imacdonn | not sure what what basic_publish_confirm message really means | 00:02 |
---|---|---|
melwitt | I don't either but just the indication that it made an attempt to send the message | 00:02 |
melwitt | I guess that implies it for some reason couldn't get through and results in the timeout? | 00:03 |
imacdonn | basic_publish_confirm is the method (that's supposed to get autoretry'ed), I guess | 00:05 |
imacdonn | I think it comes from python-amqp (?) | 00:06 |
*** zer0c00l_ has quit IRC | 00:06 | |
imacdonn | File "/usr/lib/python2.7/site-packages/amqp/channel.py", line 1787, in basic_publish_confirm | 00:07 |
imacdonn | it does seem like it calls it the second time (at 23:48:19.113), but apparently still doesn't get an answer | 00:10 |
*** slaweq has joined #openstack-nova | 00:11 | |
melwitt | yeah, that's the one I was looking at, the second that times out even after re-establishing the connection | 00:12 |
openstackgerrit | Dustin Cowles proposed openstack/nova master: Use SDK instead of ironicclient for setting instance id https://review.opendev.org/659690 | 00:13 |
openstackgerrit | Dustin Cowles proposed openstack/nova master: WIP: Use SDK instead of ironicclient for add/remove instance info from node https://review.opendev.org/659691 | 00:15 |
*** slaweq has quit IRC | 00:16 | |
*** zer0c00l has joined #openstack-nova | 00:16 | |
*** zer0c00l has quit IRC | 00:17 | |
*** zer0c00l has joined #openstack-nova | 00:17 | |
*** Guest28904 has quit IRC | 00:34 | |
openstackgerrit | Merged openstack/nova master: Stop logging traceback when skipping quiesce https://review.opendev.org/659374 | 00:34 |
*** nicolasbock has quit IRC | 00:49 | |
*** tbachman has joined #openstack-nova | 01:00 | |
imacdonn | another data-point: the retried request DOES actually get sent, because I see the compute node act upon it ... I guess somehow the client (nova-api) has some issue with the response | 01:08 |
*** hongbin has joined #openstack-nova | 01:11 | |
*** slaweq has joined #openstack-nova | 01:11 | |
*** mmethot has quit IRC | 01:12 | |
*** guozijn has joined #openstack-nova | 01:15 | |
*** slaweq has quit IRC | 01:15 | |
*** ricolin has joined #openstack-nova | 01:22 | |
*** ricolin has quit IRC | 01:30 | |
*** brinzhang has joined #openstack-nova | 01:44 | |
*** mmethot has joined #openstack-nova | 01:55 | |
*** ricolin has joined #openstack-nova | 01:56 | |
*** ricolin has quit IRC | 02:04 | |
*** tinwood has quit IRC | 02:10 | |
*** slaweq has joined #openstack-nova | 02:11 | |
*** tinwood has joined #openstack-nova | 02:12 | |
*** whoami-rajat has joined #openstack-nova | 02:12 | |
*** slaweq has quit IRC | 02:15 | |
*** JamesBenson has joined #openstack-nova | 02:31 | |
openstackgerrit | chenker proposed openstack/python-novaclient master: Optimize limit option docs string description for novaclient https://review.opendev.org/643035 | 03:02 |
openstackgerrit | Guo Jingyu proposed openstack/nova master: Skip existing VMs when hosts apply forceconfig_drive https://review.opendev.org/659703 | 03:03 |
*** guozijn has quit IRC | 03:04 | |
*** guozijn has joined #openstack-nova | 03:06 | |
*** ttsiouts has joined #openstack-nova | 03:08 | |
*** yonglihe has joined #openstack-nova | 03:12 | |
*** cfriesen has quit IRC | 03:13 | |
*** awalende has joined #openstack-nova | 03:13 | |
openstackgerrit | Guo Jingyu proposed openstack/nova master: Skip existing VMs when hosts apply force_config_drive https://review.opendev.org/659703 | 03:17 |
*** awalende has quit IRC | 03:17 | |
*** hongbin has quit IRC | 03:24 | |
*** guozijn has quit IRC | 03:27 | |
*** gaoyan has joined #openstack-nova | 03:28 | |
*** guozijn has joined #openstack-nova | 03:30 | |
*** ttsiouts has quit IRC | 03:41 | |
*** psachin has joined #openstack-nova | 03:41 | |
*** lbragstad has quit IRC | 03:53 | |
*** JamesBenson has quit IRC | 03:59 | |
*** ratailor has joined #openstack-nova | 04:01 | |
*** slaweq has joined #openstack-nova | 04:11 | |
*** slaweq has quit IRC | 04:16 | |
*** gaoyan has quit IRC | 04:31 | |
*** udesale has joined #openstack-nova | 04:34 | |
*** ykarel has joined #openstack-nova | 04:39 | |
ykarel | TheJulia, dtantsur|afk can you check comment in https://review.opendev.org/#/c/653279/1 | 04:46 |
ykarel | hmm looks like some issue is already there, as i can see https://review.opendev.org/#/c/659612/1, can you check if this ^^ is same issue or something else | 04:48 |
*** bhagyashris has joined #openstack-nova | 04:54 | |
*** slaweq has joined #openstack-nova | 05:11 | |
*** gaoyan has joined #openstack-nova | 05:14 | |
*** slaweq has quit IRC | 05:16 | |
*** pcaruana has joined #openstack-nova | 05:20 | |
*** brinzhang has quit IRC | 05:21 | |
*** brinzhang has joined #openstack-nova | 05:22 | |
*** slaweq has joined #openstack-nova | 05:24 | |
openstackgerrit | Guo Jingyu proposed openstack/nova master: Add rfb.VNC support for novncproxy https://review.opendev.org/622336 | 05:25 |
openstackgerrit | Guo Jingyu proposed openstack/nova-specs master: Proposal for a safer noVNC console with password authentication https://review.opendev.org/623120 | 05:26 |
*** gaoyan has quit IRC | 05:29 | |
*** ttsiouts has joined #openstack-nova | 05:39 | |
*** slaweq has quit IRC | 05:45 | |
*** guozijn has quit IRC | 05:50 | |
*** guozijn has joined #openstack-nova | 05:50 | |
*** ivve has quit IRC | 05:50 | |
*** Luzi has joined #openstack-nova | 05:52 | |
*** lpetrut has joined #openstack-nova | 06:05 | |
*** faizy98 has quit IRC | 06:07 | |
*** sapd1_x has joined #openstack-nova | 06:10 | |
*** ttsiouts has quit IRC | 06:12 | |
openstackgerrit | Guo Jingyu proposed openstack/nova master: Skip existing VMs when hosts apply force_config_drive https://review.opendev.org/659703 | 06:21 |
*** maciejjozefczyk has joined #openstack-nova | 06:27 | |
*** belmoreira has joined #openstack-nova | 06:32 | |
*** guozijn has quit IRC | 06:39 | |
*** guozijn has joined #openstack-nova | 06:40 | |
*** ykarel is now known as ykarel|afk | 06:41 | |
*** slaweq has joined #openstack-nova | 06:52 | |
*** belmoreira has quit IRC | 06:56 | |
*** ivve has joined #openstack-nova | 06:58 | |
*** luksky has joined #openstack-nova | 06:59 | |
*** tesseract has joined #openstack-nova | 07:08 | |
*** awalende has joined #openstack-nova | 07:11 | |
*** awalende has quit IRC | 07:13 | |
*** ykarel|afk is now known as ykarel | 07:16 | |
*** awalende has joined #openstack-nova | 07:17 | |
*** rcernin has quit IRC | 07:19 | |
*** jaosorior has quit IRC | 07:19 | |
*** rpittau|afk is now known as rpittau | 07:22 | |
*** ccamacho has joined #openstack-nova | 07:27 | |
*** dtantsur|afk is now known as dtantsur | 07:31 | |
*** xek has joined #openstack-nova | 07:43 | |
*** luksky has quit IRC | 07:45 | |
*** mcgigglier has joined #openstack-nova | 07:47 | |
*** belmoreira has joined #openstack-nova | 07:53 | |
*** belmoreira has quit IRC | 07:57 | |
*** ykarel is now known as ykarel|lunch | 08:00 | |
kashyap | aspiers: That's in queue, I'm still backlogged :-( I know it is ready to merge. If I look at, I would like to spend a 2-hour block on it. So trying to find that :P | 08:01 |
kashyap | aspiers: From your comment: | 08:01 |
kashyap | "I don't quite understand how the fix in https://www.redhat.com/archives/libvir-list/2019-April/msg01418.html would make the Python arguments optional however, since they're still missing defaults." | 08:01 |
*** ttsiouts has joined #openstack-nova | 08:01 | |
kashyap | aspiers: Just write to the upstream list (no subscription required; you can ask to be Cced) | 08:01 |
kashyap | aspiers: But as you noted yourself it's not required, given that you've obtained correct value for @emulatorbin via "Parse <emulator> elements from virConnectGetCapabilities()" | 08:02 |
kashyap | aspiers: Actually, let me give my review. You've addressed all my concerns | 08:03 |
*** lpetrut has quit IRC | 08:09 | |
*** ralonsoh has joined #openstack-nova | 08:12 | |
*** belmoreira has joined #openstack-nova | 08:13 | |
*** luksky has joined #openstack-nova | 08:19 | |
*** sapd1_x has quit IRC | 08:21 | |
aspiers | kashyap: I'm backlogged too ;-) | 08:24 |
*** awalende has quit IRC | 08:24 | |
aspiers | kashyap: So I probably owe you some reviews too. Let me know if I can make the review process easier in any way. | 08:24 |
kashyap | aspiers: Thanks for the offer! I need to do some work similar to yours -- introducing new CPU-related APIs | 08:26 |
kashyap | aspiers: For this spec: https://opendev.org/openstack/nova-specs/src/branch/master/specs/train/approved/cpu-selection-with-hypervisor-consideration.rst | 08:26 |
kashyap | See the action items at the end. | 08:27 |
*** ccamacho has quit IRC | 08:27 | |
*** wxy-xiyuan has left #openstack-nova | 08:27 | |
aspiers | ok | 08:28 |
*** lpetrut has joined #openstack-nova | 08:29 | |
*** derekh has joined #openstack-nova | 08:32 | |
*** awalende has joined #openstack-nova | 08:37 | |
*** tssurya has joined #openstack-nova | 08:38 | |
*** tkajinam has quit IRC | 08:44 | |
kashyap | aspiers: ACKed your change, FWIW | 08:45 |
aspiers | kashyap: yep, thanks! | 08:45 |
kashyap | aspiers: I think your change is ready to merge. Let's check with efried (or stephenfin) when they're about. | 08:45 |
kashyap | aspiers: Hope it won't be delayed, so that you can maintain the momentum. | 08:46 |
aspiers | Well it's already been delayed a few months ;) but thankfully git makes this much less of an issue. I remember the awful CVS days where the only place for all your pending unmerged work was mushed up together in the work tree, and you had to untangle each new commit-to-be from that | 08:49 |
aspiers | Drove me crazy | 08:49 |
*** ykarel|lunch is now known as ykarel | 08:50 | |
*** davidsha has joined #openstack-nova | 08:53 | |
jangutter | aspiers: The absolute _best_ thing about git is when you learn about rebasing. And the nice thing about gerrit is it fills in the gaps (tracking iterations of the same "patch queue"). | 08:56 |
aspiers | jangutter: agreed :) git has lots of "best" things though, e.g. I remember when I learnt about the reflog it blew my mind | 08:56 |
aspiers | ditto git-rerere | 08:56 |
aspiers | so many cool tricks | 08:57 |
jangutter | aspiers: who knew that a content-addressable file system accidentally solved the "version control" issue :-p | 08:58 |
aspiers | :D | 08:58 |
