Friday, 2023-04-28

opendevreviewGoutham Pacha Ravi proposed openstack/devstack-plugin-ceph master: [WIP] Disable msgrv2 compatibility  https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/88176700:09
opendevreviewGoutham Pacha Ravi proposed openstack/devstack-plugin-ceph master: [WIP] Disable msgrv2 compatibility  https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/88176700:10
opendevreviewGoutham Pacha Ravi proposed openstack/devstack-plugin-ceph master: [WIP] Disable msgrv2 compatibility  https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/88176700:28
dansmithgmann: I think get_server_ip() maybe doesn't work if network is already set up? seems like the sshable stuff makes that fail because it can't add another floating IP or something00:33
dansmithI'm exhausted for today, so I'll pick it up tomorrow00:33
opendevreviewGoutham Pacha Ravi proposed openstack/devstack-plugin-ceph master: [WIP] Disable msgrv2 compatibility  https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/88176703:11
*** gthiemon1e is now known as gthiemonge07:12
opendevreviewLuigi Dino Tamagnone proposed openstack/tempest master: Avoiding the clean of not tempest created resources  https://review.opendev.org/c/openstack/tempest/+/87992313:24
opendevreviewArx Cruz proposed openstack/tempest master: Update v3_endpoint_type to use public  https://review.opendev.org/c/openstack/tempest/+/88182513:36
opendevreviewArx Cruz proposed openstack/tempest master: Update v3_endpoint_type to use public  https://review.opendev.org/c/openstack/tempest/+/88182513:38
opendevreviewArx Cruz proposed openstack/tempest master: Update v3_endpoint_type to use public  https://review.opendev.org/c/openstack/tempest/+/88182513:44
opendevreviewArx Cruz proposed openstack/tempest master: Update v3_endpoint_type to use public  https://review.opendev.org/c/openstack/tempest/+/88182514:00
opendevreviewLukas Piwowarski proposed openstack/tempest master: [WIP] Catching broken tests in tempest-full-test-account-* jobs  https://review.opendev.org/c/openstack/tempest/+/88063015:21
opendevreviewDan Smith proposed openstack/tempest master: Allow wait_until=SSHABLE in scenario manager  https://review.opendev.org/c/openstack/tempest/+/88167516:19
gmanndansmith: hi17:14
dansmithgmann: mornin'17:14
gmannget_server_ip should get the fip from server but is validation_resource one is diff?17:14
dansmithgmann: get_server_ip() just blindly creates a new FIP, but then fails because we've already gotten one from the validation_resources17:15
dansmithgmann: I added a "get or create" logic to get_server_ip() which will try to find the existing one, if present before it creates17:15
gmanndansmith: right, scenario manager one seems not matching with validation one, I think we can use this https://github.com/openstack/tempest/blob/e64f4c2b07e048ea77d08a29b7afbd17448abc32/tempest/common/compute.py#L60 17:16
dansmithgmann: only if we pass validation_resources, which we don't get back from create_server() unless there's somewhere it's stored on self?17:18
dansmithgmann: so you don't want this? https://review.opendev.org/c/openstack/tempest/+/881675/10/tempest/scenario/manager.py#118917:18
gmanndansmith: this should also work. and validation resource FIP also get associated to server https://github.com/openstack/tempest/blob/master/tempest/common/compute.py#L8717:21
gmannin case ssh one using from validation_resources17:21
dansmithgmann: the thing I added to that patch does work for me locally17:21
gmannok17:21
dansmithand for the name of get_server_ip(), I expect it to get, not necessarily create, but this makes it "however you have to do it, get me an IP"17:21
gmannwe do that from API tests but scneario tests were written differently and consuming validation_resources need these type of modifucation17:22
dansmithokay17:23
gmannwe need to get it on same way as API test doing ssh with common ssh resources 17:23
gmanni mean for later free time :)17:24
dansmithokay so I'm not sure if you're asking me to change what I have.. the compute.get_server_ip() only returns from validation_resources, it doesn't create if missing, so there's more needed than just using that17:24
gmannno, all ok in your change 17:24
dansmithokay17:24
dansmiththis: https://review.opendev.org/c/openstack/tempest/+/881675/10/tempest/scenario/test_server_basic_ops.py#13717:25
dansmithused to fail because it would break the validation right below it, but with my change it now succeeds17:25
dansmithso that's not necessary to wait for sshable there, but it was a good test to make sure the get-or-create worked17:26
gmannk17:27
dansmithwoohoo, got a clean run on the cinder test plugin18:11
*** JayF is now known as Guest1244418:27
*** JasonF is now known as JayF18:27
opendevreviewDan Smith proposed openstack/tempest master: Allow wait_until=SSHABLE in scenario manager  https://review.opendev.org/c/openstack/tempest/+/88167519:55
opendevreviewDan Smith proposed openstack/tempest master: Wait until SSHABLE in volumes_extend test  https://review.opendev.org/c/openstack/tempest/+/88188819:55
opendevreviewMerged openstack/tempest master: Update "test_list_migrations_in_flavor_resize_situation" to indicate the failing request on timeout.  https://review.opendev.org/c/openstack/tempest/+/87695320:42
dansmithgmann: okay, this fixes another volumes test that does not wait for sshable, at the bottom of my current stack: https://review.opendev.org/c/openstack/tempest/+/881888/121:14
dansmithit just passed in the nova job I'm watching, so should be good to merge21:14
dansmiththat probably bites us in normal runs too, one of many volume detach fails21:14

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!