| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-01-15 | |||
| 16:42:08 | artom | Pci knowledgable folks, if I have a PciDevice (a <hostdev> in the libvirt xml), what would be the best way to find its MAC? | |
| 16:42:32 | artom | sahid, stephenfin ^^ ? | |
| 16:43:49 | sean-k-mooney | stephenfin: ill stack with the latest release of os-vif and see if i can replicate the ci failure on my end | |
| 16:44:08 | sahid | artom: not sure i understand, we do set the mac address when configuring the libvirt <interface>, no? | |
| 16:44:38 | artom | sahid, I mean when passing through an entire PF | |
| 16:44:47 | artom | It's a <hostdev> in the XML, not an <interface> | |
| 16:44:52 | artom | So there's no MAC | |
| 16:44:53 | sahid | oh i see | |
| 16:45:05 | stephenfin | sean-k-mooney: Sounds good. I'll take a look myself tomorrow | |
| 16:45:11 | stephenfin | See if we can figure out what's going on | |
| 16:45:16 | sahid | artom: no idea sorry | |
| 16:45:29 | artom | sahid, no worries :) | |
| 16:46:01 | artom | Who's the most knowledgeable PCI person we have now that vladikr and nikola aren't with us anymore? | |
| 16:47:12 | sean-k-mooney | stephenfin: it could be litally that we are not hitting the openfile limit but it could be something else too. if its ulimit then it something we need to fix in devstack | |
| 16:48:47 | sahid | artom: in such situation it's the guest os which have the device, the host can't really see it | |
| 16:48:58 | sahid | i mean i guess it can't | |
| 16:49:01 | sean-k-mooney | stephenfin: fyi this local.conf should allow you to reproduce http://paste.openstack.org/show/645256/ | |
| 16:49:09 | sahid | so probably your change is qemu-guest-agent | |
| 16:49:27 | stephenfin | sean-k-mooney: Funny how it would trigger now. Surely we were using privsep before? | |
| 16:49:29 | sahid | s/have the device/own the device | |
| 16:50:21 | sean-k-mooney | stephenfin: we were but maybe with the change to useing the python libs we are a using more filedicriptors or b we are leaking them? | |
| 16:50:58 | stephenfin | We could validate that by turning on warnings. Python 3 warns for open fds | |
| 16:54:51 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/pike: Do not attempt volume swap when guest is stopped/suspended https://review.openstack.org/533733 | |
| 16:55:23 | sean-k-mooney | stephenfin: it could possibly be a provider difference? the test that were failing on rackspace were passing on inap when run as part of the grenade multinode job. anyway ill see if i can reproduce locally | |
| 16:58:18 | sean-k-mooney | stephenfin: apparently https://review.openstack.org/508498 broke Kuryr-Kubernetes so we might need to do a 1.8.1/1.9.0 release to fix them though im not conviced the fix should not be on there end yet | |
| 16:59:21 | stephenfin | sean-k-mooney: Oh, someone else is using os-vif. Interesting... | |
| 16:59:49 | sean-k-mooney | ya i was aware kuryr was using it but i did not think they were ever saving os-vif objects | |
| 17:12:32 | openstackgerrit | Lee Yarwood proposed openstack/nova master: libvirt: QEMU native LUKS decryption for volumes https://review.openstack.org/523958 | |
| 17:23:16 | jaypipes | sean-k-mooney: I'm currently replying to mdulko about that... | |
| 17:23:51 | jaypipes | sean-k-mooney: bottom line, even if we did do obj_make_compatible(), that wouldn't have magically caused the client side to not break. | |
| 17:24:35 | jaypipes | sean-k-mooney: they still will need to advertise what versions of the objects they need (Nova does it via the nova-specific oslo versionedobject registry, etc) | |
| 17:24:42 | jaypipes | sean-k-mooney: there's no free lunch here... | |
| 17:24:52 | sean-k-mooney | jaypipes: right v2 of the patch had the make compatible code but you would have to do the down grade and i think the compat code just created an empty list of ip | |
| 17:25:03 | jaypipes | right. | |
| 17:25:07 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM driver: ovs vif https://review.openstack.org/422512 | |
| 17:25:10 | jaypipes | anyway, writing an answer... | |
| 17:25:21 | jaypipes | dansmith might be interested in this thread, too.. | |
| 17:25:55 | sean-k-mooney | the main point i wanted to get across was the lib freeze is thusday so if we can fix it by then it needs to be fixed in kuryr kubernetes | |
| 17:26:15 | jaypipes | sean-k-mooney: "fix" is relative here, which is my point :) | |
| 17:26:23 | sean-k-mooney | we could revert the patch i guess since nova is not using that field | |
| 17:26:30 | jaypipes | sean-k-mooney: "fix" might be just a flat revert, yeah | |
| 17:27:28 | sean-k-mooney | well we could certenly do that but they are using the object i a way we specifcially intoduced that patch to prevent so longterm we would have to adress the migration issue | |
| 17:31:34 | jaypipes | sean-k-mooney: did you even know that kuryr-kubernetes was consuming os-vif objects? I certainly didn't... | |
| 17:35:52 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM driver: ovs vif https://review.openstack.org/422512 | |
| 17:38:48 | melwitt | bauzas: this was previously approved by mriedem, so I think we can re-approve it after the rebase https://review.openstack.org/#/c/531407 | |
| 17:45:34 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: libvirt: create vGPU for instance https://review.openstack.org/528832 | |
| 17:45:34 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Recreate mediated devices on reboot https://review.openstack.org/533642 | |
| 17:46:13 | bauzas | melwitt: okie | |
| 17:47:04 | melwitt | thanks bauzas | |
