| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-07-09 | |||
| 11:59:04 | sean-k-mooney | lyarwood: you had another spec similar to https://review.opendev.org/c/openstack/nova-specs/+/799850 for addign the attchment id right | |
| 11:59:16 | sean-k-mooney | can we use the same microversion for both | |
| 11:59:58 | sean-k-mooney | https://review.opendev.org/c/openstack/nova-specs/+/794799 | |
| 12:00:12 | lyarwood | we could but I thought it needed something separate for the admin/policy driven part | |
| 12:00:12 | sean-k-mooney | i kid of feel like those two api cahnges shoudl be done together | |
| 12:01:19 | sean-k-mooney | i mean we could leave that to the implemation it would be nice not to have to bump twice | |
| 12:02:25 | sean-k-mooney | am one other question | |
| 12:02:39 | sean-k-mooney | we wanted to stop stashing the connection info in nova at somepoitn right | |
| 12:02:58 | sean-k-mooney | and just alwasy get it form cinder | |
| 12:04:18 | sean-k-mooney | basically if we plan to not store it in the db long term i dont think we should add it to the api respocne so that we do not need then proxy it in the future | |
| 12:09:09 | lyarwood | sorry back | |
| 12:09:40 | lyarwood | yeah I don't mind using the same microversion for both if it matters that much | |
| 12:09:55 | lyarwood | and yeah eventually I'd like to but there's no written down plan to attack that this cycle yet | |
| 12:10:16 | sean-k-mooney | it does not but im not conviced we should should add the new api | |
| 12:10:32 | sean-k-mooney | i think we should just proceed with https://review.opendev.org/c/openstack/nova-specs/+/799624 more then likely | |
| 12:11:00 | sean-k-mooney | although i have not read that yet but i think nova manage would be better if we plan to remove the cacheing eventually | |
| 12:11:46 | lyarwood | I disagree, we can always remove it in a later microversion | |
| 12:11:58 | sean-k-mooney | we could but we will have to proxy for old microverions | |
| 12:11:59 | lyarwood | and when it is there it's just going to be None once we've removed the stashing in Nova | |
| 12:12:24 | lyarwood | why would we need to proxy? | |
| 12:12:27 | sean-k-mooney | no if we remove the stashing we will have to have nova proxy to nova for the old microverion | |
| 12:12:32 | lyarwood | nope | |
| 12:12:39 | lyarwood | it's the connection_info that Nova has | |
| 12:12:41 | sean-k-mooney | it would break people if we did not | |
| 12:12:48 | lyarwood | how would it break people? | |
| 12:13:09 | sean-k-mooney | if people used the old microversion then they would expect the connection info | |
| 12:13:21 | sean-k-mooney | and it would nolonger be presnt when we stop stashing | |
| 12:13:27 | lyarwood | they'd expect the connection_info that Nova held about a volume attachment | |
| 12:13:40 | lyarwood | they're not going to use it for anything | |
| 12:13:47 | lyarwood | it's just a troubleshooting tool | |
| 12:14:01 | sean-k-mooney | which i dont think shoudl be at the api level | |
| 12:14:16 | sean-k-mooney | if we put it in the api then it will be used for other things | |
| 12:15:29 | sean-k-mooney | the fact that nova caches this info is an internal implementation detail | |
| 12:15:54 | sean-k-mooney | we also have a network info cache but we dont expose it via the api | |
| 12:16:41 | sean-k-mooney | i dont think we shoudl be treatign novas copy of the the connection info as something different from cinders at the api level | |
| 12:16:45 | sean-k-mooney | its just a cache | |
| 12:16:49 | sean-k-mooney | IMO | |
| 12:17:21 | lyarwood | Partly, there's also some additional stuff we stash in there on the Nova side at the moment | |
| 12:17:36 | lyarwood | like the device path, multipath UUIDs etc | |
| 12:17:51 | lyarwood | I still think it's entirely valid in the API | |
| 12:18:12 | lyarwood | there's nothing a caller could do with it anyway outside of calling os-brick with it | |
| 12:18:40 | sean-k-mooney | well you could use it to connect to the backend | |
| 12:18:56 | lyarwood | right I mean with our APIs | |
| 12:19:09 | lyarwood | so the fact it's going away in the future shouldn't matter | |
| 12:19:17 | lyarwood | and there's always another way of getting it | |
| 12:20:07 | sean-k-mooney | i dont know to me this just feels like adding tech debt | |
| 12:20:25 | sean-k-mooney | the admin should not really need to know or care about this | |
| 12:20:48 | sean-k-mooney | i get that its for troble shooting | |
| 12:21:16 | sean-k-mooney | but wont the nova manage command provide a better way to do that | |
| 12:21:23 | sean-k-mooney | as that will also supprot refershing it | |
| 12:21:32 | sean-k-mooney | so fixing the problem if it exitis | |
| 12:24:08 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Unify 'nova.db.api', 'nova.db.sqlalchemy.api' https://review.opendev.org/c/openstack/nova/+/799524 | |
| 12:24:09 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Post reshuffle cleanup https://review.opendev.org/c/openstack/nova/+/799526 | |
| 12:24:09 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Move remaining 'nova.db.sqlalchemy' modules https://review.opendev.org/c/openstack/nova/+/799525 | |
| 12:24:10 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Add initial alembic migration for main DB https://review.opendev.org/c/openstack/nova/+/799527 | |