*** ccamacho has joined #openstack-nova | 08:59 | |
*** ttsiouts has quit IRC | 09:04 | |
*** belmoreira has quit IRC | 09:08 | |
*** belmoreira has joined #openstack-nova | 09:10 | |
*** panda|rover|off is now known as panda|rover | 09:10 | |
*** sapd1_x has joined #openstack-nova | 09:13 | |
*** wwriverrat has quit IRC | 09:28 | |
*** bhagyashris has quit IRC | 09:36 | |
*** sapd1_x has quit IRC | 09:36 | |
*** belmoreira has quit IRC | 09:47 | |
*** awalende has quit IRC | 09:49 | |
*** sapd1_x has joined #openstack-nova | 09:49 | |
*** awalende has joined #openstack-nova | 09:49 | |
*** rcernin has joined #openstack-nova | 09:50 | |
*** awalende has quit IRC | 09:50 | |
*** mdbooth has quit IRC | 09:50 | |
*** awalende has joined #openstack-nova | 09:50 | |
*** belmoreira has joined #openstack-nova | 09:51 | |
*** mdbooth has joined #openstack-nova | 09:56 | |
*** sapd1_x has quit IRC | 09:59 | |
*** belmoreira has quit IRC | 10:00 | |
*** ykarel is now known as ykarel|afk | 10:03 | |
kashyap | aspiers: Yeah, I see what you mean. I didn't have the misfortune to use CVS, but I did start with SVN, though | 10:15 |
aspiers | I started with RCS :-o | 10:16 |
*** belmoreira has joined #openstack-nova | 10:21 | |
*** guozijn has quit IRC | 10:23 | |
*** awalende has quit IRC | 10:24 | |
*** brinzhang has quit IRC | 10:28 | |
*** cdent has joined #openstack-nova | 10:30 | |
kashyap | Hope you've fully recovered | 10:34 |
aspiers | LOL | 10:35 |
aspiers | Actually RCS can be surprisingly effective for particular use cases | 10:35 |
aspiers | but yeah, it's very limited | 10:35 |
aspiers | I actually talked a bit about the evolutionary history of version control systems in this interview http://episodes.gitminutes.com/2015/03/gitminutes-32-adam-spiers-on-git-deps.html | 10:36 |
aspiers | according to the index that starts around 07:08 | 10:37 |
*** whoami-rajat has quit IRC | 10:42 | |
*** tbachman has quit IRC | 10:47 | |
*** awalende has joined #openstack-nova | 10:48 | |
*** sridharg has joined #openstack-nova | 10:48 | |
*** belmoreira has quit IRC | 10:49 | |
cdent | What are the options for a VM after it is evacuated? I ask in relation to this new bug: https://bugs.launchpad.net/nova/+bug/1829479 | 10:52 |
openstack | Launchpad bug 1829479 in OpenStack Compute (nova) "The allocation table has residual records when instance is evacuated and the source physical node is removed" [Undecided,New] | 10:52 |
*** jaosorior has joined #openstack-nova | 10:55 | |
*** ykarel|afk is now known as ykarel | 10:56 | |
*** jaosorior has quit IRC | 11:00 | |
*** ttsiouts has joined #openstack-nova | 11:01 | |
*** guozijn has joined #openstack-nova | 11:01 | |
*** jaosorior has joined #openstack-nova | 11:09 | |
*** jcosmao has joined #openstack-nova | 11:18 | |
sean-k-mooney | cdent: in what sense | 11:20 |
sean-k-mooney | cdent: if it is evacuated sucessfully its as if ti was migrated | 11:20 |
sean-k-mooney | cdent: assuming it was on shared storage its state was not even lost | 11:21 |
sean-k-mooney | cdent: althogh if it was on epheraml stoage it wasd effectivly a new instance | 11:21 |
cdent | sean-k-mooney: So evacuate means "recreate this vm somewhere else"? | 11:21 |
sean-k-mooney | cdent: yes | 11:21 |
sean-k-mooney | but keep the same ips and volumes | 11:21 |
cdent | So shouldn't that mean that when the evacuate finishes an allocation is made which replaces the old allocations? | 11:22 |
artom | Rebuild on a different host, basically | 11:22 |
artom | So yeah, allocations need to move | 11:22 |
sean-k-mooney | so if its boot from volume or you hapeend to have the rbd image backend you keep you root disk | 11:22 |
aspiers | PLEASE can we use the word "resurrect" instead of evacuate wherever possible | 11:22 |
sean-k-mooney | aspiers: no | 11:22 |
sean-k-mooney | :P | 11:22 |
artom | aspiers, time to introduce you to this classic: http://www.danplanet.com/blog/2016/03/03/evacuate-in-nova-one-command-to-confuse-us-all/ | 11:23 |
aspiers | the word "evacuate" implies the VM is still there, but in these cases it is not | 11:23 |
sean-k-mooney | yep ^ | 11:23 |
aspiers | artom: I read that years ago ;-) | 11:23 |
sean-k-mooney | aspiers: it is for things with ephemeral storage | 11:23 |
sean-k-mooney | * non ephmeral storage | 11:23 |
sean-k-mooney | aspiers: its basically just a cold migration for that case | 11:23 |
*** awalende has quit IRC | 11:23 | |
aspiers | the implementation is like cold migration, but semantically it is not | 11:24 |
sean-k-mooney | its when you use the default image backend e.g. qcow2 and its not on shared stoage that we acrully recreate it from glance and loose data | 11:24 |
cdent | sean-k-mooney: so in that case either the person at bug https://bugs.launchpad.net/nova/+bug/1829479 is on an old version (as I thought we had fixed migration-like allocations bugs) or we've got remaining migration-like allocations bugs | 11:24 |
openstack | Launchpad bug 1829479 in OpenStack Compute (nova) "The allocation table has residual records when instance is evacuated and the source physical node is removed" [Undecided,New] | 11:24 |
sean-k-mooney | cdent: we proably have a remaing bug | 11:25 |
aspiers | artom: here's another "classic" for you ;-) https://youtu.be/lddtWUP_IKQ?t=751 | 11:26 |
sean-k-mooney | cdent: im not sure if we actullly do the whole create a migration contex use it for allocation and swap at the end | 11:26 |
artom | aspiers, oh, hah | 11:26 |
aspiers | sean-k-mooney, cdent: that video will help you understand where I'm coming from too | 11:26 |
artom | aspiers, this is like the time I naively asked about your experience with Gerrit, isn't it :P | 11:27 |
aspiers | artom: haha, maybe a bit ;-) | 11:27 |
* artom stops talking until coffee | 11:27 | |
aspiers | I've been working on compute HA in nova for about 4-5 years | 11:27 |
sean-k-mooney | cdent: or it could be that we delete the old allocation on the source node and that does not run in the evacuate case | 11:27 |
aspiers | artom: but don't feel bad, there's no good reason why you should have known that ;) | 11:27 |
cdent | that's what the bug report is saying, yes | 11:28 |
sean-k-mooney | aspiers: that was a poor life choice :P | 11:28 |
aspiers | sean-k-mooney: LOL, yeah probably ;-) | 11:28 |
artom | I see to have a lot of feet in my mouth, maybe wash them out with aforementioned coffee | 11:28 |
artom | *seem | 11:28 |
aspiers | sean-k-mooney: I mean, not *exclusively* compute HA ... I've done a lot of other stuff too ;) | 11:28 |
sean-k-mooney | aspiers: so for https://review.opendev.org/#/c/643578/3 back to the ptg session your prefere is openstack server resuerect | 11:29 |
aspiers | sean-k-mooney: yes exactly | 11:29 |
aspiers | "evacuate" and "migrate" both imply that everything is currently running (mostly) fine | 11:30 |
artom | Well | 11:30 |
aspiers | "resurrect" implies (correctly) "oh shit, something already blew up, so let's clean up after it" | 11:30 |
artom | The compute service/control plane has to be down | 11:30 |
artom | But the data plane? IIUC the instance can be running, no? | 11:31 |
sean-k-mooney | aspiers: ya that why i called the new command recreate but i can update it to resurect | 11:31 |
* artom hasn't actually checked whether we make sure to kill the instance first | 11:31 | |
sean-k-mooney | i spell recreate more consitenly however but i can live with that | 11:31 |
aspiers | artom: which scenario are you talking about exactly? there are a few here | 11:31 |
artom | aspiers, evacuate | 11:31 |
sean-k-mooney | artom: we dont | 11:31 |
sean-k-mooney | artom: but we require that you have marked the compute node as disabled before we allow it | 11:32 |
aspiers | artom: in this context you're gonna have to use more than one word to describe the scenario sorry ;-) | 11:32 |
aspiers | the whole problem is the ambiguity | 11:32 |
sean-k-mooney | artom: and we require the operator to confim that it is dead before the evac | 11:32 |
aspiers | sean-k-mooney: actually not always | 11:32 |
*** janki has joined #openstack-nova | 11:32 | |
sean-k-mooney | aspiers: that is the api contract | 11:32 |
aspiers | sean-k-mooney: OOB fencing can *ensure* that the VM is dead | 11:32 |
sean-k-mooney | sure | 11:32 |
sean-k-mooney | but the api contract is only call this if it is dead | 11:33 |
aspiers | sean-k-mooney: i.e. it can be automated. but yes we're on the same page I think | 11:33 |
aspiers | exactly | 11:33 |
aspiers | artom: there are multiple failure scenarios here to consider | 11:33 |
sean-k-mooney | if you violate that contract and you get 2 instance trying to read/write to the same non multi atach volume some how we just say "have fun" and walk away smiling | 11:34 |
*** ttsiouts has quit IRC | 11:34 | |
aspiers | sean-k-mooney: exactly, that is the main message of 3 talks I have given on this | 11:34 |
openstackgerrit | Miguel Ángel Herranz Trillo proposed openstack/nova master: Fix type error on call to mount device https://review.opendev.org/659780 | 11:34 |
aspiers | in Austin, Boston, and OpenStack Day Israel | 11:35 |
aspiers | e.g. https://youtu.be/lddtWUP_IKQ?t=366 | 11:35 |
artom | aspiers, evacuate the nova client command | 11:35 |
aspiers | (because exploding kittens is fun) | 11:35 |
artom | And my thought was with respect to resurrect, that the *instance* might not be dead, even if it's unmanageable because the host it's on *is* dead | 11:36 |
sean-k-mooney | ill convert https://review.opendev.org/#/c/643578/3 to resurect and reupload it proably on monday | 11:36 |
artom | So "resurrect" might not be the best word | 11:36 |
* artom sheds a bike | 11:36 | |
*** rcernin has quit IRC | 11:36 | |
aspiers | sean-k-mooney: ++ | 11:36 |
sean-k-mooney | artom: well its like if your heart stoped | 11:36 |
aspiers | artom: as sean-k-mooney said, the API contract is that the instance *must* be dead before calling the API | 11:36 |
artom | aspiers, ah, didn't know that, ok then | 11:37 |
sean-k-mooney | you were technically dead for a bit and this operation is like a defibulator | 11:37 |
aspiers | artom: that's why the force_down API was introduced ages ago - to get nova to realise quicker that it's dead, rather than waiting for its own RPC timeout | 11:37 |
sean-k-mooney | actully not just for that usecase | 11:38 |
aspiers | OK, that was one use case at least | 11:38 |
sean-k-mooney | there were case where for example the root disk and the disk where the vms images were seperate | 11:38 |
sean-k-mooney | so the agent and libvirt could be fine but when the backing disk died for the image and they crashed you wanted to be able to mark the host as down | 11:39 |
