Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-07-30
16:31:15 mriedem and it will log the cell mapping uuid
16:31:47 mriedem at debug i mean
16:31:53 mnaser well, I aded a LOG.error(inst_map.cell_mapping) and i see the cell mapping there on the request
16:32:11 mnaser so i guess objects.Instance.get_by_uuid is raising some exception
16:32:42 mnaser also
16:32:45 mriedem can you also log the context.db_connection right before Instance.get_by_uuid?
16:32:51 mnaser sure
16:32:52 mnaser https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2398-L2400
16:32:54 mnaser i think this is dead code
16:32:57 mriedem it won't be a string, but needs to be not None
16:33:09 melwitt usually, the only exception there would be InstanceNotFound but if deleted != 0 you wouldn't see it in 'nova list' either
16:33:17 mriedem mnaser: "# NOTE(ameade): we still need to support integer ids for ec2"
16:33:32 mnaser well the exception handles InvalidID
16:33:55 mnaser but if uuidutils.is_uuid_like is raises that then you never end up in else:
16:34:01 mnaser and if it doesn't, then InvalidID might never be raised
16:34:10 mnaser but idk, it just seems a bit weird but anyways
16:34:22 mnaser let me check context.db_connection
16:35:43 mnaser mriedem: <oslo_db.sqlalchemy.enginefacade._TransactionContextManager object at 0x955e450>
16:35:50 mriedem ok
16:35:56 mnaser i mean i could try to remember how to make sqlalchemy log sql queries
16:36:01 mnaser i remember i did that at some point
16:37:13 mriedem mnaser: it's either https://docs.openstack.org/nova/latest/configuration/config.html#database.connection_debug or https://docs.openstack.org/nova/latest/configuration/config.html#database.connection_trace i think
16:38:00 mriedem oh you know what...
16:38:04 mriedem i bet i know what this is
16:38:13 mriedem lazy-join on a deleted compute service by the instance record
16:38:18 mriedem it's not the instance record that's the problem
16:38:24 mriedem it's a related deleted service record that we can't load
16:38:31 mriedem melwitt: remember that^ ?
16:39:03 mriedem mnaser: i'd query the db that the instance is for all nova-compute services with the same name as the instance.host
16:39:07 mriedem are any of those deleted?
16:39:40 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Use oslotest CaptureOutput fixture https://review.openstack.org/587129
16:39:41 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Use a non-nova log capture fixture https://review.openstack.org/587130
16:39:42 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Use a simplified WarningsFixture https://review.openstack.org/587131
16:39:44 mriedem it could be trying to set a uuid on a deleted service record
16:39:55 melwitt yeah... thinking. the one I've got a patch up for is around generating a uuid for a deleted service record
16:40:06 mriedem but that was with listing i thought, not sohw
16:40:08 mriedem *show
16:40:36 mriedem mnaser: i'd probably put a try/except around that Instance.get_by_uuid to handle InstanceNotFound, LOG.exception and re-raise
16:40:45 mriedem mnaser: to figure out where the InstanceNotFound originates from
16:41:59 melwitt yeah, according to the bugs, that would raise ServiceNotFound or ServiceTooOld for listing
16:46:49 openstackgerrit Balazs Gibizer proposed openstack/nova master: Remove superfluous network stubbing in func test https://review.openstack.org/587134
16:49:39 anupn Hi Folks, I have deployed an instance on ironic node, but that instance is unable to ping to the external network. Can you please suggest what things should I check for? my br-ex has got an external IP, and I can ping the internet from the host but not from the vm.
16:51:28 melwitt anupn: hi, this is a development channel, not a support channel. pls see topic
16:52:49 anupn melwitt: Ah ok! Will ask in the other channel
16:59:45 mnaser sorry, in a call, ill catch up on buffer in a sec
17:07:44 openstackgerrit melanie witt proposed openstack/nova master: Set default of oslo.privsep.daemon logging to INFO level https://review.openstack.org/586643
17:13:48 mnaser ill do the try/raise, the service is actually existing
17:22:17 mnaser http://paste.openstack.org/show/726864/
17:23:13 mnaser i feel like i need to take a step back and something weird is happening
17:23:37 openstackgerrit Chris Dent proposed openstack/nova master: DNM: Explore limiting resource tracker update https://review.openstack.org/587050
17:23:38 openstackgerrit Chris Dent proposed openstack/nova master: Extract _update_to_placement method in resource tracker https://review.openstack.org/587171
17:25:18 melwitt mnaser: so the deleted column for that instance != 0? I don't get how it's showing up in 'nova list' then
17:31:33 mnaser melwitt: also btw i did some work here on trying to make it a single transaction (with a parent fix trying to be a less major change) -- https://review.openstack.org/#/c/586742/ if you have time to look at it
17:31:56 melwitt mnaser: will take a look, thanks
17:39:41 openstackgerrit Merged openstack/nova stable/pike: libvirt: add method to configure migration speed https://review.openstack.org/586965
17:54:42 openstackgerrit Merged openstack/nova stable/pike: Make host_aggregate_map dictionary case-insensitive https://review.openstack.org/585335
17:54:50 openstackgerrit Merged openstack/nova stable/pike: Fix unbound local when saving an unchanged RequestSpec https://review.openstack.org/580951
17:54:56 openstackgerrit Merged openstack/nova stable/queens: Refactor _build_device_metadata https://review.openstack.org/581398
17:55:05 openstackgerrit Merged openstack/nova stable/queens: Consider hostdev devices when building metadata https://review.openstack.org/581399
17:55:13 openstackgerrit Merged openstack/nova master: Add additional functional tests for NUMA networks https://review.openstack.org/585385
17:59:20 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Use of set_middleware_defaults https://review.openstack.org/587183
18:00:32 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Use own set_middleware_defaults https://review.openstack.org/587183
18:32:33 efried Hey folks, we're back with https://bugs.launchpad.net/nova/+bug/1737711
18:32:33 openstack Launchpad bug 1737711 in OpenStack Compute (nova) "nova boot failed when use the chinese metadata key and value" [Undecided,In progress] - Assigned to huanhongda (hongda)
18:32:59 efried They've restricted the patch https://review.openstack.org/#/c/536236/ to just metadata keys (not extra_specs keys anymore).
