Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-12-07
15:02:23 gibi we need test for tempest tests? whaaat :D
15:02:29 gibi looking ...
15:02:46 lyarwood gibi: but who is going to test the test that tests the test?!!
15:02:48 lyarwood gibi: /s
15:02:58 lyarwood gibi: yeah just basic unit tests for the waiter
15:03:10 lyarwood gibi: I was asked to do it before so I'm assuming someone will ask for them here
15:03:10 gibi yeah, I need self testing tests in a recursive way... :D
15:03:24 gibi lyarwood: sure, thanks. I will fix the patch up
15:10:37 openstackgerrit Lee Yarwood proposed openstack/nova-specs master: Image and flavor defined ephemeral storage encryption https://review.opendev.org/c/openstack/nova-specs/+/752284
15:36:53 teoobo_ gibi: I am interested in helping out!
15:38:55 kashyap lyarwood: A stupid question - upstream CI runs instance deletion (functional) tests, right? Got a pointer to a latest run for it?
15:39:06 kashyap I think in any CI run I should be able to find it
15:40:38 gibi teoobo_: so there are those nice helpers for the functional test but they are not always used in the tests, so the work would be to find where direct api calls are used for those server actions and replace it with the call to the helper
15:40:44 lyarwood kashyap: remember that we change the real meaning of functional tests in Nova
15:41:03 lyarwood kashyap: do you mean actual instance deletion including the removal of a libvirt domain etc?
15:41:19 lyarwood kashyap: if so that's actually under integration testing with tempest and yeah there should be loads of examples
15:41:35 lyarwood kashyap: each time we create and instance we should eventually clean it up
15:41:40 lyarwood create an*
15:41:48 kashyap lyarwood: Yeah, indeed - actual instance deletion including the removal
15:41:52 kashyap ... of libvirt domain
15:41:57 kashyap lyarwood: The reason why I ask is,
15:42:04 lyarwood kashyap: kk, should be loads of examples in any tempest run
15:42:21 kashyap I was debugging an NVDIMM issue with OSP QE; and Nova (upstream) code currently assumes you have 'daxio' tool installed
15:42:38 kashyap But ... some enterprise distros don't have it installed in their containers :-)
15:42:43 kashyap And it's causing a mess
15:42:48 kashyap lyarwood: Okay; thanks!
15:43:16 sean-k-mooney https://specs.openstack.org/openstack/nova-specs/specs/train/implemented/virtual-persistent-memory.html#virtual-persistent-memory-disposal
15:43:22 sean-k-mooney its called out in the sepc that its requried
15:43:32 sean-k-mooney The nova compute libvirt driver uses daxio utility (wrapped by privsep library functions) for this purpose.
15:44:05 sean-k-mooney so just like ndctl it was missed in the deployment update in ooo/kolla
15:44:09 kashyap I'm aware; that's what I pointed out earlier
15:44:32 kashyap Yeah, they need to re-update the containers (Kolla, I guess) like they had to do for 'virt-admin' package
15:44:50 lyarwood yup
15:44:57 lyarwood it's an optional dep in Nova itself
15:45:07 lyarwood so we don't enforce it to be installed in code
15:46:20 sean-k-mooney https://github.com/openstack/kolla/commit/f9ea23d940ec4ce35d503ac62f202daaa9db502d only added ndctl
15:46:44 sean-k-mooney it need to be in nova_compute not nova_libvirt by the way
15:47:41 kashyap They also need ndctl-libs; without which 'daxio' doesn't work
15:47:58 sean-k-mooney ndctl should pull that in
15:48:03 kashyap (Yeah)
15:48:23 openstackgerrit Balazs Gibizer proposed openstack/nova stable/ussuri: Use cell targeted context to query BDMs for metadata https://review.opendev.org/c/openstack/nova/+/765748
15:48:28 kashyap I'm hoping; but better be explicit than implicit
15:48:47 kashyap sean-k-mooney: Also what's the rationale for installing it in 'nova_compute' vs. 'nova_libvirt' container?
15:48:56 kashyap Just a bit ago I thought it should be in the latter
15:48:58 kashyap Container hell for the win
15:49:37 sean-k-mooney nova_libvirt just runs libvirtd
15:49:45 sean-k-mooney the compute agent that uses it is in nova_compute
15:51:00 sean-k-mooney https://rpmfind.net/linux/RPM/centos/8.3.2011/baseos/x86_64/Packages/ndctl-67-2.el8.x86_64.html
15:51:11 sean-k-mooney so it required ndctl-libs and daxctl-libs
15:51:27 sean-k-mooney but not the daxio cli im guessiing
15:51:51 kashyap (I know the container's purpose of course; but which container requires which deps to be installed isn't always apparent.)
15:51:58 sean-k-mooney ya its a seperate package https://centos.pkgs.org/8/centos-appstream-x86_64/daxio-1.6.1-1.el8.x86_64.rpm.html
15:52:02 kashyap It is
15:52:32 sean-k-mooney daxio does not depend on ndctl-libs by the way so that looks like a bug unless its an optional dep
15:52:44 kashyap It won't; daxio needs to be separately installed.
15:52:57 kashyap Either way; I'll file a tracker for TripleO to get this going
15:54:03 sean-k-mooney well we coudl reopen the bz. from an upstream ppoint of view this need 2 bugs or 1 but against 2 compnetnts
15:54:13 sean-k-mooney we need to update kolla and ooo
15:54:26 sean-k-mooney sicne we need the dep to be added to both image build systems
15:54:34 sean-k-mooney e.g. kolla and the new ooo one
15:54:58 teoobo_ gibi: do i have to open a bug on launchpad before working on it?
