| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-08-23 | |||
| 14:28:53 | jroll | ykarel: conf.scheduler.scheduler_driver | |
| 14:29:29 | sean-k-mooney | ykarel: now that you know. please never set it to anything other then the filter schduler | |
| 14:29:51 | ykarel | jroll, driver=filter_scheduler | |
| 14:29:57 | ykarel | jroll, sean-k-mooney we have ^^ | |
| 14:30:10 | jroll | cue denver: <tripleo> we rely on the caching scheduler, please don't remove it! | |
| 14:31:35 | sean-k-mooney | jroll: at this point anything we have so much stuff using placement im not even sure you could deploy a fucntional nova with the caching schduler unless you never migrate or resize anything | |
| 14:32:20 | jroll | sean-k-mooney: if anyone can hack it, it's tripleo ;D | |
| 14:33:01 | jaypipes | sean-k-mooney: "that would make the deployment unsupported downstream" wait, what!? but I thought triple-o wasn't a RH product? :P | |
| 14:33:24 | sean-k-mooney | jaypipes: i was refering to OSP | |
| 14:33:40 | jaypipes | sean-k-mooney: ok dokey. :) | |
| 14:33:40 | sean-k-mooney | e.g. osp being the downstream | |
| 14:33:56 | sean-k-mooney | tripplo has too many names | |
| 14:34:12 | jaypipes | sean-k-mooney: and you just gave it another ;) "tripplo" | |
| 14:34:18 | sean-k-mooney | its tripleo, rdo director, osp director and apex | |
| 14:34:35 | jaypipes | sean-k-mooney: oh, you mean exactly what RH did with OpenShift? | |
| 14:34:59 | jaypipes | sean-k-mooney: or at least, *the current incarnation* of OpenShift. | |
| 14:36:00 | sean-k-mooney | Openshift is (kubernetes + extentions) and i think its called Openshift container platform or somthing as down stream product. i should proable know that... | |
| 14:36:35 | jaypipes | sean-k-mooney: yes, I'm referring to the evolving meaning of openshift(.com/org) over the last 5 years... | |
| 14:41:46 | mgagne_ | mriedem: I don't think the caching scheduler will be needed anymore if placement does the job and I can migrate to it easily. we tested that already I think | |
| 14:41:46 | mriedem | mgagne_: i hoped that's what you'd say, based on earlire discussions re heal_allocations | |
| 14:41:57 | mgagne_ | mriedem: if caching scheduler can still be used in a out of tree fashion, there is a plan B if anything goes wrong. | |
| 14:43:03 | mriedem | yes, that's possible | |
| 14:45:14 | sahid | mriedem, melwitt not sure of what you can put in the release note or config doc instead of don't mix rx/tx queue size option if you don't use guests with only virtio devices :/ | |
| 14:45:37 | sahid | i commented a think we could have a patch which is limiting the impact | |
| 14:49:40 | gibi | edmondsw, mriedem: bug about the noise debug log https://bugs.launchpad.net/nova/+bug/1788634 | |
| 14:49:40 | openstack | Launchpad bug 1788634 in OpenStack Compute (nova) "Noisy DEBUG log: Defaulting the value of the field 'numa_membw_values' to None in MetricPayload" [Low,Confirmed] - Assigned to Balazs Gibizer (balazs-gibizer) | |
| 14:49:42 | sean-k-mooney | sahid: well enduses can change the interface model by uploading an imagge and setting the vif_modle in the metatdat | |
| 14:50:09 | sean-k-mooney | sahid: so the nova code needs to handel that correctly and not set the queue lenght in that case | |
| 14:50:40 | edmondsw | gibi cool, tx | |
| 14:54:59 | sahid | sean-k-mooney: yes and only virtio model works with queues, so is user is setting something else, we don't want use queues | |
| 14:55:47 | sahid | did i miss something? | |
| 14:56:18 | sean-k-mooney | sahid: right, no just makeing sure that is the current behavior. i belive we check for that in the code but if we dont its a bug | |
| 14:57:10 | sahid | a even less impacting solution is to reproduce what is done https://github.com/openstack/nova/blob/34956bea4beb8e5ba474b42ba777eb88a5eadd76/nova/virt/libvirt/vif.py#L153 | |
| 14:57:22 | sahid | for rx/tx queue block | |
| 15:00:14 | kosamara | Continuing from the meeting on https://review.openstack.org/#/c/579897 | |
| 15:01:12 | jaypipes | stephenfin: you fixing up https://review.openstack.org/#/c/595802/? | |
| 15:01:27 | stephenfin | jaypipes: as we speak | |
| 15:01:54 | kosamara | dansmith: I don't propose a new knob actually. The same condition that triggers "hide_hypervisor_id" will additionally hide the hyperv signature, if it's a windows guest. | |
| 15:02:20 | melwitt | sahid: ack | |
| 15:02:42 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Remove noisy DEBUG log https://review.openstack.org/595810 | |
| 15:02:54 | gibi | edmondsw, mriedem: ^^ | |
| 15:02:54 | dansmith | kosamara: okay I see, my mistaked | |
| 15:02:58 | kosamara | this is functionality that the "hide_hypervisor_id" feature already promises, but currently doesn't work for Windows guests. | |
| 15:03:28 | dansmith | kosamara: I think that we probably shouldn't have that feature, but that's just MHO and I'm wrong a lot. Also, I'm a terrible person. | |
| 15:03:58 | jaypipes | stephenfin: ack. cuz me likey that series. | |
| 15:04:04 | kosamara | dansmith: Self-sarcasm is usually a sign to the opposite :P | |
| 15:04:15 | mriedem | edmondsw: https://bugs.launchpad.net/nova/+bug/1788634/comments/3 | |