18:34:20 efried The bug still doesn't give a whole lot of detail as to the use case for this, but maybe it makes more sense to y'all, dansmith mriedem jaypipes
18:38:21 melwitt efried: fwiw, my initial read on that is we have more than one place where unicode keys won't work (as you mentioned extra_specs is another one) and IMHO it feels like it would be a small spec to propose to change all of them together in a consistent way to support unicode
18:38:39 melwitt is the schema validation really the only thing preventing unicode from working?
18:39:14 efried melwitt: Good question. No idea if they've tried... whatever it is they wanted to try, with their fix in place.
18:39:59 efried melwitt: Anyway, my position on this is that I don't have the background to be able to make a compelling argument to shut down the bug, or ask for it to be converted to a bp, or whatever. I need cavalry support (that would be you).
18:40:41 jaypipes efried: I'm curious why the patch includes a change to add maxLength: 255
18:41:40 efried jaypipes: Yeah, that's redundant (already covered by the regex), but not really important to the question of whether we should do it at all...
18:42:25 efried jaypipes: I added a comment for ya :)
18:42:36 jaypipes efried: right, but AFAICT, the regex doesn't change.
18:42:49 mriedem stephenfin: where does the libvirt driver actually translate hw_cpu_policy and hw_cpu_thread_policy into something that goes in the guest xml?
18:42:58 efried jaypipes: Right, that part of the delta is them making sure the extra_specs pattern does *not* change.
18:43:16 mriedem stephenfin: _get_guest_cpu_config ?
18:44:11 efried jaypipes: Actually, they've correctly carried over the redundant maxLength from the metadata key pattern...
18:44:13 openstackgerrit karim proposed openstack/nova master: Updated AggregateImagePropertiesIsolation filter illustration https://review.openstack.org/586317
18:44:45 efried jaypipes: I suppose if you were being really particular, you could say they did the right thing, because you could get back a different schema error.
18:46:49 efried jaypipes: But it's not about whether the patch is correct (it clearly isn't quite, yet). It's about whether we should allow it at all. Or if we do, if it needs to have microversion paperwork, a bp, a spec, etc.
18:47:12 melwitt efried: understood. at the very least I'd think we'd want tempest tests to cover unicode metadata key cases to verify it for now and keep it working in the future. and I'd think we'd want to change all of the user-settable keys to support unicode at the same time (rather than pick only one). and that's a bigger change and one I'd expect to have a spec about that explains all of the changes needed to properly support unicode in keys
18:47:28 dansmith and tests to cover metadata server and configdrive, amirite?
18:47:35 melwitt that's just my opinion. any other opinions?
18:47:53 dansmith personally I do not think unicode is worth the change to this, no surprises there
18:48:06 efried you anglocentric, you
18:49:03 mriedem i wouldn't ask me, i'd ask alex_xu, ken'ichi, takashi and Kevin_Zheng
18:49:20 mriedem i asked kevin long ago if huawei had any need for unicode in flavor extra specs and i believe he said no
18:49:38 mriedem passing unicode metadata values through to the guest is probably an easier case to make
18:49:49 melwitt I didn't mean opinions on the validity of the change, just whether it would need a spec and review
18:49:50 dansmith resources:<pileofpoo>=1
18:50:57 mriedem normal microversion behavior state diagram says that if i can make a request to cloud A and it works and make the same request to cloud B and it doesn't work, w/o microversions, it should probably be versioned - unless it's just a bug / something broken in existing support
18:51:15 mriedem but this has never worked as far as i know
18:51:22 mriedem so it's not a bug/regression, it's an RFE
18:51:25 dansmith what do you do for keys that are unicode if retrieved from an older microversion?
18:51:30 dansmith detect if they have unicode and exclude them?
18:51:56 mriedem i thought you couldn't do that b/c of the existing schema?
18:52:06 mriedem oh you mean,
18:52:15 efried hypothetically if this change was made, even with a microversion
18:52:16 mriedem set with unicode on newer microversion, retrieve with older microversion?

Earlier   Later