| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-02-20 | |||
| 21:33:50 | bauzas | I think we should be explicit given for example RAX uses CachingScheduler | |
| 21:34:29 | mriedem | i noted this in an earlier patchset and don't think we should be explicit about it :) | |
| 21:34:33 | mriedem | b/c we want people using placement | |
| 21:34:46 | mriedem | rax public cloud might never get to using this anyway, so who knows | |
| 21:35:05 | bauzas | well | |
| 21:35:39 | mriedem | also, the devref has the specific scheduler | |
| 21:36:34 | bauzas | sure, but that's an upgrade note | |
| 21:37:06 | bauzas | so when reading it (even I know not a lot of people are doing that :p ) people could think we would just call Placement every time | |
| #openstack-nova - 2017-02-21 | |||
| 06:57:39 | openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Don't delete neutron port when attach failed https://review.openstack.org/403568 | |
| 07:10:13 | openstackgerrit | jichenjc proposed openstack/nova master: Check instance existing before check in mapping https://review.openstack.org/433470 | |
| 07:30:33 | openstackgerrit | Sergey Nikitin proposed openstack/nova-specs master: PCI NUMA Policies https://review.openstack.org/361140 | |
| 07:54:17 | openstackgerrit | jichenjc proposed openstack/nova master: pi-ref: Body verification for os-hypervisors.inc https://review.openstack.org/383691 | |
| 09:44:40 | jaosorior | gibi: are you around? | |
| 09:44:47 | jaosorior | gibi: still working on notifications? | |
| 10:25:39 | openstackgerrit | yuhui proposed openstack/nova master: Add Apache License Content in index.rst https://review.openstack.org/436397 | |
| 11:19:05 | openstackgerrit | Sean Dague proposed openstack/nova master: Verify project id for flavor access calls https://review.openstack.org/435432 | |
| 11:19:06 | openstackgerrit | Sean Dague proposed openstack/nova master: Verify project_id when quotas are checked https://review.openstack.org/435010 | |
| 11:55:33 | openstackgerrit | Spencer Yu proposed openstack/nova master: Enable coverage report https://review.openstack.org/436423 | |
| 12:53:05 | gibi | jaosorior: hi, yes and yes | |
| 12:54:04 | jaosorior | gibi: will versioned notifications be enabled by default any time soon? | |
| 12:55:01 | jaosorior | gibi: also, I see there's no notifications for when an instance's metadata has been updated, would it be possible to add that? | |
| 12:55:19 | gibi | jaosorior: they are enabled by default from the beginning | |
| 12:56:48 | jaosorior | gibi: got docs about that? we have an application that reads nova notifications, however, as far as I had understood the versioned notifications had some encapsulation. And we're not taking that into account. | |
| 12:57:28 | gibi | jaosorior: https://docs.openstack.org/developer/nova/notifications.html | |
| 12:59:08 | jaosorior | gibi: right, I'm pretty sure we're using unversioned notifications (and it's working with the latest nova). So I was wondering if there's something we need to set to start emitting versioned notifications, so I can start changing the code base of our application to use those instead | |
| 13:00:21 | gibi | legacy notificatins are emitted to the 'notifications' topic on the message bus while versioned notifications are emitted to 'versioned_notifications' topic | |
| 13:00:33 | jaosorior | gibi: oh, that makes sense | |
| 13:01:07 | jaosorior | gibi: so, do you think is feasable to add a notification for when instance's metadata is updated? | |
| 13:02:16 | gibi | jaosorior: technically it is feasible. From priority perspective it would be better to finish transoforming the legacy notifications before we are adding new ones | |
| 13:03:57 | gibi | jaosorior: I think it is OK to propose a bp with the instance metadata update notification. We will see how far we can go with that | |
