| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-09-02 | |||
| 12:32:43 | aspiers | sean-k-mooney: Yes, that's why image_meta is a parameter to get_machine_type_no_default() in the code I just linked | |
| 12:33:08 | kashyap | aspiers: Thinking out loud: do we need an additional function for handling SEV-specific quirks for machine types, instead of squeezing it into the generic guest_machine_type()? | |
| 12:33:10 | aspiers | https://review.opendev.org/#/c/644565/49/nova/virt/libvirt/utils.py@539 | |
| 12:33:25 | sean-k-mooney | yes but overrideing an explcit machine_type in the image is not correct | |
| 12:33:43 | aspiers | sean-k-mooney: like I said, we could bail there instead | |
| 12:33:49 | aspiers | that would probably be fine | |
| 12:33:58 | sean-k-mooney | i would bail on conflict with the image | |
| 12:34:07 | aspiers | Yes that's what I'm saying | |
| 12:34:24 | sean-k-mooney | that leave the question of what to do if its set to pc in nova.conf | |
| 12:34:39 | sean-k-mooney | your could bail then also or override | |
| 12:35:01 | kashyap | aspiers: One more: It seems like if machine type is _not_ set, then we're defaulting to 'q35' -- https://review.opendev.org/#/c/644565/49/nova/virt/libvirt/driver.py@5099 | |
| 12:35:07 | sean-k-mooney | i think raising an exception in both case is less surpising | |
| 12:35:23 | sean-k-mooney | but i am ment ot be getting food so i should go do that | |
| 12:35:26 | sean-k-mooney | brb | |
| 12:35:32 | aspiers | kashyap: Yes of course, otherwise SEV won't work without explicit hw_machine_type every time | |
| 12:36:06 | kashyap | aspiers: Okay, the function name confused me, then | |
| 12:36:21 | aspiers | Which name? | |
| 12:36:41 | kashyap | The _guest_machine_type() --> should this be named more SEV-specific | |
| 12:37:03 | kashyap | Something like ... _guest_machine_type_for_sev()? | |
| 12:37:13 | aspiers | Yes sure I could extract the big chunk out into _guest_machine_type_for_sev() | |
| 12:37:27 | aspiers | then _guest_machine_type() would call that | |
| 12:37:45 | aspiers | another non-inlined function to annoy stephenfin ;-) | |
| 12:38:14 | kashyap | aspiers: Yeah, thank you. So in _guest_machine_type() -- we can call the _for_sev() based on some SEV-specific conditional | |
| 12:38:27 | aspiers | It would use the exact same conditional which is already there | |
| 12:38:35 | aspiers | on line 5090 | |
| 12:38:41 | kashyap | Right, sev_enabled | |
| 12:38:49 | aspiers | and X86_64 | |
| 12:38:57 | aspiers | We're not supporting SEV on other architectures ;-) | |
| 12:38:57 | kashyap | Yep, agreed | |
| 12:39:01 | kashyap | Hehe | |
| 12:39:16 | aspiers | OK I'm going for a run to preserve sanity | |
| 12:39:21 | aspiers | biab | |
| 12:40:11 | kashyap | +1 | |
| 12:40:22 | kashyap | (On a run) | |
| 12:46:56 | openstackgerrit | Matthew Booth proposed openstack/nova master: libvirt: Fix service-wide pauses caused by un-proxied libvirt calls https://review.opendev.org/677736 | |
| 12:47:16 | openstackgerrit | Merged openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420 | |
| 12:47:27 | openstackgerrit | Merged openstack/nova master: trivial: Rewrap definitions of 'NUMACell' https://review.opendev.org/674395 | |
| 12:52:00 | openstackgerrit | Merged openstack/os-vif master: Fix code bug in document https://review.opendev.org/679382 | |
| 12:58:45 | kashyap | aspiers: When you're back, on a similar vein, I also to introduce such a method for Secure Boot: _guest_machine_type_for_secure_boot() | |
| 13:00:25 | openstackgerrit | Luyao Zhong proposed openstack/nova master: db: Add resources column in instance_extra table https://review.opendev.org/678447 | |
| 13:00:26 | openstackgerrit | Luyao Zhong proposed openstack/nova master: object: Introduce Resource and ResouceList objs https://review.opendev.org/678448 | |
| 13:00:26 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Add resources dict into _Provider https://review.opendev.org/678449 | |
| 13:00:27 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Retrieve the allocations early https://review.opendev.org/678450 | |
| 13:00:28 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Claim resources in resource tracker https://review.opendev.org/678452 | |
| 13:00:28 | openstackgerrit | Luyao Zhong proposed openstack/nova master: libvirt: Enable driver discovering PMEM namespaces https://review.opendev.org/678453 | |
| 13:00:28 | openstackgerrit | Luyao Zhong proposed openstack/nova master: libvirt: report VPMEM resources by provider tree https://review.opendev.org/678454 | |
| 13:00:29 | openstackgerrit | Luyao Zhong proposed openstack/nova master: libvirt: Support VM creation with vpmems and vpmems cleanup https://review.opendev.org/678455 | |
| 13:00:30 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Parse vpmem related flavor extra spec https://review.opendev.org/678456 | |
| 13:00:30 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Add functional tests for virtual persistent memory https://review.opendev.org/678470 | |
