| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-07-09 | |||
| 07:39:53 | openstackgerrit | Luyao Zhong proposed openstack/nova master: object: Add VirtualPMEM and VirtualPMEMList objs https://review.opendev.org/634548 | |
| 07:50:04 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: nova-manage: heal port allocations https://review.opendev.org/637955 | |
| 07:50:59 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Move consts from neutronv2/api to constants module https://review.opendev.org/668945 | |
| 08:02:26 | openstackgerrit | Shilpa Devharakar proposed openstack/nova master: Support filtering of hosts by forbidden aggregates https://review.opendev.org/667952 | |
| 08:04:04 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use neutron contants in cmd/manage.py https://review.opendev.org/668946 | |
| 08:04:04 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add 'resource_request' to neutronv2/constants https://review.opendev.org/668947 | |
| 08:04:43 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Translatable output strings in heal allocation https://review.opendev.org/668925 | |
| 08:42:45 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use the safe get_binding_profile https://review.opendev.org/669817 | |
| 08:59:41 | openstackgerrit | Balazs Gibizer proposed openstack/nova stable/rocky: Add support for osprofiler in placement wsgi https://review.opendev.org/667832 | |
| 09:55:57 | openstackgerrit | Matthew Booth proposed openstack/nova master: Whether to use config drive depends on instance.configdrive https://review.opendev.org/485930 | |
| 11:43:50 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: nova-manage: heal port allocations https://review.opendev.org/637955 | |
| 11:44:18 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Move consts from neutronv2/api to constants module https://review.opendev.org/668945 | |
| 11:44:58 | openstackgerrit | Ivaylo Mitev proposed openstack/nova master: VMware: Attach volumes using adapter type from instance https://review.opendev.org/616599 | |
| 11:52:09 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use neutron contants in cmd/manage.py https://review.opendev.org/668946 | |
| 11:52:39 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use the safe get_binding_profile https://review.opendev.org/669817 | |
| 11:52:59 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add 'resource_request' to neutronv2/constants https://review.opendev.org/668947 | |
| 11:53:16 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Translatable output strings in heal allocation https://review.opendev.org/668925 | |
| 12:53:52 | openstackgerrit | Arthur Dayne proposed openstack/nova-specs master: Proposal for a safer noVNC console with password authentication https://review.opendev.org/623120 | |
| 13:01:29 | openstackgerrit | sean mooney proposed openstack/nova master: make config drives sticky bug 1835822 https://review.opendev.org/669738 | |
| 13:01:29 | openstack | bug 1835822 in OpenStack Compute (nova) "vms loose acess to config drive with CONF.force_config_drive=True after hard reboot" [Medium,In progress] https://launchpad.net/bugs/1835822 - Assigned to sean mooney (sean-k-mooney) | |
| 13:07:27 | yonglihe | sean-k-mooney: I had finished a new logic per our discuss. Code released without unit test, so you could give advice in advance, https://review.opendev.org/#/c/627765/ | |
| 13:08:47 | sean-k-mooney | i just finsihed ^ so i can take a look before i start my next thing | |
| 13:10:23 | openstackgerrit | huanhongda proposed openstack/nova master: Fix for auth during live-migration (2) https://review.opendev.org/669867 | |
| 13:16:59 | openstackgerrit | Merged openstack/nova master: Remove assumption of http error if consumer not exists https://review.opendev.org/669188 | |
| 13:23:38 | yonglihe | sean-k-mooney, thanks, looking forward your comment. The 'reap_unknown' options seems not suitable which make that logic mess. I just collapse the orphan instance clean up use existing 'poweroff' and 'reap' option. Cause we had filter out instance not created by nova, this might acceptably. | |
| 13:30:49 | yonglihe | @melanie witt, I had your comments addressed mostly, please take a look, thanks. | |
