| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-04-16 | |||
| 23:19:57 | artom | melwitt, yeah, but apparently they need to do that for *every* job? As in, there's no common deployment that covers all jobs? | |
| 23:20:31 | melwitt | that said, I don't really have an opinion about whether SameHost and DifferentHost belong in default filters because TBH I don't know if those are really commonly used by end users or not | |
| 23:20:34 | artom | https://review.gerrithub.io/#/c/407794/ is the infrared change | |
| 23:20:53 | artom | Which apparently got -2 by Arie? I have no idea what's going on | |
| 23:21:23 | artom | melwitt, heh, don't we have those sosreports we can use for insights? At least for OSP users? | |
| 23:21:36 | melwitt | yeah, we could I would think | |
| 23:21:44 | artom | Man, the effort:complexity ratio of this thing is through the roof | |
| 23:22:00 | artom | There's, like, 5 of us, just to enable a filter | |
| 23:22:22 | melwitt | well, enabling a filter is supposed to be all pluggable, like you can do it in your nova.conf | |
| 23:23:41 | melwitt | this is starting a discussion about "what belongs in defaults?" "should everything be in defaults?" "what's the downside of putting everything in defaults?" and I don't really know the answers to those questions off the top of my head | |
| 23:24:29 | artom | Right, but the original problem is still "a filter that we need isn't enabled" | |
| 23:26:06 | melwitt | by default, no. there are a lot of filters we could say that about. so to me, it's more like, should we just add all filters to defaults and is there any reason to not do that? I'd be asking scheduler subteam folk about it | |
| 23:36:05 | artom | melwitt, yeah, it might be better to rephrase this as 2 questions: 1. do we have data that justifies adding same/different hostfilter to the defaults? | |
| 23:36:23 | artom | 2. Should we just enable all filters by default? | |
| 23:36:45 | artom | I can poke around our sosreports for 1 tomorrow, for 2... I guess tomorrow as well | |
| 23:37:48 | cfriesen | melwitt: https://review.openstack.org/#/c/560690/ has been updated to address your final comment | |
| 23:37:50 | melwitt | artom: yeah. the point of the default filters is to give someone a config for a typical cloud. if same/different are typically used, then they should be in defaults I think. CI are a different beast IMHO because they're aiming to run and verify as many features and capabilities as possible | |
| 23:38:36 | melwitt | so you would expect them to enable everything they possibly could. that said, IIUC there's some mismatch between what tempest defaults to and what nova defaults to and that is what's actually causing this problem | |
| 23:39:49 | melwitt | cfriesen: thanks, will take another look | |
| 23:40:25 | cfriesen | melwitt: the default set of enabled filters is actually pretty small | |
| 23:40:42 | melwitt | cfriesen: yes. I have learned that recently | |
| 23:41:38 | melwitt | or re-learned it. I've seen default filter list in the past and eventually forgot it | |
| 23:55:27 | openstackgerrit | Merged openstack/nova master: Remove RequestContext.instance_lock_checked https://review.openstack.org/554378 | |
| #openstack-nova - 2018-04-17 | |||
| 01:06:32 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient master: Microversion 2.61 - support extra_specs in flavor API https://review.openstack.org/561520 | |
| 01:12:57 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient master: Microversion 2.61 - support extra_specs in flavor API https://review.openstack.org/561520 | |
| 01:16:47 | arvindn051 | how do we get host uuid from hostname? | |
| 01:17:07 | arvindn051 | i am trying to get the host uuid for host specified in the request_spec | |
| 01:40:11 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all resources in provider_summaries https://review.openstack.org/558045 | |
| 01:40:12 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all nested providers in tree https://review.openstack.org/559480 | |
| 01:40:13 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return resource providers without inventories https://review.openstack.org/559554 | |
| 01:54:33 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient master: Microversion 2.62 - List/Show all server migration types https://review.openstack.org/430839 | |
