| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-03-08 | |||
| 15:51:33 | dansmith | jaypipes: yes it does | |
| 15:52:14 | cdent | mriedem: I suspect to avoid depending on ovo | |
| 15:52:23 | mriedem | cdent: we already depend on o.voc | |
| 15:52:25 | mriedem | *o.vo | |
| 15:52:38 | cdent | nova does | |
| 15:52:44 | cdent | other things need resource classes | |
| 15:52:46 | mriedem | and placement will | |
| 15:52:46 | jaypipes | dansmith: oh, sorry, you are referring to the resource **classes** | |
| 15:52:53 | cdent | besides placement | |
| 15:52:54 | jaypipes | dansmith: not nova.objects.resource_provider | |
| 15:53:09 | dansmith | jaypipes: no I mean the objects | |
| 15:53:10 | dansmith | Rnova/api/openstack/placement/objects/resource_provider.py | |
| 15:53:41 | jaypipes | oh ffs. | |
| 15:53:54 | jaypipes | I did not notice that moved the file location. :( | |
| 15:53:56 | jaypipes | sorry dansmith | |
| 15:53:58 | dansmith | cdent: I gave you like three seams in my snarky snarkage above :) | |
| 15:55:09 | cdent | dansmith: where have I split things into separate files (beyond for resource classes)? Or do you just mean that? | |
| 15:55:30 | dansmith | cdent: I'm commenting, I'll put some ideas in there | |
| 15:56:19 | jaypipes | cdent: I think dansmith is just wanting to segregate the changes that put the resource classes stuff into a separate file from the changes that moved the location of nova/objects/resource_provider.py | |
| 15:56:40 | jaypipes | dansmith: amirite? | |
| 15:56:47 | dansmith | at least that | |
| 15:57:26 | cdent | mriedem: As I said in the intro to http://lists.openstack.org/pipermail/openstack-dev/2018-March/128004.html the idea is to put everyting in the existing placement directory because a) the entire system is the api, so that directory will be _the_ directory in the future, b) enable a lift and shift with presered history from just one dir | |
| 15:58:36 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: cover migration cases with functional tests https://review.openstack.org/493865 | |
| 15:58:45 | cdent | thanks dansmith | |
| 15:58:54 | dansmith | man, gerrit is really (really) confused by this change | |
| 16:00:23 | mriedem | cdent: so when we have a separate placement repo, | |
| 16:00:30 | mriedem | there will be a single root dir and it's called 'api'? | |
| 16:00:42 | cdent | no it will be called placement, and there will be no dir called api | |
| 16:00:43 | mriedem | well, presumably 2 dirs, api and test | |
| 16:00:59 | cdent | in placement will be tests | |
| 16:01:30 | mriedem | so chop of the nova/api/openstack prefix and just everything below that becomes the new root | |
| 16:01:40 | cdent | yes | |
| 16:01:42 | mriedem | ok | |
| 16:01:59 | cdent | placement is "simple" enough for that to work | |
| 16:02:12 | cdent | (at least initially) | |
| 16:02:16 | mriedem | yeah that's fine | |
| 16:03:25 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Transform instance.exists notification https://review.openstack.org/403660 | |
| 16:05:41 | Kevin_Zheng | dansmith hi, according to https://github.com/openstack/nova/blob/3a3b0f09db318faf1a1ea711a73bb365cab8b233/nova/context.py#L376 context.db_connection & context.mq_connection will be set to None, will that be a problem when access DB? or if it is None we will fallback to use the option in config file? | |
| 16:06:24 | dansmith | Kevin_Zheng: no, if they're none then it uses the global config, same as you have now | |
| 16:07:03 | Kevin_Zheng | OK, thanks | |
| 16:13:22 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add sample test for instance audit https://review.openstack.org/480955 | |
| 16:14:26 | cdent | dansmith: the resource provider objects were pulled out of a test_objects long ago when we stopped versioning them. Or do you mean something else? | |
| 16:14:50 | dansmith | cdent: ah, right, we register_if(False)'d them right? | |
| 16:14:58 | cdent | yeah | |
| 16:15:13 | dansmith | okay then ignore the fingerprint comments :) | |
| 16:15:22 | cdent | roger :) | |
| 16:15:31 | cdent | rest of them make sense, will think on this a bit later | |
| 16:15:39 | dansmith | ack | |
| 16:20:28 | mriedem | stephenfin: is there a way to give different names for the :oslo.config:option: stuff? | |
| 16:20:42 | mriedem | because i have to put the group in the link but then i don't want that actually showing up in the name | |
| 16:20:49 | mriedem | i.e. | |
| 16:20:50 | mriedem | :oslo.config:option:`api.metadata_cache_expiration` = 15 | |
| 16:20:57 | mriedem | shows up with "api.metadata_cache_expiration" as the name | |
| 16:21:26 | stephenfin | :oslo.config:option:`foobar <api.metadata_cache_expiration>` should work | |