*** awalende has joined #openstack-nova | 11:39 | |
sean-k-mooney | or similar for network issues | 11:39 |
aspiers | right | 11:39 |
sean-k-mooney | aspiers: were you working with emma foley on the collectd/doctor/vitriage demoes? | 11:40 |
aspiers | artom: https://youtu.be/lddtWUP_IKQ?t=648 covers some of the failure modes, but you and sean-k-mooney are both right that there are other failure modes based on different network failures | 11:40 |
aspiers | sean-k-mooney: nope not heard of that, but with my self-healing SIG hat on, I'd really like to learn more about it! | 11:41 |
aspiers | sean-k-mooney: got any URLs? | 11:41 |
sean-k-mooney | there have been at least 3 summit presentation on it | 11:41 |
sean-k-mooney | the effort was driven via opnfv | 11:41 |
aspiers | damn, thought I'd been to all the vitrage presentations | 11:41 |
aspiers | so I'm just looking through my old decks and found this https://aspiers.github.io/openstack-day-israel-2017-compute-ha/#/comparison | 11:43 |
aspiers | had forgotten about the "needs improvements in nova" bit | 11:43 |
aspiers | trying to remember the details of that | 11:43 |
aspiers | I think it was around the reliability of the evacuate API | 11:44 |
*** jaosorior has quit IRC | 11:44 | |
cdent | that's some can of worms I opened | 11:44 |
aspiers | bad cdent! | 11:45 |
aspiers | keep quiet in future ;-) | 11:45 |
cdent | unpossible! | 11:45 |
aspiers | j/k, glad to be reminded about this stuff | 11:45 |
cdent | I hope that somehow this leads to that bug getting more attention. people get squeamish about orphaned providers | 11:45 |
cdent | especially because we have that damn uniq constraint on rp names | 11:46 |
cdent | which is the source of a ton of support requests | 11:46 |
cdent | "I was messing around and one of my compute nodes got rejiggered and it says can't create a resource provider" | 11:46 |
cdent | turns out the old one with the same name still exists | 11:46 |
aspiers | cdent: sorry to hijack your discussion with a big tangent | 11:47 |
* cdent is a fan of tangents | 11:48 | |
*** yan0s has joined #openstack-nova | 11:48 | |
aspiers | sean-k-mooney, artom: I think this 4-year-old blog post is still very relevant http://blog.clusterlabs.org/blog/2015/living-intersection-pets-cattle | 11:48 |
cdent | I tend to think the reason tangents happen is because there is insufficient shared understanding and that the tangents are necessary to get it | 11:48 |
aspiers | yeah | 11:48 |
cdent | but there is so little shared understanding in nova that is sometimes seems like we are on a tangent all the time | 11:49 |
aspiers | haha true | 11:49 |
cdent | tangents on tangets on tangents | 11:49 |
aspiers | beekhof and I did a lot of work a few years back considering all the failure cases of nova compute HA | 11:49 |
aspiers | he was definitely the thought-leader in this space, but he escaped to greener pastures | 11:49 |
sean-k-mooney | isnt a tangent to a tangent called a normal | 11:50 |
sean-k-mooney | nova embrases this :P | 11:50 |
aspiers | if a tangent is a 90 degree turn, then 4 tangents get you back in the original direction, right? | 11:50 |
sean-k-mooney | aspiers: for what its worth a lot of the presentation i was refering to were framed in the context fo service assurance not self healing or instance ha | 11:51 |
aspiers | sean-k-mooney: got it | 11:52 |
cdent | at the limit, N tangents is a circle, no matter the angle of turn | 11:52 |
cdent | or something | 11:52 |
aspiers | :D | 11:52 |
aspiers | https://aspiers.github.io/openstack-day-israel-2017-compute-ha/#/nova-host-alerter is the future converged architecture beekhof and I agreed on, in case anyone is interested | 11:52 |
sean-k-mooney | cdent: are you suggesting we would go around in circles on things, im shocked :) | 11:53 |
aspiers | coupled with http://blog.clusterlabs.org/blog/2015/reliable-notifications and masakari adding a layer of retries on top of nova's somewhat lossy workflow, the hope was that this would be more robust | 11:53 |
aspiers | s/more/sufficiently/ | 11:53 |
sean-k-mooney | anyway i better file my expense report for the ptg or finance will get annoyed at me | 11:54 |
aspiers | sean-k-mooney: IIRC one of the concerns was that if scheduler or conductor crashed while handling evacuate API, it would get lost forever - is that still true? | 11:54 |
cdent | sean-k-mooney: hush | 11:54 |
sean-k-mooney | aspiers: ya maybe | 11:54 |
*** hamzy has quit IRC | 11:55 | |
sean-k-mooney | it woudl depend on when it failed i guess | 11:55 |
sean-k-mooney | if we have not deleted it form the souce node yet then you should be fine | 11:55 |
aspiers | I think the idea was maybe that masakari could reinvoke the API in those circumstances | 11:55 |
sean-k-mooney | ya i think if the db says its still on the source node then you can evac again | 11:55 |
aspiers | but is there even a way to query the API for the state of an evacuate workflow? | 11:55 |
sean-k-mooney | but depending on the failure mode the vm could be paused/running on another node at that point | 11:56 |
aspiers | I vaguely remember that being discussed at past summits | 11:56 |
aspiers | yeah, the devil is in the details I guess | 11:56 |
aspiers | I don't think there's ever been enough corporate sponsorship from RH or SUSE to spend time really cleaning up all those corner cases | 11:57 |
sean-k-mooney | if your contol plane if failing you shoudl really stop what your doing and fix that first then go back to fixing your broken compute nodes | 11:57 |
aspiers | oh sure, that's a totally different kettle of fish | 11:57 |
* artom imagines fish being boiled alive | 11:58 | |
aspiers | compute HA is impossible without a resilient control plane | 11:58 |
aspiers | same for most workflows | 11:58 |
aspiers | sean-k-mooney: mriedem makes a good point on https://review.opendev.org/#/c/643578/3//COMMIT_MSG that resurrect is an admin-only command, so "openstack server resurrect" could mislead non-admin users into thinking they can call it | 12:01 |
aspiers | OTOH I don't agree with his statement "evacuate moves the server to another host (actually it rebuilds it on another host)" | 12:01 |
sean-k-mooney | aspiers: migrate is an admin command too | 12:02 |
sean-k-mooney | and that is openstack server migrate | 12:02 |
aspiers | since a) you can't "move" a server which has already died, and b) it only rebuilds if it was using ephemeral storage | 12:02 |
sean-k-mooney | the only non admin migration is resize | 12:02 |
aspiers | sean-k-mooney: right, that's why he was suggesting nesting under "openstack server migrate" | 12:03 |
sean-k-mooney | no | 12:03 |
sean-k-mooney | migrate implies it will preserve you data | 12:03 |
aspiers | but I really don't like that cos resurrect is not a form of migration | 12:03 |
sean-k-mooney | so that would be worse then evac | 12:03 |
*** awalende has quit IRC | 12:03 | |
aspiers | yes, I think we are saying the same thing in different ways :) | 12:03 |
sean-k-mooney | what im saying is migrate is not an option because migrate implies we move data and we dont in the most common form for evac | 12:04 |
sean-k-mooney | most deployment have an ephermeal root disk | 12:04 |
aspiers | yes, that's one good reason to avoid migrate | 12:04 |
aspiers | and it's not the only one | 12:04 |
aspiers | at a higher level of abstraction, what you are saying could be reframed as "migrate" implies moving something from A to B without significantly changing it | 12:05 |
sean-k-mooney | we also discussed having two commands one that guarnetees it preserves your data and one tha that always destroys it | 12:05 |
*** awalende has joined #openstack-nova | 12:05 | |
*** tbachman has joined #openstack-nova | 12:06 | |
aspiers | but you start with something dead on A, and end with something alive on B, that is a big change in state | 12:06 |
sean-k-mooney | e.g. recreate woudl always destoy any instance and recreate it loasing data | 12:06 |
aspiers | therefore does not deserve to be called a form of migration | 12:06 |
artom | Data destruction/preservation also depends on where said data is | 12:07 |
sean-k-mooney | openstack server migrate --evacuate would guarentte that your data was moved or would not try to do dit | 12:07 |
aspiers | sean-k-mooney: yeah that might be a good approach | 12:07 |
aspiers | I mean, having two different commands | 12:07 |
aspiers | but I don't like migrate --evacuate, for reasons just explained | 12:07 |
aspiers | I don't really like migrate --resurrect either | 12:08 |
aspiers | although it bothers me less than migrate --evacuate | 12:08 |
aspiers | "openstack server recreate" for the ephemeral case and "openstack server resurrect" for the shared storage case feels OK to me | 12:09 |
openstackgerrit | Surya Seetharaman proposed openstack/python-novaclient master: Allow passing negative values for the locked search_opt in server list https://review.opendev.org/659783 | 12:09 |
aspiers | but I can see the concern around non-admin users expecting to be able to use it | 12:10 |
sean-k-mooney | aspiers: well wel could allow openstack server recreate to be non admin | 12:10 |
sean-k-mooney | its basically a rebuild but optionally to another host | 12:11 |
aspiers | sean-k-mooney: in what scenarios would non-admin users use that? not if the server is already running right? cos it reuses the old name | 12:12 |
sean-k-mooney | aspiers: i was implying that openstack server recreate would kill the server if it was running and rescudle it | 12:13 |
aspiers | oh | 12:13 |
sean-k-mooney | they would use it if the image or flavor had been updated and they wanted to create the sever with the updated version | 12:13 |
sean-k-mooney | but keep the same ports and data volumes | 12:14 |
aspiers | yeah I guess that could be useful | 12:14 |
aspiers | or if the instance got screwed up somehow, e.g. accidental rm -rf ;-) | 12:14 |
*** ccamacho has quit IRC | 12:15 | |
sean-k-mooney | yep if you instacne are vnfs it would be a neat way to do a software update | 12:15 |
aspiers | yep | 12:15 |
sean-k-mooney | kindof like in k8s land | 12:15 |
sean-k-mooney | e.g. just kill the pod and recreate with the latest version of it | 12:15 |
aspiers | right | 12:17 |
aspiers | so are you thinking to update https://review.opendev.org/#/c/643578 to split into two separate commands recreate/resurrect? | 12:18 |
aspiers | that would work for me, not sure what mriedem will think though | 12:18 |