| 02:29:42 | gmann | alex_xu: ping | |
| 02:35:24 | openstackgerrit | Dinesh Bhor proposed openstack/nova master: WIP: PCPU: Define numa dedicated CPU resource class https://review.openstack.org/561770 | |
| 02:37:03 | openstackgerrit | Dinesh Bhor proposed openstack/nova master: WIP: PCPU, VCPU: Add respective conf options https://review.openstack.org/561771 | |
| 02:40:31 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient master: Fix comments in novaclient/tests/unit/fakes.py https://review.openstack.org/561772 | |
| 02:44:50 | openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add host/hostId to instance action events API https://review.openstack.org/557145 | |
| 02:52:02 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova-specs master: Return resources of entire trees in Placement https://review.openstack.org/559466 | |
| 03:11:35 | openstackgerrit | jichenjc proposed openstack/nova master: Enhance api-guide general info https://review.openstack.org/561773 | |
| 03:30:12 | openstackgerrit | Nguyen Hai proposed openstack/nova-specs master: Follow the new PTI for document build https://review.openstack.org/551802 | |
| 03:30:54 | openstackgerrit | Nguyen Hai proposed openstack/nova-specs master: Follow the new PTI for document build https://review.openstack.org/551802 | |
| 03:57:54 | jichen | melwitt: hi regarding your comments in the ML about z/VM driver | |
| 03:59:21 | jichen | melwitt: the z/VM driver behaves unlike any other in-tree driver [0-5] ---> we will make our code exactly as you pointed , we will NOT do anything special | |
| 04:14:03 | openstackgerrit | Nguyen Hai proposed openstack/nova-specs master: Follow the new PTI for document build https://review.openstack.org/551802 | |
| 04:24:50 | openstackgerrit | Nguyen Hai proposed openstack/nova-specs master: Follow the new PTI for document build https://review.openstack.org/551802 | |
| 06:20:24 | openstackgerrit | Merged openstack/nova master: remove ec2 in service and cmd https://review.openstack.org/556778 | |
| 06:53:04 | openstackgerrit | Naichuan Sun proposed openstack/nova master: xenapi: Use XAPI pool instead of aggregate pool for shared SR migration https://review.openstack.org/554154 | |
| 06:57:02 | gokhan | hi , I have HA openstack pike environment. for 2 weeks there is not any problem about environment, but for 2 days I am getting rabbitmq timeout errors on some compute nodes. there are errors on nova compute and rabbitmq cluster. on nova compute there are timeouts errors like that : http://paste.openstack.org/show/719338/ and I see on rabbitmq there are error reports like that: http://paste.openstack.org/show/719339/ what can be the reason of this a | |
| 06:57:02 | gokhan | nd how can I solve this problem ? now I can not create multiple instances and some instances are waiting in build statement for hours | |
| 07:16:57 | Kevin_Zheng | gibi Hi, could you check this series https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:refactor-notification-samples | |
| 07:17:35 | gibi | Kevin_Zheng: Hi! It is in my queue. Thanks for working on it | |
| 07:18:20 | Kevin_Zheng | gibi, Thanks, it is related to the bp I working on anyway :) | |
| 07:44:43 | alex_xu | gmann: pong | |
| 07:45:30 | openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: [WIP] ServerActionsSampleJsonTest refactor https://review.openstack.org/561819 | |
| 07:47:50 | gmann | alex_xu: could not catch you since long. i want to discuss about API meeting or office hour | |
| 07:48:09 | alex_xu | gmann: yea | |
| 07:48:32 | gmann | alex_xu: idea is we continue on that and do mainly 2 things 1. we track priority items mainly for review and 2. do bug triage | |
| 07:48:56 | alex_xu | gmann: I think it is cool | |
| 07:50:05 | alex_xu | gmann: what time we choice? | |
| 07:50:51 | gmann | alex_xu: not decided actually. i feel may be during our day time as not much people attended past API meeting | |
| 07:51:02 | alex_xu | yes | |
| 07:51:18 | alex_xu | gmann: and if there is trouble problem we can bring it to the nova weekly meeting | |
| 07:51:24 | gmann | alex_xu: and i can send a summary of updates every week after that | |
| 07:51:36 | alex_xu | gmann: that will be cool | |
| 07:51:39 | gmann | alex_xu: yes, we give same updated there too | |
| 07:53:17 | alex_xu | gmann: should we start a email to ask abot the idea first? | |
