| *** yamamoto has quit IRC | 00:00 | |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Add 'servers_links' attribute in server list schema https://review.openstack.org/159037 | 00:02 |
|---|---|---|
| openstackgerrit | xu-haiwei proposed openstack/tempest: Add 'fault' attributes in server schema https://review.openstack.org/159360 | 00:02 |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Rename compute response schema dir to v2_1 https://review.openstack.org/162511 | 00:02 |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Block additionalProperties on Nova API tests https://review.openstack.org/156130 | 00:02 |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Adding common header attributes in image header schema https://review.openstack.org/159008 | 00:02 |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Merge availability_zone response schema into one file https://review.openstack.org/162816 | 00:02 |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Add 'images_links' attribute in image list schema https://review.openstack.org/159044 | 00:02 |
| *** achanda has quit IRC | 00:03 | |
| openstackgerrit | xu-haiwei proposed openstack/tempest: Merge hosts response schema into one file https://review.openstack.org/162821 | 00:04 |
| *** dimtruck is now known as zz_dimtruck | 00:09 | |
| *** Noice has quit IRC | 00:10 | |
| *** dmorita has joined #openstack-qa | 00:15 | |
| *** Longgeek has joined #openstack-qa | 00:18 | |
| *** woodster_ has quit IRC | 00:20 | |
| *** woodster_ has joined #openstack-qa | 00:30 | |
| *** sputnik13 has quit IRC | 00:31 | |
| *** dims has joined #openstack-qa | 00:32 | |
| *** Longgeek has quit IRC | 00:32 | |
| *** yamamoto has joined #openstack-qa | 00:41 | |
| *** pal has joined #openstack-qa | 00:42 | |
| *** marun has quit IRC | 00:42 | |
| *** jaypipes has quit IRC | 00:45 | |
| openstackgerrit | Ian Wienand proposed openstack-dev/devstack: Add defaults for yum proxy variables https://review.openstack.org/164040 | 00:51 |
| *** topshare has joined #openstack-qa | 00:52 | |
| *** mwagner_lap has quit IRC | 00:53 | |
| *** liusheng has quit IRC | 00:56 | |
| openstackgerrit | Morgan Fainberg proposed openstack-dev/devstack: Add response time to keystone access log https://review.openstack.org/164042 | 00:56 |
| *** cody-somerville has quit IRC | 00:56 | |
| morganfainberg | jogo, ^ the other items i'm poking at will come down the line | 00:57 |
| *** kevinbenton has quit IRC | 00:59 | |
| *** zz_dimtruck is now known as dimtruck | 01:09 | |
| *** rwsu is now known as rwsu-afk | 01:13 | |
| *** annashen has quit IRC | 01:19 | |
| *** ibiris is now known as ibiris_away | 01:20 | |
| *** dkranz has joined #openstack-qa | 01:24 | |
| *** enikanorov__ has joined #openstack-qa | 01:24 | |
| *** enikanorov_ has quit IRC | 01:26 | |
| openstackgerrit | Merged openstack-dev/devstack: Shut down ironic services in the modern way https://review.openstack.org/163813 | 01:32 |
| openstackgerrit | Merged openstack/tempest: Add scenario 'test_preserve_preexisting_port' https://review.openstack.org/160924 | 01:33 |
| *** yamahata has quit IRC | 01:33 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/137817 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1372670 | 01:34 |
| openstack | Launchpad bug 1372670 in OpenStack Compute (nova) "libvirtError: operation failed: cannot read cputime for domain" [High,In progress] - Assigned to Eduardo Costa (ecosta) | 01:34 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/145619 failed gate-nova-python27 in the gate queue with an unrecognized error | 01:34 |
| *** annashen has joined #openstack-qa | 01:39 | |
| *** amotoki has joined #openstack-qa | 01:39 | |
| *** ChuckC has joined #openstack-qa | 01:42 | |
| *** amotoki has quit IRC | 01:44 | |
| openstackrecheck | openstack/heat change: https://review.openstack.org/163717 failed gate-heat-python27 in the gate queue with an unrecognized error | 01:46 |
| *** ChuckC has quit IRC | 01:46 | |
| *** ChuckC has joined #openstack-qa | 01:47 | |
| *** tsekiyama has joined #openstack-qa | 01:52 | |
| *** luqas has quit IRC | 01:55 | |
| *** kevinbenton has joined #openstack-qa | 01:57 | |
| openstackgerrit | Merged openstack-dev/devstack: install missing elasticsearch client https://review.openstack.org/163693 | 02:00 |
| *** patrickeast has quit IRC | 02:02 | |
| openstackgerrit | yuntongjin proposed openstack-dev/devstack: Copy etc/nova/policy.d to $NOVA_CONF_DIR https://review.openstack.org/160068 | 02:06 |
| *** dpaterson has quit IRC | 02:07 | |
| *** harlowja is now known as harlowja_away | 02:07 | |
| *** asselin has quit IRC | 02:09 | |
| *** asselin has joined #openstack-qa | 02:10 | |
| *** dustins has quit IRC | 02:11 | |
| *** dustins has joined #openstack-qa | 02:12 | |
| *** dims has quit IRC | 02:14 | |
| *** marun has joined #openstack-qa | 02:16 | |
| *** mlavalle has quit IRC | 02:17 | |
| *** Longgeek has joined #openstack-qa | 02:18 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/162850 failed gate-grenade-dsvm-partial-ncpu in the gate queue with an unrecognized error | 02:18 |
| *** marun has quit IRC | 02:22 | |
| openstackgerrit | Masayuki Igawa proposed openstack/tempest: Remove python-ironicclient from requirements https://review.openstack.org/164057 | 02:23 |
| *** tsekiyama has quit IRC | 02:26 | |
| *** tsekiyama has joined #openstack-qa | 02:27 | |
| *** dims has joined #openstack-qa | 02:34 | |
| *** Poornima has joined #openstack-qa | 02:34 | |
| *** Longgeek has quit IRC | 02:35 | |
| *** tsekiyama has quit IRC | 02:36 | |
| *** apevec has quit IRC | 02:38 | |
| *** pal has quit IRC | 02:39 | |
| *** dims has quit IRC | 02:39 | |
| *** mlavalle has joined #openstack-qa | 02:41 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Merge agent response schema into one file https://review.openstack.org/162516 | 02:41 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Rearrange certificates response schema into one file https://review.openstack.org/162519 | 02:41 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Rearrange hypervisors response schemas into one file https://review.openstack.org/162523 | 02:42 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Fix interfaces response schema https://review.openstack.org/162530 | 02:42 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Rearrange keypairs response schema into one file https://review.openstack.org/162551 | 02:42 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Arrange quotas response schema into one file https://review.openstack.org/162841 | 02:42 |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest: Merge server response schema into one file https://review.openstack.org/162875 | 02:42 |
| *** mpopow_ has quit IRC | 02:43 | |
| *** achanda has joined #openstack-qa | 02:44 | |