| 13:41:12 | sean-k-mooney | yonglihe: ya i think that is much closer to what i was expecting | |
| 13:41:22 | sean-k-mooney | i left some comments inline | |
| 13:43:57 | openstackgerrit | sean mooney proposed openstack/nova master: Whether to use config drive depends on instance.configdrive https://review.opendev.org/485930 | |
| 13:54:42 | mriedem | gmann: i'll update the policy rule name in https://review.opendev.org/#/c/645520/33 | |
| 14:10:17 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Test heal port allocations in nova-next https://review.opendev.org/669879 | |
| 14:11:55 | mriedem | i've got a series of pretty simple "implement update_provider_tree" changes for xen and hyperv here with +2s https://review.opendev.org/#/q/status:open+project:openstack/nova+branch:master+topic:update_provider_tree which eventually allows us to deprecate the non-upt compact code in the RT and lets us clean up that RT mess in U | |
| 14:13:38 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Add host and hypervisor_hostname flag to create server https://review.opendev.org/645520 | |
| 14:17:16 | gibi | mriedem: do we have somebody from xen and hyperv to ack those changes? I looked at the xen one and it seems straitforward to me but I could wait for a xen developer to ack | |
| 14:18:07 | sean-k-mooney | gibi: im not sure we have anyone maintaining xen support at the moment | |
| 14:18:14 | sean-k-mooney | out side of the core team that is | |
| 14:20:36 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Update AZ admin doc to mention the new way to specify hosts https://review.opendev.org/666767 | |
| 14:23:57 | gibi | sean-k-mooney: OK, I left my +2 on the patch. | |
| 14:26:16 | mriedem | gibi: lpetrut from hyperv already acked the hyperv one and the CI passed, and it's a cherry pick from the out-of-tree hyperv driver | |
| 14:26:22 | mriedem | we do'nt have any xen people nor ci | |
| 14:26:31 | mriedem | but the xen change itself is very straightforward - just a port of get_inventory | |
| 14:26:51 | gibi | mriedem: I'm +A on the hyperv as I saw lpetrut vote | |
| 14:26:55 | mriedem | thanks | |
| 14:27:09 | gibi | mriedem: OK, then I push the +A on the xen patch as well | |
| 14:27:11 | mriedem | will be nice to eventually remove the spaghetti for that stuff from the RT | |
| 14:28:53 | gibi | mriedem: btw, I made my first stab at adding port healing integration test to nova-next https://review.opendev.org/669879 , Eagerly waiting for the gate to tell me why it won't work :) | |
| 14:29:39 | mriedem | interesting | |
| 14:31:00 | mriedem | gibi: i think you put the changes on the wrong job in zuul.yaml | |
| 14:31:36 | gibi | mriedem: you are right. :facepalm: | |
| 14:33:02 | mriedem | left another comment, but nova-next isn't zuul v3 native yet, | |
| 14:33:15 | mriedem | but i think you could hack the devstack plugin/services/local.conf into the nova-next job using DEVSTACK_GATE_* variables | |
| 14:33:21 | mriedem | until we convert nova-next to zuul v3 | |
| 14:34:31 | gibi | mriedem: thanks. I will do the translation to the old format | |
| 14:34:44 | mriedem | that's probably easier for now | |
| 14:35:11 | mriedem | i might tinker with converting nova-next to zuul v3 separately, i've been meaning to do that anyway | |
| 14:39:13 | gibi | mriedem: cool, thanks for the quick review, you spared me some CI runs | |
| 14:46:31 | sean-k-mooney | mriedem: by the way we talked about actully enfroceing our policy of deprecating and removing virt drivers/backend without ci. have we pushed a patch to mark xen as deprected yet? | |
| 14:47:09 | mriedem | we merged a patch to say xen is experimental but not yet deprecated | |
| 14:47:35 | mriedem | https://github.com/openstack/nova/commit/92c7e0e0b4fcc596e86b493aafb28ee656cb046c#diff-775c4605e52aa43e2ad8c90489216dc8 | |
| 14:48:02 | sean-k-mooney | i still hope to either spend a day and fix the libvirt/lxc driver or push a patch to deprecate it. | |
| 14:48:25 | sean-k-mooney | but ya we will still have them for train in anycase | |
| 14:49:58 | mriedem | if you care about lxc then maybe follow / talk to this guy https://review.opendev.org/#/q/owner:miguel%2540midokura.com+(status:open+OR+status:merged) | |