| 07:53:43 | sean-k-mooney | stephenfin: jaypipes sahid i was talking to dansmith about the native vhost-user support in the libvirt driver yesterday, do you have any objection to me moving this to os-vif as an in tree plugin? the impmentation in os-vif will effectivly be an no-op but it means we can remove that code path from nova. the nataiv vhost-user support is used by snab switch, vpp and i belive it was used by contrail at | |
| 07:53:43 | gmann | alex_xu: sure. that will be good and if we get more people join that | |
| 07:53:45 | sean-k-mooney | one point. | |
| 07:54:05 | alex_xu | gmann: yea, can you start a email :) | |
| 07:54:33 | gmann | alex_xu: yea, i will send tomorrow and get the opinion on idea and time also | |
| 07:54:45 | alex_xu | gmann: cool, thanks a lot | |
| 07:54:47 | gmann | alex_xu: thanks :) | |
| 07:59:01 | sean-k-mooney | stephenfin: jaypipes sahid. as the native support is used by multiple drivers and since it is always expected to be present in existing installs keeping it in os-vif makes sense from an upgrade point of view as packagers do not need to include an addtional package to maintain current behavior | |
| 08:02:14 | sean-k-mooney | stephenfin: jaypipes sahid i have created a bug to track this https://bugs.launchpad.net/os-vif/+bug/1764665 but if ye have objections let me know and i can take a different path | |
| 08:02:14 | openstack | Launchpad bug 1764665 in os-vif "Add generic-vhost-user pluging to os-vif" [Wishlist,Triaged] - Assigned to sean mooney (sean-k-mooney) | |
| 08:06:42 | openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: ServerActionsSampleJsonTest refactor https://review.openstack.org/561819 | |
| 08:11:05 | openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: Initial change set of z/VM driver https://review.openstack.org/523387 | |
| 08:11:06 | openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: Spawn and destroy function of z/VM driver https://review.openstack.org/527658 | |
| 08:11:07 | openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add snapshot function https://review.openstack.org/534240 | |
| 08:11:08 | openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add power actions https://review.openstack.org/543340 | |
| 08:11:09 | openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add get console output https://review.openstack.org/543344 | |
| 08:26:54 | openstackgerrit | sahid proposed openstack/nova master: libvirt: add support for virtio-net rx/tx queue sizes https://review.openstack.org/484997 | |
| 08:27:44 | sahid | sean-k-mooney: i just updated the patch | |
| 08:28:10 | sahid | if you can have a look. I really would like limiting the scope of the change if possible | |
| 08:28:15 | sahid | i added some comments | |
| 08:28:23 | sean-k-mooney | sahid: https://review.openstack.org/#/c/484997 ? | |
| 08:28:38 | sahid | yes | |
| 08:29:28 | sahid | sean-k-mooney: about your ovs-vif asking | |
| 08:29:42 | sahid | so you want to create an other pluging in the tree for vhostuser? | |
| 08:30:02 | sean-k-mooney | sahid: specifically for vhost user when not using ovs | |
| 08:30:46 | sean-k-mooney | the plug an unplug methods will be no ops as in the non ovs case nova/os-vif does not need to plug the vhost-user interface into anything | |
| 08:31:19 | sahid | but we will still have to keep the nova part to generate the native domain xml | |
| 08:31:20 | sean-k-mooney | it will however allow use to remove the fallback code in the nova/virt/libvirt/vif.py | |
| 08:31:53 | sean-k-mooney | sahid: yes just the xml generation but that is comon between ovs and non ovs vhost-user interfaces | |
| 08:32:11 | sahid | right, so i'm cool with you plan :) | |
| 08:32:12 | sean-k-mooney | currently we have that twice in the nova tree but we can converge them | |
| 08:33:10 | sahid | yes if we can remove 'get_config_vhost' to ony user the ovsvif one that would be perfect | |
| 08:33:22 | sahid | get_config_vhostuser | |
| 08:37:07 | sean-k-mooney | sahid: ya we would remove the get_config_vhostuser fucntion and then always use the os-vif version instead. im going to try and set up a vpp test env later this week so i can validate it all still works correctly. | |
| 08:40:54 | openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add host/hostId to instance action events API https://review.openstack.org/557145 | |