| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-10-27 | |||
| 08:05:14 | melwitt | kashyap: oh, in case you didn't know, the reason we were moving so fast on that change is because newton eol was being held back on it | |
| 08:05:42 | kashyap | melwitt: Ah, I recall dansmith saying during a call last week | |
| 08:06:03 | kashyap | melwitt: Thank you for the unit test fixes. I struggle a bit with it | |
| 08:06:32 | melwitt | np. writing the tests is the most challenging part for me most of the time | |
| 08:07:36 | kashyap | melwitt: Any tips you have (write later, when you rested) to get better at it are welcome :-) | |
| 08:07:50 | melwitt | heh, k | |
| 08:07:56 | kashyap | Maybe just: keep reading and tinkering more and write more of them :-) | |
| 08:08:41 | melwitt | yeah, that's most of it. gets better with practice | |
| 08:11:38 | melwitt | alright yall, seeya in some hours | |
| 08:14:52 | kashyap | Get some sleep. | |
| 08:14:55 | openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add instance action record for attach/detach interface https://review.openstack.org/512190 | |
| 08:19:17 | openstackgerrit | Yikun Jiang proposed openstack/python-novaclient master: Use utils.prepare_query_string instead of duplicated code https://review.openstack.org/515640 | |
| 08:21:19 | openstackgerrit | Merged openstack/nova master: libvirt: Don't disregard cache mode for instance boot disks https://review.openstack.org/514339 | |
| 08:33:25 | openstackgerrit | Jianghua Wang proposed openstack/nova master: vgpu: add enabled white list https://review.openstack.org/512580 | |
| 08:33:25 | openstackgerrit | Jianghua Wang proposed openstack/nova master: vGPU: XenAPI: get vgpu stats from hypervisor https://review.openstack.org/512965 | |
| 09:10:54 | openstackgerrit | Chen Hanxiao proposed openstack/python-novaclient master: inject file: add description of injecting multiple files. https://review.openstack.org/515655 | |
| 09:27:49 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Remove 'vendordata_driver' opt https://review.openstack.org/397835 | |
| 09:28:05 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Validate '[api] vendordata_providers' options https://review.openstack.org/513318 | |
| 09:29:29 | stephenfin | mikal: Could you have a look at https://review.openstack.org/#/c/513318 and the preceding patch? | |
| 09:54:20 | openstackgerrit | Radoslav Gerganov proposed openstack/nova master: VMware: fix memory stats https://review.openstack.org/515666 | |
| 10:02:33 | openstackgerrit | Yikun Jiang proposed openstack/python-novaclient master: Use utils.prepare_query_string instead of duplicated code https://review.openstack.org/515640 | |
| 10:07:50 | openstackgerrit | Merged openstack/nova master: Complete dostring of live_migration related methods https://review.openstack.org/490807 | |
| 10:13:40 | openstackgerrit | Radoslav Gerganov proposed openstack/nova master: VMware: implement get_inventory() driver method https://review.openstack.org/506175 | |
| 10:19:44 | openstackgerrit | Jianghua Wang proposed openstack/nova master: XenAPI: resolve the error of VM_MISSING_PV_DRIVERS for PV VM https://review.openstack.org/514902 | |
| 10:43:36 | jianghuaw | bauzas, could you help to look at this patch? https://review.openstack.org/#/c/339204 | |
| 11:30:09 | openstackgerrit | Merged openstack/python-novaclient master: Use generic user for both zuul v2 and v3 https://review.openstack.org/512515 | |
| 11:42:26 | openstackgerrit | Radoslav Gerganov proposed openstack/nova master: VMware: implement get_inventory() driver method https://review.openstack.org/506175 | |
| 12:01:48 | jmccarthy | Any folks notice an issue where on a xen compute, in horizon, you may get a working console, or working console logging (depending on the instance you spin up) - but not both ? | |
| 12:25:59 | jaypipes | bauzas, stephenfin, gibi: series starting at https://review.openstack.org/#/c/509027/ should be relatively easy reviews. just cleanups in the resource_provider.py module. | |
