| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-07-17 | |||
| 09:27:19 | openstackgerrit | Alex Xu proposed openstack/nova master: Moves the allocation retrieving early https://review.opendev.org/670783 | |
| 09:27:20 | openstackgerrit | Alex Xu proposed openstack/nova master: [WIP] Using the claim/unclaim_for_instance for mdevs https://review.opendev.org/671222 | |
| 09:27:20 | openstackgerrit | Alex Xu proposed openstack/nova master: [WIP] Populate the existing mediated devices in the libvirt device manager https://review.opendev.org/670787 | |
| 09:29:24 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Provide HW_CPU_X86_AMD_SEV trait when SEV is supported https://review.opendev.org/638680 | |
| 09:29:25 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Extract SEV-specific bits on host detection https://review.opendev.org/636334 | |
| 09:29:25 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420 | |
| 09:29:26 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Apply SEV-specific guest config when SEV is required https://review.opendev.org/644565 | |
| 09:29:26 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Add |
|
| 09:29:27 | openstackgerrit | Boris Bobrov proposed openstack/nova master: Enable booting of libvirt guests with AMD SEV memory encryption https://review.opendev.org/666616 | |
| 09:29:47 | openstackgerrit | Chris Dent proposed openstack/nova master: Restore RT.old_resources if ComputeNode.save() fails https://review.opendev.org/668263 | |
| 09:31:37 | cdent | stephenfin: https://review.opendev.org/#/c/668263/ went into merge conflict, so I fixed it since mriedem is away. Wanna drop your +W back on? | |
| 09:31:51 | stephenfin | Sure | |
| 09:32:55 | cdent | thanks | |
| 11:44:41 | kashyap | coreycb: Hi, just as a follow-up on that distro-specific work for EDK2/OVMF: | |
| 11:44:50 | kashyap | coreycb: My Fedora patch is merged: https://src.fedoraproject.org/rpms/edk2/c/674b3c8a27a8 | |
| 11:45:10 | kashyap | coreycb: And I've filed the bugs for Debian and Ubuntu: | |
| 11:45:17 | openstack | Launchpad bug 1836859 in edk2 (Ubuntu) "RFE: Ship the firmware "descriptor files" as part of the 'ovmf' package" [Undecided,New] | |
| 11:45:17 | kashyap | - Ubuntu: https://bugs.launchpad.net/ubuntu/+source/edk2/+bug/1836859 | |
| 11:45:26 | openstack | Debian bug 932269 in ovmf "Ship the firmware "descriptor files" as part of the 'ovmf' package" [Normal,Open] | |
| 11:45:26 | kashyap | - Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=932269 | |
| 12:41:13 | coreycb | kashyap: thank you very much for doing that. i'll let cpaelzer know. | |
| 12:42:08 | kashyap | coreycb: Cool; can you Cc me when an Ubuntu build? | |
| 12:42:32 | coreycb | kashyap: i'd recommend syncing with cpaelzer in #ubuntu-server | |
| 12:42:48 | coreycb | kashyap: the bug will get updated though | |
| 12:43:08 | kashyap | coreycb: Okay, I'll follow the bug. I don't want to spend 100% of the time doing all distro work :-) | |
| 12:43:13 | kashyap | Thanks! | |
| 12:43:33 | coreycb | kashyap: completely understandable :) | |
| 12:43:55 | kashyap | There's a cap on the no. yaks I could trim in a week :D | |
| 13:14:54 | Bidwe_jay | Hi all | |
| 13:16:49 | Bidwe_jay | want to discuss regarding how to create placementclient in horizon if anyone have any idea please help.. | |
| 13:26:03 | ralonsoh | stephenfin, show should I do this? https://review.opendev.org/#/c/641670/4/specs/train/approved/vm-boot-with-unaddressed-port.rst@198 | |