| 17:54:17 | itlinux | hello nova team.. I have a OOO deployment looking to implement cells2, by default cells2 has one already any tips on how to expand that? Thanks | |
| 17:55:13 | sean-k-mooney | jaypipes: i knew that kuryr was for port plugging but i did not think they were ever storing if the os-vif objects | |
| 17:56:11 | openstackgerrit | Jackie Truong proposed openstack/nova master: Add trusted_image_certificates to REST API https://review.openstack.org/486204 | |
| 17:56:21 | sean-k-mooney | jaypipes: the ips field that we moved form the subnet object to the vif object is never used in os-vif or nova hence why i taught it was safe to move it | |
| 18:01:39 | sean-k-mooney | stephenfin: by the way looking at the nova unit test failuer in https://review.openstack.org/#/c/532987 i dont think we should ever be comparein the serialised form of os-vif objects in nova. we can add the datapath_type field to allow this to pass but i think we should reduce the specificity of the test in the long run | |
| 18:03:19 | stephenfin | sean-k-mooney: Agreed | |
| 18:03:32 | stephenfin | Just about to leave the office, but I can tackle that first thing in the morning | |
| 18:03:54 | stephenfin | Soon as I finish mriedem's comments on the websocket security series | |
| 18:03:56 | stephenfin | o/ | |
| 18:04:19 | sean-k-mooney | stephenfin: ill give it a go tonight but if i have not pushed anything then please do | |
| 18:09:52 | melwitt | itlinux: do you mean you want to know how to add an additional cells v2 cell to a deployment? | |
| 18:10:24 | itlinux | thanks melwitt: I need to add other cell2 to the env. | |
| 18:10:39 | itlinux | since OOO has a default but I need to include other DC | |
| 18:13:32 | melwitt | itlinux: sorry what's DC? | |
| 18:13:39 | itlinux | Datacenter. | |
| 18:14:02 | melwitt | itlinux: we have docs for adding a new cell here https://docs.openstack.org/nova/latest/user/cells.html#adding-a-new-cell-to-an-existing-deployment | |
| 18:14:17 | itlinux | thanks | |
| 18:14:32 | itlinux | this is for cell2 right? | |
| 18:14:47 | melwitt | you'll have to set up a database, message queue, etc for the new cell first, then you'll use nova-manage commands to create_cell and discover_hosts | |
| 18:15:19 | melwitt | yes. the doc doesn't cover how to deploy the new message queue etc | |
| 18:15:20 | itlinux | ok so I could run another OOO and then merge them? | |
| 18:15:39 | itlinux | but it also says . Note, however, that this command is deprecated as of the 16.0.0 Pike release. | |
| 18:15:44 | itlinux | I am in Pike.. | |
| 18:15:50 | melwitt | hm, no, I don't think you'd want to do that because OOO is going to create new controllers and you don't want that | |
| 18:16:17 | itlinux | ok | |
| 18:16:19 | itlinux | that's right.. | |
| 18:16:42 | itlinux | so do we have steps on how to create the Q | |
| 18:17:28 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] use simple FaultWrapper https://review.openstack.org/533752 | |
| 18:27:02 | cdent | do I recall correctly that changing etc/nova/api-paste.ini is not something we can do easily? | |
| 18:27:18 | melwitt | itlinux: no, I don't think so. and I don't think OOO yet has the ability to add a new cells v2 cell | |
| 18:28:17 | itlinux | ahh I heard that cells2 is enabled .. so just want to check how to find out | |
| 18:29:57 | melwitt | it's enabled for one cell already, I just don't think they have a way of adding more cells | |
| 18:30:17 | itlinux | ok.. | |
| 18:30:26 | itlinux | yes one is there.. | |
| 18:30:44 | itlinux | so how do we find that out? thanks melwitt: | |
| 18:32:22 | melwitt | itlinux: I would try asking in #tripleo about if there's support for multiple cells or if it's in the works | |
| 18:46:59 | itlinux | how about regions.. | |
| 18:48:01 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM Driver: SEA https://review.openstack.org/523216 | |
| 18:55:16 | openstackgerrit | Jackie Truong proposed openstack/python-novaclient master: Microversion 2.60 - Add trusted_image_certificates https://review.openstack.org/500396 | |
| 19:11:22 | openstackgerrit | Jay Pipes proposed openstack/nova master: func tests for nested providers in alloc candidate https://review.openstack.org/531443 | |
| 19:11:23 | openstackgerrit | Jay Pipes proposed openstack/nova master: add _has_provider_trees() utility function https://review.openstack.org/531474 | |
| 19:11:23 | openstackgerrit | Jay Pipes proposed openstack/nova master: placement: _get_trees_matching_all() https://review.openstack.org/531512 | |
| 19:11:24 | openstackgerrit | Jay Pipes proposed openstack/nova master: add tests for _get_trees_matching_all() with trait https://review.openstack.org/531899 | |
| 19:11:51 | melwitt | jaypipes: o/ hope you're feeling better | |
| 19:13:16 | jaypipes | melwitt: thanks Melanie, I am :) | |
| 19:13:29 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM driver: ovs vif https://review.openstack.org/422512 | |
| 19:15:39 | melwitt | jaypipes: glad to hear it :) | |
| 19:15:42 | prometheanfire | sean-k-mooney: for the os-vif thing, is it gating related (open files) or neutron related (db deadlock)? | |
| 19:16:26 | melwitt | itlinux: you'll want to consult tripleo folks for questions about managing regions in a OOO deployment. this is a nova development channel | |
| 19:16:35 | sean-k-mooney | prometheanfire: i ran the senario test locally and all the failing test passed so im guessing the openfile issue is the real reason for the tempest failure | |
| 19:16:56 | sean-k-mooney | prometheanfire: ill be pushing a patch for the nova unit tests in a few seconds | |