| openstackgerrit | Merged openstack/tempest: Skip boto tests when auth_version is v3 https://review.openstack.org/163792 | 02:53 |
| *** amotoki has joined #openstack-qa | 02:55 | |
| openstackgerrit | Armando Migliaccio proposed openstack/tempest: Add API tests for Neutron DVR extension https://review.openstack.org/98235 | 02:59 |
| *** amotoki has quit IRC | 03:00 | |
| openstackgerrit | Merged openstack-infra/devstack-gate: add oslo.versionedobjects https://review.openstack.org/162218 | 03:02 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/164045 failed gate-tempest-dsvm-neutron in the gate queue with an unrecognized error | 03:02 |
| *** blogan has quit IRC | 03:05 | |
| *** annashen has quit IRC | 03:05 | |
| *** annashen has joined #openstack-qa | 03:07 | |
| *** blogan has joined #openstack-qa | 03:12 | |
| openstackgerrit | Ian Wienand proposed openstack-dev/devstack: Remove packaged rpm python libraries https://review.openstack.org/161455 | 03:15 |
| *** annashen has quit IRC | 03:18 | |
| *** jaypipes has joined #openstack-qa | 03:18 | |
| *** ChanServ sets mode: +o jaypipes | 03:18 | |
| *** annashen has joined #openstack-qa | 03:19 | |
| *** blogan has quit IRC | 03:20 | |
| *** blogan has joined #openstack-qa | 03:24 | |
| *** dimtruck is now known as zz_dimtruck | 03:24 | |
| *** blogan has quit IRC | 03:27 | |
| *** annashen has quit IRC | 03:30 | |
| *** Longgeek has joined #openstack-qa | 03:31 | |
| *** Poornima has quit IRC | 03:34 | |
| *** blogan has joined #openstack-qa | 03:36 | |
| *** cody-somerville has joined #openstack-qa | 03:36 | |
| *** cody-somerville has quit IRC | 03:36 | |
| *** cody-somerville has joined #openstack-qa | 03:36 | |
| *** diegows has quit IRC | 03:38 | |
| *** annashen has joined #openstack-qa | 03:40 | |
| *** Longgeek has quit IRC | 03:41 | |
| *** annashen has quit IRC | 03:43 | |
| *** annashen has joined #openstack-qa | 03:44 | |
| *** annashen has quit IRC | 03:47 | |
| *** tsekiyama has joined #openstack-qa | 03:49 | |
| *** annashen has joined #openstack-qa | 03:49 | |
| *** Longgeek has joined #openstack-qa | 03:52 | |
| *** tsekiyama has quit IRC | 03:53 | |
| *** annashen has quit IRC | 03:59 | |
| *** rushiagr_away is now known as rushiagr | 03:59 | |
| *** mpopow_ has joined #openstack-qa | 04:02 | |
| *** mpopow__ has joined #openstack-qa | 04:03 | |
| *** annashen has joined #openstack-qa | 04:04 | |
| *** annashen has quit IRC | 04:06 | |
| *** liusheng has joined #openstack-qa | 04:06 | |
| *** mpopow_ has quit IRC | 04:07 | |
| *** amotoki has joined #openstack-qa | 04:11 | |
| *** annashen has joined #openstack-qa | 04:11 | |
| *** BharatK has joined #openstack-qa | 04:12 | |
| *** mpopow has joined #openstack-qa | 04:13 | |
| *** mpopow_ has joined #openstack-qa | 04:13 | |
| *** mpopow__ has quit IRC | 04:15 | |
| *** cody-somerville has quit IRC | 04:15 | |
| *** amotoki has quit IRC | 04:15 | |
| *** pal has joined #openstack-qa | 04:16 | |
| *** mpopow has quit IRC | 04:17 | |
| openstackrecheck | openstack/glance change: https://review.openstack.org/132717 failed gate-glance-python27, gate-tempest-dsvm-full, gate-tempest-dsvm-postgres-full, gate-tempest-dsvm-neutron-full, gate-grenade-dsvm, gate-tempest-dsvm-large-ops, gate-tempest-dsvm-neutron-large-ops, gate-devstack-dsvm-cells in the gate queue with an unrecognized error | 04:21 |
| *** annashen has quit IRC | 04:23 | |
| *** dboik_ has quit IRC | 04:34 | |
| openstackgerrit | Yuriy Taraday proposed openstack-dev/devstack: Add rootwrap daemon mode support for Neutron https://review.openstack.org/107386 | 04:35 |
| *** zz_dimtruck is now known as dimtruck | 04:36 | |
| *** chandankumar has joined #openstack-qa | 04:39 | |
| *** annashen has joined #openstack-qa | 04:39 | |
| *** woodster_ has quit IRC | 04:40 | |
| *** sputnik13 has joined #openstack-qa | 04:43 | |
| *** mtanino has quit IRC | 04:46 | |
| *** Longgeek has quit IRC | 04:47 | |
| openstackgerrit | Al Miller proposed openstack-dev/devstack: Remove LBaaS support from devstack https://review.openstack.org/161436 | 04:48 |
| *** annashen has quit IRC | 04:54 | |
| *** annashen has joined #openstack-qa | 04:56 | |
| *** mpopow_ has quit IRC | 05:00 | |
| *** sputnik13 has quit IRC | 05:11 | |
| *** coolsvap|afk is now known as coolsvap | 05:14 | |
| *** swamireddy is now known as swamireddy_sick_ | 05:16 | |
| openstackgerrit | Adam Gandelman proposed openstack/tempest: Skip network scenario tests for baremetal https://review.openstack.org/163934 | 05:20 |
| adam_g | mtreinish, gahh that needed to be updated with another that landed today ^ | 05:21 |
| *** Poornima has joined #openstack-qa | 05:24 | |
| *** Poornima has quit IRC | 05:24 | |
| *** vigneshvar has joined #openstack-qa | 05:24 | |
| *** Poornima has joined #openstack-qa | 05:25 | |
| *** dimtruck is now known as zz_dimtruck | 05:28 | |
| *** chandankumar has quit IRC | 05:36 | |
| *** annashen has quit IRC | 05:37 | |
| *** chandankumar has joined #openstack-qa | 05:40 | |
| *** anshul has joined #openstack-qa | 05:40 | |
| *** annashen has joined #openstack-qa | 05:40 | |
| *** annashen has quit IRC | 05:44 | |
| *** Longgeek has joined #openstack-qa | 05:48 | |
| *** chandankumar has quit IRC | 05:52 | |
| *** annashen has joined #openstack-qa | 05:53 | |
| *** ajo has joined #openstack-qa | 05:53 | |
| *** mlavalle has quit IRC | 06:00 | |
| *** yfried|afk is now known as yfried | 06:02 | |
| openstackgerrit | Abhishek G M proposed openstack/tempest: Test creation of server attached to created port https://review.openstack.org/162886 | 06:02 |
| *** mlavalle has joined #openstack-qa | 06:07 | |
| *** achanda has quit IRC | 06:10 | |
| *** annashen has quit IRC | 06:11 | |
| *** achanda has joined #openstack-qa | 06:13 | |
| *** Sukhdev has joined #openstack-qa | 06:18 | |
| *** vigneshvar has quit IRC | 06:19 | |
| *** Longgeek has quit IRC | 06:21 | |
| *** amotoki has joined #openstack-qa | 06:23 | |
| *** achanda has quit IRC | 06:28 | |
| *** ajo has quit IRC | 06:31 | |
| *** ajo has joined #openstack-qa | 06:32 | |
| *** achanda has joined #openstack-qa | 06:34 | |
| *** yfried is now known as yfried|afk | 06:35 | |
| *** SridharG1 has joined #openstack-qa | 06:35 | |
| *** achanda has quit IRC | 06:38 | |
| *** Sukhdev has quit IRC | 06:43 | |
| *** rushiagr is now known as rushiagr_away | 06:45 | |
| *** pal has quit IRC | 06:49 | |
| *** k4n0 has joined #openstack-qa | 06:50 | |
| *** chandankumar has joined #openstack-qa | 06:52 | |
| *** pal has joined #openstack-qa | 06:53 | |
| *** coolsvap is now known as coolsvap|afk | 06:54 | |
| *** yfried|afk is now known as yfried | 06:55 | |
| *** thingee has joined #openstack-qa | 07:02 | |
| *** ibiris_away is now known as ibiris | 07:04 | |
| *** afazekas has joined #openstack-qa | 07:04 | |
