| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-06-18 | |||
| 16:36:23 | stephenfin | mriedem: Do you mean it's working and you're happy or that it shouldn't be working because the older compute node should be rejected? | |
| 16:36:30 | mriedem | just shows that when we're doing a live migration over n-1 and n level computes, if one is file-backed memory and the other isn't, we fail as the code is doing | |
| 16:36:35 | mriedem | i'm happy | |
| 16:36:58 | stephenfin | OK, that link was what I need. I'll re-review that in the morning so | |
| 16:37:03 | stephenfin | *needed | |
| 16:38:14 | mriedem | gibi: stephenfin: thanks for hitting https://review.openstack.org/#/c/561284/ | |
| 17:03:41 | openstackgerrit | Eric Fried proposed openstack/nova master: Handle agg generation conflict in report client https://review.openstack.org/556669 | |
| 17:07:37 | openstackgerrit | Eric Fried proposed openstack/nova master: Nix unused raise_if_custom_resource_class_pre_v1_1 https://review.openstack.org/575847 | |
| 17:33:40 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Spec: Handling Reshaped Provider Trees https://review.openstack.org/572583 | |
| 17:43:13 | zcorneli | dansmith, mriedem: For the file backed memory work, when going to implement the discard option (libvirt 4.4.0+), for migration XML generation we need to pass a second value back to the source from the destination. Is there a preference toward making the existing field into an Enum (or similar) vs creating a new field for discard? | |
| 17:45:57 | mriedem | the dst_wants_file_backed_memory field? | |
| 17:46:08 | zcorneli | yes | |
| 17:46:19 | mriedem | discard is a different thing and would be a different version of the migrate data object | |
| 17:46:32 | mriedem | so probably easiest to model that as a different field | |
| 17:47:54 | zcorneli | The only reason I would have against that would be that dst_wants_file_backed_memory=False + discard=True doesn't make sense. | |
| 17:48:11 | zcorneli | And I'm not sure if there's a way (or even really a will) to validate that in the object. | |
| 17:48:28 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: DNM: Use claim context during live migration https://review.openstack.org/576222 | |
| 17:49:23 | mriedem | would either of the is_shared_* fields make sense with dst_wants_file_backed_memory? | |
| 17:49:29 | mriedem | nvm, wrong type of thing | |
| 17:49:58 | mriedem | anyway, i don't think it's that big a deal, you'd just have a comment by the discard field saying it only makes sense when dst_wants_file_backed_memory=True | |
| 17:50:12 | zcorneli | Yep, works for me. Thanks for the input | |
| 17:52:36 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/queens: Add policy rule to block image-backed servers with 0 root disk flavor https://review.openstack.org/563692 | |
| 18:01:53 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: Add policy rule to block image-backed servers with 0 root disk flavor https://review.openstack.org/563700 | |
| 18:16:13 | crazik | huh, finally. all services in pike. | |
| 18:16:25 | crazik | i was a looong and hard journey | |
| 18:16:27 | dansmith | mriedem: hmm, did you know a migration record can go into "pre-migrating" state because of the resource tracker? | |
| 18:17:05 | crazik | dansmith: or other issue | |
| 18:17:11 | crazik | like DB schema mismatch | |
| 18:17:17 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/ocata: Add policy rule to block image-backed servers with 0 root disk flavor https://review.openstack.org/563719 | |
| 18:17:18 | dansmith | huh? | |
| 18:17:48 | dansmith | mriedem: I mean "will go" by the way | |
| 18:17:54 | mriedem | dansmith: i'm fairly sure that's on page 374 of the nova core handbook | |
| 18:18:34 | mriedem | why, what is it breaking? | |
| 18:18:49 | dansmith | mriedem: since you updated the comment above the list of filters, I challenge the fact that you knew that :P | |
| 18:19:11 | dansmith | mriedem: well, a migration can be in pre-migrating state, and thus neither of accepted or done, | |
| 18:19:18 | mriedem | what filters? | |
| 18:19:21 | dansmith | so the source compute may not find it if it comes up at the exact time | |
| 18:19:39 | dansmith | mriedem: https://review.openstack.org/#/c/500144/ | |
| 18:19:41 | mriedem | i assume you're referring to something else, like evac on startpu | |
| 18:20:03 | dansmith | yes, that's what I'm talking about | |
| 18:21:16 | mriedem | ok yeah i guess rebuild_claim will put the migration into pre-migrating status for evac | |
| 18:21:18 | dansmith | so it will go accepted->pre-migrating -> done -> completed | |
| 18:21:20 | dansmith | yeah | |
| 18:21:34 | mriedem | sorry i didn't have the full evac migration task state in my head when writing that comment... | |
| 18:21:45 | dansmith | well, I never knew it would do that | |
| 18:21:55 | dansmith | although I wonder if it would when I wrote that, because we did rebuild_claim recently, IIRC | |
| 18:22:30 | mriedem | as in 2015? https://review.openstack.org/#/c/200485/ | |
| 18:22:39 | dansmith | oh | |
| 18:22:58 | dansmith | I thought we added that while cleaning things up for placement, but maybe not | |
