| *** jparker is now known as jparker|afk | 01:48 | |
| *** bhagyashris is now known as bhagyashris|ruck | 05:46 | |
| *** amoralej|off is now known as amoralej | 06:05 | |
| opendevreview | ZhouYanbing proposed openstack/nova master: correct the wrong content in the notes https://review.opendev.org/c/openstack/nova/+/838205 | 06:30 |
|---|---|---|
| bauzas | good morning Nova | 07:34 |
| gibi | bauzas: o/ | 07:34 |
| * bauzas is a bit on and off this morning, going back from Lyon by train | 07:34 | |
| elodilles | o/ | 07:38 |
| bauzas | gibi: damn, we forgot to ask to add the bp link for https://specs.openstack.org/openstack/nova-specs/specs/zed/approved/owner-nova-trait-usage.html | 07:55 |
| gibi | just push a fup and I can approve it | 07:56 |
| bauzas | https://blueprints.launchpad.net/nova/+specs?searchtext=trait | 07:57 |
| bauzas | can't see any | 07:57 |
| bauzas | maybe wenping forgot to create one | 07:57 |
| bauzas | https://launchpad.net/~wenping1/+specs?role=assignee | 07:58 |
| bauzas | none of them are for the spec | 07:58 |
| bauzas | ok, then I'll create a bp and i'll assign it to wenping | 08:01 |
| gibi | everybody loves paperwork ;) | 08:01 |
| bauzas | gibi: should I say I'm French ? | 08:07 |
| gibi | no need :) | 08:07 |
| bauzas | ok, https://blueprints.launchpad.net/nova/+spec/owner-nova-trait-usage is created | 08:10 |
| * bauzas works on a FUP for the spec adding the link | 08:11 | |
| opendevreview | Sylvain Bauza proposed openstack/nova-specs master: Add the BP link to owner-nova-trait-usage spec https://review.opendev.org/c/openstack/nova-specs/+/839159 | 08:16 |
| bauzas | gibi: sean-k-mooney: ^ | 08:16 |
| gibi | bauzas: +A | 08:19 |
| bauzas | thanks | 08:27 |
| *** tkajinam is now known as tkajinam|away | 08:33 | |
| opendevreview | Merged openstack/nova-specs master: Add the BP link to owner-nova-trait-usage spec https://review.opendev.org/c/openstack/nova-specs/+/839159 | 08:43 |
| sean-k-mooney1 | ah already merged | 09:10 |
| *** sean-k-mooney1 is now known as sean-k-mooney | 09:10 | |
| sean-k-mooney | bauzas: congratulations on still having a republic by the way | 09:11 |
| bauzas | sean-k-mooney: well, indeed even if we have far-left and far-right for more than 60%... | 09:15 |
| bauzas | actually, if we were not having a republic, then we would only have one party, I guess... | 09:17 |
| sean-k-mooney | ya but fortunetly your electerit is able to put there difference aside when they know infiting will result in the extream end winning | 09:17 |
| bauzas | sean-k-mooney: at least, for one month | 09:18 |
| bauzas | sean-k-mooney: we will have a new election for parliament MPs in June | 09:19 |
| bauzas | if the parliament changes, this would mean that we would have a new government | 09:19 |
| bauzas | even if the president is the same | 09:19 |
| sean-k-mooney | thats true well best of luck in june so | 09:21 |
| Uggla | hello folks, little question, what is the difference between an instance in pause vs suspend ? | 09:49 |
| vbel | good morning/evening, I'm looking into various ways to migrate VM disks from one storage to another (Ceph/LVM/NFS/...) and currently I see only two possibilities - 1) cinder retype of Available disks 2) cinder retype of In-use with VMs in Active state. I wonder why shut off VMs cannot be retyped | 10:13 |
| gibi | Uggla: as a starting point: https://github.com/openstack/nova/blob/1ff89a09d45efadb592e6becab872ec1f7ac9894/nova/virt/driver.py#L779-L786 https://github.com/openstack/nova/blob/1ff89a09d45efadb592e6becab872ec1f7ac9894/nova/virt/driver.py#L812-L818 | 10:13 |
| gibi | Uggla: I think in case of the libvirt virt driver pause is just stop the cpus but does not move stuff out from ram. In the other hand the suspend virt driver action will call libvirts managedSave capability https://libvirt.org/sources/virshcmdref/html/sect-managedsave.html | 10:16 |
| gibi | Uggla: just to complicate thing the pause virt driver action calls libvirt's suspend call https://libvirt.org/docs/libvirt-appdev-guide-python/en-US/html/libvirt_application_development_guide_using_python-Guest_Domains-Lifecycle-Save.html | 10:16 |
| sean-k-mooney | Uggla: tl;dr is susped is like hybernate. it suspends the vm to disk saving its ram to a file | 10:17 |
| sean-k-mooney | Uggla: pause just stops the vcpu process | 10:17 |