| 13:26:35 | ralonsoh | stephenfin, is there a way to write those references correctly? I say that because I want those references to be there | |
| 13:26:45 | ralonsoh | stephenfin, thanks in advance! | |
| 13:28:52 | stephenfin | ralonsoh: If you did care, I think you need to use the footnote style of reference https://thomas-cokelaer.info/tutorials/sphinx/rest_syntax.html#footnote | |
| 13:29:06 | stephenfin | Personally though, I'd just say 'None' or remove the 'Reference' section if you don't need it | |
| 13:32:28 | stephenfin | dansmith: Care to look at this https://review.opendev.org/#/c/665626/ ? | |
| 13:33:33 | ralonsoh | stephenfin, I don't understand. This is the way used in all specs in Stein | |
| 13:34:00 | sean-k-mooney | ralonsoh: by the way was https://review.opendev.org/#/c/640258/ what you were asking me to do | |
| 13:35:03 | ralonsoh | sean-k-mooney, get_vifs_by_ids is used in https://github.com/openstack/neutron/blob/3b2521a89496a9857ed3873e5a0172a41b2f8a9f/neutron/plugins/ml2/drivers/openvswitch/agent/ovs_dvr_neutron_agent.py#L401 | |
| 13:35:09 | sean-k-mooney | yes | |
| 13:35:17 | sean-k-mooney | i filtered the ids it passed in | |
| 13:35:24 | ralonsoh | sean-k-mooney, and you are modifying the behaviour in your patch | |
| 13:35:31 | ralonsoh | allowing ofport=-1 | |
| 13:35:35 | sean-k-mooney | https://review.opendev.org/#/c/640258/9/neutron/plugins/ml2/drivers/openvswitch/agent/ovs_dvr_neutron_agent.py | |
| 13:35:51 | stephenfin | ralonsoh: If that's the case, feel free to ignore me :) This is what I mean though, fwiw https://bit.ly/2Y5AjUR | |
| 13:35:59 | sean-k-mooney | i also forgot a space after the comma which is why pep8 is unhappy | |
| 13:36:10 | ralonsoh | ok, that's PS9, I reviewed PS8 | |
| 13:36:22 | sean-k-mooney | yep i was asking is that all you wanted me to fix | |
| 13:36:34 | sean-k-mooney | ill fix the pep8 issue shortly | |
| 13:36:48 | ralonsoh | sean-k-mooney, yes, exactly | |
| 13:36:55 | stephenfin | (vs. https://bit.ly/2XN4D7f) | |
| 13:37:02 | ralonsoh | and please, use the constants for invalid ports | |
| 13:37:19 | ralonsoh | stephenfin, ahhhhhhh | |
| 13:37:26 | ralonsoh | stephenfin, now I understand!! | |
| 13:37:28 | sean-k-mooney | i prefer the second for what its worth | |
| 13:37:32 | ralonsoh | stephenfin, thanks!! | |
| 13:37:40 | stephenfin | no problemo | |
| 13:37:54 | sean-k-mooney | i think it looks better in pdf too | |
| 13:38:26 | sean-k-mooney | not that i expect nova specs to rendered to pdf like some of the other docs but its nice to be consitent | |
| 13:38:40 | ralonsoh | sean-k-mooney, you mean https://bit.ly/2XN4D7f ?? | |
| 13:39:15 | sean-k-mooney | ya i think that is nicer to read and eisier to serach for if you are reading the rst file directly | |
| 13:39:20 | ralonsoh | okidoki | |
| 13:39:41 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: func test for migrate server with ports having resource request https://review.opendev.org/655113 | |
| 13:40:20 | sean-k-mooney | i almost exclisively read teh docs and sepc in gerrit/locally without rendering so i have a bias to that form. | |
| 13:42:54 | openstackgerrit | Stephen Finucane proposed openstack/nova master: doc: Note use of 'nova-manage db sync --config-file' https://review.opendev.org/671298 | |
| 13:43:21 | stephenfin | sean-k-mooney: You'll probably like that ^ | |
| 13:43:27 | stephenfin | Also | |
| 13:44:13 | stephenfin | I forget just how good we've got it in OpenStack whenever I look at community projects | |
| 13:44:33 | stephenfin | "Why are half your tests skipped?" "Oh, they just stopped working so we disabled them" | |