| 12:24:10 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Avoid use of ALTER in initial migration https://review.opendev.org/c/openstack/nova/+/800076 | |
| 12:24:11 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Trivial style changes https://review.opendev.org/c/openstack/nova/+/799529 | |
| 12:24:11 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Add initial alembic migration for API DB https://review.opendev.org/c/openstack/nova/+/799528 | |
| 12:24:12 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Normalize migrations tests https://review.opendev.org/c/openstack/nova/+/799684 | |
| 12:24:13 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Enable auto-generation of migrations https://review.opendev.org/c/openstack/nova/+/800077 | |
| 12:24:13 | opendevreview | Stephen Finucane proposed openstack/nova master: db: Integrate alembic https://review.opendev.org/c/openstack/nova/+/799530 | |
| 12:24:15 | opendevreview | Stephen Finucane proposed openstack/nova master: docs: Add documentation on database migrations https://review.opendev.org/c/openstack/nova/+/800078 | |
| 12:34:48 | viks__ | hi, i'm facing some issue in network interface naming... after i create an ubuntu instance, i assign a private interface, and then ubuntu instance gets an interface named `ens7`... but when i power off and then start the instance again, the interface name changes to `ens4`. I'm not sure how to avoid this name changing? Can someone plz guide? | |
| 12:39:30 | sean-k-mooney | viks__: that is because of two things one the pci adddreess of the instance likely change and 2 hotplug name are normally not stable in general | |
| 12:39:58 | sean-k-mooney | one way to adress this is with udev rules that mach on the mac to set the name | |
| 12:40:50 | sean-k-mooney | viks__: in generally if you shoudl leverage device role tagging https://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/virt-device-role-tagging.html | |
| 12:41:49 | opendevreview | Merged openstack/nova stable/ussuri: [neutron] Get only ID and name of the SGs from Neutron https://review.opendev.org/c/openstack/nova/+/787253 | |
| 12:44:30 | viks__ | sean-k-mooney: if we tag, the names will persist after instance comes up again? | |
| 12:45:41 | sean-k-mooney | no | |
| 12:45:53 | sean-k-mooney | the name is out of the contol of openstack | |
| 12:46:19 | sean-k-mooney | but if you tag it will provide a way for you to lookup the device and not depend on the name | |
| 12:46:48 | sean-k-mooney | in the metadata api you will get info like this | |
| 12:46:51 | sean-k-mooney | { | |
| 12:46:53 | sean-k-mooney | "type": "nic", | |
| 12:46:55 | sean-k-mooney | "bus": "pci", | |
| 12:46:57 | sean-k-mooney | "address": "0000:00:03.0", | |
| 12:46:59 | sean-k-mooney | "mac": "01:22:22:42:22:21", | |
| 12:47:01 | sean-k-mooney | "tags": ["nfvfunc2"] | |
| 12:47:03 | sean-k-mooney | }, | |
| 12:47:11 | sean-k-mooney | and you can have a script at boot name you interfaces as you wish | |
| 12:47:54 | sean-k-mooney | viks__: the nic names should not change again provide you dont add or remvoe any nics or volumes to the vm | |
| 12:49:58 | viks__ | sean-k-mooney: Thanks a lot.. will try out... | |
| 13:18:05 | sean-k-mooney | lyarwood: +1 on https://review.opendev.org/c/openstack/nova-specs/+/799624 i an alternitive approch inline and just want you to confirm it wont work | |
| 13:18:30 | opendevreview | Elod Illes proposed openstack/nova stable/train: [neutron] Get only ID and name of the SGs from Neutron https://review.opendev.org/c/openstack/nova/+/787316 | |
| 13:54:51 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor assertPortMatchesAllocation https://review.opendev.org/c/openstack/nova/+/792458 | |
| 13:54:52 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] ports with both bw and pps resources https://review.opendev.org/c/openstack/nova/+/792394 | |
| 13:54:52 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor asserts in qos tests https://review.opendev.org/c/openstack/nova/+/798930 | |
| 13:54:53 | opendevreview | Balazs Gibizer proposed openstack/nova master: Parse extended resource request from the port https://review.opendev.org/c/openstack/nova/+/800085 | |
| 14:00:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: Transfer RequestLevelParams from ports to scheduling https://review.opendev.org/c/openstack/nova/+/791506 | |
| 14:01:19 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 14:03:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test]Refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 14:03:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 14:03:01 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 14:06:51 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 15:02:48 | opendevreview | Kashyap Chamarthy proposed openstack/nova master: libvirt: Switch the default video model from 'cirrus' to 'virtio' https://review.opendev.org/c/openstack/nova/+/798680 | |
| 16:17:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 16:17:00 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 16:17:01 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 16:17:01 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test]Refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 16:19:36 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |