| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-06-29 | |||
| 17:01:50 | sean-k-mooney | in this case though virtio will work in all configuration i belive | |
| 17:02:02 | sean-k-mooney | so we really just need to check if tis build and if tis set in the image | |
| 17:02:10 | sean-k-mooney | if not set it in the image metada copy we have | |
| 17:03:59 | dansmith | well, I'd really say we should avoid breaking sensitive windows vms by changing anything much | |
| 17:04:03 | sean-k-mooney | so basiclly image_meta.properties.set('hw_video_model', image_meta.properties.get('hw_video_model')) | |
| 17:04:26 | sean-k-mooney | * image_meta.properties.set('hw_video_model', image_meta.properties.get('hw_video_model', 'virtio')) | |
| 17:04:53 | dansmith | https://www.howson.pro/content/images/2016/07/sound-popped-up-after-fi.png | |
| 17:04:57 | dansmith | don't want that on your cloud instance :P | |
| 17:05:23 | sean-k-mooney | hehe no that would be awkward | |
| 17:06:00 | sean-k-mooney | i mean you can attach a cinder volume as a driver disk but it sucks | |
| 17:06:56 | dansmith | let us not go there :) | |
| 17:08:03 | sean-k-mooney | you can kind of do that today actully but its really inovlved and convulted | |
| 17:09:11 | dansmith | lol | |
| 17:10:16 | sean-k-mooney | dansmith: so 1.) set new default in image metata copy on new spawn. 2.) keep current logic in driver which will use the image vaule preferencally if set. 3.) recored value if not in image after its calulated for exising instance? | |
| 17:10:38 | sean-k-mooney | 3.) would only run once per instance the first time they hard reboot | |
| 17:11:12 | sean-k-mooney | and after that it just uses what in the insance_system_metadata | |
| 17:11:45 | sean-k-mooney | then rise and repeat taht for any default we want to change like this in the future | |
| 17:12:11 | sean-k-mooney | is that about right ^ if so it would be nice to add to the contibutors docs | |
| 17:12:41 | sean-k-mooney | i can proably submit a patch for that | |
| 17:13:11 | dansmith | well, I was going to say just always assume cirrus if it's not set | |
| 17:13:31 | dansmith | instead of "fixing" all the existing instances, but either works, as long as we assume cirrus if not set there | |
| 17:13:48 | sean-k-mooney | well assume exsiting behavior | |
| 17:13:55 | sean-k-mooney | which will be cirrus on x86 | |
| 17:14:04 | opendevreview | Merged openstack/nova master: Add test coverage for API version headers in CORS https://review.opendev.org/c/openstack/nova/+/796580 | |
| 17:14:17 | sean-k-mooney | its vga on power and virtio on arm | |
| 17:14:25 | dansmith | yeah | |
| 17:14:57 | opendevreview | Merged openstack/nova master: Fix typos in minimum version policy docs https://review.opendev.org/c/openstack/nova/+/795575 | |
| 17:15:30 | sean-k-mooney | kashyap: does ^ work/make sense to you | |
| 17:15:49 | opendevreview | Merged openstack/nova master: Make test_refresh_associations_* deterministic https://review.opendev.org/c/openstack/nova/+/794396 | |
| 17:16:49 | sean-k-mooney | dansmith: we may have other default like this we want to change but i cant rememebr them at present which is why i want to document the workflow for this type of change | |
| 17:16:58 | dansmith | ack | |
| 17:49:40 | opendevreview | Merged openstack/nova master: Remove PROJECT_ADMIN limitation from zero-disk and external-network policy https://review.opendev.org/c/openstack/nova/+/794360 | |
| 17:50:23 | opendevreview | Merged openstack/nova master: Improve policy doc for supported scope info https://review.opendev.org/c/openstack/nova/+/762013 | |
| 20:57:23 | slaweq | Hi nova-stable-cores, can You check https://review.opendev.org/c/openstack/nova/+/787253? Thx in advance | |
| #openstack-nova - 2021-06-30 | |||
| 01:32:02 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - new vnic type https://review.opendev.org/c/openstack/nova/+/771363 | |