| 13:05:10 | jaosorior | gibi: also, I don't see floating ip related notifications in that list. Is that coming soon? | |
| 13:09:15 | gibi | jaosorior: yeah, we have a long todo list here https://vntburndown-gibi.rhcloud.com/index.html | |
| 13:10:08 | gibi | jaosorior: btw are you in Atlanta for the PTG? | |
| 13:10:13 | jaosorior | gibi: not this time | |
| 13:10:39 | jaosorior | gibi: just came from Mexico; enough long travels for me for a while haha | |
| 13:11:11 | gibi | jaosorior: I see. This was a long flight for me as well | |
| 13:11:47 | gibi | jaosorior: which module you are working with needing these notifications? | |
| 13:12:25 | jaosorior | gibi: it's a nova vendordata plugin https://github.com/openstack/novajoin/blob/master/novajoin/notifications.py | |
| 13:13:02 | jaosorior | gibi: registers nova instances as FreeIPA clients | |
| 13:14:41 | jaosorior | gibi: is there a time plan where we should migrate the unversioned notifications? | |
| 13:14:45 | gibi | thanks for the info it is always good to know somebody trys to use versioned notifications :) | |
| 13:16:08 | jaosorior | gibi: I'll let you know when we do. But we would need pretty much these ones https://github.com/openstack/novajoin/blob/master/novajoin/notifications.py#L56 before we can migrate | |
| 13:20:57 | gibi | jaosorior: you can always help us transforming the floating ip notifications | |
| 13:21:21 | gibi | jaosorior: there are lot of examples already merged or up on review | |
| 13:21:55 | jaosorior | gibi: pass me some patchsets and I'll see what I can do. my plate is quite full at the moment but I could try to make some time. | |
| 13:23:04 | gibi | jaosorior: for example https://review.openstack.org/#/c/394511/ | |
| 13:23:55 | jaosorior | gibi: thanks | |
| 13:24:14 | gibi | jaosorior: here is the ocata topic with all the patches https://review.openstack.org/#/q/topic:bp/versioned-notification-transformation-ocata | |
| 13:31:14 | openstackgerrit | Moshe Levi proposed openstack/nova-specs master: change pci-passthrough-whitelist-regex spec to 644 https://review.openstack.org/436451 | |
| 13:34:11 | openstackgerrit | Moshe Levi proposed openstack/nova-specs master: User-controlled SR-IOV ports allocation https://review.openstack.org/182242 | |
| 13:50:02 | mriedem | dansmith: https://review.openstack.org/#/c/426926/ | |
| 13:58:52 | frickler | is there a placement-api client planned? from a deployers perspective, it would seem nice to have an easy way to check the data in the db | |
| 14:06:55 | mriedem | edmondsw: http://lists.openstack.org/pipermail/openstack-dev/2016-August/100586.html | |
| 14:07:02 | mriedem | for the nova-network deprecation/removal talk at the newton midcycle | |
| 14:08:06 | mriedem | which reminds me that we need to remove the network CLIs from novaclient | |
| 14:18:59 | openstackgerrit | Gyorgy Szombathelyi proposed openstack/nova master: Use the keystone session loader in the placement reporting https://review.openstack.org/436475 | |
| 14:23:33 | openstackgerrit | sahid proposed openstack/nova-specs master: Libvirt driver emulator threads placement policy https://review.openstack.org/427066 | |
| 14:34:56 | openstackgerrit | sahid proposed openstack/nova master: numa: add numa constraints for emulator threads policy https://review.openstack.org/385023 | |
| 14:34:59 | openstackgerrit | sahid proposed openstack/nova master: libvirt: return a CPU overhead if isolate emulator threads requested https://review.openstack.org/385364 | |
| 14:35:00 | openstackgerrit | sahid proposed openstack/nova master: numa: fit instance topology with cpus reserved onto cores https://review.openstack.org/400875 | |
