| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-02-08 | |||
| 21:34:19 | mriedem | not all entries are a list of dicts | |
| 21:34:24 | mriedem | 'root_device_name': instance.root_device_name, | |
| 21:34:25 | mriedem | AND | |
| 21:34:30 | melwitt | :| | |
| 21:34:32 | mriedem | the 'block_device_mapping' entry is not a list of dicts | |
| 21:34:35 | mriedem | no ma'am | |
| 21:34:44 | melwitt | FINE | |
| 21:34:47 | melwitt | :) | |
| 21:34:49 | mriedem | it's a list of DriverVolumeBlockDevice objects, which wrap a BlockDeviceMapping object | |
| 21:34:55 | mriedem | and proxy all of the bdm object stuff to the db | |
| 21:35:15 | mriedem | DriverVolumeBlockDevice at some level might extend dict, so maybe you're right :) | |
| 21:35:34 | mriedem | it does https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L81 | |
| 21:35:35 | mriedem | heh | |
| 21:35:38 | mriedem | shows what i know | |
| 21:36:59 | melwitt | from this, they're probably not all lists | |
| 21:37:01 | melwitt | https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L1566-L1569 | |
| 21:37:25 | melwitt | or maybe swap's not. I dunno | |
| 21:38:07 | mriedem | swap, | |
| 21:38:18 | mriedem | is a list of DriverSwapBlockDevice objects which extend DriverVolumeBlockDevice (i think) | |
| 21:38:20 | jaypipes | cdent: nice article. | |
| 21:38:27 | mriedem | https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L202 | |
| 21:38:31 | melwitt | hah | |
| 21:39:16 | mriedem | or not, wtf knows | |
| 21:39:20 | mriedem | it's bdm roulette | |
| 21:39:55 | mriedem | melwitt: +W if you want to start backports | |
| 21:40:35 | melwitt | thanks | |
| 21:41:31 | cdent | thanks jaypipes | |
| 21:43:53 | mriedem | general observation, | |
| 21:44:07 | mriedem | i feel like this rc1 period has been much less of a clusterfuck than pike and ocata | |
| 21:44:17 | mriedem | right? | |
| 21:44:17 | mriedem | so i think that's progress | |
| 21:46:06 | cdent | mriedem: it has kind of felt like this cycle has been more to do with either fixing stuff or laying ground work for later stuff, so perhaps that means less exposure? but I agree things do seem a bit more chilled out | |
| 21:54:56 | openstackgerrit | Takashi NATSUME proposed openstack/nova stable/ocata: Add 'delete_host' command in 'nova-manage cell_v2' https://review.openstack.org/513721 | |
| 21:55:37 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: List/show all server migration types (1/2) https://review.openstack.org/430608 | |
| 21:56:10 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: List/show all server migration types (2/2) https://review.openstack.org/459483 | |
| 21:56:52 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Parameter verification for servers.inc https://review.openstack.org/528201 | |
| 21:57:19 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add functional tests for traits API https://review.openstack.org/524094 | |
| 21:57:53 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Example verification for servers.inc https://review.openstack.org/529520 | |
| 21:58:09 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient master: Microversion 2.61 - List/Show all server migration types https://review.openstack.org/430839 | |
| 21:58:49 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Fix parameter order in rebuild https://review.openstack.org/529718 | |
| 21:59:21 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Move body examples to an isolated directory https://review.openstack.org/529998 | |
| 21:59:23 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add a nova-caching-scheduler job to the experimental queue https://review.openstack.org/539260 | |
| 22:00:52 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Fix 500 error while passing 4-byte unicode data https://review.openstack.org/407514 | |
| 22:03:00 | mriedem | hmm, we seem to be running both legacy-tempest-dsvm-neutron-full and tempest-full on changes | |
| 22:03:04 | mriedem | which are redundant | |
| 22:03:41 | mriedem | gmann_: ^ right? | |
| 22:05:08 | mriedem | ah yes i see the problem | |
| 22:06:08 | mriedem | oh hmm | |
| 22:09:46 | mriedem | melwitt: i've got the rc1 patch up with depends-on to the 2 remaining things we need to get into rc1 https://review.openstack.org/#/c/542459/ | |
| 22:09:53 | mriedem | plus cycle highlights pulled from the release note prelude | |
| 22:10:05 | mriedem | i've gotta run to parent/teacher conferences (yay) but will be back online later | |
| 22:10:14 | melwitt | ack | |
| 22:17:55 | openstackgerrit | Merged openstack/nova master: Fix the wrong description https://review.openstack.org/540265 | |
| 22:37:04 | openstackgerrit | Arvind Nadendla proposed openstack/nova-specs master: Update to push image traits to ironic https://review.openstack.org/542468 | |
| 22:41:55 | openstackgerrit | melanie witt proposed openstack/nova master: Make scheduler.utils.setup_instance_group query all cells https://review.openstack.org/540258 | |
| 22:49:10 | openstackgerrit | Arvind Nadendla proposed openstack/nova-specs master: Support traits in Glance https://review.openstack.org/541507 | |