| *** yfried is now known as yfried|afk | 07:05 | |
| *** pkoniszewski has joined #openstack-qa | 07:07 | |
| *** bkopilov has quit IRC | 07:08 | |
| *** coolsvap has joined #openstack-qa | 07:09 | |
| *** Longgeek has joined #openstack-qa | 07:18 | |
| *** dims has joined #openstack-qa | 07:23 | |
| *** dims has quit IRC | 07:27 | |
| *** bkopilov has joined #openstack-qa | 07:28 | |
| openstackgerrit | afazekas proposed openstack-dev/devstack: f22 default to dnf instead of yum https://review.openstack.org/163741 | 07:29 |
| *** jamielennox has quit IRC | 07:31 | |
| *** bkopilov has quit IRC | 07:32 | |
| *** jamielennox has joined #openstack-qa | 07:34 | |
| *** yfried|afk is now known as yfried | 07:39 | |
| *** jamielennox has quit IRC | 07:39 | |
| *** Longgeek has quit IRC | 07:41 | |
| *** Longgeek has joined #openstack-qa | 07:41 | |
| *** topshare has quit IRC | 07:45 | |
| *** topshare has joined #openstack-qa | 07:46 | |
| *** ajo has quit IRC | 07:47 | |
| *** yfried is now known as yfried|afk | 07:49 | |
| *** thingee has quit IRC | 07:51 | |
| *** pal has quit IRC | 07:52 | |
| *** jamielennox|away has joined #openstack-qa | 07:55 | |
| *** pal has joined #openstack-qa | 07:55 | |
| *** jamielennox|away is now known as jamielennox | 07:55 | |
| openstackgerrit | Merged openstack/tempest-lib: Updated from global requirements https://review.openstack.org/157803 | 07:57 |
| *** coolsvap has quit IRC | 07:59 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Copy the default params dict to avoid race https://review.openstack.org/163895 | 07:59 |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Use the configured version of identity in stress https://review.openstack.org/161259 | 08:00 |
| *** markus_z has joined #openstack-qa | 08:00 | |
| *** topshare has quit IRC | 08:04 | |
| *** topshare has joined #openstack-qa | 08:06 | |
| *** safchain has joined #openstack-qa | 08:19 | |
| *** openstackgerrit has quit IRC | 08:21 | |
| *** openstackgerrit has joined #openstack-qa | 08:21 | |
| *** gfidente has joined #openstack-qa | 08:23 | |
| *** gfidente has quit IRC | 08:23 | |
| *** gfidente has joined #openstack-qa | 08:23 | |
| *** topshare has quit IRC | 08:25 | |
| openstackgerrit | Adam Gandelman proposed openstack-infra/elastic-recheck: Add query for 1431724 https://review.openstack.org/164106 | 08:26 |
| *** topshare has joined #openstack-qa | 08:26 | |
| *** pal has quit IRC | 08:28 | |
| *** ihrachyshka has joined #openstack-qa | 08:30 | |
| openstackgerrit | Adam Gandelman proposed openstack/tempest: Skip network scenario tests for bug 1431724 + baremetal https://review.openstack.org/163934 | 08:33 |
| openstack | bug 1431724 in tempest "test_preserve_preexisting_port failing against icehouse" [Undecided,New] https://launchpad.net/bugs/1431724 | 08:33 |
| adam_g | mriedem_away, mtreinish ^ | 08:33 |
| *** pal has joined #openstack-qa | 08:36 | |
| *** jlanoux has joined #openstack-qa | 08:38 | |
| *** yfried|afk is now known as yfried | 08:46 | |
| *** ajo has joined #openstack-qa | 08:47 | |
| *** ihrachyshka has quit IRC | 08:48 | |
| *** andreykurilin_ has joined #openstack-qa | 08:53 | |
| *** amotoki has quit IRC | 08:57 | |
| *** dkusidlo has joined #openstack-qa | 08:59 | |
| *** luqas has joined #openstack-qa | 09:03 | |
| *** yfried is now known as yfried|afk | 09:05 | |
| *** SridharG1 has left #openstack-qa | 09:09 | |
| *** e0ne has joined #openstack-qa | 09:11 | |
| *** e0ne has quit IRC | 09:16 | |
| *** jordanP has joined #openstack-qa | 09:16 | |
| *** Longgeek has quit IRC | 09:16 | |
| *** yfried|afk is now known as yfried | 09:18 | |
| openstackgerrit | Andrea Frittoli proposed openstack-infra/elastic-recheck: Add query for bug 1431724 https://review.openstack.org/164120 | 09:19 |
| openstack | bug 1431724 in tempest "test_preserve_preexisting_port failing against icehouse" [Undecided,New] https://launchpad.net/bugs/1431724 | 09:19 |
| *** mhorban has quit IRC | 09:20 | |
| *** amotoki has joined #openstack-qa | 09:20 | |
| *** e0ne has joined #openstack-qa | 09:20 | |
| andreaf | 43 hits in the last ~7h ^^^^ | 09:20 |
| *** bnemec has quit IRC | 09:24 | |
| *** bnemec has joined #openstack-qa | 09:25 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Revert "Add scenario 'test_preserve_preexisting_port'" https://review.openstack.org/164121 | 09:26 |
| andreaf | and the proposed fix ^^^ | 09:27 |
| *** yfried is now known as yfried|afk | 09:28 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 09:29 | |
| *** SridharG1 has joined #openstack-qa | 09:29 | |
| *** andreykurilin_ has quit IRC | 09:31 | |
| *** andreykurilin_ has joined #openstack-qa | 09:32 | |
| *** e0ne has quit IRC | 09:37 | |
| *** dims has joined #openstack-qa | 09:42 | |
| *** yfried|afk is now known as yfried | 09:46 | |
| *** dims has quit IRC | 09:47 | |
| *** vigneshvar has joined #openstack-qa | 09:47 | |
| *** bkopilov has joined #openstack-qa | 09:49 | |
| *** dims has joined #openstack-qa | 09:54 | |
| *** bkopilov has quit IRC | 09:55 | |
| *** SridharG has quit IRC | 09:58 | |
| *** SridharG has joined #openstack-qa | 10:00 | |
| *** luqas has quit IRC | 10:04 | |
| *** yfried is now known as yfried|afk | 10:06 | |
| *** ihrachyshka has joined #openstack-qa | 10:08 | |
| *** yamamoto has quit IRC | 10:11 | |
| *** eugene_bag has quit IRC | 10:14 | |
| *** tkammer has joined #openstack-qa | 10:17 | |
| *** Longgeek has joined #openstack-qa | 10:17 | |
| *** e0ne has joined #openstack-qa | 10:17 | |
| *** e0ne is now known as e0ne_ | 10:17 | |
| *** vigneshvar has quit IRC | 10:19 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Support v3 in credential providers and subclasses https://review.openstack.org/133029 | 10:20 |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Use the configured version of identity in stress https://review.openstack.org/161259 | 10:20 |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Copy the default params dict to avoid race https://review.openstack.org/163895 | 10:20 |
| *** e0ne_ is now known as e0ne | 10:22 | |
| andreaf | mtreinish: it turns out that the faiures in https://review.openstack.org/133029 on the auth_v3 job are related to https://review.openstack.org/163895 - DEFAULT_PARAMS contained an unrelated tenant_name which clashed with the provided project_name | 10:24 |
| *** dkusidlo_ has joined #openstack-qa | 10:24 | |
| *** dkusidlo has quit IRC | 10:28 | |
| *** dkusidlo_ is now known as dkusidlo | 10:28 | |
| *** e0ne_ has joined #openstack-qa | 10:28 | |
| *** e0ne has quit IRC | 10:29 | |
| *** yfried|afk is now known as yfried | 10:29 | |
| *** pc_m has joined #openstack-qa | 10:29 | |
| *** Longgeek has quit IRC | 10:32 | |
| *** luqas has joined #openstack-qa | 10:35 | |
| *** yamamoto has joined #openstack-qa | 10:37 | |