| sean-k-mooney | so the qemu isntastance is still running | 10:17 |
| sean-k-mooney | but the cpus are not so all memory and devices are still attached to teh qemu process | 10:17 |
| sean-k-mooney | Uggla: thinks of pause like your laptops sleep or "suspend to ram" in linux | 10:18 |
| sean-k-mooney | Uggla: in general neither should be used in a cloud you shoud generally shelve the vm or just stop it | 10:19 |
| sean-k-mooney | the usecase for using suspend and pause in a cloud exist but are very limitted | 10:19 |
| vbel | commit b40d949b3137473227949b597b2a61da41752ee5 explicitly prohibits any swap volume of supspended/stopped/resized/soft-deleted | 10:20 |
| Uggla | gibi, sean-k-mooney thanks | 10:21 |
| sean-k-mooney | vbel: in some cases i coudl see it being a problem depenidn on if the connection mechanisuem changes | 10:23 |
| sean-k-mooney | stopped i think shoudl be ok | 10:23 |
| sean-k-mooney | im kind of suprised we allow live retype but not stopped | 10:23 |
| sean-k-mooney | when you say resized do you mena resize_verify? | 10:24 |
| sean-k-mooney | resized is nto a vm state | 10:24 |
| sean-k-mooney | wehn we stop the vm i think the volume goes into reserved | 10:25 |
| sean-k-mooney | which is likely why we cant retype then | 10:25 |
| sean-k-mooney | same for suspend i suspect | 10:25 |
| *** amoralej is now known as amoralej|lunch | 12:27 | |
| gibi | artom__: I'm +2 on https://review.opendev.org/c/openstack/nova/+/839091 now with a hint for an extra assert if you ever have to respin this | 12:34 |
| elodilles | bauzas sean-k-mooney: just a heads up, victoria-em deadline is in 2 days & there's no new merged patches in stable/victoria (except the l-c job drop patch, but that is CI-only)... so I think we could just approve the transition patch as there is hardly any time to merge and release from stable/victoria | 13:00 |
| bauzas | wfm | 13:01 |
| gibi | bauzas: the neutron side of this bugfix now merged, could you look at the nova side? https://review.opendev.org/c/openstack/nova/+/829248 sean-k-mooney is already +2 on it | 13:03 |
| * bauzas looks | 13:04 | |
| bauzas | ouch. | 13:04 |
| bauzas | gibi: going first to the coffee machine then :p | 13:04 |
| gibi | make a strong one :) | 13:04 |
| * bauzas woke up at 5.30am this morning and just wants to make sure his mind is still good :) | 13:05 | |
| elodilles | bauzas: thanks, then please review the victoria-em patch o:) this one >>> https://review.opendev.org/c/openstack/releases/+/837953 | 13:05 |
| kashyap | bauzas: gibi: I'm trying decaf (I know, I know) in the afternoons | 13:05 |
| kashyap | Some days it works, other days not so. | 13:06 |
| gibi | I just finished my cold brew so no more coffein to me today | 13:06 |
| bauzas | kashyap: surely it can work for some people, but not me :) | 13:06 |
| kashyap | gibi: I should learn to make cold brews, the for the reminder | 13:06 |
| bauzas | I mean, I can pee for sure if I try decaf | 13:06 |
| bauzas | but then, I won't have the same for a coffee :) | 13:07 |
| gibi | kashyap: sadly it was from a coffee shop, I also only plan to learn how to make a good cold one at home | 13:07 |
| kashyap | gibi: If you haven't tried yet, get an AeroPress! It's the "most successful gadget in my home" award - at a paltry $45 | 13:08 |
| kashyap | I first dimissed it as cheap plastic crap; but it's been my daily driver for a couple of years | 13:09 |
| gibi | it is on my list too but now your recommendation boosted it up on the list :L) | 13:09 |
| kashyap | gibi: It's like French Press, but without the soot at the bottom | 13:10 |
| bauzas | fwiw, I love my Mignifica S coffee machine :) | 13:11 |
| kashyap | gibi: A single AeroPress in a cafe costs about 5 euros here! If I buy one. In 10 days the device will start paying for itself! | 13:11 |
| kashyap | :) | 13:11 |
| bauzas | Magnifica sorry | 13:11 |
| gibi | I talked the barista in the coffee shop at the corner they do it with soaking during the night | 13:11 |
| kashyap | gibi: Yeah, that's the "real" cold brew, indeed | 13:11 |
| kashyap | I only make hot-brew w/ AeroPress, but cold is also possible, as the instructions point it out | 13:12 |
| gibi | kashyap: totally true about the cost of shop coffee. I rearly visit the shop. I have my bialetti for mokka | 13:12 |
