| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-06-05 | |||
| 13:25:08 | kashyap | artom: An email needs to be written to discuss on the list | |
| 13:25:28 | kashyap | [nova] Test 'q35' machine type in the Gate | |
| 13:25:31 | kashyap | Or something like that | |
| 13:25:39 | artom | kashyap, sounds like a plan | |
| 13:25:53 | artom | Who's doing the needful? | |
| 13:25:58 | kashyap | I can do that. But not now | |
| 13:26:19 | kashyap | One of the main reasons is that upstream and (nor a certain major Linux distro) QEMU doesn't "fix" bugs or add additional features to the 'pc' machine type | |
| 13:26:42 | kashyap | Not least because it's considered legacy (20+ years old), and difficult to extend | |
| 13:27:09 | artom | That's an argument for outright switching the default, no? | |
| 13:27:41 | kashyap | It is. We've had approximately 240 email-long threads about it. | |
| 13:27:53 | artom | Hah :( | |
| 13:28:00 | kashyap | Sorry, by "we" as in QEMU and libvirt. | |
| 13:28:18 | kashyap | Some fear the backwards compat that it "might" break something | |
| 13:28:20 | artom | Well, I meant in Nova, but in QEMU works too | |
| 13:28:24 | kashyap | But we can't ever be paralyzed | |
| 13:28:44 | kashyap | artom: Yes: https://review.openstack.org/#/c/631154/ | |
| 13:28:52 | kashyap | "Gracefully handle QEMU machine types for guests" | |
| 13:29:38 | artom | kashyap, oh, had no idea that existed, thanks | |
| 13:29:42 | kashyap | artom: For now, installer tools are supposed to configure the default. | |
| 13:29:50 | artom | Will add to the pile of specs to review | |
| 13:30:23 | artom | ... which I guess makes sense (installer configuring the machine type) | |
| 13:30:58 | kashyap | That's one reason I put on the back burner for this cycle | |
| 13:33:10 | yaawang | johnthetubaguy: mriedem Could you please take a look at auto-converge/post-copy spec? I've updated it. https://review.opendev.org/#/c/651681/ | |
| 14:05:35 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Remove global state from the FakeDriver https://review.opendev.org/656709 | |
| 14:05:35 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Enhance service restart in functional env https://review.opendev.org/512552 | |
| 14:05:36 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add functional test coverage for bug 1724172 https://review.opendev.org/512553 | |
| 14:05:37 | openstack | bug 1724172 in OpenStack Compute (nova) rocky "Allocation of an evacuated instance is not cleaned on the source host if instance is not defined on the hypervisor" [Medium,Confirmed] https://launchpad.net/bugs/1724172 | |
| 14:06:20 | efried | cfriesen: Thanks for the update. | |
| 14:09:14 | aspiers | https://review.opendev.org/#/c/638680/ got a +1 from Zuul finally | |
| 14:10:05 | efried | yup, \o/ | |
| 14:10:35 | aspiers | is it just me or are we getting some semi-spam +1 reviews? | |
| 14:11:38 | efried | happens all the time | |
| 14:11:41 | aspiers | ah ok | |
| 14:11:58 | efried | stat padders | |
| 14:13:30 | aspiers | :-( | |
| 14:14:22 | aspiers | kashyap: funny this should just emerge after we were talking about it the other day https://itnext.io/pyramid-of-doom-the-signs-and-symptoms-of-a-common-anti-pattern-c716838e1819 | |
| 14:14:38 | kashyap | Hiya | |
| 14:15:24 | kashyap | aspiers: Hehe, "pyrmaid of doom" | |
| 14:16:55 | gibi | aspiers: they are coming in batches | |
| 14:17:01 | gibi | aspiers: https://www.stackalytics.com/?metric=marks&module=nova-group&company=awcloud | |
| 14:17:03 | kashyap | aspiers: On "semi-spam +1s", I wonder if they would dare to do that had it were an e-mail | |
| 14:17:15 | kashyap | Imagine sending random +1s to e-mail based patch workflows. | |
| 14:17:26 | aspiers | right | |
| 14:17:41 | aspiers | perhaps someone in the foundation can have a word? | |
| 14:17:41 | kashyap | Each Gerrit change is an island, nobody notices it (beside those who do). So who cares who craps on it | |
| 14:18:03 | kashyap | aspiers: This is something we have to live with, as a community-based open source project, afraid. | |
| 14:18:15 | aspiers | kashyap: not sure I agree | |
| 14:18:28 | aspiers | there's nothing to stop us raising complaints about this behaviour | |
| 14:18:44 | kashyap | aspiers: Oh, certainly; there was a recent thread about it on the list, too | |
| 14:18:48 | kashyap | (Not sure you noticed it) | |
| 14:19:06 | aspiers | I think I saw it | |
| 14:19:09 | lyarwood | mdbooth / kashyap / sean-k-mooney ; https://review.opendev.org/#/c/663011/ is ready for another round of reviews btw | |
| 14:19:11 | kashyap | aspiers: I was just implying we can't eliminate this problem entirely. | |
| 14:19:46 | aspiers | kashyap: sure, but "This is something we have to live with" sounded more like a statement of resignation | |
| 14:20:08 | aspiers | We can probably eliminate 80% of spam reviews with small effort | |
| 14:20:15 | kashyap | aspiers: Heh, let me "phrasing guard" down for a bit there. | |
| 14:20:27 | aspiers | It's an odd choice of changes to review they picked | |
| 14:21:44 | kashyap | lyarwood: Will look in a bit. | |
| 14:21:48 | aspiers | https://www.stackalytics.com/?company=awcloud&metric=marks&release=train <- 25 contributors with 100% approval ratio | |