aspiers | with both I'd prefer to avoid mentioning the word "migrate" anywhere | 12:19 |
aspiers | since "migrate" implies avoiding any significant change in state (except the location) | 12:19 |
*** nicolasbock has joined #openstack-nova | 12:23 | |
*** mchlumsky has joined #openstack-nova | 12:24 | |
aspiers | sean-k-mooney: BTW which timezone are you in? | 12:26 |
aspiers | I'm guessing somewhere in Ireland? | 12:26 |
sean-k-mooney | yes | 12:27 |
sean-k-mooney | im shannon on the west coast of ireland | 12:27 |
aspiers | nice | 12:27 |
sean-k-mooney | what timezone i am awake in is not alway the same as where im located | 12:28 |
aspiers | true for me too ;) | 12:28 |
aspiers | for some reason I still expect all nova devs to be in north america | 12:28 |
aspiers | I'm in London | 12:28 |
openstackgerrit | Gorka Eguileor proposed openstack/nova master: Use os-brick locking for volume attach and detach https://review.opendev.org/614190 | 12:29 |
aspiers | got a lot of family in Dublin but never been further west than there yet :-( | 12:29 |
*** udesale has quit IRC | 12:30 | |
*** udesale has joined #openstack-nova | 12:31 | |
*** dikonoor has joined #openstack-nova | 12:36 | |
*** janki has quit IRC | 12:40 | |
*** janki has joined #openstack-nova | 12:40 | |
*** janki has quit IRC | 12:41 | |
*** dikonoor has quit IRC | 12:46 | |
*** mdbooth_ has joined #openstack-nova | 12:47 | |
*** cdent has quit IRC | 12:48 | |
*** dklyle has joined #openstack-nova | 12:50 | |
*** mdbooth has quit IRC | 12:50 | |
*** slaweq has quit IRC | 12:51 | |
*** slaweq has joined #openstack-nova | 12:52 | |
*** ratailor has quit IRC | 12:54 | |
*** ratailor has joined #openstack-nova | 12:56 | |
*** cdent has joined #openstack-nova | 13:01 | |
*** jaosorior has joined #openstack-nova | 13:11 | |
*** jmlowe has quit IRC | 13:11 | |
*** awalende has quit IRC | 13:15 | |
efried | kashyap, aspiers: what's up now? | 13:15 |
kashyap | efried: It's a change of scenery :-) | 13:16 |
aspiers | efried: ah, you missed the almight bike osc shed ;-) | 13:16 |
kashyap | efried: SEV-related, uncontroversial SEV-related | 13:16 |
kashyap | Err | 13:16 |
aspiers | oh, that too | 13:16 |
kashyap | "SEV-related, uncontroversial infra patch" | 13:16 |
kashyap | aspiers: That's what I assumed efried meant, when he pinged both of us together :D | 13:16 |
efried | I refer to ==> 3:45:56 AM - kashyap: aspiers: I think your change is ready to merge. Let's check with efried (or stephenfin) when they're about. | 13:17 |
aspiers | yeah, that one | 13:17 |
kashyap | Yep, I correctly guessed. | 13:17 |
kashyap | It is this one: https://review.opendev.org/#/c/655268/ -- "Add infrastructure for invoking libvirt's getDomainCapabilities API" | 13:17 |
efried | Okay. This isn't really in my, ahem, "domain" of expertise | 13:18 |
efried | but I'll take a look. | 13:18 |
ganso | melwitt, dansmith, tonyb: Hi! could you please review https://review.opendev.org/#/c/659338 when you have some time available? Thanks in advance! =) | 13:19 |
aspiers | efried: it should be pretty simple and not require special expertise | 13:19 |
*** belmoreira has joined #openstack-nova | 13:20 | |
*** mriedem has joined #openstack-nova | 13:21 | |
*** tbachman has quit IRC | 13:21 | |
efried | aspiers, kashyap: Does this also have to do with video models? | 13:22 |
efried | https://blueprints.launchpad.net/nova/+spec/libvirt-video-device-models | 13:22 |
kashyap | efried: No, that getDomainCapabilities() isn't related to video models. | 13:23 |
efried | ight | 13:23 |
efried | I see video models in your fake xml, so... | 13:23 |
efried | uh, I thought I did... | 13:23 |
aspiers | I guess it could be a future way of detecting available video models? | 13:24 |
efried | yeah, there. (Chrome's text search has been way slow for me lately) | 13:24 |
aspiers | if anything needed to do that | 13:24 |
aspiers | but that's certainly not what's driving the change right now | 13:24 |
kashyap | efried: You're right -- they _are_ present, though. But not relevant in this context | 13:24 |
kashyap | Indeed, what aspiers said. | 13:24 |
efried | k | 13:24 |
aspiers | It's been driven by SEV and secure boot | 13:24 |
kashyap | efried: Nothing like jumping right into the bowels of libvirt when you wake up, is it? :D | 13:24 |
aspiers | I haven't woken up yet | 13:24 |
kashyap | s/wake up/get started/ | 13:24 |
kashyap | aspiers: Haha | 13:25 |
aspiers | I'm sure sean-k-mooney could comment on whether this change would be useful for that bp | 13:25 |
efried | something something bowels when I wake up in general | 13:25 |
*** ccamacho has joined #openstack-nova | 13:25 | |
kashyap | aspiers: You got attacked by ZombieLoad | 13:25 |
aspiers | efried: TMI X-D | 13:25 |
efried | no doubt | 13:25 |
aspiers | on the plus side, I actually managed to get out of bed before 9am for the first time since Denver | 13:26 |
kashyap | aspiers: It's definitely useful -- as that _allows_ querying for the models. | 13:26 |
aspiers | but getting out of bed doesn't imply waking up | 13:26 |
aspiers | kashyap: right | 13:26 |
kashyap | aspiers: Wauw, I thought it was just me, whose sleep cycle got completely botched since Denver. | 13:26 |
aspiers | nope | 13:27 |
efried | my sleep cycle has been fine, thank you. | 13:27 |
aspiers | mine's been botched ever since SUSECON :-( | 13:27 |
* aspiers -> late lunch | 13:27 | |
kashyap | aspiers: I kept thinking: "No, it can't be Denver, it's been more than 8 days") | 13:27 |
efried | Shanghai will be a different story | 13:27 |
aspiers | oh god | 13:27 |
kashyap | efried: Right, enjoy :D | 13:27 |
aspiers | funny thing is, I was fine for Sydney | 13:27 |
aspiers | but then I made sleep spreadsheets for that trip, and stuck to them religiously | 13:27 |
aspiers | and with that teaser hanging, I'm really off for lunch | 13:28 |
*** lbragstad has joined #openstack-nova | 13:28 | |
*** tbachman has joined #openstack-nova | 13:29 | |
* cdent shakes fist (warily) at efried | 13:29 | |
cdent | I got the jetlag so hard | 13:29 |
sean-k-mooney | aspiers: which change was that? | 13:30 |
efried | sean-k-mooney: We're asking whether https://review.opendev.org/#/c/655268/ might be relevant to https://blueprints.launchpad.net/nova/+spec/libvirt-video-device-models | 13:30 |
sean-k-mooney | efried: oh not really | 13:30 |
efried | k | 13:30 |
efried | aslo | 13:31 |
efried | also* | 13:31 |
sean-k-mooney | that is not reported in the capablities api | 13:31 |
efried | could you review that patch please? | 13:31 |
efried | :) | 13:31 |
sean-k-mooney | ya | 13:31 |
efried | thanks | 13:31 |
sean-k-mooney | actully it looks like it has been added since i last looked | 13:32 |
sean-k-mooney | so i guess i could use it | 13:32 |
sean-k-mooney | i had planned to just to version checks | 13:32 |
sean-k-mooney | basically reusing the tables we already have in the code | 13:32 |
sean-k-mooney | like this one https://github.com/openstack/nova/blob/master/nova/virt/libvirt/vif.py#L67-L100 | 13:33 |
sean-k-mooney | so for disk bus and video yes i can use it but vif models are not repored and it will have to use the table approach | 13:35 |
stephenfin | kashyap, efried, aspiers: I can take a look at that now. Was Bugzilla wrangling this morning | 13:35 |
kashyap | stephenfin: Completely understand; aspiers diligently worked through all the feedback. I can't punch any holes in the change | 13:36 |
sean-k-mooney | hum actully i will have to use both as it look that info is not there for older libvirt | 13:36 |
kashyap | stephenfin: And it's a _very_ useful infrastructure patch that helps other features as well. So that can be merged regardless of SEV | 13:36 |
kashyap | (Although, aspiers might be shaking his fist at me, as I said "regardless of SEV") | 13:37 |
*** xek_ has joined #openstack-nova | 13:37 | |
sean-k-mooney | kashyap: aspiers wehre did the fake libvirt capablites xml come form? | 13:37 |
sean-k-mooney | aspiers: did you take it form you machine? | 13:37 |
stephenfin | sean-k-mooney: I was just about to ask the same question :) | 13:38 |
kashyap | sean-k-mooney: I'd guess so, from his test box | 13:39 |
sean-k-mooney | this is what i get on my desktop | 13:39 |
sean-k-mooney | http://paste.openstack.org/show/751512/ | 13:39 |
*** xek has quit IRC | 13:39 | |
stephenfin | Hmm, Fedora doesn't come with libvirt out-of-the-box | 13:39 |
sean-k-mooney | running libvirt 5.2.0 | 13:39 |
stephenfin | sean-k-mooney: That's capabilities, not domain capabilities | 13:40 |
stephenfin | There's a difference in command and output but I'm trying to remember what it is | 13:40 |
sean-k-mooney | ah i have this for that http://paste.openstack.org/show/751514/ | 13:40 |
*** tbachman has quit IRC | 13:41 | |
sean-k-mooney | can we ask them to add interface to that list | 13:41 |
sean-k-mooney | i get similar output on 4.0.0 | 13:41 |
stephenfin | Those are available under a subcommand, but I don't recall what that's called either | 13:42 |
sean-k-mooney | so i guess it been around for a while | 13:42 |
* stephenfin had to use it months when trying to figure out if his i350 NIC was dead or not | 13:42 | |
sean-k-mooney | its sudo virsh domcapabilities | 13:42 |
stephenfin | no, for the interface capabilities | 13:42 |
sean-k-mooney | oh well there are two things | 13:43 |
sean-k-mooney | there are the physical nic feature flags | 13:43 |
sean-k-mooney | and what virtual nics qemu can emulate | 13:43 |
sean-k-mooney | the nic feature flags are under the nodedev-dumpxml command | 13:44 |
kashyap | sean-k-mooney: It renders the same data without `sudo`, too. (With the ultra-minor diff being: with `sudo` it links to actual emulator path /usr/bin/qemu-system-x86_64; instead of the wrapper /usr/bin/qemu-kvm) | 13:44 |
stephenfin | ah, I was referring to the former in this case actually, my bad | 13:44 |
sean-k-mooney | the models that can be emulated are not available | 13:44 |
openstackgerrit | Guo Jingyu proposed openstack/nova master: Skip existing VMs when hosts apply force_config_drive https://review.opendev.org/659703 | 13:44 |
stephenfin | Remind me, what's the "is my system set up correctly for KVM" command? | 13:44 |
sean-k-mooney | kvm-ok | 13:45 |
stephenfin | ta | 13:45 |
sean-k-mooney | its part of the cpu-its part of the cpu-checker package if you dont have it installed | 13:46 |
stephenfin | ta x2 :) | 13:46 |
sean-k-mooney | its also only available on debian/ubuntu i think :) | 13:47 |
sean-k-mooney | so if you use somethin else all its doing is checking for /dev/kvm and checik lsmod for kvm | 13:47 |