| *** andreykurilin_ has quit IRC | 10:39 | |
| *** luqas has quit IRC | 10:40 | |
| *** vigneshvar has joined #openstack-qa | 10:40 | |
| *** aix has joined #openstack-qa | 10:44 | |
| *** dkusidlo_ has joined #openstack-qa | 10:45 | |
| *** dkusidlo has quit IRC | 10:48 | |
| *** dkusidlo_ is now known as dkusidlo | 10:48 | |
| *** apevec has joined #openstack-qa | 10:49 | |
| *** apevec has quit IRC | 10:49 | |
| *** apevec has joined #openstack-qa | 10:49 | |
| *** topshare_ has joined #openstack-qa | 10:53 | |
| *** jamielennox has quit IRC | 10:55 | |
| *** topshare has quit IRC | 10:55 | |
| *** dmorita has quit IRC | 10:55 | |
| *** yfried is now known as yfried|afk | 10:58 | |
| *** tosky has joined #openstack-qa | 10:58 | |
| *** harlowja_away has quit IRC | 10:59 | |
| *** erlon has joined #openstack-qa | 11:01 | |
| *** jamielennox|away has joined #openstack-qa | 11:02 | |
| *** jamielennox|away is now known as jamielennox | 11:02 | |
| *** mpaolino has joined #openstack-qa | 11:06 | |
| *** aysyd has joined #openstack-qa | 11:13 | |
| *** jamielennox has quit IRC | 11:14 | |
| *** yfried|afk is now known as yfried | 11:18 | |
| *** aysyd has quit IRC | 11:20 | |
| *** jamielennox|away has joined #openstack-qa | 11:24 | |
| *** jamielennox|away is now known as jamielennox | 11:24 | |
| openstackgerrit | Joe Hakim Rahme proposed openstack/tempest: Deprecates 'format' in Javelin image description https://review.openstack.org/162963 | 11:24 |
| *** pal has quit IRC | 11:35 | |
| openstackgerrit | Marian Horban proposed openstack/tempest: API test for instance's migration added https://review.openstack.org/153005 | 11:35 |
| openstackgerrit | Merged openstack/tempest: Revert "Add scenario 'test_preserve_preexisting_port'" https://review.openstack.org/164121 | 11:39 |
| *** Poornima has quit IRC | 11:40 | |
| *** cdent has joined #openstack-qa | 11:44 | |
| *** topshare_ has quit IRC | 11:47 | |
| *** vigneshvar has quit IRC | 11:51 | |
| *** bkopilov has joined #openstack-qa | 11:53 | |
| *** eugene_bag has joined #openstack-qa | 11:56 | |
| *** diegows has joined #openstack-qa | 11:57 | |
| *** mwagner_lap has joined #openstack-qa | 11:57 | |
| *** mpaolino has quit IRC | 12:00 | |
| *** nicodemos has quit IRC | 12:01 | |
| kashyap | I notice -- $ openstack ec2 credentials list --user 33c44ff28ba847b683913758ac1bba27 with a minimal Nova config, is there a way I can avoid having EC2 configured? | 12:01 |
| kashyap | Probably explicitly disabling them would do I guess. | 12:02 |
| kashyap | afazekas, Probably you know something above the above? ^ | 12:02 |
| *** yfried is now known as yfried|afk | 12:02 | |
| afazekas | kashyap, NOVA_ENABLED_APIS=osapi_compute,metadata # just remove the ec2 | 12:04 |
| *** eugene_bag has quit IRC | 12:04 | |
| *** aysyd has joined #openstack-qa | 12:04 | |
| kashyap | afazekas, Ah, right, thank you. | 12:04 |
| andreaf | jamielennox: hi - are you around? | 12:08 |
| openstackgerrit | Andrea Frittoli proposed openstack-dev/devstack: Adding tempest_roles when auth_version is v3 https://review.openstack.org/164155 | 12:11 |
| *** apevec has quit IRC | 12:11 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Support v3 in credential providers and subclasses https://review.openstack.org/133029 | 12:12 |
| *** dkusidlo has quit IRC | 12:13 | |
| kashyap | afazekas, My IRC proxy died, not sure if you saw my previous message, let me repost: | 12:14 |
| kashyap | (died briefly, i.e.) | 12:14 |
| kashyap | <kashyap> Also, if you've seen this w/ current DevStack git? | 12:14 |
| kashyap | <kashyap> "2015-03-13 12:00:49.730 | Authorization Failed: type object 'HTTPHeaderDict' has no attribute 'from_httplib'" | 12:14 |
| kashyap | <kashyap> Keystone fails to get token. | 12:14 |
| *** Longgeek has joined #openstack-qa | 12:15 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Use the configured version of identity in stress https://review.openstack.org/161259 | 12:16 |
| *** bkopilov has quit IRC | 12:20 | |
| *** bknudson has joined #openstack-qa | 12:21 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Support v3 in credential providers and subclasses https://review.openstack.org/133029 | 12:21 |
| *** jhenner has quit IRC | 12:25 | |
| openstackgerrit | Filip HubÃk proposed openstack/tempest: Add keystone v2.0 and v3 api discovery checks https://review.openstack.org/162710 | 12:27 |
| *** tochi has quit IRC | 12:27 | |
| pc_m | Hi all! Seeing function test breakage and need some advice. Anyone able to advise? | 12:29 |
| *** Longgeek has quit IRC | 12:31 | |
| *** yamamoto has quit IRC | 12:37 | |
| *** chandankumar has quit IRC | 12:40 | |
| *** yamamoto has joined #openstack-qa | 12:41 | |
| *** k4n0 has quit IRC | 12:46 | |
| afazekas | kashyap: no | 12:50 |
| *** weshay has joined #openstack-qa | 12:52 | |
| kashyap | afazekas, Np, I'll re-provision my env | 12:53 |
| mriedem_away | adam_g: ihrachyshka: ah, that sucks | 12:55 |
| mriedem_away | funny we don't run an icehouse job against tempest changes on master https://review.openstack.org/#/c/160924/ | 12:55 |
| mriedem_away | looks like we do, so why didn't this fail? http://logs.openstack.org/24/160924/8/check/check-tempest-dsvm-full-icehouse/d33a0aa/ | 12:56 |
| mriedem_away | actually http://logs.openstack.org/24/160924/8/check/check-tempest-dsvm-neutron-icehouse/3d4b6d0/ | 12:56 |
| mriedem_away | ah, the test class was skipped on icehouse in the original change: http://logs.openstack.org/24/160924/8/check/check-tempest-dsvm-neutron-icehouse/3d4b6d0/console.html#_2015-03-11_22_13_06_088 | 12:57 |
| mriedem_away | so why would it fails stable/icehouse? this doesn't make sense. | 12:57 |
| ihrachyshka | mriedem_away, maybe gate job for tempest is not configured to support l3 extension | 12:58 |
| ihrachyshka | which is concerning... | 12:58 |
| *** mriedem_away is now known as mriedem | 12:58 | |
| ihrachyshka | since l3 is pretty base | 12:59 |
| mriedem | so there was one hit in the gate on stable/icehouse | 13:00 |
| mriedem | http://goo.gl/WiB32O | 13:00 |
| mriedem | the tests aren't skipped on icehouse: http://logs.openstack.org/45/164045/1/gate/gate-tempest-dsvm-neutron/0fccaed/console.html#_2015-03-13_02_43_19_464 | 13:01 |
| mriedem | so gate-tempest-dsvm-neutron isn't configured the same way that check-tempest-dsvm-neutron-icehouse is | 13:02 |
| mriedem | check-tempest-dsvm-neutron-icehouse doesn't have the router extension enabled, gate-tempest-dsvm-neutron does | 13:03 |
| *** mlavalle has quit IRC | 13:03 | |
| mriedem | why didn't this fail for juno i wonder? | 13:03 |
| mriedem | on master the tests are skipped on juno b/c the router extension isn't enabled http://logs.openstack.org/24/160924/8/check/check-tempest-dsvm-neutron-full-juno/3f2ecf0/console.html#_2015-03-11_23_23_53_312 | 13:04 |