| bauzas | I can do cold brew with my machine... I just need to wait for 10 mins after creating a coffee :p | 13:13 |
| kashyap | LOL, it's not the same as overnight soaking | 13:13 |
| bauzas | I know | 13:13 |
| bauzas | but, | 13:13 |
| kashyap | gibi: I work a lot from coffee shops (to beat the cabin fever) ... and probably spend more than I should on coffees. | 13:13 |
| gibi | bauzas: you can try waiting 12 hours after you brew with a machine :) | 13:13 |
| bauzas | I prefer to wait 12 hours for bbq :) | 13:13 |
| gibi | kashyap: btw here a good cold brew costs ~2.5 EUR, but still :D | 13:15 |
| * kashyap nods | 13:15 | |
| artom__ | gibi, ack, I saw the review email come in, cheers! | 13:17 |
| *** artom__ is now known as artom | 13:17 | |
| *** amoralej|lunch is now known as amoralej | 13:18 | |
| bauzas | elodilles: you left me puzzled https://review.opendev.org/c/openstack/releases/+/837953 | 13:31 |
| elodilles | bauzas: well, i heard in the past that unreleased .gitreview and TOX_CONSTRAINTS update patches caused problems in RH downstream. in general that is not worth to create a release (no change in package at all), but with a 'patch' version bump maybe it does not hurt | 13:37 |
| bauzas | elodilles: honestly, we don't really care of stable/victoria :) | 13:38 |
| bauzas | we = my employer | 13:38 |
| bauzas | sean-k-mooney: already heard of this ^ ? | 13:39 |
| sean-k-mooney | ooo does not care | 13:40 |
| * sean-k-mooney but one sec reads back | 13:40 | |
| sean-k-mooney | elodilles: downstream we do override the .gitignore but i dont think it should break us | 13:41 |
| sean-k-mooney | elodilles: if it did it would be trivial to fix so i would always do the right thing upstream | 13:41 |
| elodilles | bauzas sean-k-mooney : ack, then i think they are good as they are :) | 13:42 |
| sean-k-mooney | elodilles: i think we can safely merge the .gitreview change and _not_ release it then apply the EM tag based on the sha but if we want the EM tag to point to the last release that is also ok | 13:42 |
| bauzas | sean-k-mooney: the patch is already merged | 13:43 |
| elodilles | sean-k-mooney: *-em tags always tha *last* release from the branch | 13:43 |
| bauzas | the question is, should we include it as part of the -em tag, that's it | 13:43 |
| bauzas | as we can't release a package after, like elodilles said | 13:44 |
| bauzas | while I don't care of the CI changes as elodilles noted, I'm a bit torn whether we should include the propre gitreview target in the last build | 13:45 |
| bauzas | "build" rather with quotes | 13:45 |
| elodilles | from package point of view they are not included (pypi, tarball) so the release would just be unnecessary fuss for package consumers | 13:45 |
| sean-k-mooney | elodilles: right but teh way the em tag is ment to work is it is ment ot allow you to restore the branch | 13:46 |
| sean-k-mooney | so the em tag shoudl be on the last patch in the stable branch | 13:46 |
| bauzas | then we should include the gitreview patch | 13:46 |
| elodilles | sean-k-mooney: well, no, that's the *-eol you mean i think :) | 13:47 |
| sean-k-mooney | oh sorry you are right | 13:47 |
| sean-k-mooney | i was thinking of eol | 13:47 |
| sean-k-mooney | for em we are not removing the branch | 13:47 |
| sean-k-mooney | then ya its fine as is | 13:47 |
| sean-k-mooney | no need to do another release | 13:47 |
| sean-k-mooney | bauzas: do you agree? | 13:48 |
| bauzas | hmmmm | 13:48 |
| bauzas | I'm not *that* opiniated, I just want to be convinced | 13:49 |
| bauzas | what's the purpose of the -em tag ? | 13:49 |
| sean-k-mooney | really just to signal that no more releases will be made | 13:49 |
| elodilles | yepp | 13:52 |
| bauzas | that's my understanding | 13:54 |
| bauzas | now, looking at https://review.opendev.org/q/project:openstack/python-novaclient+branch:stable/victoria | 13:55 |
| bauzas | we only have 2 patches in the branch :) | 13:55 |
| bauzas | one isn't required | 13:55 |
| bauzas | the other one just updates the gitreview target | 13:55 |
| elodilles | yepp, those are not part of the packages (even if we tag and release them) | 14:01 |
| bauzas | elodilles: OK, then I'll change my vote | 14:07 |