stephenfin | So it would seem. Damn. I thought libvirt provided some tool for this but maybe I'm mistaking it for something | 13:47 |
sean-k-mooney | on and /proc/cpu-info for vmx or svm | 13:48 |
sean-k-mooney | you can tell form virsh capablityes | 13:48 |
stephenfin | Yeah, I went and did it manually instead. All good now | 13:48 |
sean-k-mooney | it list the emulator that can be run | 13:49 |
sean-k-mooney | if you dont se a hvm section then you cant use kvm | 13:49 |
stephenfin | I was seeing "error: invalid argument: unable to find any emulator to serve 'x86_64' architecture" because I hadn't installed qemu-kvm yet, assuming libvirt would do it by default | 13:49 |
stephenfin | *installing libvirt | 13:49 |
*** mmethot has quit IRC | 13:50 | |
sean-k-mooney | or do sudo virsh capabilities | grep "<domain type='kvm'>" | 13:50 |
sean-k-mooney | stephenfin: libvirt does not install qemu by defualt | 13:50 |
stephenfin | In any case, I'm seeing something similar to you but I've got references to the OVMF stuff | 13:50 |
stephenfin | Yup, I know that now :) | 13:50 |
sean-k-mooney | and annoying installing qemu-kvm does not install qemu anymore... | 13:51 |
* sean-k-mooney shakes fist at fedora | 13:51 | |
stephenfin | and I'm missing the virtio disk models, oddly | 13:51 |
sean-k-mooney | strange | 13:52 |
sean-k-mooney | you have installed qemu and qemu kvm packages and restarted libvirtd after | 13:52 |
sean-k-mooney | what os are you testing by the way | 13:52 |
stephenfin | Yeah. Not the end of the world though | 13:53 |
stephenfin | Fedora 30 | 13:53 |
stephenfin | Hold your "popOS!" comments now :P | 13:53 |
sean-k-mooney | stephenfin: they have been makign the qemu packages modular | 13:53 |
sean-k-mooney | have you tried gentoo :P | 13:54 |
stephenfin | No, I figured I should give Slackware a shot first | 13:54 |
sean-k-mooney | gentoo is the most painful disto i have used even more painful then linux form scratch | 13:54 |
sean-k-mooney | kashyap: is the virito support for qemu disk part of a sepereate package on fedroa 30? | 13:55 |
kashyap | sean-k-mooney: What do you mean? | 13:55 |
sean-k-mooney | kashyap: i woudl guess as that should be part of qemu by default | 13:55 |
kashyap | sean-k-mooney: Yeah | 13:55 |
kashyap | There has been a split of RPM packages for QEMU | 13:56 |
kashyap | Especially the block drivers | 13:56 |
sean-k-mooney | kashyap: stephenfin is not seing vitio in the domcapablities output on f30 | 13:56 |
stephenfin | I'm referring to lines 101-105 from here http://paste.openstack.org/show/751514/ | 13:56 |
stephenfin | It doesn't matter much. More curiosity than anything | 13:56 |
kashyap | stephenfin: Can you post your `rpm -qa | grep -i qemu` output? | 13:56 |
kashyap | sean-k-mooney: Check this one that i did recently: https://github.com/rdo-packages/nova-distgit/commit/4e490d2348cfe99b68d70c4b95408348d0b54492 (" Use granular libvirt and QEMU RPMs for EL-8 Nova") | 13:57 |
stephenfin | kashyap: http://paste.openstack.org/show/751515/ | 13:57 |
sean-k-mooney | stephenfin: well its novas default for qemu | 13:57 |
sean-k-mooney | and kvm | 13:57 |
sean-k-mooney | so if it really is not there your vms wont boot | 13:57 |
kashyap | stephenfin: Can you install 'qemu-kvm-core'? | 13:58 |
sean-k-mooney | unless you set hw_disk_bus=sata in the image or something | 13:58 |
sean-k-mooney | kashyap: waith how is qemu-kvm install without qemu-kvm-core | 13:58 |
kashyap | stephenfin: The domainCapabilities output you posted -- is that from F30? | 13:58 |
sean-k-mooney | that seam like a packageing bug | 13:58 |
stephenfin | kashyap: Done. No good. That paste is from sean-k-mooney | 13:58 |
*** mlavalle has joined #openstack-nova | 13:59 | |
kashyap | sean-k-mooney: Hang on -- I find it strange, too many things flying around :-) | 13:59 |
stephenfin | Here's mine, fwiw http://paste.openstack.org/show/751516/ | 13:59 |
stephenfin | 5.0.16-300.fc30.x86_64 | 13:59 |
*** roukoswarf has joined #openstack-nova | 14:00 | |
sean-k-mooney | kashyap: :) that ws from manjor/arch with the latest libvirt available. | 14:00 |
*** jmlowe has joined #openstack-nova | 14:00 | |
sean-k-mooney | stephenfin: i see it on my popos system too with libvirt 4.0.0 so its not a libvirt issue i dont think | 14:00 |
roukoswarf | any recommendations on how to get nova scheduler to anti-affinity a bigger unit than a "host"? i have many "hosts" in a single chassis, or sometimes a single rack, and i need the ability to get something like octavia to schedule things to different chassis/racks for HA. | 14:02 |
roukoswarf | masses of cells seems like... the wrong solution? | 14:03 |
*** JamesBenson has joined #openstack-nova | 14:04 | |
kashyap | sean-k-mooney: It is _not_ a packaging bug: in stephenfin's paste: you _do_ have: qemu-system-x86-core-3.1.0-8.fc30.x86_64 | 14:06 |
kashyap | It is 'qemu-system-$ARCH-core' | 14:06 |
*** lpetrut has quit IRC | 14:06 | |
kashyap | (Also note: 'qemu-kvm-core' is a _dummy_ package that pulls in 'qemu-system-x86' on Fedora) | 14:08 |
mriedem | efried: remind me, mock spec is a string right? https://review.opendev.org/#/c/659726/2/watcher/tests/datasources/test_ceilometer_helper.py | 14:11 |
kashyap | stephenfin: You need libvirt-5.2.0 or newer to get the virtio-* enum values reported. | 14:11 |
efried | mriedem: no, it is *not* a string. Unless you want to spec a string. | 14:11 |
mriedem | gdi | 14:11 |
efried | hold on, I have a commit that explains it clearly (I hope) | 14:12 |
kashyap | stephenfin: It's noted "v5.2.0 (2019-04-03) | 14:12 |
efried | mriedem: https://review.opendev.org/#/c/650370/ | 14:12 |
kashyap | stephenfin: Err. I meant: It's noted as part of the release notes of libvirt "v5.2.0 (2019-04-03)" here: https://libvirt.org/news.html | 14:13 |
*** mmethot has joined #openstack-nova | 14:13 | |
mriedem | so i just changed to this (added autospec=True): | 14:13 |
mriedem | @mock.patch.object(ceilometer_helper.CeilometerHelper, | 14:13 |
mriedem | 'statistic_aggregation', autospec=True) | 14:13 |
mriedem | and it blew up as expected | 14:13 |
mriedem | so why not just autospec? | 14:13 |
mriedem | i think the answer is "sometimes it doesn't work how you'd expect" is the answer | 14:15 |
mriedem | like everything with specs and mock | 14:15 |
*** ykarel is now known as ykarel|afk | 14:16 | |
*** luksky has quit IRC | 14:19 | |
*** dpawlik has quit IRC | 14:24 | |
*** Luzi has quit IRC | 14:25 | |
stephenfin | mriedem: claudiub has a series to convert everything to autospec. Would it be worth reviving that? | 14:25 |
stephenfin | aspiers: Reviewed https://review.opendev.org/#/c/655268/. One concern about logging but otherwise it looks good | 14:25 |
openstackgerrit | Mark Goddard proposed openstack/nova master: [ironic] Don't remove instance info twice in destroy https://review.opendev.org/659809 | 14:30 |
efried | aspiers, kashyap: reviewed same, soft-ish -1 on some refactoring to make things cleaner. | 14:30 |
efried | stephenfin: If those guys aren't around rn, maybe you could look and tell me if I'm way off base with my comments; maybe there's some reason we put in extra fluff. | 14:31 |
stephenfin | sure | 14:31 |
kashyap | efried: Just skimmed (in a call); thanks for the detailed review. I'll pay full attention to your comments on Monday. | 14:32 |
efried | thanks both | 14:33 |
*** alex_xu has joined #openstack-nova | 14:33 | |
kashyap | Thank _you_ for spending a block of time on it. | 14:35 |
mriedem | stephenfin: this isn't a question for nova really | 14:35 |
mriedem | it's a generic dev thing | 14:35 |
mriedem | trying to apply it to a bug fix in watcher | 14:36 |
stephenfin | efried: Yup, they all look valid. I was ignoring the added config opts because I saw them used in later patches, but adding them then might make more sense, aye | 14:37 |
stephenfin | *config classes | 14:37 |
stephenfin | mriedem: Ah, fair. I didn't catch that | 14:37 |
*** jaosorior has quit IRC | 14:41 | |
*** jmlowe has quit IRC | 14:41 | |
*** guozijn has quit IRC | 14:42 | |
*** macza has joined #openstack-nova | 14:45 | |
efried | stephenfin: I'm less concerned about that than about the unnecessary middleman class | 14:45 |
cdent | mriedem: are your wips on https://review.opendev.org/#/q/topic:bug/1790204 still a going concern? | 14:47 |
*** macza has quit IRC | 14:47 | |
mriedem | cdent: as in vmware needs them now? | 14:47 |
cdent | no, I'm writing the pupdate | 14:47 |
mriedem | oh | 14:47 |
*** kmalloc is now known as kmalloc_away | 14:47 | |
mriedem | it's very complicated so i'm not working on it regularly no, maybe once per month before i write up my internal monthly report :) | 14:48 |
cdent | i'll keep them on the list as why not... | 14:48 |
mriedem | efried: this is why spec baffles me http://paste.openstack.org/show/751523/ | 14:51 |
mriedem | note expected and actual are identical | 14:51 |
mriedem | not sure wtf "b'TypeError: too many positional arguments'" is about | 14:51 |
mriedem | this being the test http://paste.openstack.org/show/751524/ | 14:52 |
*** dtantsur is now known as dtantsur|afk | 14:52 | |
efried | mriedem: what's the method signature for statistic_aggregation? | 14:52 |
*** jamesdenton has quit IRC | 14:53 | |
mriedem | def statistic_aggregation(self, resource_id=None, meter_name=None, | 14:53 |
mriedem | period=300, granularity=300, dimensions=None, | 14:53 |
mriedem | aggregation='avg', group_by='*'): | 14:53 |
mriedem | is it complaining that kwargs aren't used for all params? | 14:54 |
efried | This looks to be a buglet in mock itself; it's confusingly triggering a failure on assert_called_once_with | 14:54 |
efried | yes | 14:54 |
efried | that's a little weird, but I'm a fan. | 14:54 |
*** belmoreira has quit IRC | 14:54 | |
efried | kwargs should be named | 14:54 |
mriedem | i've seen the tattoo | 14:54 |
*** yan0s has quit IRC | 14:55 | |
*** shilpasd has quit IRC | 14:56 | |
aspiers | sean-k-mooney: yes, I got the sample XML from an SEV-capable test box | 14:58 |
*** jmlowe has joined #openstack-nova | 14:58 | |
* efried needs to chauffeur... | 14:58 | |
*** belmoreira has joined #openstack-nova | 14:58 | |
mriedem | changing to use all kwargs results in an even weirder error, so i'll just not try spec | 14:59 |
mriedem | oh nvm | 15:00 |
*** cfriesen has joined #openstack-nova | 15:04 | |
*** mcgigglier has quit IRC | 15:05 | |
artom | mriedem, yey, we can talk to you again! | 15:13 |