| 13:00:30 | openstackgerrit | Luyao Zhong proposed openstack/nova master: libvirt: Enable driver configuring PMEM namespaces https://review.opendev.org/679640 | |
| 13:26:24 | openstackgerrit | Chris Dent proposed openstack/nova master: single pass instance info fetch in host manager https://review.opendev.org/623558 | |
| 13:32:58 | cdent | efried, mriedem: I've still not managed to unit test that ^ but I did do some tweaks to the loop so that it yields more simply, as efried suggested. As I said on the patch, if someone else wants to poke at it that would be welcome, I've got other things leaking into my priorities :( | |
| 13:35:13 | openstackgerrit | Chris Dent proposed openstack/nova master: Tune up db.instance_get_all_uuids_by_hosts https://review.opendev.org/679627 | |
| 13:35:14 | openstackgerrit | Chris Dent proposed openstack/nova master: single pass instance info fetch in host manager https://review.opendev.org/623558 | |
| 13:52:25 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Refactor MigrationTask._execute https://review.opendev.org/679651 | |
| 13:55:12 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: unit test: do not fill rp mapping for failed re-schedule https://review.opendev.org/679653 | |
| 13:55:59 | bauzas | gibi: to make it clear, I'm about to +W https://review.opendev.org/#/c/655110/ but do you want me to hold until we fully review the series ? | |
| 13:56:20 | bauzas | gibi: I think we can safely merge the above given we don't call it yet | |
| 13:56:50 | gibi | bauzas:I have no objection against merging that :) | |
| 13:56:56 | bauzas | heh ok | |
| 13:57:19 | gibi | bauzas: and thank you | |
| 13:57:28 | luyao | dansmith: Are you around? | |
| 13:57:54 | bauzas | gibi: well, this one is easy | |
| 13:57:58 | bauzas | all the credits go to you | |
| 14:04:39 | bauzas | luyao: today's bank holiday for US folks https://en.wikipedia.org/wiki/Labor_Day | |
| 14:05:08 | openstackgerrit | sean mooney proposed openstack/nova master: [WIP] multi numa nfv testing job https://review.opendev.org/679656 | |
| 14:07:37 | luyao | bauzas: ok, thanks | |
| 14:08:38 | openstackgerrit | sean mooney proposed openstack/nova master: [WIP] multi numa nfv testing job https://review.opendev.org/679656 | |
| 14:10:41 | sean-k-mooney | donnyd: not sure if your around today but ^ should run against the FN multi numa lable and do some basic testing | |
| 14:11:06 | donnyd | Oh nice | |
| 14:11:21 | donnyd | I will be in and out | |
| 14:11:43 | sean-k-mooney | its basicaly my fedora job but swaped to use the new lable with some minor tweaks. | |
| 14:12:36 | sean-k-mooney | donnyd: no worries its a holiday in the us right. im going to spend some time today tweaking the job and hopefully it will be working when people get back tomorow | |
| 14:14:53 | donnyd | Please LMK if something isn't working right on my end | |
| 14:15:13 | sean-k-mooney | it seams to be workign fine so far http://zuul.openstack.org/stream/4778ebe053804dbb976917fd40c1a860?logfile=console.log | |
| 14:59:04 | sean-k-mooney | donnyd: so in the plus side some of the test that spawn vms are passing. on the minus side i for got to turn on the live migration tests | |
| 15:03:52 | donnyd | Do you think the extra memory helps? | |
| 15:05:28 | sean-k-mooney | am it does to a degree in that i can more easily allocate memory to hugepages | |
| 15:05:37 | sean-k-mooney | but i could make it work with 8G | |
| 15:06:11 | sean-k-mooney | donnyd: looks like the job is almost done | |
| 15:12:56 | sean-k-mooney | donnyd: uploading logs to swift takes a while | |
| 15:15:03 | sean-k-mooney | donnyd: https://d14c0e495b00eea27855-0acfe5e696fda6aec28b6c718c3c4841.ssl.cf2.rackcdn.com/679656/2/check/tempest-nfv-multi-numa-multinode/4778ebe/testr_results.html.gz :) thats not bad for a first try. | |
| 15:16:00 | aspiers | kashyap: what exception should be raised? | |
| 15:16:06 | donnyd | That's not too bad at all | |
| 15:17:21 | sean-k-mooney | i have a few bugs in my job definition mainly stuff i remvoed that i thought was not needed but is | |
| 15:17:29 | sean-k-mooney | once i fix that i think everythin should pass | |
| 15:18:20 | kashyap | aspiers: In a call. But how about an error on "Invalid MachineType" or some such? | |
| 15:18:35 | sean-k-mooney | the contoler node was deployed the way i wanted with kvm but the compute node is missing some of the setting | |
| 15:18:47 | aspiers | kashyap: you mean a new exception class? | |
| 15:19:06 | aspiers | InvalidMachineType? | |
| 15:19:23 | kashyap | aspiers: I just used CamelCase as an example. But but a simple error should do, no? | |
| 15:19:40 | aspiers | kashyap: I need to decide on an exception class | |
| 15:19:57 | aspiers | I can't just raise a string | |
| 15:20:01 | aspiers | TypeError: exceptions must derive from BaseException | |
| 15:20:10 | kashyap | aspiers: Right, InvalidMachineType sounds reasonable, then? | |
| 15:20:14 | aspiers | so I'm not sure what you mean by "simple error" | |
| 15:20:24 | aspiers | Yeah I could add that | |
| 15:20:32 | kashyap | Forgive; split attention | |
| 15:22:41 | kashyap | aspiers: Mind alerting me when you post that? | |
| 15:33:54 | aspiers | sure | |