| 15:04:15 | openstack | Launchpad bug 1788634 in OpenStack Compute (nova) "Noisy DEBUG log: Defaulting the value of the field 'numa_membw_values' to None in MetricPayload" [Low,In progress] - Assigned to Balazs Gibizer (balazs-gibizer) | |
| 15:04:17 | jaypipes | kosamara: like dansmith I am also a terrible person. | |
| 15:04:30 | jaypipes | kosamara: and no, we're not being sarcastic, :P | |
| 15:06:20 | mriedem | edmondsw: tl;dr the object changes were added for the numa bw metric but the implementation wasn't | |
| 15:06:39 | mriedem | b/c we decided around the time of the newton midcycle meeting that nova as a stats gatherer was dumb | |
| 15:06:50 | edmondsw | mriedem got it | |
| 15:06:55 | edmondsw | tx for the history | |
| 15:07:00 | mriedem | and that field likely shouldn't be in the notification payload | |
| 15:07:04 | mriedem | now that you mention it | |
| 15:07:20 | mriedem | but dropping it would be a major version bump i think - gibi agree? | |
| 15:07:31 | edmondsw | mriedem gibi so should the solution be to remove the log statement or to remove numa_membw_values from the schema? | |
| 15:07:38 | gibi | mriedem: yepp, I was in the process of typing this comment | |
| 15:07:52 | edmondsw | or to keep it in the schema but start setting it to None explicitly | |
| 15:08:24 | gibi | mriedem, edmondsw: I think the notification change should be tight to the cleanup of this half done feature | |
| 15:08:27 | kosamara | jaypipes, dansmith: terrible or not, thanks a lot for the discussion. | |
| 15:08:34 | mriedem | gibi: edmondsw: looking at http://logs.openstack.org/97/579897/4/check/tempest-full/53d05dc/controller/logs/screen-n-cpu.txt.gz | |
| 15:08:41 | mriedem | "Defaulting the value of the field" shows up over 1K times | |
| 15:08:52 | mriedem | so i think the debug log should go away regardless | |
| 15:09:20 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Move mocking to setUp https://review.openstack.org/595802 | |
| 15:09:21 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Create functional libvirt test base class https://review.openstack.org/407055 | |
| 15:09:39 | mriedem | Flavor.projects is a huge one | |
| 15:09:44 | mriedem | since flavor is in all instance payloads | |
| 15:09:58 | gibi | mriedem: I agree that we need to drop the debug log | |
| 15:10:37 | gibi | mriedem: regardless of the notificaton change | |
| 15:10:44 | mriedem | +2 | |
| 15:10:48 | gibi | mriedem: or more like independently | |
| 15:11:27 | edmondsw | +1 from me | |
| 15:11:31 | ykarel | mriedem, jaypipes jroll sean-k-mooney Thanks, i commented https://bugs.launchpad.net/tripleo/+bug/1787910/comments/16, please correct if i added something wrong there | |
| 15:11:31 | openstack | Launchpad bug 1787910 in tripleo "OVB overcloud deploy fails on nova placement errors" [Critical,Triaged] - Assigned to Marios Andreou (marios-b) | |
| 15:11:33 | gibi | awesome :) | |
| 15:12:00 | jroll | ykarel: thanks! looks good to me | |
| 15:12:17 | openstackgerrit | Merged openstack/nova master: Fix nits in resource_provider.py https://review.openstack.org/588470 | |
| 15:12:30 | sean-k-mooney | mriedem: ok i finally got to a point where i could test https://review.openstack.org/#/c/594508/2. the error is still being thrown from qemu but the newroking on the vm is not broken anymore | |
| 15:13:11 | sean-k-mooney | mriedem: so that looks good to me. ill +1 it upstream | |
| 15:13:17 | mriedem | sean-k-mooney: awesome, can you note in the review so gibi can "put his love on it" or whatever he said earlier? | |
| 15:13:20 | mriedem | sean-k-mooney: thanks | |
| 15:13:36 | mriedem | i'm pretty sure gibi sexually harassed my patch | |
| 15:13:40 | edmondsw | efried https://docs.openstack.org/nova/latest/contributor/policies.html#metrics-gathering says there is "not yet a clear replacement"... is that still true, or does the placement service fill that role now? | |
| 15:13:51 | mriedem | placement doesn't gather metrics | |
| 15:14:21 | mriedem | but you could probably report metrics type things into placement to be used during nova scheduling | |
| 15:14:31 | gibi | mriedem: I think you started it ;) "* mriedem assumes gibi will love it immediately and want to +W so hard" | |
| 15:14:32 | mriedem | like the nw bw stuff gibi is working on | |
| 15:14:44 | mriedem | gibi: ha | |
| 15:15:33 | efried | "report metrics type things into placement"? mind blown | |
| 15:16:54 | gibi | I'm not sure how the available and used nw bw in placement relates to metrics but yeah | |
| 15:16:57 | sean-k-mooney | efried: metric type things? | |
| 15:17:10 | efried | sean-k-mooney: Don't look at me. It was mriedem's zany idea | |
| 15:17:46 | sean-k-mooney | i obvioulsy missed somthing while debuging my devstack setup | |
| 15:17:54 | edmondsw | efried you can probably blame me there | |
| 15:18:21 | edmondsw | I don't know that either of us were suggesting it should be done, just wondering | |
| 15:18:25 | sean-k-mooney | edmondsw: placement does not store metrics no | |
| 15:18:33 | sean-k-mooney | edmondsw: also it likely never will | |
| 15:18:47 | edmondsw | I don't really know why nova ever used metrics for scheduling | |