artom | Had some scary legals emails about the Huawei embargo thing | 15:13 |
aspiers | artom: :-( | 15:14 |
aspiers | what a crappy situation | 15:14 |
aspiers | also I'm guessing pretty unprecedented and nonsensical in an open source context | 15:14 |
jangutter | I am so glad South Africa is no longer under sanctions. | 15:14 |
*** ykarel|afk is now known as ykarel|away | 15:15 | |
mriedem | artom: i've heard | 15:16 |
*** priteau has joined #openstack-nova | 15:16 | |
mriedem | realize that i already know all of your secrets before you tell me | 15:16 |
artom | If you knew all my secrets you'd be in an asylum | 15:16 |
*** ratailor has quit IRC | 15:17 | |
artom | aspiers, it's definitely weird... though I suspect if there's on legal department that can make sense of it, it's RH's, given our open source history | 15:19 |
artom | *one legal | 15:19 |
aspiers | or SUSE's - a few months older than RH even ;-) | 15:19 |
aspiers | although we're no longer an American company | 15:20 |
aspiers | so maybe we don't need to care as much | 15:20 |
aspiers | I'm sure RH has a much more impressive army of lawyers than we do too | 15:20 |
*** ivve has quit IRC | 15:21 | |
*** sapd1_x has joined #openstack-nova | 15:21 | |
artom | True, though being German (albeit with US... affiliates?) they may not be subjected to the same laws | 15:22 |
aspiers | We're not German | 15:22 |
aspiers | Owned by EQT which is Swedish | 15:22 |
aspiers | but we have companies in many countries around the world, just like RH | 15:23 |
* mriedem always thought there were DE now too | 15:23 | |
aspiers | so yeah, there'll be a SUSE Inc. or whatever | 15:23 |
aspiers | and SUSE GmbH, SUSE UK Ltd. etc. | 15:23 |
aspiers | each of those subject to the whims of the respective country | 15:24 |
openstackgerrit | Merged openstack/python-novaclient master: Optimize limit option docs string description for novaclient https://review.opendev.org/643035 | 15:24 |
aspiers | I wonder if it could result in situations where Huawei-related work could get taken off the hands of American members of a team and given to colleagues in other countries | 15:24 |
artom | aspiers, in our case even our non-US companies are subject to the embargo | 15:24 |
aspiers | interesting, why's that? | 15:25 |
artom | IANAL | 15:25 |
aspiers | because the parent company is US? or some other reason | 15:25 |
artom | I guess | 15:25 |
aspiers | if so, SUSE wouldn't have the same issue | 15:25 |
artom | But yeah, we got a pretty strongly worded email that we essentially need to stop all dealings with Huawei, all over the world | 15:25 |
aspiers | crazy stuff | 15:25 |
artom | Apparently open source upstream communities are exempt from that, tho | 15:26 |
artom | Which is why mriedem is still my friend | 15:26 |
aspiers | hah | 15:26 |
* aspiers quickly backtracks and claims he's not allowed to work on Huawei stuff | 15:26 | |
artom | Oh man, you saw your chance and you went for it XD | 15:26 |
aspiers | ;P | 15:26 |
* artom wonders off to get a hair cut and lunch | 15:27 | |
*** gyee has joined #openstack-nova | 15:34 | |
*** tssurya has quit IRC | 15:36 | |
aspiers | kashyap: remind me, will your secure boot code care about <features> at all? | 15:37 |
*** jmlowe has quit IRC | 15:38 | |
*** wwriverrat has joined #openstack-nova | 15:41 | |
*** ykarel|away has quit IRC | 15:58 | |
*** rpittau is now known as rpittau|afk | 16:01 | |
sean-k-mooney | artom: that is tecnically a gray area | 16:04 |
sean-k-mooney | but most multintionall play it safe | 16:04 |
sean-k-mooney | efried: im going to finish earlish today. i have libvirt domcapabilies review on my todo for today is there anything else you want me to look at before monday | 16:05 |
efried | sean-k-mooney: I'm sure nothing is urgent, thanks for asking. | 16:06 |
*** panda|rover is now known as panda|rover|off | 16:06 | |
kashyap | aspiers: The code-that-doesn't-exist-yet will care about it | 16:07 |
sean-k-mooney | cool ya im going to drop off irc soon which is why i asked | 16:07 |
kashyap | aspiers: I need to be AFK: catch up on Monday. Sorry | 16:07 |
kashyap | Have a good weekend, folks (in CEST) | 16:07 |
aspiers | kashyap: cya | 16:07 |
aspiers | sean-k-mooney: I'm just about to push a new patchset | 16:07 |
aspiers | in response to reviews from efried and stephenfin | 16:07 |
aspiers | efried: and one of the comments is "please can we discuss in IRC" ;-) | 16:08 |
sean-k-mooney | cool ill finish updating the downstream but then ill look at it then | 16:08 |
sean-k-mooney | *bug | 16:08 |
efried | aspiers: I'm here, discuss away. | 16:09 |
aspiers | efried: I don't want to repeat the long essay I've drafted in Gerrit ;) | 16:09 |
aspiers | efried: give me a few mins to finish the response | 16:09 |
efried | ack | 16:09 |
*** mdbooth_ has quit IRC | 16:11 | |
*** belmoreira has quit IRC | 16:11 | |
mriedem | gibi_off: i'm guessing we're not going to be able to backport the fix for https://review.opendev.org/#/c/512623/ without dropping the functional test as well because of all the underlying refactor to make the functional test work? | 16:12 |
*** tbachman has joined #openstack-nova | 16:12 | |
*** cdent has quit IRC | 16:13 | |
*** tesseract has quit IRC | 16:17 | |
*** psachin has quit IRC | 16:17 | |
*** tesseract has joined #openstack-nova | 16:18 | |
*** whoami-rajat has joined #openstack-nova | 16:20 | |
*** macza has joined #openstack-nova | 16:21 | |
*** sridharg has quit IRC | 16:22 | |
*** jangutter has quit IRC | 16:22 | |
*** macza has quit IRC | 16:26 | |
stephenfin | mriedem: You think a patch to exclude the 'user_data' field for each instance logged in this log sounds reasonable? https://opendev.org/openstack/nova/src/commit/ed5b7c7fc311c2af8c8be09044bb2723141f6f0d/nova/scheduler/host_manager.py#L180 | 16:29 |
sean-k-mooney | stephenfin: we woudl have to modify the __repr__ function in the ovo to not incude that right | 16:30 |
stephenfin | I'm looking at a log where the user has bulk created some instances and it's just pages and pages of base64 (?) data | 16:30 |
stephenfin | I guess? I know we do some filtering of objects on security grounds | 16:30 |
stephenfin | I was hoping to reuse some of that, if that's possible | 16:30 |
sean-k-mooney | ah then ya it could be simple so | 16:31 |
*** jcosmao has left #openstack-nova | 16:31 | |
sean-k-mooney | well looks like you have the instace info as a dict so you would have to make a copy and just remove the user_data section from teh copy or replace it with a message saying it was remvoed | 16:32 |
stephenfin | Or that <?> marker that gets inserted by...something | 16:33 |
*** luksky has joined #openstack-nova | 16:33 | |
stephenfin | o.vo normally, I guess | 16:33 |
sean-k-mooney | yes that thing | 16:33 |
sean-k-mooney | e.g. so you can tell there was user data but not log it | 16:33 |
mriedem | stephenfin: you're looking at an old log then, | 16:35 |
mriedem | because that was fixed to only log instance uuids | 16:35 |
mriedem | https://opendev.org/openstack/nova/commit/4fd7c93726eff4cc0b010741ea1772cf19c314fc | 16:35 |
mriedem | https://review.opendev.org/#/q/I0eda1c58a7eb54121230c880818b4b1d0fdf4893 | 16:36 |
mriedem | take that back to pike and ocata if you want | 16:36 |
stephenfin | I'm going to do just that. Thanks, mriedem | 16:36 |
mriedem | np | 16:36 |
openstackgerrit | Stephen Finucane proposed openstack/nova stable/pike: Do not dump all instances in the scheduler https://review.opendev.org/659832 | 16:36 |
sean-k-mooney | oh the list grabs the keys that are the uuids neat | 16:37 |
openstackgerrit | Stephen Finucane proposed openstack/nova stable/ocata: Do not dump all instances in the scheduler https://review.opendev.org/659833 | 16:37 |
mriedem | it's a dict, the uuids are the keys | 16:37 |
sean-k-mooney | yep | 16:37 |
sean-k-mooney | so going from inst_dict to list(inst_dict) when form full dict content to just keys whic is cool | 16:38 |
sean-k-mooney | or we could have dont inst_dict.keys() to be explicit but same effect | 16:39 |
stephenfin | sean-k-mooney: Python be cool like that | 16:39 |
sean-k-mooney | yes and no | 16:40 |
stephenfin | list(dict) has the advantage that you'll always get a list back | 16:40 |
sean-k-mooney | i know why python iterate over the keys not the items but it always trips me on things like that | 16:40 |
stephenfin | whereas with dict.keys(), you'll get an iterator on Python3 | 16:40 |
sean-k-mooney | also true | 16:40 |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add infrastructure for invoking libvirt's getDomainCapabilities API https://review.opendev.org/655268 | 16:40 |
stephenfin | and I don't know how oslo.log handles that | 16:40 |
sean-k-mooney | i just would have expected list(dict) to be the same as list(dict.itmes()) but i also know why that is not the case | 16:41 |
aspiers | efried, sean-k-mooney: new patchset 4 and response to patchset 3 ^^^ | 16:41 |
efried | stephenfin: would you mind taking a quick look at this? https://review.opendev.org/#/c/659682/ | 16:42 |
efried | The tox failure is weird, it's claiming it can't find deps, but they're being set in [testenv] | 16:42 |
stephenfin | looking | 16:42 |
efried | stephenfin: This is a throwaway patch, so if it's fixable by adding explicit deps lines back in, that's fine. | 16:43 |
stephenfin | efried: They're being set in 'testenv', not 'testenv:functional' | 16:43 |
efried | no inheritance there? | 16:43 |
efried | okay | 16:43 |
stephenfin | Nope. tox isn't that clever | 16:43 |
stephenfin | Thankfully, tbh. It's already too clever | 16:44 |
efried | so just copy them into those envs | 16:44 |
efried | also would have thought having used envdir would be enough | 16:44 |
efried | but apparently something is *requiring* deps? | 16:45 |
openstackgerrit | Eric Fried proposed openstack/nova stable/stein: DNM: Revert "Use external placement in functional tests" https://review.opendev.org/659682 | 16:45 |
openstackgerrit | Eric Fried proposed openstack/nova stable/stein: Skip _exclude_nested_providers() if not nested https://review.opendev.org/659206 | 16:45 |
efried | thanks for the look stephenfin | 16:45 |
*** xek_ has quit IRC | 16:46 | |
*** davidsha has quit IRC | 16:48 | |
*** sapd1_x has quit IRC | 16:48 | |
*** mriedem is now known as mriedem_away | 16:49 | |
*** maciejjozefczyk has quit IRC | 16:55 | |
*** tesseract has quit IRC | 16:58 | |
efried | aspiers: thanks for the updates. Convince me that LibvirtConfigDomainCapsFeature needs to exist *ever*. | 16:58 |