| 22:54:45 | bkero- | Hey all, can anyone recall seeing a nova-libvirt bug in 2.5 about logfile chardev not being supported? | |
| 22:59:30 | cfriesen | this is maybe a weird question, but is the ec2-id returned by the instance metadata server (in the format "i-00000004") expected to match the "instance-xxxx" value as returned in the OS-EXT-SRV-ATTR:instance_name field of "nova show"? | |
| 23:02:42 | openstackgerrit | Jianghua Wang proposed openstack/nova master: VGPU: Modify the example of vgpu white_list set https://review.openstack.org/539183 | |
| 23:22:22 | smcginnis | mriedem: Hows that RC coming along? :) | |
| 23:23:47 | melwitt | smcginnis: he's away atm, the release patch is up https://review.openstack.org/#/c/542459/ and depends-on the things we have left, which I'm going to finish reviewing in a few | |
| 23:23:58 | melwitt | he'll be back later, he said | |
| 23:24:10 | smcginnis | melwitt: OK, so just a couple things to push through yet? | |
| 23:24:34 | melwitt | yeah, one is in the gate already, the other's need a +W and I'm gonna review those now | |
| 23:24:55 | mriedem | i'm back | |
| 23:25:26 | mriedem | bkero-: sounds familiar but not sure, might have been something markus zoeller worked on in ocata? | |
| 23:25:41 | mriedem | bkero-: https://specs.openstack.org/openstack/nova-specs/specs/ocata/implemented/libvirt-virtlogd.html ? | |
| 23:25:45 | smcginnis | melwitt: Thanks! | |
| 23:30:06 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Check for leaked server resource allocations in post_test_hook https://review.openstack.org/538510 | |
| 23:34:11 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add functional tests for traits API https://review.openstack.org/524094 | |
| 23:38:09 | melwitt | mriedem: do you know why this test is removed instead of setting up the ComputeV4Proxy like the others? https://review.openstack.org/#/c/541005/6/nova/tests/unit/compute/test_compute_mgr.py@a6834 | |
| 23:38:52 | mriedem | because _rollback_live_migration doesn't do that thing anymore | |
| 23:39:00 | mriedem | and _do_live_migration calls _rollback_live_migration | |
| 23:39:13 | mriedem | and _do_live_migration already has the non-dict form of migrate_data by the time it calls _rollback_live_migration | |
| 23:39:17 | mriedem | i think that's why anyway | |
| 23:39:48 | mriedem | https://review.openstack.org/#/c/541005/5/nova/compute/manager.py@6394 | |
| 23:41:50 | melwitt | yeah, I mean, it looks like the other "handles_dict" tests were adapted to use the _ComputeV4Proxy because the V4Proxy version has to do the dict compat thing, right? | |
| 23:42:59 | mriedem | yeah, but there isn't a compute v4 proxy for that rollback method | |
| 23:43:07 | mriedem | the entry point / proxy method is live_migration | |
| 23:43:17 | melwitt | oh, I see. okay | |
| 23:43:20 | mriedem | which does the massageroo | |
| 23:52:19 | gmann_ | mriedem: yea it seems both running. it has been replaced with tempest-full in integrated job(and fix for stable branches etc) | |
| 23:52:46 | ianw | in case anyone finds my comment in logs about aarch64 config-drive ... yes it does work, but ensure you've got all the right kernel cdrom drivers, which the default xenial kernel may not | |
| 23:53:02 | mriedem | gmann_: https://review.openstack.org/542484 | |
| 23:53:04 | mriedem | think i figured it out | |
| 23:53:06 | gmann_ | mriedem: legacy-tempest-dsvm-neutron-ful should be running on stable branches only but it seem we have it running on master too in many place | |
| 23:53:19 | mriedem | ianw: hrw is the aarch64 expert around these parts | |
| 23:53:35 | mriedem | https://review.openstack.org/#/c/541728/ | |
| 23:54:29 | mriedem | ianw: maybe you can provide some details here? https://review.openstack.org/#/c/541728/9/doc/source/user/support-matrix.ini@872 | |
| 23:54:39 | mriedem | hrw is working on documenting aarch64+libvirt support in the feature matrix | |
| 23:54:46 | mriedem | so that would be good info since we can put notes in these doc entries | |
| 23:54:49 | ianw | mriedem: yes, wouldn't have got nearly so far without his help ;) | |
| 23:54:50 | gmann_ | mriedem: yea but that should not needed i guess as it is handled in integrated-gate template - http://git.openstack.org/cgit/openstack-infra/openstack-zuul-jobs/tree/zuul.d/zuul-legacy-project-templates.yaml#n107 | |
| 23:55:11 | mriedem | gmann_: see the response from corvus in -infra, | |
| 23:55:23 | mriedem | apparently b/c nova re-defines it in project-config w/o the branch specifier, it messes that up | |
| 23:55:57 | gmann_ | mriedem: yes, i think we should remove it all from project-config (nova, tempest, swift etc) | |
| 23:56:13 | cfriesen | mriedem: that earlier issue with libvirt and specifying the disks to migrate....I think it was from before we rebased to 16.0.4. So likely has gone away now. | |
| 23:56:19 | gmann_ | mriedem: anyways i ll check that. thanks for this | |
| 23:57:18 | ianw | mriedem: oh cool! thanks for pointing to that. yeah, we're working on bringing up dib/nodepool/infra etc on the linaro cloud. all things considered, it's going well i'd say! soon you should be able to easily add aarch64 testing (modulo cloud capacity ... it will be a soft start) | |
| 23:57:21 | mriedem | cfriesen: ok | |