| 14:50:36 | sean-k-mooney | oh he fixed the bug i was thinking of fixing | |
| 14:50:38 | sean-k-mooney | cool | |
| 14:51:29 | sean-k-mooney | am im not sure how much i care about it but i am somewhat nostalgic in that lxc was second technically container tech i used and it has some usecases still | |
| 14:51:50 | sean-k-mooney | the two thing i was thinking of doing was fixing that bug and trying to add a ci job | |
| 14:52:25 | sean-k-mooney | basically if we still offically support it i would like to ci it but if no one is useing it then i dont mind if its removed. | |
| 14:54:00 | mriedem | i had a non-voting nova-lxc ci job for it long ago and had to blacklist a lot of tempest tests and even then it would fail and the failures would snowball hard on the compute | |
| 14:54:33 | mriedem | nbd mounts and such would get messed up from what i remember | |
| 14:55:08 | mriedem | we don't officially support it btw, | |
| 14:55:16 | mriedem | there is a warning that gets logged on the start of the driver saying it's not tested | |
| 14:55:48 | sean-k-mooney | ah ok well its still there and i think its more useful the uml | |
| 14:56:05 | sean-k-mooney | or some of the other non qmeu/kvm libvirt backends | |
| 14:57:07 | sean-k-mooney | anyway its a low priority | |
| 14:57:55 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Add host and hypervisor_hostname flag to create server https://review.opendev.org/645520 | |
| 14:57:55 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Update AZ admin doc to mention the new way to specify hosts https://review.opendev.org/666767 | |
| 14:59:26 | mriedem | speaking of priorities, how is that numa aware live migration series coming? | |
| 15:04:32 | mriedem | cdent: did i dream this or did you at some point get nova's code coverage job working with unit + functional tests? it only runs unit tests today so we're not reporting all of the coverage. | |
| 15:05:29 | cdent | I think I played with it after I got it working for placement but don't remember if it ever went beyond that. I can play with it again if that seems like a useful thing to do | |
| 15:06:50 | mriedem | i played with it at some point too and seem to remember things hanging or crashing somehow | |
| 15:07:55 | cdent | let's blame eventlet | |
| 15:08:02 | cdent | which might mean things are different now | |
| 15:10:58 | mriedem | right, i think i tried configuring coverage with concurrency=eventlet but that didn't help | |
| 15:11:04 | mriedem | https://coverage.readthedocs.io/en/v4.5.x/config.html#run | |
| 15:14:13 | sean-k-mooney | mriedem: its proably failing due to our use of fixture. one of the oslo libs (oslo.fixtures?) i think calles exec directly with a code object and that breaks my debugger | |
| 15:14:25 | sean-k-mooney | the coverage jobs hooks the debug entry point | |
| 15:14:31 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Fix live migration break group policy simultaneously https://review.opendev.org/651969 | |
| 15:15:10 | sean-k-mooney | its break on from oslo_utils.fixture import uuidsentinel as uuids | |
| 15:25:14 | mriedem | i'm actually hitting errors with it trying to find the placement fixture | |
| 15:27:06 | sean-k-mooney | yep | |
| 15:27:22 | sean-k-mooney | that was the error i endup having | |
| 15:27:42 | sean-k-mooney | its in the rewrite module when its actully trying to patch stuff | |
| 15:28:49 | sean-k-mooney | the point in the code it raise the exception is actully kind of missleading but the final exception you get makes more sense | |
| 15:31:18 | sean-k-mooney | im not really sure why "from oslo_utils.fixture import uuidsentinel as uuids" has to do with placement but im guessing we are messing with some gloabl state and its not happy | |
| 15:32:09 | cdent | look at the functional job defs: you have to require placement explicitly | |
| 15:32:23 | cdent | i've done that and I'm running with some moderate form of success | |
| 15:32:30 | cdent | but it's not done yet | |
| 15:32:37 | sean-k-mooney | how do you mean | |
| 15:32:41 | mriedem | deps = | |
| 15:32:41 | mriedem | -r{toxinidir}/test-requirements.txt | |