| mriedem | so i guess i don't know how to handle this besides a config option in tempest, which would be useless after kilo. we could disable the router extension in the icehouse job so it actually acts like the job on master and stable/juno and skips those tests. | 13:05 |
| *** dpaterson has joined #openstack-qa | 13:06 | |
| *** mlavalle has joined #openstack-qa | 13:07 | |
| openstackgerrit | Zu Qiang proposed openstack-dev/devstack: Reset no_proxy when the node is installed behind a proxy https://review.openstack.org/94988 | 13:07 |
| *** eugene_bag has joined #openstack-qa | 13:10 | |
| ihrachyshka | mriedem, I would prefer actually testing router "extension" instead of disabling all tests for it | 13:14 |
| mriedem | ihrachyshka: yeah, me too, but i'd also prefer that we don't skip tests on master changes which is basically lying about what's going to be tested with master tempest on stable/icehouse | 13:14 |
| mriedem | we don't run those tests on stable/juno changes either | 13:15 |
| *** vigneshvar has joined #openstack-qa | 13:15 | |
| mriedem | http://logs.openstack.org/56/163456/2/check/check-tempest-dsvm-neutron-full/a41f094/console.html#_2015-03-11_20_25_11_856 | 13:15 |
| mriedem | that's a stable/juno nova patch | 13:15 |
| *** rfolco has joined #openstack-qa | 13:15 | |
| mriedem | that or we need these compat jobs that run against tempest changes to actually run the tests that we have on stable today | 13:16 |
| *** dpaterson has quit IRC | 13:18 | |
| *** dpaterson has joined #openstack-qa | 13:18 | |
| *** dboik has joined #openstack-qa | 13:21 | |
| mriedem | mtreinish: when you're around, i'd like to know why check-tempest-dsvm-neutron-full on stable/icehouse runs tests that aren't run in check-tempest-dsvm-neutron-full-icehouse on master | 13:23 |
| *** mpaolino has joined #openstack-qa | 13:24 | |
| mriedem | i might need a config option for this preexisting ports test | 13:24 |
| *** yfried|afk has quit IRC | 13:24 | |
| ihrachyshka | mriedem, keep me informed. but overall, I'm against reducing test coverage for stable branches. | 13:27 |
| *** yamamoto has quit IRC | 13:27 | |
| openstackgerrit | Zu Qiang proposed openstack-dev/devstack: Reset no_proxy when the node is installed behind a proxy https://review.openstack.org/94988 | 13:28 |
| *** mpaolino has quit IRC | 13:31 | |
| openstackgerrit | Andrea Frittoli proposed openstack-infra/elastic-recheck: Add query for bug 1431724 https://review.openstack.org/164120 | 13:32 |
| openstack | bug 1431724 in tempest "test_preserve_preexisting_port failing against icehouse" [High,Fix released] https://launchpad.net/bugs/1431724 - Assigned to Andrea Frittoli (andrea-frittoli) | 13:32 |
| andreaf | mriedem: I fixed the query ^^^ | 13:32 |
| mriedem | andreaf: +W | 13:33 |
| *** BharatK has quit IRC | 13:39 | |
| *** dboik has quit IRC | 13:50 | |
| *** dboik has joined #openstack-qa | 13:50 | |
| *** obutenko_ has joined #openstack-qa | 13:54 | |
| *** rbak has joined #openstack-qa | 13:59 | |
| *** dansmith is now known as superdan | 13:59 | |
| *** jecarey has joined #openstack-qa | 14:03 | |
| *** dims has quit IRC | 14:06 | |
| *** jaypipes is now known as leakypipes | 14:07 | |
| *** dims has joined #openstack-qa | 14:07 | |
| mtreinish | mriedem: they shouldn't be different, icehouse neutron wasn't capable of doing a full tempest run in parallel | 14:10 |
| mtreinish | so the set of tests should be very limited | 14:10 |
| mtreinish | just smoke and scenario (which should cover your new test) | 14:10 |
| mriedem | mtreinish: i looked for about a half hour to try and figure out where the api_extensions are filtered in the icehouse jobs but couldn't figure that out | 14:11 |
| *** yamamoto has joined #openstack-qa | 14:11 | |
| mtreinish | mriedem: oh is it that, the api extensions are hardcoded on icehouse and juno | 14:11 |
| mtreinish | in lib/tempest | 14:11 |
| mtreinish | there's a bug with a fix pending | 14:12 |
| *** pkoniszewski has quit IRC | 14:12 | |
| mtreinish | one sec I'll dig it up | 14:12 |
| mriedem | http://logs.openstack.org/24/160924/8/check/check-tempest-dsvm-neutron-icehouse/3d4b6d0/logs/tempest_conf.txt.gz | 14:12 |
| mriedem | [network-feature-enabled] api_extensions = agent, | 14:12 |
| mriedem | so that job only runs with the neutron agent extension, whatever that is | 14:12 |
| mtreinish | mriedem: https://review.openstack.org/#/c/160572/ | 14:12 |
| mriedem | http://logs.openstack.org/45/164045/1/gate/gate-tempest-dsvm-neutron/0fccaed/logs/tempest_conf.txt.gz | 14:12 |
| mriedem | has them all: api_extensions = agent, allowed-address-pairs, binding, dhcp_agent_scheduler, ext-gw-mode, external-net, extra_dhcp_opt, extraroute, fwaas, l3_agent_scheduler, lbaas, lbaas_agent_scheduler, metering, multi-provider, provider, quotas, router, security-group, service-type, vpnaas | 14:13 |
| mtreinish | mriedem: it was a time based thing | 14:13 |
| mtreinish | mriedem: the fix landed between your patch going through check and being approved | 14:13 |
| mriedem | yeah, so now that https://review.openstack.org/#/c/160572/ is in, | 14:13 |
| mriedem | i should be good to restore my new test? | 14:13 |
| mtreinish | mriedem: yes | 14:13 |
| mtreinish | and it'll should fail on icehouse | 14:14 |
| mriedem | i verified that stable/juno neutron jobs are skipping test_network_basic_ops.py also | 14:14 |
| mtreinish | mriedem: https://review.openstack.org/#/c/160576/ | 14:14 |
| *** dboik has quit IRC | 14:15 | |
| *** Longgeek has joined #openstack-qa | 14:15 | |
| *** Ananci has joined #openstack-qa | 14:16 | |
| *** amotoki has quit IRC | 14:16 | |
| *** dboik has joined #openstack-qa | 14:16 | |
| *** ChuckC_ has joined #openstack-qa | 14:16 | |
| *** ChuckC has quit IRC | 14:17 | |
| *** afazekas has quit IRC | 14:18 | |
| openstackgerrit | Matt Riedemann proposed openstack/tempest: Restore "Add scenario 'test_preserve_preexisting_port'" https://review.openstack.org/164190 | 14:18 |
| mriedem | there we go | 14:18 |
| *** vigneshvar has quit IRC | 14:18 | |
| openstackrecheck | openstack/horizon change: https://review.openstack.org/159541 failed because of: gate-grenade-dsvm: https://bugs.launchpad.net/bugs/1286818 | 14:18 |
| openstack | Launchpad bug 1286818 in OpenStack-Gate "Ubuntu package archive periodically inconsistent causing gate build failures" [Low,In progress] - Assigned to Jeremy Stanley (fungi) | 14:18 |
| *** SridharG has quit IRC | 14:20 | |
| *** dboik has quit IRC | 14:21 | |
| *** bkopilov has joined #openstack-qa | 14:21 | |
| *** dboik has joined #openstack-qa | 14:21 | |
| openstackrecheck | openstack/glance change: https://review.openstack.org/164048 failed because of: gate-tempest-dsvm-full: https://bugs.launchpad.net/bugs/1286818 | 14:21 |