| 18:23:15 | mriedem | no, it was part of this https://review.openstack.org/#/q/topic:bug/1417667+(status:open+OR+status:merged) | |
| 18:23:42 | dansmith | yep, okay | |
| 18:24:00 | mriedem | i semi remember looking at that recently because of https://review.openstack.org/#/c/560467/ | |
| 18:24:23 | mriedem | and because we log a warning during every live migratoin | |
| 18:25:08 | mriedem | artom_: artom__: you might care about ^ | |
| 18:28:28 | artom | Sorry, moving around between rooms in the office :( | |
| 18:28:43 | artom | mriedem, thanks for the tip, I *would* care about that, but what's the context? | |
| 18:29:23 | mriedem | artom: RT claims code with live migration | |
| 18:29:27 | mriedem | which is now your thing | |
| 18:30:36 | artom | Sorta (see http://lists.openstack.org/pipermail/openstack-dev/2018-June/131605.html) | |
| 18:32:58 | crazik | :D | |
| 18:39:08 | openstackgerrit | Zack Cornelius proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876 | |
| 19:04:33 | openstackgerrit | Giridhar Jayavelu proposed openstack/nova master: Avoid redundant compute node update https://review.openstack.org/576235 | |
| 19:11:22 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Spec: Handling Reshaped Provider Trees https://review.openstack.org/572583 | |
| 19:15:06 | openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Compute: Handle reshaped provider trees https://review.openstack.org/576236 | |
| 19:52:46 | openstackgerrit | Eric Fried proposed openstack/nova master: Nix unused raise_if_custom_resource_class_pre_v1_1 https://review.openstack.org/575847 | |
| 20:18:29 | openstackgerrit | Merged openstack/nova master: Skip ServerShowV263Test.test_show_update_rebuild_list_server for cellsv1 https://review.openstack.org/576194 | |
| 20:24:15 | anupn | Hi Folks, I am trying to schedule an instance over my baremetal node which is in "Available" state, but "host not mapped to any cell" error is coming upon scheduling an instance. Could you please point me how to map the host to the nova_cell1? | |
| 20:25:03 | anupn | I tried running "nova-manage db sync", and "nova-manage cell_v2 simple_cell_setup" but no luch | |
| 20:25:06 | anupn | *luck | |
| 20:25:31 | mriedem | anupn: see the discover_hosts command https://docs.openstack.org/nova/latest/cli/nova-manage.html#nova-cells-v2 | |
| 20:27:31 | anupn | mriedem: Yes I did, and I don't see any hosts there | |
| 20:27:55 | mriedem | did you run it with the --by-service option? | |
| 20:28:20 | openstackgerrit | Dan Smith proposed openstack/nova master: Fix MigrateData object tests for compat routines https://review.openstack.org/576256 | |
| 20:28:21 | dansmith | zcorneli: ^ | |
| 20:29:01 | zcorneli | dansmith: Was just looking at jaypipes suggestion for that. | |
| 20:29:43 | dansmith | zcorneli: basically just what I did | |
| 20:29:44 | anupn | mriedem: Nope, i ran without it and I got a user warning "UserWarning: The psycopg2 wheel package will be renamed from release 2.8......" | |
| 20:29:45 | dansmith | except better | |
| 20:30:00 | mriedem | anupn: well, try running with it, since it's for ironic nodes | |
| 20:30:08 | zcorneli | Yep. I like the lambda, makes it more clear, I think. | |
| 20:30:19 | anupn | mriedem: still the same | |
| 20:30:29 | dansmith | zcorneli: I spent my time writing code, else I'd have committed my comment before jaypipes :P | |
| 20:30:53 | anupn | mriedem: Should i run "nova-manage cell_v2 map_cell1"? | |
| 20:30:53 | mriedem | anupn: did you run it with --verbose to see if it actually mapped anything? | |
| 20:30:59 | mriedem | no | |
| 20:31:12 | zcorneli | dansmith: Do we want to follow his other suggestion as well, and do the associated AssertIn's? | |
| 20:31:14 | mriedem | unless the host is actually in some other db | |
| 20:31:22 | dansmith | zcorneli: I did | |
| 20:31:47 | dansmith | zcorneli: well, what I meant was, I added a positive assertion in each case | |
| 20:31:55 | anupn | mriedem: Okay so with --verbose, I see it finds two cells - cell0 and cell1 | |
| 20:32:00 | dansmith | zcorneli: no need to assert the presence of all the things for the latest version, IMHO | |
| 20:32:07 | dansmith | zcorneli: just one to make sure you're not looking at the wrong level of the structure | |
| 20:32:07 | anupn | it is skips cell0 as no hosts in that | |
| 20:32:42 | anupn | " | |
| 20:32:42 | anupn | Found 0 unmapped computes in cell: 182677b0-4e7d-416b-a47b-57c2f33e090e | |
| 20:32:42 | anupn | and from cell1 I see - "Getting computes from cell 'cell1': 182677b0-4e7d-416b-a47b-57c2f33e090e | |
| 20:33:01 | zcorneli | Seems like the positive assertion for every field is "clear" - It should be covered elsewhere in the tests as well, but ensures "It was there before making compatible, and not there after" | |
| 20:33:10 | mriedem | anupn: what does this give you? nova-manage cell_v2 list_hosts 182677b0-4e7d-416b-a47b-57c2f33e090e | |
| 20:33:19 | mriedem | does the compute host that's hosting that baremetal node show up in the list? | |
| 20:33:41 | zcorneli | dansmith: The single definitely helps fix the previous issue we had through with the false positives | |