| 12:28:36 | gibi | jaypipes: opening... | |
| 12:58:17 | gibi | jaypipes: what is the reason you pass a whole RP object to get_all_by_resource_provider ? It only uses the id of that RP today so it would be enough to pass that id. | |
| 12:58:50 | jaypipes | gibi: because I don't then have to reconstruct a ResourceProvider object for each allocation record. | |
| 12:59:21 | jaypipes | gibi: and that would entail more joins and/or queries to the DB | |
| 13:01:56 | bauzas | jianghuaw: jaypipes: I had a comment in https://review.openstack.org/#/c/339204/8 | |
| 13:03:08 | leakypipes | bauzas: nice catch :) | |
| 13:04:25 | bauzas | oh man, friday | |
| 13:04:46 | gibi | leakypipes: I'm not sure we are looking at the same patch. I try to understand TraitList.get_all_by_resource_provider() here https://review.openstack.org/#/c/509027/5/nova/objects/resource_provider.py@2289 | |
| 13:05:10 | gibi | leakypipes: and I don't see how this connects to the allocation records you refer to | |
| 13:05:31 | leakypipes | gibi: lol. sorry, thought you were referring to the AllocationList.get_all_by_provider() | |
| 13:05:33 | bauwser | leakypipes: honestly, it's a known problem, and we should maybe discuss about how to modify the RC to use bytes and not GB but meh | |
| 13:05:59 | bauwser | changing that should be a very huge modification unfortunately | |
| 13:06:18 | leakypipes | gibi: I have no idea why that isn't get_by_provider_id()... | |
| 13:06:26 | leakypipes | gibi: lemme fix. | |
| 13:06:37 | openstackgerrit | Hironori Shiina proposed openstack/nova-specs master: Ironic: Resize and cold migration support https://review.openstack.org/449155 | |
| 13:06:38 | leakypipes | gibi: might just be copy-pasta | |
| 13:07:13 | gibi | leakypipes: thanks | |
| 13:07:20 | leakypipes | gibi: sorry about that! | |
| 13:08:08 | gibi | leakypipes: no problem | |
| 13:08:30 | leakypipes | gibi: although... | |
| 13:08:41 | leakypipes | gibi: based on the commit message | |
| 13:09:12 | leakypipes | gibi: looks like this was intentional to make the calling interfaces consistent | |
| 13:09:20 | leakypipes | gibi: from the commit message: | |
| 13:09:20 | leakypipes | In following patches, I change similar method on List objects like UsageList.get_all_by_resource_provider_uuid() and AllocationList.get_all_by_resource_provider_uuid() to match this new pattern of get_all_by_resource_provider() and passing in a ResourceProvider object. | |
| 13:09:49 | leakypipes | gibi: those "following patches" were re-ordered and are actually already merged. | |
| 13:10:35 | gibi | leakypipes: that removes some of my confusion :) anyhow if there is pattern and it is intentional then I'm OK with it | |
| 13:10:43 | leakypipes | gibi: so, long story short, I did it for consistency with the other calls | |
| 13:11:18 | gibi | leakypipes: make sense now, thanks for the explanation | |
| 13:25:57 | belmoreira | mriedem, dansmith I continue to test how to migrate from newton to ocata with cellsV1 | |
| 13:26:15 | belmoreira | If I have multiple nova_api(s) DBs/placements I will need to merge the allocation tables when moving to pike and cellsV2 | |
| 13:26:30 | belmoreira | This is not easy to achieve because "allocations" table uses the "resource_provider_id" instead "resource_provider_uuid" | |
| 13:26:36 | jmccarthy | Question about using libvirt + xen on xen compute, anyone familiar with how to get changes to 'stick' when virsh editing the xml ? I'm trying to add another serial device to file, for console logging for example: http://paste.openstack.org/show/624831/ | |
| 13:26:42 | belmoreira | There is any way to recreate the "allocations" table from scratch when I'm in pike? | |
| 13:26:58 | jmccarthy | (I have what is on the right, but I think I need what is on the left) | |
| 13:27:43 | mriedem | belmoreira: would likely have to write a tool to migrate the entries out of the separate placement/api db's into the global one when you upgrade | |