| openstack | Launchpad bug 1286818 in OpenStack-Gate "Ubuntu package archive periodically inconsistent causing gate build failures" [Low,In progress] - Assigned to Jeremy Stanley (fungi) | 14:21 |
| *** dims is now known as dimsum__ | 14:21 | |
| *** ChuckC_ has quit IRC | 14:23 | |
| mtreinish | EmilienM: I was looking at https://review.openstack.org/#/c/163622/ again and I was wondering is there always going to be a floating ip on each network label? | 14:24 |
| EmilienM | mtreinish: oh... very good question | 14:24 |
| EmilienM | mtreinish: let me check in Nova | 14:25 |
| mtreinish | because if there is a network without a floating ip the check loop will fail | 14:25 |
| EmilienM | yeah | 14:25 |
| openstackgerrit | Joseph Lanoux proposed openstack/tempest: WIP - Add create_test_server function https://review.openstack.org/151601 | 14:26 |
| openstackgerrit | Joseph Lanoux proposed openstack/tempest: Define validation_resources function for ssh validation https://review.openstack.org/149229 | 14:26 |
| openstackgerrit | Joseph Lanoux proposed openstack/tempest: WIP - Add new multi ssh configuration options https://review.openstack.org/146535 | 14:26 |
| *** eugene_bag has quit IRC | 14:27 | |
| openstackgerrit | Zengfa Gao proposed openstack/tempest: :Add API tests for Neutron subnetpool https://review.openstack.org/159644 | 14:28 |
| EmilienM | mtreinish: looking at http://developer.openstack.org/api-ref-compute-v2.html | 14:29 |
| EmilienM | mtreinish: it seems you're right. | 14:29 |
| EmilienM | mtreinish: I'm checking now in my platform | 14:30 |
| *** aysyd has quit IRC | 14:32 | |
| *** aysyd has joined #openstack-qa | 14:36 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for bug 1431724 https://review.openstack.org/164120 | 14:37 |
| openstack | bug 1431724 in tempest "test_preserve_preexisting_port failing against icehouse" [High,Fix released] https://launchpad.net/bugs/1431724 - Assigned to Andrea Frittoli (andrea-frittoli) | 14:37 |
| *** sabeen has joined #openstack-qa | 14:37 | |
| *** cbader has joined #openstack-qa | 14:40 | |
| openstackgerrit | Matthew Treinish proposed openstack/tempest: Skip test_connectivity_between_vms_on_different_networks for baremetal https://review.openstack.org/163934 | 14:41 |
| mtreinish | adam_g: ^^^ I just went back to rev 2 since the existing port test was reverted | 14:41 |
| EmilienM | mtreinish: I confirm I don't see floating when no floating is associated. I'm updating my patch. | 14:41 |
| *** pal has joined #openstack-qa | 14:45 | |
| *** amotoki has joined #openstack-qa | 14:46 | |
| EmilienM | mtreinish: so in fact I think we consider if use_floatingip_for_ssh is at True, the servers have floating IP. Since we try to create one in Javelin/Create | 14:47 |
| EmilienM | mtreinish: am I missing something? | 14:47 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/164045 failed because of: gate-tempest-dsvm-neutron: https://bugs.launchpad.net/bugs/1286818 | 14:47 |
| openstack | Launchpad bug 1286818 in OpenStack-Gate "Ubuntu package archive periodically inconsistent causing gate build failures" [Low,In progress] - Assigned to Jeremy Stanley (fungi) | 14:47 |
| *** Longgeek has quit IRC | 14:49 | |
| mtreinish | dkranz: if you've got a sec a couple of quick doc patches: https://review.openstack.org/#/c/163578/ , https://review.openstack.org/163587 , and https://review.openstack.org/158017 | 14:54 |
| *** SridharG1 has quit IRC | 14:54 | |
| dkranz | mtreinish: sure, I was just starting to do some reviews | 14:55 |
| dkranz | mtreinish: BTW, the issue I fixed with not filling creds when called from is_admin_available was the actual source of the TypeError issues I was seeing | 14:55 |
| dkranz | mtreinish: So I don't think we need to add a catch for that | 14:56 |
| mtreinish | dkranz: cool, yeah I saw that patch go through | 14:56 |
| *** SridharG has joined #openstack-qa | 14:56 | |
| mtreinish | dkranz: it might still be worth adding unit tests for the is_admin_available() function though | 14:56 |
| dkranz | mtreinish: I started to do that but ran into some nasty problems | 14:56 |
| *** pal_ has joined #openstack-qa | 14:57 | |
| dkranz | mtreinish: I need some help from a mock expert | 14:57 |
| *** pal has quit IRC | 14:57 | |
| dkranz | mtreinish: I actually have a complete test that I thought should work but it fails in strange ways | 14:57 |
| mtreinish | dkranz: well I'm no expert w/ mock but I can take a look at it. Can you upload it as a wip | 14:58 |
| dkranz | mtreinish: That is what I was thinking | 14:58 |
| dkranz | mtreinish: One thing was that the behavior of the unit tests seemed to change depending on if there was an actual tempest.conf or not. Is that expected? I was not expecting it. | 14:59 |
| dkranz | mtreinish: I used the same mock for PrivateConfig as the other tests use. | 14:59 |
| mtreinish | dkranz: hmm, if you mock out config then it shouldn't ever use the tempest.conf | 15:00 |
| dkranz | mtreinish: that's what I thought. The other strange thing was that I got different behavior depending on whether I ran all the test cases or just one. Strange. ANyway I will pu;sh it up today | 15:01 |
| *** openstackrecheck has quit IRC | 15:04 | |
| *** mriedem has quit IRC | 15:04 | |
| *** openstackrecheck has joined #openstack-qa | 15:04 | |
| *** ChuckC_ has joined #openstack-qa | 15:07 | |
| *** erlon has quit IRC | 15:11 | |
| *** mtanino has joined #openstack-qa | 15:13 | |
| *** bknudson has quit IRC | 15:13 | |
| *** jgrimm has quit IRC | 15:13 | |
| *** jgrimm has joined #openstack-qa | 15:14 | |
| *** yfried|afk has joined #openstack-qa | 15:21 | |
| *** tsekiyama has joined #openstack-qa | 15:28 | |
| *** erlon has joined #openstack-qa | 15:29 | |
| *** sputnik13 has joined #openstack-qa | 15:30 | |
| *** zz_dimtruck is now known as dimtruck | 15:31 | |
| *** david-lyle_afk is now known as david-lyle | 15:33 | |
| *** mriedem has joined #openstack-qa | 15:34 | |
| *** ibiris is now known as ibiris_away | 15:35 | |
| *** DaSchab has joined #openstack-qa | 15:35 | |
| *** yfried|afk has quit IRC | 15:36 | |
| *** cody-somerville has joined #openstack-qa | 15:37 | |
| *** sputnik13 has quit IRC | 15:38 | |
| openstackgerrit | gordon chung proposed openstack-dev/devstack: switch to use ceilometermiddleware https://review.openstack.org/155473 | 15:41 |
| *** dboik_ has joined #openstack-qa | 15:43 | |
| *** dboik has quit IRC | 15:43 | |
| *** ibiris_away is now known as ibiris | 15:43 | |
| openstackgerrit | David Kranz proposed openstack/tempest: Add test for the various config cases of is_admin_available() https://review.openstack.org/164225 | 15:44 |
| *** mpopow has joined #openstack-qa | 15:44 | |
| *** bknudson has joined #openstack-qa | 15:45 | |
| *** ajo has quit IRC | 15:48 | |
| *** ChuckC_ has quit IRC | 15:49 | |
| *** DaSchab has quit IRC | 15:51 | |