| 01:32:02 | opendevreview | Yongli He proposed openstack/nova master: Smartnic support - cyborg drive https://review.opendev.org/c/openstack/nova/+/771362 | |
| 01:32:03 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - create arqs https://review.opendev.org/c/openstack/nova/+/758944 | |
| 01:32:04 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - cleanup arqs https://review.opendev.org/c/openstack/nova/+/798054 | |
| 01:32:04 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - build instance with smartnic arqs https://review.opendev.org/c/openstack/nova/+/798249 | |
| 01:32:06 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - functional tests https://review.opendev.org/c/openstack/nova/+/780147 | |
| 01:32:06 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - reject server move and suspend https://review.opendev.org/c/openstack/nova/+/779913 | |
| 07:37:38 | ralonsoh | sean-k-mooney, hi good morning! Just a heads-up: the idea for OVN (and OVS because is the same VIF type) has been approved. Well, at least the RFE. Now I need to push a spec. | |
| 07:37:47 | ralonsoh | but the idea was understood and accepted | |
| 07:38:32 | ralonsoh | context (for anyone else): https://bugs.launchpad.net/neutron/+bug/1933517 | |
| 07:54:33 | opendevreview | Yongli He proposed openstack/nova master: Smartnic support - cyborg drive https://review.opendev.org/c/openstack/nova/+/771362 | |
| 07:54:34 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - new vnic type https://review.opendev.org/c/openstack/nova/+/771363 | |
| 07:54:36 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - create arqs https://review.opendev.org/c/openstack/nova/+/758944 | |
| 07:54:37 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - cleanup arqs https://review.opendev.org/c/openstack/nova/+/798054 | |
| 07:54:37 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - build instance with smartnic arqs https://review.opendev.org/c/openstack/nova/+/798249 | |
| 07:54:38 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - reject server move and suspend https://review.opendev.org/c/openstack/nova/+/779913 | |
| 07:54:39 | opendevreview | Yongli He proposed openstack/nova master: smartnic support - functional tests https://review.opendev.org/c/openstack/nova/+/780147 | |
| 08:43:48 | yonglihe | Hi gibi, alex had finished the core logic review and +2 already. I split original 3rd patch to 3 one, that make review easier. Seems lots of review work, thanks. | |
| 10:56:05 | kashyap | sean-k-mooney: I'll read back the discussion and respond. | |
| 10:58:26 | sean-k-mooney | kashyap: tl;dr dansmith proposed a way that shoudl allow you to do the change of default this cycle by not changing the default in the code that currently sets it and instead populating the image metadata copy we have with the new default if its an intial boot | |
| 10:59:49 | kashyap | sean-k-mooney: Also, I was wondering why not also set the default for x86 in _add_video_driver()? | |
| 11:00:04 | kashyap | sean-k-mooney: The image metadat copy we have on initial boot: got a code pointer here? | |
| 11:00:17 | sean-k-mooney | because we dont want to change existing image | |
| 11:00:46 | kashyap | sean-k-mooney: So you're saying: we don't want to change the default for existing images on cold-boot? | |
| 11:00:54 | sean-k-mooney | what we can do outside of _add_video_driver is update the exisitng vms to store what they have | |
| 11:01:00 | sean-k-mooney | kashyap: correct | |
| 11:01:18 | sean-k-mooney | we do not want exisating vms to stop using cirrus automatically | |
| 11:01:23 | kashyap | sean-k-mooney: Sorry, if it was already discussed - but why not? | |
| 11:01:33 | kashyap | What does it break? | |
| 11:02:00 | sean-k-mooney | other then rhel6 we are not sure if it will break anything | |