aspiers | efried: just saw your comment on the follow-up change about that | 16:59 |
aspiers | efried: think I see where you're coming from now, maybe | 16:59 |
*** tesseract has joined #openstack-nova | 16:59 | |
aspiers | so your point is that a base class isn't needed here | 16:59 |
aspiers | even though we expect there to be other feature classes in the future | 16:59 |
aspiers | or at least, have no reason to believe that there definitely won't be | 17:00 |
aspiers | right? | 17:00 |
efried | Right, because the only thing you're doing with that base class is converting a positional 'name' arg into the 'root_name' kwarg. | 17:00 |
aspiers | Yeah, that's a fair comment | 17:00 |
efried | which the subclasses can do just as easily. | 17:00 |
aspiers | but couldn't the same argument apply to LibvirtConfigGuestDeviceAddress? | 17:01 |
aspiers | I was trying to follow the precedent which already seemed to exist in config.py | 17:01 |
*** derekh has quit IRC | 17:02 | |
aspiers | Anyway I'm not religious about the need for that base class | 17:03 |
efried | Looking... but I tend to be in the camp of "follow precedent thoughtfully" | 17:03 |
aspiers | We could always reintroduce a base class later if need be | 17:03 |
aspiers | I'm in the same camp for sure :) | 17:03 |
efried | no, **DeviceAddress does some actual things in its overrides. | 17:03 |
aspiers | I don't see any method in the DeviceAddress base class which adds any value | 17:04 |
aspiers | the subclass overrides do add value, naturally | 17:04 |
aspiers | but actually there's a much better example which I remember following | 17:05 |
aspiers | LibvirtConfigGuestFeature | 17:05 |
efried | https://review.opendev.org/#/c/633855/13/nova/virt/libvirt/config.py@1351 ? | 17:05 |
aspiers | I was just copying that approach | 17:05 |
efried | Yes, LibvirtConfigGuestFeature is similarly worthless and could be removed imo. | 17:06 |
efried | (separately) | 17:06 |
aspiers | OK, so LibvirtConfigGuestDeviceAddress.__init__() does *one* thing of use: setting self.type | 17:06 |
aspiers | so not entirely redundant | 17:07 |
efried | yeah, but the other methods do real things. | 17:07 |
aspiers | they are overridden by both subclasses though | 17:07 |
*** igordc has joined #openstack-nova | 17:07 | |
aspiers | oh, super() | 17:07 |
aspiers | OK | 17:07 |
aspiers | missed that | 17:07 |
efried | but for LibvirtConfigGuestFeature, there's no reason e.g. the next one couldn't just be | 17:07 |
efried | class LibvirtConfigGuestFeatureACPI(LibvirtConfigObject): | 17:07 |
efried | def __init__(self, **kwargs): | 17:07 |
efried | super(LibvirtConfigGuestFeatureACPI, self).__init__(root_name="acpi", | 17:07 |
efried | **kwargs) | 17:07 |
efried | similarly for the subsequent two. | 17:08 |
openstackgerrit | Doug Wiegley proposed openstack/nova master: Improve metadata server performance with large security groups https://review.opendev.org/656084 | 17:08 |
aspiers | Now that this conversation jogged my memory, I remember I was definitely emulating LibvirtConfigGuestFeature | 17:08 |
efried | etc etc | 17:08 |
aspiers | Yeah | 17:08 |
aspiers | Well like I said, I'm not religious about the base class at all | 17:08 |
aspiers | I originaly thought you were saying it was pointless for other reasons | 17:08 |
efried | yeah, if you were emulating... you could have just used that. | 17:08 |
efried | but I'd rather kill it. | 17:08 |
aspiers | fine | 17:09 |
aspiers | coming right up in PS5 | 17:09 |
efried | In fact, I'm going to propose a patch to get rid of all that other cruft and see if it breaks the world. | 17:09 |
sean-k-mooney | im expecting an rm -rf of the nova subdiretory | 17:10 |
aspiers | lol | 17:10 |
aspiers | sean-k-mooney: that would be one way of getting rid of all the bugs ;) | 17:10 |
aspiers | "Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away." | 17:10 |
aspiers | not sure St Exupery meant it quite like that though | 17:11 |
sean-k-mooney | well we have almost reached perfrection of cellsv1 in that case | 17:12 |
*** lbragstad has quit IRC | 17:12 | |
*** lbragstad has joined #openstack-nova | 17:12 | |
* stephenfin -> 🏡 🍻 | 17:12 | |
aspiers | efried: I can remove that cruft if you want | 17:14 |
efried | aspiers: I'm already working on it. | 17:14 |
aspiers | ok | 17:14 |
*** udesale has quit IRC | 17:15 | |
aspiers | efried: didn't anyone tell you PTLs aren't allowed to code? ;-p | 17:15 |
efried | sore point | 17:15 |
aspiers | strictly no fun allowed | 17:15 |
aspiers | just feeling smug cos I temporarily escaped endless meetings | 17:15 |
aspiers | I'm sure it won't last | 17:15 |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add infrastructure for invoking libvirt's getDomainCapabilities API https://review.opendev.org/655268 | 17:16 |
aspiers | and now a rebase for good measure ... | 17:17 |
*** ykarel|away has joined #openstack-nova | 17:17 | |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add infrastructure for invoking libvirt's getDomainCapabilities API https://review.opendev.org/655268 | 17:17 |
*** tesseract has quit IRC | 17:17 | |
aspiers | efried: I'm around for a while in case there are any remaining things you want addressed in that | 17:21 |
efried | aspiers: 2 mins and I'll have another look. | 17:21 |
aspiers | kewl | 17:22 |
openstackgerrit | Eric Fried proposed openstack/nova master: Remove no-op intermediary libvirt config classes https://review.opendev.org/659848 | 17:23 |
efried | aspiers: ^ | 17:23 |
*** ralonsoh has quit IRC | 17:24 | |
efried | aspiers: +2. Now you get to rebase the others on top :) | 17:24 |
efried | shouldn't be too hard | 17:24 |
aspiers | efried: already started | 17:24 |
aspiers | nah it's not | 17:25 |
*** ttsiouts has joined #openstack-nova | 17:26 | |
*** ivve has joined #openstack-nova | 17:30 | |
*** mrch_ has quit IRC | 17:40 | |
sean-k-mooney | aspiers: v6 of https://review.opendev.org/#/c/655268 looks good to me. if i want to reuse this for the device model based schduleing i will need to extend the config parsing logic accordingly but tha that shoudl be quite clean to do | 17:41 |
aspiers | sean-k-mooney: cool | 17:41 |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add detection of SEV support from QEMU/AMD-SP/libvirt on AMD hosts https://review.opendev.org/633855 | 17:42 |
aspiers | efried: there's the first rebase ^^^ | 17:42 |
aspiers | sean-k-mooney: ^^^ in case you want to see the first example of a real customer of that API | 17:43 |
sean-k-mooney | yep ill try an review the full sev series next week. | 17:43 |
sean-k-mooney | but for now time to go get food | 17:43 |
sean-k-mooney | o/ | 17:43 |
aspiers | sean-k-mooney: thanks, it's still incomplete but working on it | 17:43 |
aspiers | o/ | 17:43 |
efried | aspiers: Once kashyap and/or sean-k-mooney have +1ed https://review.opendev.org/#/c/633855/ I'll send it. | 17:51 |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add new "supports_amd_sev" capability to libvirt driver https://review.opendev.org/638680 | 17:51 |
aspiers | efried: thanks, and there's the next one on top of that ^^^ | 17:51 |
sean-k-mooney | efried: by the way really small change but can you comment on my comment in the commit message regarding bug vs specless blueprint | 17:51 |
sean-k-mooney | https://review.opendev.org/#/c/658785/4 | 17:51 |
efried | sean-k-mooney: looking | 17:52 |
efried | sean-k-mooney: I would need a better explanation of wtf this patch is doing. | 17:56 |
*** jdillaman has quit IRC | 17:56 | |
sean-k-mooney | efried: we support hardware offlwoad ovs with kernel ovs today | 17:56 |
efried | keeping in mind that my eyes glaze over any time I hear ovs or dpdk | 17:56 |
sean-k-mooney | this is part fo making it work with ovs-dpdk too | 17:56 |
efried | Let me put it this way: what would the reno say? | 17:57 |
sean-k-mooney | so this is just passign the ovs datapath type wich we have in teh vif to os-vif | 17:57 |
efried | What will the user notice? | 17:57 |
sean-k-mooney | opereators will now by able to deploy ovs-dpdk with hardwar offloads for nvf/telco deployment that need hi data rates and a fast fallback path when a feature is not supported in hardware | 17:58 |
sean-k-mooney | we could skip the reno i guess but it would be nice to call this out in 1 of nova, os-vif or neutron | 17:59 |
sean-k-mooney | the patch to enable it in all 3 repos are equally short however | 17:59 |
efried | sean-k-mooney: Okay, so a bug definitely doesn't seem appropriate; it sounds like a feature to me. | 18:00 |
sean-k-mooney | this prettymuch almost works already | 18:00 |
sean-k-mooney | ya i was thinking that too but its also a tiny change | 18:00 |
sean-k-mooney | hence sepcless blueprint | 18:00 |
efried | Right, so if it's truly tiny, a specless blueprint would be fine. | 18:00 |
efried | imo | 18:01 |
sean-k-mooney | well that patch is the total cahnge needed to nova | 18:01 |
efried | what would the operator need to do to make things happen? | 18:01 |
efried | (is the kind of thing I would want the bp to say) | 18:01 |
efried | is there an API change? | 18:01 |
efried | configs that affect? | 18:02 |
efried | docs that need to be updated? | 18:02 |
sean-k-mooney | just install ovs-dpdk on a host with a melonox nic that support hardware offloand and have the right version of nova and neutron | 18:02 |
sean-k-mooney | there is not api change or change to iamge or flaovr | 18:02 |
sean-k-mooney | you use the feature the same way as hardware offloaded kernel ovs | 18:02 |
openstackgerrit | Adam Spiers proposed openstack/nova master: Extract SEV-specific bits on host detection https://review.opendev.org/636334 | 18:03 |
sean-k-mooney | i am not sure if we have docs for that but assuming we do the might want to be updated | 18:03 |
efried | neat. So yeah, a specless blueprint sounds appropriate to me. Though I guess the problem is that you can't file a single blueprint against multiple projects? | 18:03 |
aspiers | ALL YOUR REBASE ARE BELONG TO US | 18:03 |
sean-k-mooney | efried: https://docs.openstack.org/neutron/rocky/admin/config-ovs-offload.html shoudl be updated | 18:03 |
sean-k-mooney | but that is in neutron | 18:03 |
aspiers | that's probably the best joke I ever made </tragically_proud_moment> | 18:04 |
sean-k-mooney | efried: yep which is why i was questioning should it be an rfe bug | 18:04 |
efried | aspiers: Cause you know I'm all about rebase, 'bout rebase... | 18:04 |