| *** ChuckC has joined #openstack-qa | 15:53 | |
| *** marun has joined #openstack-qa | 15:54 | |
| *** ajo has joined #openstack-qa | 15:55 | |
| openstackgerrit | Merged openstack/tempest: Create test nodes for test_baremetal_nodes https://review.openstack.org/162773 | 15:55 |
| *** amotoki has quit IRC | 15:56 | |
| *** yamamoto has quit IRC | 15:57 | |
| sdague | mtreinish: what keeps merge conflicting this - https://review.openstack.org/#/c/158852/ | 15:57 |
| mtreinish | sdague: that was what we were just talking about in infra | 15:58 |
| mtreinish | there was a corrupt ceilo clone which confused zuul | 15:58 |
| sdague | ah | 15:58 |
| sdague | ok, thanks for looking into it | 15:59 |
| *** Sukhdev has joined #openstack-qa | 15:59 | |
| openstackgerrit | Merged openstack/tempest: Cleanup the tempest readme https://review.openstack.org/163587 | 15:59 |
| *** e0ne_ has quit IRC | 15:59 | |
| *** amotoki has joined #openstack-qa | 16:00 | |
| mtreinish | sdague: sure, np. I actually only noticed because the zuul graph looked cool with like 7 merge conflicts on all different projects in the gate | 16:03 |
| *** mlavalle has quit IRC | 16:03 | |
| sdague | heh | 16:04 |
| *** apevec has joined #openstack-qa | 16:04 | |
| openstackgerrit | Emilien Macchi proposed openstack/tempest: javelin: implement floating-ip support https://review.openstack.org/163622 | 16:06 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for detach_volume NoneType bug 1327218 https://review.openstack.org/164235 | 16:08 |
| openstack | bug 1327218 in OpenStack Compute (nova) "Volume detach failure because of invalid bdm.connection_info" [High,Confirmed] https://launchpad.net/bugs/1327218 | 16:08 |
| *** amotoki has quit IRC | 16:10 | |
| *** mpopow has quit IRC | 16:10 | |
| mtreinish | sdague: so for the subunit2sql thing, you were saying I should change that migration to just make the time columns ints and be done with it | 16:11 |
| *** Noice has joined #openstack-qa | 16:11 | |
| sdague | that's my feeling | 16:12 |
| sdague | I think a lot of things get simpler then | 16:12 |
| sdague | and with mysql you can get pretty formatted times back with the functions I pointed you at | 16:13 |
| mriedem | mtreinish: so we're reusing volumes in tempest runs right? | 16:13 |
| mtreinish | sdague: ok cool, that's what I'll do then | 16:13 |
| mtreinish | mriedem: hmm, maybe for some tests, but I thought most tests just create and delete | 16:14 |
| *** SridharG has left #openstack-qa | 16:14 | |
| mriedem | i was tracing a volume id through the n-cpu logs and seeing the same volume showing up in multiple tests | 16:14 |
| mriedem | and one of them blew up b/c on detach the bdm.connection_info was gone | 16:14 |
| mriedem | i'm guessing b/c another test did something to wipe it out (update the db) | 16:15 |
| mriedem | not sure though, it's been a long-standing bug | 16:15 |
| *** rwsu-afk is now known as rwsu | 16:15 | |
| mtreinish | mriedem: was they all tests in one test class? | 16:16 |
| mtreinish | man I can't speak today... | 16:16 |
| *** woodster_ has joined #openstack-qa | 16:16 | |
| mriedem | maybe they was... | 16:16 |
| mriedem | :) | 16:16 |
| mtreinish | ok, because I can see that being a reuse case, especially because volumes are kinda expensive to create for every test | 16:17 |
| mtreinish | but between test classes if you saw that, something is really going wrong | 16:17 |
| openstackgerrit | Merged openstack/tempest-lib: Add api docs to the rest client https://review.openstack.org/158017 | 16:20 |
| *** amotoki has joined #openstack-qa | 16:22 | |
| *** jogo is now known as flashgordon | 16:22 | |
| *** mpopow has joined #openstack-qa | 16:22 | |
| *** tkammer has quit IRC | 16:26 | |
| *** amotoki has quit IRC | 16:27 | |
| *** Longgeek has joined #openstack-qa | 16:34 | |
| *** dboik_ has quit IRC | 16:36 | |
| *** dboik has joined #openstack-qa | 16:37 | |
| *** mlavalle has joined #openstack-qa | 16:37 | |
| openstackgerrit | Filip HubÃk proposed openstack/tempest: Add keystone v2.0 and v3 api discovery checks https://review.openstack.org/162710 | 16:38 |
| *** mlavalle has quit IRC | 16:38 | |
| *** pkoniszewski has joined #openstack-qa | 16:39 | |
| mtreinish | sdague: I was looking more at the unix_timestamp stuff, and it looks like it's still only second resolution as an int. How were you able to deal with the microseconds? | 16:39 |
| mriedem | mtreinish: well, this at least is really confusing when you're stepping through the n-cpu logs: | 16:40 |
| mriedem | http://git.openstack.org/cgit/openstack/tempest/tree/tempest/scenario/test_encrypted_cinder_volumes.py#n38 | 16:40 |
| mriedem | for the cryptsetup test we still name the volume_type with 'luks' | 16:40 |
| mriedem | so i was thinking the luks test failed while digging through the compute logs, but it was the cryptsetup test that failed | 16:40 |
| mtreinish | mriedem: oh, that's a clear bug | 16:41 |
| mriedem | i'll push a change for that | 16:41 |
| mtreinish | resources should never be created with a static name | 16:41 |
| *** melwitt has joined #openstack-qa | 16:41 | |
| mtreinish | I think that's even written down somewhere :) | 16:41 |
| mriedem | mtreinish: well the utility creates a random name: randomized_name = data_utils.rand_name('scenario-type-' + name + '-') | 16:41 |
| mriedem | but name is always 'luks' | 16:41 |
| mriedem | even when not using luks | 16:41 |
| mtreinish | http://docs.openstack.org/developer/tempest/HACKING.html#parallel-test-execution | 16:42 |
| mtreinish | mriedem: it should just call tempest_lib.common.utils.data_utils.rand_name('luks') on L39 | 16:43 |
| mtreinish | instead of just 'luks' | 16:43 |
| mtreinish | mriedem: oh, heh I didn't even notice that one wasn't luks | 16:43 |
| mriedem | "volume_type": "scenario-type-luks--1230229905" | 16:43 |
| mriedem | it still gets a random name | 16:44 |
| mriedem | but it's not only the luks test | 16:44 |
| mriedem | anyway, easy change | 16:44 |
| mriedem | i'll push a thing | 16:44 |
| *** mpopow has quit IRC | 16:44 | |
| mtreinish | oh, ok. I guess the create_volume_type() call is calling that inside | 16:44 |
| mriedem | the other interesting thing is that detach_volume explodes in nova but apparently doesn't return a failure code | 16:44 |
| mriedem | yeah | 16:45 |
| mtreinish | oh, that's fun | 16:45 |
| *** pkoniszewski has quit IRC | 16:46 | |
| mriedem | yeah, we're just stuck in 'detaching' state until that times out | 16:47 |
| mtreinish | mriedem: do you have a link with the explosion? | 16:47 |
| mriedem | you know it | 16:47 |
| mriedem | http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_17_567 | 16:48 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1327218 | 16:48 |
