| gmann | dansmith: I do not think we have any such mechanism. testscenarios and ddt are some way but as you mentioned those are static generated tests instead of during test runtime | 01:42 |
|---|---|---|
| gmann | dansmith: what we have done in such cases is defined the logic in base method _test_* and defined all cases tests explicitly calling the base method | 01:43 |
| gmann | dansmith: but I think in your case, those different cases depends on config option value so ^^ will not work as well | 01:45 |
| gmann | dansmith: one way is to test all those different format in single test case if that does not become very long running test | 01:46 |
| opendevreview | Merged openstack/devstack stable/2024.1: Replacing usage of rdo-release rpm https://review.opendev.org/c/openstack/devstack/+/925400 | 03:38 |
| opendevreview | Jens Harbott proposed openstack/devstack stable/2023.1: Drop Fedora support https://review.opendev.org/c/openstack/devstack/+/925837 | 06:21 |
| frickler | gmann: regarding rbac, devstack is still using the devstack-system-admin cloud in a lot of places, shouldn't this be obsolete now and reverted back to using the normal admin account? or maybe I'm misunderstanding the whole situation | 06:35 |
| frickler | weird unrelated question: why does the "cinder" service have type "block-storage", but "cinderv3" has "volumev3"? | 06:38 |
| *** tkajinam is now known as Guest2415 | 12:52 | |
| *** tkajinam is now known as Guest2416 | 13:00 | |
| dansmith | gmann: I figured out a way to do it.. my most recent set splits them all out during load time (and works very nice) | 13:37 |
| dansmith | so we get a bunch of short-running tests and proper failures for a given scenario/image combination | 13:37 |
| *** whoami-rajat_ is now known as whoami-rajat | 13:56 | |
| opendevreview | Stephen Finucane proposed openstack/hacking master: Bump flake8 version https://review.opendev.org/c/openstack/hacking/+/925380 | 14:10 |
| opendevreview | Stephen Finucane proposed openstack/hacking master: Remove unnecessary setup.cfg config https://review.opendev.org/c/openstack/hacking/+/897008 | 15:15 |
| opendevreview | Stephen Finucane proposed openstack/hacking master: Remove unnecessary setup.cfg config https://review.opendev.org/c/openstack/hacking/+/897008 | 15:15 |
| dansmith | gmann: I'm failing to add testscenarios to the requirements here https://review.opendev.org/c/openstack/tempest/+/925431/5/requirements.txt | 15:53 |
| dansmith | I'm adding it in the same way as it appears in other projects | 15:53 |
| dansmith | it's not really required if there's some reason I can't use that, it's just a shortcut to doing what I want | 15:53 |
| dansmith | I guess I'll try >= there and see if that makes it happy | 15:53 |
| opendevreview | Dan Smith proposed openstack/tempest master: Test image formats https://review.opendev.org/c/openstack/tempest/+/925431 | 15:54 |
| opendevreview | Dan Smith proposed openstack/tempest master: Test image formats with import https://review.opendev.org/c/openstack/tempest/+/925490 | 15:54 |
| opendevreview | Dan Smith proposed openstack/tempest master: Add compute image format test https://review.opendev.org/c/openstack/tempest/+/925617 | 15:54 |
| dansmith | gmann: otherwise, I'd sure appreciate review of this ^ the oslo port is almost completely landed and they're going to do a special release for me, so I'd like to have these in-tree before I go flipping nova, glance and cinder to using the oslo version so we have before/after coverage | 15:55 |
| gmann | frickler: I think we still need devstack-system-admin for ironic (keystone if there are tests) as they both need system admin | 17:10 |
| gmann | frickler: services alias mismatch ("cinder", "block-storage", "cinderv3" "volumev3") is there for historic reason and to avoid breaking things we kept those same way. I think 'block-storage' matches with other service pattern. Nova did v2 API to compute_legacy and v2.1 as 'compute' | 17:15 |
| gmann | maybe same way we should have done. | 17:15 |
| gmann | dansmith: ok, i will check those review. not sure why testscenarios did not work in requirements let me have a look | 17:42 |
| dansmith | gmann: changing to >= seemed to work, I dunno why | 17:43 |
| gmann | dansmith: yeah, just saw PS5->PS6. I am not sure why == did not work as it match with u-c | 17:43 |
| dansmith | right | 17:43 |
| opendevreview | Brian Haley proposed openstack/devstack master: Move the check of "rpc_workers" after the post-config phase https://review.opendev.org/c/openstack/devstack/+/925412 | 19:36 |
| opendevreview | Merged openstack/hacking master: Bump flake8 version https://review.opendev.org/c/openstack/hacking/+/925380 | 22:24 |
| opendevreview | Merged openstack/hacking master: Remove unnecessary setup.cfg config https://review.opendev.org/c/openstack/hacking/+/897008 | 22:29 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!