| 13:44:34 | stephenfin | :) | |
| 13:45:29 | cdent | sweet | |
| 13:47:59 | sean-k-mooney | stephenfin: +1'd what had you running that manually anwyway? just trying to avoid restacking? | |
| 13:48:39 | stephenfin | I'm still working on the cpu-resources spec. I'd stacked with master and then wanted to work through an upgrade | |
| 13:48:52 | stephenfin | I foresee a lot of restacks in the coming week or so | |
| 13:49:24 | stephenfin | But! http://paste.openstack.org/show/754499/ | |
| 13:50:26 | Nick_A | Hello - we have instance images and snapshots set to qcow2 in nova.conf. Is there a way to bypass what appears to be an unnecessary conversion step before it uploads the snapshot to glance? Am I right in understanding that it's trying to convert qcow2 > qcow2? | |
| 14:09:36 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/rocky: [DNM] testing bug/1813789 revert resize events https://review.opendev.org/671303 | |
| 14:12:49 | openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/nova-specs master: Boot a VM with an unaddressed port https://review.opendev.org/641670 | |
| 14:45:27 | Zara | hullo! I'm back with my '`nova list` slow when run with member user but not admin user' question-- narrowed it down a bit; the slow requests are the ones via the neutronclient to ports.json (as part of nova's servers/detail request). 10s apiece for member; 200ms apiece for admin. maybe a neutron or keystone issue? | |
| 14:46:45 | sean-k-mooney | Zara: did you check if the delay was in the neuron api or in nova | |
| 14:59:42 | Zara | ahhhh, listing ports alone is also slow, so almost certainly neutron (so I'm even more in-the-wrong-channel than I thought; sorry!) | |
| 15:00:04 | sean-k-mooney | haha no worries | |
| 15:01:09 | Zara | :) | |
| 15:28:46 | alex_xu | dansmith: I faced a problem. Seems the dst host's libvirt won't keep the dst domain xml when live-migration failed due to the dst host down. So when the dst host's nova-compute startup again, it can't know the target vpmem which may dirty by the live-migration, then can't do a cleanup for that vpmem. | |
| 15:28:51 | alex_xu | efried: ^ | |
| 15:29:25 | dansmith | um | |
| 15:29:38 | dansmith | you mean live migration, which fails in the middle because the compute host goes down or something? | |
| 15:29:44 | efried | yes | |
| 15:29:52 | alex_xu | yes, the dst compute host goes down | |
| 15:30:36 | sean-k-mooney | well that should be reverted right. so i guess the problem is some of the data could have been copied | |
| 15:30:51 | alex_xu | yes | |
| 15:31:08 | sean-k-mooney | alex_xu: could we optionally clean the unsed namespaces on agent start | |
| 15:31:09 | dansmith | yeah, I mean, that's a serious failure, and all kinds of things could go wrong there | |
| 15:31:23 | dansmith | sean-k-mooney: yeah was just going to say.. that's probably safest | |
| 15:31:39 | alex_xu | ok, but it should be slow | |
| 15:31:51 | sean-k-mooney | we could proably store a file somewhere of things we are currently programing and delete the file if we completed properly | |
| 15:32:03 | alex_xu | it is ok to cleanup, since the live-migration won't be continue again | |
| 15:32:10 | dansmith | because even if you were tracking them some other way, that failure scenario, followed by a db archive before the compute node starts up again would result in not knowing about the previous failure | |
| 15:32:34 | dansmith | sean-k-mooney: yeah, there may be residue just from the incoming disk images and stuff actually | |
| 15:32:38 | dansmith | also, | |