| 16:21:34 | mriedem | yeah that's what i'm trying now, thanks | |
| 16:21:36 | mriedem | like any other ref | |
| 16:38:04 | openstackgerrit | Zhenyu Zheng proposed openstack/nova master: nova-manage db archive_deleted_rows is not multi-cell aware https://review.openstack.org/507486 | |
| 16:55:39 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Re-work the metadata service docs https://review.openstack.org/550250 | |
| 16:56:27 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Replace raise with self.fail() in unit test https://review.openstack.org/550914 | |
| 17:15:23 | openstackgerrit | Merged openstack/nova master: Add simple db purge command https://review.openstack.org/550171 | |
| 17:15:39 | dansmith | hoo boy | |
| 17:18:01 | openstackgerrit | Dan Smith proposed openstack/nova master: Add --purge helper flag to archive_deleted_rows https://review.openstack.org/550182 | |
| 17:18:01 | openstackgerrit | Dan Smith proposed openstack/nova master: Make nova-manage db purge take --all-cells https://review.openstack.org/550502 | |
| 17:18:14 | dansmith | mriedem: had to rebase ^ and fixed your comments on the final one | |
| 17:19:36 | openstackgerrit | Merged openstack/nova stable/pike: Lazy-load instance attributes with read_deleted=yes https://review.openstack.org/545988 | |
| 17:22:39 | openstackgerrit | Stephen Finucane proposed openstack/nova master: hardware: Rework '_get_cpu_topology_constraints' https://review.openstack.org/407173 | |
| 17:22:40 | openstackgerrit | Stephen Finucane proposed openstack/nova master: hardware: Rework get_number_of_serial_ports https://review.openstack.org/407174 | |
| 17:23:06 | stephenfin | jaypipes: I made a silly typo in the first of those patches. Any chance you could re-add it to your review queue 🙈 | |
| 17:23:31 | stephenfin | (You can't do all(a, b, c) - needs to be all((a, b, c)) ) | |
| 17:25:46 | mriedem | dansmith: small thing in https://review.openstack.org/#/c/550502/9 | |
| 17:26:11 | dansmith | ack | |
| 17:26:58 | openstackgerrit | Dan Smith proposed openstack/nova master: Make nova-manage db purge take --all-cells https://review.openstack.org/550502 | |
| 17:27:03 | dansmith | mriedem: ^ | |
| 17:27:25 | mriedem | +5 | |
| 17:28:28 | dansmith | I'm going through Kevin_Zheng's revision now | |
| 17:29:37 | mriedem | i'm going to go through the cupboard | |
| 17:47:29 | dansmith | jaypipes: if you want to send this I can rebase Kevin_Zheng's conflicting patch on master for him when it merges: https://review.openstack.org/#/c/550502 | |
| 17:47:45 | dansmith | it's just a minor --all-cells iteration of what you've already reviewed | |
| 17:48:22 | dansmith | and after that, we'll be able to actually purge cruft from our database, just eight short years after we started | |
| 18:05:55 | mriedem | just in time for spring | |
| 18:07:01 | openstackgerrit | Surya Seetharaman proposed openstack/nova master: Make archive_deleted_rows handle a missing CONF.api_database.connection https://review.openstack.org/550967 | |
| 18:08:42 | stephenfin | mriedem: Am I correct in thinking that all filters (e.g. NUMATopologyFilter) run on whatever host is running 'nova-scheduler'? | |
| 18:09:15 | mriedem | umm, they don't care about the actual physical host that nova-scheduler is running on | |
| 18:09:24 | mriedem | but yes they all run in the nova-scheduler process | |
| 18:09:29 | mriedem | based on the enabled_filters config list | |
| 18:10:06 | mriedem | iow you can run nova-scheduler in a separate host, vm, venv, container whatever | |
| 18:10:24 | stephenfin | Coolness | |
| 18:10:32 | mriedem | just not a function in lambda yet | |
| 18:10:45 | mriedem | but intel is working on it | |
| 18:11:03 | stephenfin | (I'd defined a host-specific option on a nova-compute node and was wondering why the NUMATopologyFilter couldn't see it) | |
| 18:11:23 | mriedem | you'd have to pump that information up to the scheduler via the compute node | |
| 18:11:33 | mriedem | the HostState object in the scheduler is really a wrapper over a compute node | |
| 18:11:45 | mriedem | and the compute node record has all sorts of hook points for crap, like the stats dict | |
| 18:12:03 | stephenfin | Yup. I'm targeting the HostState object now | |
| 18:12:03 | mriedem | or, via traits or something in placement | |
| 18:12:08 | mriedem | if you want the new hotness | |
| 18:12:22 | stephenfin | jaypipes is going to _hate_ v1 of this NUMA-aware vSwitch code | |
| 18:12:26 | mriedem | but since we don't model numa nodes yet... | |
| 18:12:44 | mriedem | s/jaypipes/everyone/ | |
| 18:12:44 | stephenfin | Nor network-NUMA affinity (from neutron) | |