| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-11-15 | |||
| 09:10:08 | bauzas | anyway, it doesn't expose a standard PCI address | |
| 09:12:49 | bauzas | I just feel that stephenfin is absolutely right, nova punts the address because of the internal conditional he said | |
| 09:15:21 | awalende | someone rebooted the server =.= one more minute | |
| 09:17:19 | awalende | ok, its spawning :D fingers crossed | |
| 09:20:53 | awalende | Ok, the device shows up in the guest | |
| 09:20:54 | awalende | hooray | |
| 09:21:29 | awalende | But why do we explicitly have to specify device_type. In all of our other devices it wasn't really needed.....is it because pci is default? | |
| 09:21:52 | stephenfin | awalende: https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L289-L297 | |
| 09:22:43 | stephenfin | If you have a device that supports SR-IOV, the assumption we've made is that you'll generally want to use that capability because of the scalability it offers | |
| 09:23:33 | awalende | Uhhh just a minor complaint....this isn't documented :D :D. But nice, the T4 is now there in the guest. Thanks all for helping us! | |
| 09:23:44 | stephenfin | So by default we don't allow you to use the parent/root/PF device because if you do, you won't be able to use the child/VF device(s) | |
| 09:24:11 | stephenfin | Agreed. I'm working on a patch to do that now. I'll stick you on the review once it's up, if you like | |
| 09:24:24 | awalende | sure thing :D | |
| 09:27:09 | awalende | irc name is launchpad name | |
| 09:33:30 | huaqiang | sean-k-mooney: about the per-vm-pci-NUMA-policy spec, I want to confirm that it defines a policy applied to all PCI devices for whole VM, right? | |
| 09:43:51 | huaqiang | sean-k-mooney: seems I have sent my questions we have drawn a conclusion agian by mistake, pls ignore it. | |
| 10:01:53 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP block_device: Use original volume_type when creating volumes from snapshots https://review.opendev.org/694497 | |
| 10:21:26 | openstackgerrit | Liang Fang proposed openstack/nova-specs master: Support volume local cache https://review.opendev.org/689070 | |
| 11:01:22 | openstackgerrit | Lee Yarwood proposed openstack/nova-specs master: Virtual instance rescue with stable disk devices https://review.opendev.org/693849 | |
| 11:01:23 | openstackgerrit | Lee Yarwood proposed openstack/nova-specs master: Boot from volume instance rescue https://review.opendev.org/694063 | |
| 11:46:01 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Change order of PCI configuration steps https://review.opendev.org/694521 | |
| 11:46:01 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Clarify configuration steps for PF devices https://review.opendev.org/694522 | |
| 11:46:04 | stephenfin | awalende: ^ | |
| 11:46:14 | stephenfin | Also sean-k-mooney, adrianc: ^ | |
| 11:46:30 | stephenfin | I'd like to backport that as far as we can go | |
| 11:48:18 | openstackgerrit | Merged openstack/nova master: Reset vm_state to original value if rebuild claim fails https://review.opendev.org/692185 | |
| 11:48:51 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Clarify configuration steps for PF devices https://review.opendev.org/694522 | |
| 11:53:52 | openstackgerrit | Merged openstack/nova master: FUP for Ib62ac0b692eb92a2ed364ec9f486ded05def39ad https://review.opendev.org/693556 | |
| 12:19:02 | sean-k-mooney | stephenfin: im not sure if that is correct | |
| 12:19:23 | sean-k-mooney | if you dont set the device_type in the alias at all i think it will work with type-PF | |
| 12:19:44 | sean-k-mooney | the device_type is not required in the alias as far as i remember | |
| 12:19:56 | sean-k-mooney | if it is present it definetly needs to match | |
| 12:20:24 | sean-k-mooney | i need to test some sriov stuff later today so ill try this if i have time otherwise ill do it on monday | |
| 12:21:25 | shilpasd | gibi: hi, can you please elaborate on point 'add compute nodes to the aggregates of shared RP aggregates' regarding Shared NFS, so that will think on designing the same | |
| 12:22:56 | sean-k-mooney | the product id will be differente for the PF vs VF so if you have the correct product id and no VF are allocated i think it just works but again that is just going off memory i have not tried this in a year or two | |
| 12:25:28 | shilpasd | gibi: at driver level, how ww will get aggragtes associated to shared RP and will map those to compute node? And if compute node already has aggregates (not the same as of shared RP) are we really bother of them? | |
| 12:37:59 | openstackgerrit | Lee Yarwood proposed openstack/nova master: docs: Extract rescue from reboot https://review.opendev.org/694529 | |
| 12:39:23 | gibi | shilpasd: there are two types of aggregates: there are placement aggregates https://docs.openstack.org/api-ref/placement/#resource-provider-aggregates and there are nova host aggregates https://docs.openstack.org/api-ref/compute/#host-aggregates-os-aggregates | |
| 12:39:50 | gibi | the disk sharing RP needs to be in the same placement aggregate as the compute RP it shares disk with | |
| 12:40:34 | gibi | we agreed that a new nova config option will tell the nova compute the uuid of the placement aggregate | |
| 12:40:49 | gibi | in which the admin added the sharing disk RP | |
| 12:40:58 | openstackgerrit | Lee Yarwood proposed openstack/nova master: docs: Extract rescue from reboot https://review.opendev.org/694529 | |
| 12:41:26 | gibi | so when the compute start up it can change if its own RP (the compute PR) is already part of the placement aggregate configured in it's nova.conf | |
| 12:41:42 | gibi | and if not then call PUT | |