| 11:02:45 | sean-k-mooney | you could try and test that. grab a windows image boot with cirrus use virt manager to change it and reboot | |
| 11:02:54 | kashyap | (I'm not sure even if RHEL6 actually breaks here. Needs to be tested.) | |
| 11:02:55 | sean-k-mooney | it proably will be fine | |
| 11:03:11 | kashyap | Yeah; needs testing. | |
| 11:03:22 | sean-k-mooney | kashyap: have you confirm if virtio works on all architectures | |
| 11:03:32 | sean-k-mooney | its the default on arm but powerpc defaults to vga | |
| 11:03:56 | sean-k-mooney | do we need to check the arch of the image | |
| 11:04:06 | sean-k-mooney | or just use virtio always | |
| 11:04:07 | kashyap | sean-k-mooney: "all" archs, I don't know. What do we care about? | |
| 11:04:28 | kashyap | For x86 and Arm, 'virtio' disply works fine | |
| 11:04:34 | sean-k-mooney | power8/9 aarch64 and x86_64 really | |
| 11:05:12 | kashyap | I can check for power8/9 - although I doubt how many non-topy deployments actually are there for it | |
| 11:52:07 | sean-k-mooney | kashyap: the power kvm ci should test that config for us | |
| 12:02:19 | kashyap | sean-k-mooney: I see; is the Power KVM CI is maintianed? | |
| 12:03:07 | sean-k-mooney | its a requirement for in tree dirver to remain so i hope so | |
| 12:03:11 | sean-k-mooney | i think it is | |
| 12:20:11 | kashyap | Ack | |
| 13:10:03 | ganso | bauzas, lyarwood: hi! when you have a minute could you please take a look at the train backport now (it is the same one as yesterday): https://review.opendev.org/c/openstack/nova/+/798717 ? Thanks in advance | |
| 13:30:25 | dansmith | sean-k-mooney: kashyap I thought you were already talking about annotating instances to avoid changing the default on existing ones, but with inverse logic and a waiting cycle | |
| 13:30:51 | dansmith | I definitely think that when we can avoid switching the ABI on an instance suddenly without a lot of other work, we should do that | |
| 13:31:21 | dansmith | vmware maintains a hardware reference version and while it prompts you after every upgrade to upgrade the guest ABI, it doesn't force you (until it's very old) | |
| 13:31:51 | dansmith | given how hard it might be to rescue a cloud instance with broken video, and how easy it is to avoid it and not even require a waiting cycle, I think it's worth doing | |
| 13:32:05 | kashyap | dansmith: Hi, reading back | |
| 13:32:21 | kashyap | dansmith: I'm a bit foggy here, what's the "inverse logic" referred to here? | |
| 13:32:42 | kashyap | dansmith: Nothing is "settled" here, BTW. I'm going to do a bunch of tests to come back with concrete results | |
| 13:33:02 | dansmith | IIUC the original proposal was to annotate all instances as "old video" in this release, so that next release we could switch anything not annotated to the new video | |
| 13:33:11 | dansmith | that's the opposite of what sean-k-mooney was describing from our conversation above | |
| 13:33:35 | kashyap | I see | |
| 13:33:37 | dansmith | kashyap: sure, but you won't be able to test every type of guest with every config, so just not changing the ABI is the safest thing, IMHO | |
| 13:34:58 | kashyap | dansmith: Yeah; I agree. I just wanted to get a sense of what breaks, if anything. So if I got you right: you're indeed saying to have a deprecation cycle - yes? | |
| 13:35:07 | dansmith | no | |
| 13:35:21 | dansmith | I'm saying if we invert the logic where we would need one, we can get away without one | |
| 13:35:22 | sean-k-mooney | dansmith: yes i was suggesting annotating the instance so they dont change | |
| 13:35:39 | dansmith | ack | |
| 13:35:54 | sean-k-mooney | so if the image meta is not set when we do a hard reboot use old logic then recored it so next time we just use teh db value | |
| 13:36:15 | dansmith | and record the new model on ambiguous new boots, so they take the new default | |