| 13:27:53 | mriedem | we mentioned this as one of the trade offs the other day of doing placement per cell for now | |
| 13:28:37 | dansmith | what mriedem said | |
| 13:28:51 | dansmith | belmoreira: the use of integer ids is for performance there | |
| 13:30:27 | mriedem | bauwser: what is your -1 for specifically on this? https://review.openstack.org/#/c/515530/ | |
| 13:31:37 | bauwser | mriedem: on a meeting | |
| 13:34:58 | fried_rice | mriedem superdan Can we please move the ksa adapter series along a little bit? It's been stagnating. Here's one that involves placement https://review.openstack.org/#/c/492247/ -- sdague was +2 6 weeks ago and no actual changes since. | |
| 13:37:05 | sdague | fried_rice: it has a gerrit cannot merge on it atm | |
| 13:37:21 | fried_rice | sdague Aaargh, missed that. | |
| 13:37:33 | fried_rice | rebasing... | |
| 13:37:47 | superdan | also, I'm not really the best person to look at that | |
| 13:37:52 | superdan | probably mriedem and sdague though | |
| 13:38:26 | mriedem | fried_rice: you're just doing all of these separately right? | |
| 13:38:38 | mriedem | for some reason i thought they were all lined up behind the glance one, which looked scary | |
| 13:38:39 | fried_rice | mriedem Yes | |
| 13:38:49 | fried_rice | mriedem Oh, some of them are lined up behind the glance one. | |
| 13:38:53 | fried_rice | But this one isn't. | |
| 13:41:08 | openstackgerrit | OpenStack Proposal Bot proposed openstack/os-vif stable/pike: Updated from global requirements https://review.openstack.org/493146 | |
| 13:41:39 | sdague | realized that after going from Ubuntu Unity -> Gnome 3 in 17.10, I lost my mechanism of knowing people had pinged me on irc. Think I got a solution for that today. Which hopefully means noticing when mriedem pokes at me | |
| 13:43:24 | belmoreira | mriedem, dansmith in newton when placement is enabled the resource_providers/allocations tables are populated. This doesn't happen for pike. Could this be achieved somehow? would be better than having a tool to merge the dbs | |
| 13:44:26 | mriedem | belmoreira: once all computes are upgraded to pike, the compute resource tracker stops reporting allocations in the update_available_resource periodic task | |
| 13:44:28 | mriedem | on purpose, | |
| 13:44:43 | mriedem | because in pike the scheduler creates allocations during create and move operations like migrate, | |
| 13:44:52 | mriedem | and the ocata computes would overwrite the allocations created by the pike scheduler | |
| 13:46:17 | superdan | what mriedem said again :) | |
| 13:47:11 | superdan | belmoreira: I think we should probably be on the hook to help with said merging tool because you won't be the only one that needs it | |
| 13:49:14 | openstackgerrit | Eric Fried proposed openstack/nova master: Use ksa adapter for placement conf & requests https://review.openstack.org/492247 | |
| 13:49:18 | fried_rice | sdague mriedem ^ | |
| 13:49:19 | mriedem | superdan: is there any specific reason that nova.objects.base.obj_to_primitive doesn't convert datetimes to primitives? | |
| 13:49:39 | superdan | mriedem: um | |
| 13:49:46 | belmoreira | mriedem if in ocata the resource tracker still reports the allocations, means that in my upgrade procedure I move to the global placement just before upgrade to pike. In that case I would not need to merge the dbs. | |
| 13:51:08 | mriedem | belmoreira: so you're going to update nova.conf on all of the computes to point at global placement, take down the control plane and wait for the periodic task to do the inventory and allocation population? | |
| 13:51:19 | mriedem | s/control plane/api | |
| 13:51:56 | mriedem | belmoreira: to be clear, the resource tracker will continue to update allocations in placement as long as there is at least 1 <pike compute | |
| 13:52:02 | superdan | oh _that_ obj_to_primitive | |
| 13:52:11 | mriedem | it's just once all computes are >= pike we stop doing that | |
| 13:52:21 | mriedem | superdan: yeah i expected that to turn the entire object and it's fields into primitives | |