15:54:59 sean-k-mooney kolla for 16.1 and ooo for 16.2
15:58:29 gibi teoobo_: no need for a bug, just add me and stephenfin to the patch as reviewer
15:59:20 gibi teoobo_: I suggest to propose one patch per helper if possible, and add the same gerrit topic to each patch
16:00:02 teoobo_ gibi: thanks again for the help
16:00:15 gibi teoobo_: I'm happy to help :)
16:00:27 gibi teoobo_: thanks for picking up some of these work
16:00:47 gibi stephenfin: you were +2 on https://review.opendev.org/c/openstack/nova-specs/+/752284 now I'm +2 too, so you can send it through if you wish
16:01:48 stephenfin sure, will take a look shortly :)
16:08:03 openstackgerrit Lee Yarwood proposed openstack/nova master: lower-constraints: Bump packaging to 20.4 https://review.opendev.org/c/openstack/nova/+/765824
16:08:06 lyarwood ^ should fix openstack-tox-lower-constraints
16:10:27 openstackgerrit Lee Yarwood proposed openstack/nova master: lower-constraints: Bump packaging to 20.4 https://review.opendev.org/c/openstack/nova/+/765824
16:14:27 gibi lyarwood: +2
17:46:10 lyarwood stephenfin / melwitt / bauzas ; https://review.opendev.org/c/openstack/nova/+/765824 - This should unblock the lc job if anyone has time this evening.
17:54:47 melwitt lyarwood: done, thanks for fixing that!
17:55:13 stephenfin lyarwood: done
17:55:21 stephenfin oh, Mel beat me to it
17:56:52 lyarwood np thanks both :)
20:11:24 openstackgerrit Merged openstack/nova master: lower-constraints: Bump packaging to 20.4 https://review.opendev.org/c/openstack/nova/+/765824
20:14:01 gmann stephenfin: you are not covering the detail-False cases in this as it pass True for detail param - https://review.opendev.org/c/openstack/nova/+/764039/1/nova/tests/unit/api/openstack/compute/test_hypervisors.py#285
20:22:40 openstackgerrit Lee Yarwood proposed openstack/nova stable/victoria: lower-constraints: Bump packaging to 20.4 https://review.opendev.org/c/openstack/nova/+/765749
20:26:35 openstackgerrit Lee Yarwood proposed openstack/nova master: nova-evacuate: Remove leftover playbook from standalone job https://review.opendev.org/c/openstack/nova/+/764618
20:29:52 spatel johnsom: Hey!
20:30:04 johnsom Hello
20:30:07 spatel My octivia lb working great now :)
20:30:16 johnsom Great to hear!
20:30:40 spatel I am going to do some benchmarking to certify performance.
20:30:45 openstackgerrit Rafael Weingartner proposed openstack/nova master: Enhance simple usage API documentation https://review.opendev.org/c/openstack/nova/+/765871
20:30:59 johnsom spatel This is the nova channel, we should talk about Octavia in #opendstack-lbaas
20:31:10 spatel damn!! sorry
20:31:40 johnsom NP, just being courteous to our nova friends.
20:34:46 openstackgerrit Rafael Weingartner proposed openstack/nova master: Nova's `os-simple-tenant-usage` has the following statement. https://review.opendev.org/c/openstack/nova/+/765871
20:36:22 openstackgerrit Rafael Weingartner proposed openstack/nova master: Enhance simple usage API documentation https://review.opendev.org/c/openstack/nova/+/765871
20:39:14 openstackgerrit Rafael Weingartner proposed openstack/nova master: Enhance simple usage API documentation https://review.opendev.org/c/openstack/nova/+/765871
21:44:16 openstackgerrit Lee Yarwood proposed openstack/nova stable/victoria: libvirt: Skip encryption metadata lookups if secret already exists on host https://review.opendev.org/c/openstack/nova/+/765769
22:07:29 openstackgerrit Lee Yarwood proposed openstack/nova master: block_device: Use initialize APIs to refresh when reported as idempotent https://review.opendev.org/c/openstack/nova/+/720769
23:15:53 openstackgerrit melanie witt proposed openstack/nova master: Disallow CONF.compute.max_disk_devices_to_attach = 0 https://review.opendev.org/c/openstack/nova/+/755360
#openstack-nova - 2020-12-08
00:50:11 brinzhang_ gmann: hi, why the legency_admin's context contain tenant and project_id at the same time?
00:53:44 brinzhang_ gmann: this is the test report, if I change the compute_api:detail:get_all_tenants to compute_api:detail:get_all_projects, the legacy_admin context cannot authorize, and it always return the {'project_id':'xxxx'}
00:53:47 brinzhang_ Testing unauthorized context: <Context {'user': 'legacy_admin', 'tenant': '0269077f-e8ec-4b72-8890-d6fe35e47913', 'system_scope': None, 'project': '0269077f-e8ec-4b72-8890-d6fe35e47913', 'domain': None, 'user_domain': None, 'project_domain': None, 'is_admin': True, 'read_only': False, 'show_deleted': False, 'auth_token': None, 'request_id': 'req-f3225348-8288-45a8-810b-38d1464e1c0d', 'global_request_id': None, 'resource_uuid': None, 'roles':

Earlier   Later