| 12:41:42 | gibi | /resource_providers/{uuid}/aggregatesPUT | |
| 12:41:47 | gibi | /resource_providers/{uuid}/aggregates | |
| 12:42:05 | gibi | to put it's compute PR into the that aggregate | |
| 12:43:15 | gibi | shilpasd: an PR can be in any number or placement aggregates | |
| 12:43:44 | gibi | so if the compute RP is already in other aggregates that does not really matter to us | |
| 12:45:48 | openstackgerrit | Lee Yarwood proposed openstack/nova-specs master: Virtual instance rescue with stable disk devices https://review.opendev.org/693849 | |
| 12:45:49 | openstackgerrit | Lee Yarwood proposed openstack/nova-specs master: Boot from volume instance rescue https://review.opendev.org/694063 | |
| 12:46:07 | shilpasd | gibi: new nova config option which holds uuid of the placement aggregate or uuid of the shared RP? | |
| 12:47:37 | shilpasd | and it will at compute level or libvirt level? | |
| 12:49:41 | gibi | shilpasd: uuid of the placement aggregate | |
| 12:49:51 | gibi | shilpasd: and I think the logic can be in the compute manager level | |
| 12:51:15 | gibi | shilpasd: have you seen this mailthread ? http://lists.openstack.org/pipermail/openstack-discuss/2019-November/010624.html | |
| 12:52:38 | shilpasd | gibi: no, will go through, thnaks for sharing | |
| 12:52:55 | gibi | shilpasd: I think that summarized most of what we talked above | |
| 12:53:36 | shilpasd | gibi: yes, that's clear | |
| 12:53:56 | shilpasd | gibi: thank you | |
| 12:54:54 | gibi | shilpasd: you're welcome | |
| 12:56:14 | openstackgerrit | Merged openstack/nova master: Add functional recreate test for bug 1852610 https://review.opendev.org/694351 | |
| 12:56:14 | openstack | bug 1852610 in OpenStack Compute (nova) "API allows source compute service/node deletion while instances are pending a resize confirm/revert" [Medium,In progress] https://launchpad.net/bugs/1852610 - Assigned to Matt Riedemann (mriedem) | |
| 13:45:59 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove duplicate ServerMovingTests._resize_and_check_allocations https://review.opendev.org/694538 | |
| 14:18:11 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/train: Add functional recreate test for bug 1852610 https://review.opendev.org/694544 | |
| 14:18:11 | openstack | bug 1852610 in OpenStack Compute (nova) "API allows source compute service/node deletion while instances are pending a resize confirm/revert" [Medium,In progress] https://launchpad.net/bugs/1852610 - Assigned to Matt Riedemann (mriedem) | |
| 14:18:11 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/train: Add functional recreate revert resize test for bug 1852610 https://review.opendev.org/694545 | |
| 14:18:11 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/train: Block deleting compute services with in-progress migrations https://review.opendev.org/694546 | |
| 14:22:38 | artom | "Before posting a comment to any patch, a third party testing system must contact the project they wish to test and get approval to post comments on their patches. This can be done by attending the project’s meeting." | |
| 14:22:43 | artom | From https://docs.openstack.org/infra/system-config/third_party.html | |
| 14:22:49 | artom | Is that an actual thing for Nova? | |
| 14:23:15 | dansmith | it's for every project I think | |
| 14:23:41 | dansmith | we used to have problems with people setting up their own CI and it going haywire and spraying comments everywhere | |
| 14:23:42 | dansmith | (in nova) | |
| 14:23:51 | artom | dansmith, so context is http://post-office.corp.redhat.com/archives/rh-openstack-dev/2019-October/msg00060.html (sorry for the internal-only link) | |
| 14:25:48 | artom | dansmith, and http://post-office.corp.redhat.com/archives/rh-openstack-dev/2019-October/msg00209.html that's sort of a summary of where we ended up after initial discussions | |
| 14:26:14 | dansmith | you could probably just say "I want to set up a CI system" | |
| 14:27:10 | artom | I want to set up a CI system | |
| 14:27:13 | artom | :D | |
| 14:27:15 | dansmith | but anyway, yes, there are hoops to jump through | |
| 14:28:07 | artom | Yeah | |
| 14:28:12 | artom | Internally as well as here, *sigh* | |
| 14:28:22 | artom | Hopefully it'll be worth it | |
| 14:28:31 | artom | I also noticed Mellanox has their own SRIOV CI | |
| 14:28:40 | artom | But... presumably that only tests their hardware | |
| 14:28:47 | artom | And I have no idea what the status is | |
| 14:29:09 | artom | adrianc, ^^ ? | |
| 14:29:42 | artom | And the point of RHEx (Red Hat Exotic hardware CI) would be more than just SRIOV, SRIOV is just the initial MVP scope | |
| 14:29:47 | artom | GPUs come to mind | |
| 14:29:48 | dansmith | artom: are you looking for help setting it up? If so, I'm sure the infra people are who you want to talk to | |
| 14:30:02 | artom | dansmith, yeah, that conversation is already happening - migi as well | |
| 14:30:12 | dansmith | ack | |
| 14:30:42 | mriedem | i wish the dell emc people would have asked b/c their's comments and always fails | |
| 14:32:01 | bauzas | gibi: I'm about to provide a new revision for the audit command, would it be possible for you to check it with the bandwidth-aware instances ? | |
| 14:32:15 | bauzas | (at least once I'm done with reno) | |
| 14:40:58 | gibi | bauzas: sure | |
| 14:41:28 | gibi | bauzas: If I time out on it today then I will do it next week | |
| 14:41:36 | bauzas | gibi: thanks | |
| 14:41:46 | bauzas | hopefully, I'll push it in 20 mins | |
| 14:47:31 | openstackgerrit | Merged openstack/nova master: Add functional recreate revert resize test for bug 1852610 https://review.opendev.org/694364 | |