifications new in queens https://review.openstack.org/#/c/518018/ | |
| 20:11:30 | mriedem | cfriesen: did your perf testing show any issues with the counting quotas stuff that was new in pike? | |
| 20:24:19 | cfriesen | mriedem: don't know if we've gotten that far. I'm pretty sure it was on a mostly idle system so I wouldn't expect that to be a major factor. | |
| 20:35:10 | openstackgerrit | Ameed Ashour proposed openstack/nova master: detach instance volumes when VM creation fails https://review.openstack.org/528385 | |
| 20:37:58 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Re-arrange cells docs for modern deployments https://review.openstack.org/539674 | |
| 20:38:25 | mriedem | dansmith: melwitt: we had talked at one point a few weeks ago about re-arranging the cells docs to move the cells v2 setup stuff to the top since that's mostly what people care about now when upgrading to ocata ^ | |
| 20:38:47 | mriedem | tried to move that stuff around and still leave the historical manifesto and cells v1 stuff in there to still make sense | |
| 20:39:18 | melwitt | ack | |
| 20:44:37 | efried | mriedem: "the in-tree powervm driver doesn't support gpu passthrough right?" Not yeeeeeetttt.... | |
| 20:49:00 | hrw | does someone uses hw_pointer_model option? with something else than 'usbtablet' there? | |
| 20:49:38 | hrw | and if you do then does it work for you? | |
| 21:05:35 | ameeda | mriedem: replied | |
| 21:05:52 | mriedem | hrw: i thought it was end of day for you? | |
| 21:07:05 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Re-arrange cells docs for modern deployments https://review.openstack.org/539674 | |
| 21:17:18 | mriedem | dansmith: can you hit this backport so i can get the one below it? https://review.openstack.org/#/c/527560/ | |
| 21:17:24 | ameeda | mriedem: please check the patch there is important comment there https://review.openstack.org/#/c/507486 | |
| 21:18:14 | dansmith | mriedem: doneski | |
| 21:23:09 | mriedem | ameeda: i'm not sure what you want me to look at in that change | |
| 21:24:55 | mriedem | ugh i guess we can't backport this fix now https://review.openstack.org/#/c/524755/4/nova/objects/host_mapping.py@152 | |
| 21:25:47 | mriedem | even though nothing uses it over rpc? | |
| 21:27:49 | dansmith | mriedem: well like I said I don't think that needed to be a bump anyway | |
| 21:28:03 | dansmith | mriedem: you could backport without the version bump and be fine I think.. seems scary, but should be alright | |
| 21:28:37 | openstackgerrit | Claudiu Belu proposed openstack/nova master: hyper-v: autospec classes before they are instantiated https://review.openstack.org/342211 | |
| 21:32:05 | mriedem | efried: see what you think about my 'experimental' wording idea in https://review.openstack.org/#/c/539266/ | |
| 21:32:43 | mriedem | dansmith: might not be necessary, i was just thinking about it b/c the cells faq page says you can use this to list hosts in a cell, but there are other options for doing that as well, so not super important | |
| 21:36:57 | ameeda | mriedem: Surya Seetharaman -> The same change of adding "--until-complete" is been done here I think ? : https://review.openstack.org/#/c/507486/29/doc/source/cli/nova-manage.rst | |
| 21:37:17 | ameeda | mriedem: I will upload new patch now | |
| 21:37:59 | mriedem | ameeda: your patch is much more targeted to just updating the man page | |
| 21:38:15 | openstackgerrit | Ameed Ashour proposed openstack/nova master: add "--until-complete" option for nova-manage db archive_deleted_rows. https://review.openstack.org/539654 | |