sean-k-mooney | but specless blueprint for nova and rfe bugs that is shared for os-vif/nueton is fine too | 18:04 |
efried | sean-k-mooney: Is that a thing? Then sure. | 18:04 |
aspiers | damn, instantly upstaged, just like that | 18:04 |
sean-k-mooney | efried: its what neutron uses for all feature requests | 18:05 |
efried | heh, okay. | 18:05 |
efried | wfm | 18:05 |
sean-k-mooney | we dont use rfe bugs in nova | 18:05 |
sean-k-mooney | its litrally just a rfe tag on the bug | 18:05 |
sean-k-mooney | cool anyway really going this time o/ | 18:06 |
aspiers | haha the number of times I've said that | 18:07 |
artom | aspiers, you have no chance to survive, merge your time | 18:08 |
aspiers | artom: somebody set up us the branch | 18:10 |
artom | aspiers, move TAG | 18:11 |
aspiers | CI was beginning | 18:11 |
aspiers | We get review | 18:12 |
aspiers | You are on the way to -2 | 18:12 |
aspiers | What you say !! | 18:12 |
artom | You've gone and rewritten the story. | 18:13 |
* artom bows. | 18:13 | |
aspiers | :D | 18:14 |
artom | Not just random puns, a coherent narrative | 18:14 |
aspiers | I try my best | 18:14 |
artom | You definitely know what you doing. | 18:16 |
aspiers | For great fun | 18:17 |
aspiers | OK, I think it's time for the weekend | 18:17 |
aspiers | Main screen turn off | 18:17 |
*** ykarel|away has quit IRC | 18:20 | |
*** mriedem_away is now known as mriedem | 18:23 | |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add <launchSecurity> element to libvirt guest XML for AMD SEV https://review.opendev.org/636318 | 18:31 |
efried | aspiers: https://review.opendev.org/#/c/638680/ reviewed, lemme know if we need to Talk. | 18:37 |
*** itlinux has joined #openstack-nova | 18:38 | |
*** whoami-rajat has quit IRC | 18:39 | |
aspiers | efried: looking | 18:41 |
aspiers | totally fell into the same trap as sean-k-mooney | 18:41 |
aspiers | Main screen did not turn off :-/ | 18:42 |
*** macza has joined #openstack-nova | 18:44 | |
*** maciejjozefczyk has joined #openstack-nova | 18:45 | |
openstackgerrit | Adam Spiers proposed openstack/nova master: Add detection of SEV support from QEMU/AMD-SP/libvirt on AMD hosts https://review.opendev.org/633855 | 18:46 |
*** macza has quit IRC | 18:49 | |
*** mvkr has joined #openstack-nova | 18:49 | |
aspiers | efried: thanks, makes sense to move it to u_p_t() | 18:52 |
aspiers | I'll do that next week | 18:52 |
*** maciejjozefczyk has quit IRC | 18:53 | |
efried | cool, thanks. | 18:54 |
aspiers | efried: so that would mean it's no longer called from init_host(), right? | 18:55 |
openstackgerrit | Eric Fried proposed openstack/nova master: Introduces the openstacksdk to nova https://review.opendev.org/643664 | 18:57 |
openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Use OpenStack SDK for placement https://review.opendev.org/656023 | 18:57 |
openstackgerrit | Eric Fried proposed openstack/nova master: Introduces the openstacksdk to nova https://review.opendev.org/643664 | 18:57 |
openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Use OpenStack SDK for placement https://review.opendev.org/656023 | 18:57 |
*** tjgresha has quit IRC | 18:59 | |
aspiers | ComputeManager.pre_start_hook() -> .update_available_resource() -> ._update_available_resource_for_node() -> ResourceTracker.update_available_resource() -> ._update_available_resource() -> ._update() -> ._update_to_placement() -> driver.updated_provider_tree() | 19:00 |
aspiers | efried: but it seems that there are many other code paths which could reach update_provider_tree(), so your previous question about locking of the region modifying _domain_caps now will have a different answer | 19:01 |
efried | aspiers: I don't have a problem with the call chain through init_host as you've currently got it setting *some* instance variable on the libvirt driver. | 19:02 |
efried | which upt can then just go read | 19:02 |
aspiers | although ResourceTracker._update_available_resource has a semaphore, so maybe that's OK anyway? | 19:02 |
efried | it just shouldn't be in the capability dict. | 19:02 |
aspiers | OK, just wondering where's the best place to put it | 19:02 |
efried | self._supports_sev | 19:03 |
efried | whatever | 19:03 |
aspiers | yeah I guess the way you suggest works | 19:03 |
*** jmlowe has joined #openstack-nova | 19:03 | |
efried | you can put a comment saying we do this during init_host for efficiency and because it doesn't change | 19:03 |
aspiers | right - if I put it in resource_tracker.py then it would probably get called too often (i.e. more than once) | 19:03 |
efried | right. And the singleton-ness you've got set up on it would partially mitigate that - but there's no reason to even have the extra call stacks in the way. | 19:04 |
efried | memoizing, not singleton | 19:04 |
aspiers | yep | 19:04 |
efried | how's that whole stopping-working-because-it's-eod-friday thing going for you? | 19:05 |
aspiers | really great, just fantastic | 19:05 |
efried | You got a taste for it in Denver, now you can't get enough. | 19:06 |
aspiers | sad but true | 19:07 |
*** dannins has quit IRC | 19:08 | |
openstackgerrit | Eric Fried proposed openstack/nova stable/stein: DNM: Revert "Use external placement in functional tests" https://review.opendev.org/659682 | 19:23 |
openstackgerrit | Eric Fried proposed openstack/nova stable/stein: Skip _exclude_nested_providers() if not nested https://review.opendev.org/659206 | 19:23 |
aspiers | efried: last question before I go, you're saying that I don't need a driver capability at all, and u_p_t() would just provide the trait directly? | 19:25 |
efried | aspiers: Correct | 19:25 |
aspiers | OK good, thought so | 19:25 |
efried | aspiers: From the operator's point of view: | 19:25 |
efried | if they want mem encryption, they request resource MEM_ENCRYPTION_CONTEXT=1 | 19:26 |
efried | if they want SEV specifically, they request required=SEV | 19:26 |
aspiers | BTW the kernel module parameter file is readable non-root, so I'm assuming privsep not required | 19:26 |
aspiers | right | 19:26 |
efried | okay. | 19:26 |
aspiers | IIRC, what you just wrote is captured in the spec | 19:26 |
aspiers | I remember us going over that before | 19:26 |
efried | I think so, yes | 19:26 |
aspiers | sweet | 19:26 |
efried | but I guess we never talked about driver capabilities | 19:27 |
aspiers | right | 19:27 |
aspiers | OK, *really* going now :) | 19:27 |
aspiers | have a great w/e! | 19:27 |
efried | yeah, it just seems silly for PowerVM, which will never ever ever have AMD SEV support, to have to say no every time. | 19:27 |
efried | later o/ | 19:27 |
aspiers | yes totally, felt wrong to have to add to powervm driver | 19:27 |
aspiers | cya! o/ | 19:27 |
*** priteau has quit IRC | 19:31 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: api-ref: mention default project filtering when listing servers https://review.opendev.org/659873 | 19:32 |
*** hamzy has joined #openstack-nova | 19:35 | |
*** efried has quit IRC | 19:43 | |
*** efried has joined #openstack-nova | 19:43 | |
*** imacdonn has quit IRC | 19:45 | |
*** raub has joined #openstack-nova | 19:45 | |
*** raub has left #openstack-nova | 19:45 | |
*** slaweq has quit IRC | 19:47 | |
*** mchlumsky has quit IRC | 19:47 | |
*** ircuser-1 has quit IRC | 19:50 | |
*** imacdonn has joined #openstack-nova | 19:52 | |
*** ircuser-1 has joined #openstack-nova | 19:53 | |
*** ttsiouts has quit IRC | 19:59 | |
*** pcaruana has quit IRC | 20:02 | |
*** slaweq has joined #openstack-nova | 20:04 | |
*** itlinux has quit IRC | 20:06 | |
*** eandersson has joined #openstack-nova | 20:14 | |
*** ttsiouts has joined #openstack-nova | 20:15 | |
*** ttsiouts has quit IRC | 20:20 | |
*** slaweq has quit IRC | 20:32 | |
*** ttsiouts has joined #openstack-nova | 20:51 | |
*** ttsiouts has quit IRC | 20:56 | |
*** igordc has quit IRC | 21:02 | |
*** xek_ has joined #openstack-nova | 21:02 | |
*** igordc has joined #openstack-nova | 21:03 | |
*** xek_ has quit IRC | 21:08 | |
*** xek_ has joined #openstack-nova | 21:08 | |
*** xek_ has quit IRC | 21:10 | |
*** xek has joined #openstack-nova | 21:11 | |
*** xek has quit IRC | 21:13 | |
*** hongbin has joined #openstack-nova | 21:17 | |
*** ttsiouts has joined #openstack-nova | 21:28 | |
*** whoami-rajat has joined #openstack-nova | 21:31 | |
*** xek has joined #openstack-nova | 21:31 | |
*** igordc has quit IRC | 21:33 | |
*** igordc has joined #openstack-nova | 21:33 | |
*** JamesBen_ has joined #openstack-nova | 21:48 | |
openstackgerrit | Eric Fried proposed openstack/nova stable/stein: Skip _exclude_nested_providers() if not nested https://review.opendev.org/659206 | 21:48 |
openstackgerrit | Merged openstack/nova-specs master: Follow up for I9200c2b18369da8b9194a8ccbbf6f1baf6a9613c https://review.opendev.org/659300 | 21:49 |
efried | mriedem, melwitt: ---^ stein & rocky backports are ready for your stable eyes now. | 21:50 |
*** JamesBe__ has joined #openstack-nova | 21:50 | |
melwitt | ack | 21:50 |
*** JamesBenson has quit IRC | 21:51 | |
mriedem | my eyeballs are currently embargoed | 21:52 |
mriedem | my sweet dance moves, however, are not | 21:52 |
*** JamesBe__ has quit IRC | 21:54 | |
*** JamesBen_ has quit IRC | 21:54 | |
*** JamesBenson has joined #openstack-nova | 21:56 | |
*** itlinux has joined #openstack-nova | 21:58 | |
*** JamesBenson has quit IRC | 22:00 | |
*** hongbin has quit IRC | 22:01 | |
*** rchurch_ has joined #openstack-nova | 22:01 | |
*** ttsiouts has quit IRC | 22:01 | |
*** hongbin has joined #openstack-nova | 22:02 | |
*** rchurch has quit IRC | 22:02 | |
openstackgerrit | Matt Riedemann proposed openstack/python-novaclient master: Allow searching for hypervisors and getting back details https://review.opendev.org/659886 | 22:09 |
*** mriedem has quit IRC | 22:20 | |
*** jcosmao has joined #openstack-nova | 22:36 | |
*** jcosmao has left #openstack-nova | 22:37 | |
*** dave-mccowan has quit IRC | 22:50 | |
*** luksky has quit IRC | 22:50 | |
*** itlinux_ has joined #openstack-nova | 22:51 | |
*** hongbin has quit IRC | 22:52 | |
*** itlinux has quit IRC | 22:52 | |
*** tbachman has quit IRC | 23:08 | |
*** macza has joined #openstack-nova | 23:11 | |
*** macza has quit IRC | 23:15 | |
*** mvkr has quit IRC | 23:21 | |
*** gyee has quit IRC | 23:25 | |
*** francoisp has quit IRC | 23:27 | |
*** whoami-rajat has quit IRC | 23:40 | |
*** mmethot has quit IRC | 23:53 | |
*** artom has quit IRC | 23:53 | |
*** ttsiouts has joined #openstack-nova | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!