| 14:27:20 | openstackgerrit | Merged openstack/nova master: conf: Remove cells v1 options, group https://review.opendev.org/651310 | |
| 14:35:10 | boxiang | https://review.opendev.org/#/c/649963/ need someone to review it, give me some comments, thanks :) | |
| 14:36:51 | mriedem | the pci passthrough whitelist config is not mutable so changing it requires a restart, and that means it would be the same for all computes processed by the scheduler yeah? meaning when we create a HostState object per compute node per scheduling request, we're re-parsing that pci passthrouh whitelist spec per node per request, right? https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L227 | |
| 14:37:29 | mriedem | if you have several hundred nodes to process per request i'd imagine that eventually adds up | |
| 14:39:34 | efried | aspiers: I'm a bit fuzzy yet this morning, but I think there's a hole in https://review.opendev.org/#/c/662105/ - please see comments. | |
| 14:40:06 | aspiers | efried: OK | |
| 14:40:38 | sean-k-mooney | mriedem: that is done on the compute node but yes | |
| 14:41:03 | mriedem | sean-k-mooney: this is done in the scheduler | |
| 14:41:31 | sean-k-mooney | the pci whitelist is on each the compute nodes and can be different | |
| 14:41:37 | sean-k-mooney | the alias is parsed on the schduler | |
| 14:42:22 | sean-k-mooney | there was a patch to cache this too at one point but i think it was blocked because peopel didnt feel it was enough of a perfomace imporvemnt | |
| 14:42:30 | openstackgerrit | Dan Smith proposed openstack/nova master: Make nova-next archive using --before https://review.opendev.org/661002 | |
| 14:43:31 | sean-k-mooney | mriedem: https://review.opendev.org/#/c/427145/ | |
| 14:44:12 | sean-k-mooney | mriedem: not sure if stephenfin still feels its not worth it | |
| 14:44:45 | sean-k-mooney | mriedem: but i certenly tried to get it landed 2 years ago :) | |
| 14:47:20 | sean-k-mooney | mriedem: by the way https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L227 is loading the pci info frmo the database not parsing the whitelist | |
| 14:48:46 | sean-k-mooney | well that is not quite true its contcutiono the PciDeviceStats object form the compute.pci_device_pools | |
| 14:49:52 | mriedem | https://bugs.launchpad.net/nova/+bug/1831758 | |
| 14:49:53 | openstack | Launchpad bug 1831758 in OpenStack Compute (nova) "passthrough_whitelist is parsed per compute node per scheduling request even though the whitelist doesn't change between requests" [Low,Confirmed] | |
| 14:50:01 | mriedem | constructing PciDeviceStats parses the whitelist | |
| 14:50:49 | sean-k-mooney | the whitelist does not have to be set on the the schduler node | |
| 14:51:29 | sean-k-mooney | and it is and i can be different per compute node but you are refering to https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L66-L67 | |
| 14:51:49 | openstackgerrit | Adrian Chiris proposed openstack/nova stable/rocky: Move get_pci_mapping_for_migration to MigrationContext https://review.opendev.org/661499 | |
| 14:51:49 | openstackgerrit | Adrian Chiris proposed openstack/nova stable/rocky: Allow driver to properly unplug VIFs on destination on confirm resize https://review.opendev.org/661500 | |
| 14:52:58 | sean-k-mooney | mriedem: i guess that makes sense on the compute node as this class is used in the pci manager but we should not need to do it in the schduler | |
| 14:55:15 | sean-k-mooney | mriedem: we proably should add a use_whitelist kwarg or something so we can disable this on the schduler | |
| 14:55:27 | openstackgerrit | Lee Yarwood proposed openstack/nova master: DNM dump image_meta when attempting to validate image signatures https://review.opendev.org/663348 | |
| 14:56:44 | sean-k-mooney | the pci whitelist shoudl only be used in the compute node and the alias should only be used on teh api or schduler nodes. | |
| 15:00:44 | efried | can we not save the date stamp of the file? | |
| 15:00:57 | efried | or use fsnotify (or whatever that thing is called)? | |
| 15:07:46 | mriedem | we're parsing from the config option value which is not mutable | |
| 15:08:31 | sean-k-mooney | mriedem: and as i mention on the bug the config value is only used on teh compute node and not in the schduler | |
| 15:08:34 | efried | oh, sorry, yeah, I was thinking of something quite different. Maybe I need more coffee | |
| 15:09:11 | mriedem | sean-k-mooney: i would not doubt that config mgmt tooling - if it's setting it - is setting it everywhere because people wouldn't know which services care about it | |
| 15:09:29 | mriedem | sean-k-mooney: nor do you know what out of tree filters could be doing with that thing in the HostState object | |
| 15:09:51 | sean-k-mooney | mriedem: well the value can and typicaly iss different per compute node | |
| 15:10:05 | sean-k-mooney | so if they are doing anything with the config value its proably wronge | |
| 15:10:11 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: nova-manage: heal port allocations https://review.opendev.org/637955 | |
| 15:10:15 | mriedem | yeah that's true | |
| 15:12:46 | mriedem | tssurya: why do you have a -1 on https://review.opendev.org/#/c/623558/ ? | |
| 15:13:21 | mriedem | it looks like from your numbers the patch makes a noticeable improvement in scheduling time, | |