| opendevreview | Balazs Gibizer proposed openstack/nova master: Simulate bug 1969496 https://review.opendev.org/c/openstack/nova/+/838554 | 14:25 |
| opendevreview | Balazs Gibizer proposed openstack/nova master: Allow claiming PCI PF if child VF is unavailable https://review.opendev.org/c/openstack/nova/+/838555 | 14:25 |
| gibi | sean-k-mooney: finalized the fix ^^ | 14:26 |
| opendevreview | Merged openstack/nova master: VMware: Split out VMwareAPISession https://review.opendev.org/c/openstack/nova/+/832156 | 14:27 |
| sean-k-mooney | gibi: ack ill add that to my review list so for this week | 14:27 |
| gibi | thanks! | 14:28 |
| sean-k-mooney | ah nice you have all 3 parts. the repoduce, allowing it to claim and also fixing the inconsitent state | 14:28 |
| gibi | yepp | 14:28 |
| sean-k-mooney | gibi: is there a reason you did the simulation second in the series | 14:36 |
| sean-k-mooney | i woudl have expected the startup patch to be last | 14:37 |
| sean-k-mooney | so simulate, claim then start up | 14:37 |
| sean-k-mooney | it works this way too just wondering | 14:37 |
| gibi | the reason was that I started with teh cleanup path as it was easier to get to it. but no technical dependency so I can reorder if needed | 14:38 |
| sean-k-mooney | no thats ok in principal once the start up check is merged you dont need the others | 14:39 |
| sean-k-mooney | so that is why i was suggesting reordering | 14:40 |
| sean-k-mooney | but i like having all 3 | 14:40 |
| sean-k-mooney | (you would have to restart the agent to get this code hench the startup cleanup is enough) | 14:40 |
| *** tkajinam|away is now known as tkajinam | 14:42 | |
| gibi | the current cleanup logic only works if the admin removes the VFs from the hypervisor first, then restart the agent. So if that removal (of the restart) is not possible then the claim-while-child-unavailable patch helps | 14:42 |
| sean-k-mooney | ah you are right | 14:42 |
| gibi | or I could create a wider cleanup logic that does not require the VF removal from the hypervisor | 14:42 |
| gibi | just push the VF to available at startup | 14:42 |
| gibi | regardles if it is removed or not | 14:43 |
| gibi | and do this before the existing removed device cleanup logic | 14:43 |
| gibi | we have full of options :) | 14:43 |
| sean-k-mooney | im reviewing it now so unless others object lets keep this simple and go with what you have | 14:45 |
| sean-k-mooney | we might want to do a larger cleanup later but this looks pretty backportable to me | 14:45 |
| gibi | yepp it is pretty self contains now | 14:46 |
| gibi | I will do the backports too | 14:46 |
| gibi | once it lands on master | 14:46 |
| opendevreview | Artom Lifshitz proposed openstack/nova master: Reproduce live migration rollback w/o multi port bindings error https://review.opendev.org/c/openstack/nova/+/839091 | 14:50 |
| opendevreview | Artom Lifshitz proposed openstack/nova master: Fix LM rollback w/o multi port bindings extension https://review.opendev.org/c/openstack/nova/+/839227 | 14:50 |
| *** amoralej is now known as amoralej|off | 15:49 | |
| opendevreview | Andre Aranha proposed openstack/nova master: Test setting the nova job to centos-9-stream https://review.opendev.org/c/openstack/nova/+/831844 | 15:56 |
| sean-k-mooney | ah fips | 16:13 |
| sean-k-mooney | i was like which job | 16:14 |
| opendevreview | Dan Smith proposed openstack/nova master: DNM: Run against performance.json patch https://review.opendev.org/c/openstack/nova/+/838934 | 16:30 |
| melwitt | gmann: this is a small patch involving deprecation of [api] config option that you might be interested in reviewing https://review.opendev.org/c/openstack/nova/+/836253 | 17:45 |
| gmann | melwitt: thanks, +A. lgtm. | 18:05 |
| melwitt | thanks gmann | 18:13 |
| opendevreview | Jakub Libosvar proposed openstack/os-vif master: Check for hybrid plugging in OVS https://review.opendev.org/c/openstack/os-vif/+/839102 | 18:32 |
| opendevreview | Merged openstack/nova master: Deprecate [api] use_forwarded_for https://review.opendev.org/c/openstack/nova/+/836253 | 19:39 |
| *** dasm is now known as dasm|off | 22:06 | |
| opendevreview | Ghanshyam proposed openstack/nova master: Move centos stream testing to centos-9-stream https://review.opendev.org/c/openstack/nova/+/839275 | 23:15 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!