| 14:35:01 | openstackgerrit | sahid proposed openstack/nova master: virt: update overhead to take into account vCPUs https://review.openstack.org/385024 | |
| 14:35:04 | openstackgerrit | sahid proposed openstack/nova master: libvirt: make emulator threads to run on the reserved pCPU https://review.openstack.org/400878 | |
| 14:35:05 | openstackgerrit | sahid proposed openstack/nova master: numa: update numa usage incording the cpu reserved https://review.openstack.org/400877 | |
| 14:35:05 | openstackgerrit | sahid proposed openstack/nova master: numa: take into account cpus reserved https://review.openstack.org/400876 | |
| 14:54:14 | openstackgerrit | Dinesh Bhor proposed openstack/nova-specs master: Policy support for specific metadata keys https://review.openstack.org/436493 | |
| 14:54:52 | openstackgerrit | Matt Riedemann proposed openstack/python-novaclient master: WIP: Remove deprecated image commands/API bindings https://review.openstack.org/436494 | |
| 14:58:19 | mriedem | frickler: mlavalle has one somewhere, sec | |
| 14:59:20 | mriedem | frickler: https://github.com/miguellavalle/python-placementclient | |
| 15:00:07 | mriedem | i don't think that's the one i was thinking of, it doesn't have any code in it | |
| 15:00:52 | mriedem | cdent: do you remember where miguel's osc placement client plugin lives? ^ | |
| 15:01:33 | cdent | mriedem: yeah, one sec | |
| 15:01:41 | mriedem | https://github.com/miguellavalle/python-placementclient/tree/adding-grp-support | |
| 15:01:43 | mriedem | found it | |
| 15:01:49 | cdent | https://github.com/miguellavalle/python-placementclient | |
| 15:01:57 | cdent | more coke | |
| 15:02:10 | frickler | mriedem: cdent: ah, nice, thanks | |
| 15:02:18 | mriedem | cdent: yeah the branch was the key | |
| 15:02:21 | mriedem | master is mostly empty | |
| 15:03:38 | cdent | frickler: what we found in the resource tracker was that a client wasn't all that needed. keystoneauth makes it pretty easy | |
| 15:04:52 | mriedem | i think admins are still going to want something in osc | |
| 15:05:03 | mriedem | for linking aggregates to resource providers | |
| 15:05:15 | cdent | mriedem: yeah, sure, just noting it | |
| 15:05:20 | tittywai | hi... what country are you people from | |
| 15:06:04 | mriedem | tittywai: that's a very random question, is there context? | |
| 15:06:13 | edmondsw | tittywai lots of different ones | |
| 15:06:19 | frickler | it may be easy, but for an operator it would still be nice to have a command at the end of https://docs.openstack.org/developer/nova/placement.html that confirms that all computes have successfully placed their data into the api | |
| 15:07:00 | tittywai | no what I mean is that.. where you are right now.. to be more specific... :P | |
| 15:07:26 | mriedem | tittywai: many people are currently at the PTG in Atlanta, GA in the US | |
| 15:08:12 | bauzas | frickler: I think it's a bit premature to have that, given we also need to cut the current methods we use into a different module at least | |
| 15:08:36 | bauzas | frickler: a separate try was https://review.openstack.org/#/c/417111/ | |
| 15:08:56 | tittywai | mriedem: thanks for the info | |
| 15:09:01 | tittywai | im from sg | |
| 15:09:06 | mriedem | bauzas: frickler just wants to know they are fully upgraded i believe | |
| 15:09:16 | mriedem | so that all computes are reporting into placement so the filter scheduler starts using placement | |
| 15:09:27 | bauzas | um ok | |
| 15:09:56 | mriedem | so i guess nova hypervisor-list wc matches placement resource-provider-list | |
| 15:10:02 | mriedem | or something like that | |
| 15:10:05 | mriedem | frickler: please confirm | |
| 15:10:23 | mriedem | frickler: also, check out the "nova-status upgrade check" command | |
| 15:10:44 | mriedem | frickler: https://docs.openstack.org/developer/nova/man/nova-status.html#upgrade | |