| openstack | Launchpad bug 1327218 in OpenStack Compute (nova) "Volume detach failure because of invalid bdm.connection_info" [High,Confirmed] | 16:48 |
| Noice | Hello, to finalize the installatin of swift, what node is the swift.conf downloaded to? Controller? | 16:49 |
| notmyname | Noice: swift.conf should be on every swift node | 16:49 |
| *** mestery has quit IRC | 16:50 | |
| *** mpaolino has joined #openstack-qa | 16:50 | |
| openstackgerrit | Matt Riedemann proposed openstack/tempest: Distinguish between luks and cryptsetup volume types https://review.openstack.org/164259 | 16:51 |
| mriedem | mtreinish: ^ | 16:51 |
| *** mestery has joined #openstack-qa | 16:51 | |
| *** safchain has quit IRC | 16:51 | |
| openstackgerrit | Filip HubÃk proposed openstack/tempest: Add keystone v2.0 and v3 api discovery checks https://review.openstack.org/162710 | 16:51 |
| mriedem | also https://review.openstack.org/#/c/164235/ | 16:52 |
| mriedem | that's the e-r query for this failure ^ | 16:52 |
| mtreinish | mriedem: yeah that's weird, I would have expected the traceback to result in setting something as an error state or an error resp | 16:56 |
| *** Longgeek has quit IRC | 16:56 | |
| mtreinish | but, thinking about it more, it's async so the resposne cant be sent back to the user, and nova doesn't own the volume state so it probably can't set it as error | 16:56 |
| mtreinish | mriedem: nova doesn't have a mechanism to tell cinder things went bad does it? | 16:56 |
| *** harlowja has joined #openstack-qa | 16:57 | |
| mriedem | it's a cast, yeah | 16:58 |
| mriedem | damn | 16:58 |
| mriedem | well nova calls cinderclient.begin_detaching and i'm not sure what that does | 16:59 |
| mriedem | assuming it sets some state, like don't use it | 16:59 |
| *** apevec has quit IRC | 16:59 | |
| *** dustins_ has joined #openstack-qa | 16:59 | |
| mriedem | oh yeah, sets volume status to 'detaching' | 16:59 |
| mriedem | i saw that in the tempest console logs | 16:59 |
| *** anshul has quit IRC | 16:59 | |
| mriedem | it's 'detaching' until tempest times out and deletes the volume | 17:00 |
| sdague | mtreinish: honestly, do the 2 columns | 17:01 |
| mtreinish | mriedem: yeah I just found that in the cinder code: http://git.openstack.org/cgit/openstack/cinder/tree/cinder/volume/api.py#n508 | 17:02 |
| mriedem | mtreinish: ok, so at a minimum, we need to call roll_detaching when detach_volume explodes in the compute manager | 17:03 |
| mriedem | to put the volume status back in 'in-use' | 17:03 |
| *** dustins has quit IRC | 17:03 | |
| mriedem | roll_detaching would be better named 'rollback_detaching', but whatevs | 17:03 |
| *** sputnik13 has joined #openstack-qa | 17:03 | |
| *** jlanoux has quit IRC | 17:03 | |
| mtreinish | sdague: oh ok, well then I'm not sure that changes things for the migration I have. I would probably just add a follow-on one to convert the datetime column to an int | 17:04 |
| sdague | mtreinish: sure | 17:04 |
| sdague | mriedem: you saw folks had to revert your revert right? it cratered icehouse jobs | 17:04 |
| mtreinish | mriedem: are you sure that's what you want to do that. Is the volume still usable after the failed detach? | 17:04 |
| mriedem | sdague: ? | 17:05 |
| mtreinish | sdague: yeah, we got the bottom of how it slipped in too | 17:05 |
| mriedem | https://review.openstack.org/#/c/164190/ | 17:05 |
| mriedem | mtreinish: good question, i'm not sure if it's usable | 17:05 |
| mriedem | nova doesn't have the connection_info anymore... | 17:05 |
| mriedem | i'm wondering if nova can somehow refresh the connection info | 17:05 |
| mriedem | mtreinish: actually, heh, we already try to rollback the detach in the nova compute manager, but where we're failing is outside that try/except block | 17:07 |
| mtreinish | mriedem: haha, nice | 17:07 |
| sdague | mriedem: that patch still fails icehouse, fwiw | 17:07 |
| mriedem | sdague: hmm, i'm confused then | 17:08 |
| mriedem | sdague: we shouldn't merge it if it's going to break stable/icehouse jobs again | 17:08 |
| mtreinish | sdague: it should the issue was we fixed the tempest extension list on icehouse devstack | 17:08 |
| mriedem | mtreinish: the icehouse compat job is failing on the restore https://review.openstack.org/#/c/164190/ | 17:08 |
| mtreinish | mriedem: yeah, that was expected. Because we have the right extension list on icehouse now | 17:09 |
| mriedem | oh | 17:09 |
| mriedem | well we shouldn't merge this test if it's going to break icehouse then... | 17:09 |
| mtreinish | it only skipped before because the extension list was wrong | 17:09 |
| mriedem | it's skipped in juno fwiw | 17:09 |
| mtreinish | mriedem: juno devstack still has the broken extension list (there is a patch in progress to fix it, it's blocked by another broken job) | 17:10 |
| mtreinish | mriedem: we just need to fix your patch to have the proper skip conditions so it doesn't run where it can't run | 17:10 |
| mriedem | mtreinish: yeah, that's going to require a compute-feature-enabled flag then | 17:10 |
| *** cody-somerville has quit IRC | 17:11 | |
| *** mpaolino has quit IRC | 17:11 | |
| mriedem | just marked the restore patch WIP | 17:11 |
| *** cody-somerville has joined #openstack-qa | 17:11 | |
| mtreinish | sigh, well if that's what it takes. It's something beyond the scope of backport I'm guessing? | 17:11 |
| *** Nic has joined #openstack-qa | 17:13 | |
| *** annashen has joined #openstack-qa | 17:15 | |
| *** patrickeast has joined #openstack-qa | 17:16 | |
| *** ajo has quit IRC | 17:16 | |
| *** mpaolino has joined #openstack-qa | 17:16 | |
| *** jordanP has quit IRC | 17:18 | |
| *** markus_z has quit IRC | 17:20 | |
| openstackgerrit | Merged openstack-dev/devstack: Fail if run in POSIX compatability mode https://review.openstack.org/163214 | 17:27 |
| adam_g | dkranz, hey sorry to bother you again, https://review.openstack.org/#/c/163934/ got jammed up in the gate yesterday and needs a reapproval, if you have a sec | 17:27 |
| openstackrecheck | openstack/cinder change: https://review.openstack.org/158829 failed gate-cinder-pep8, gate-cinder-python27 in the gate queue with an unrecognized error | 17:29 |
| openstackgerrit | Merged openstack/tempest: Remove neutron CLI tests https://review.openstack.org/162850 | 17:30 |
| *** erlon has quit IRC | 17:31 | |
| dkranz | adam_g: looks like some one beat me to it | 17:32 |
| *** achanda has joined #openstack-qa | 17:33 | |
| *** dustins_ has quit IRC | 17:33 | |
| adam_g | dkranz, ah! :) | 17:34 |
| *** bkopilov has quit IRC | 17:34 | |
| *** bkopilov has joined #openstack-qa | 17:35 | |
| Noice | This is a noob question, but the swift hash path suffix and prefix can be anything right? There is no defined standard beyond the organization running the cloud? | 17:36 |
| openstackgerrit | Merged openstack/tempest: Replace hardcoded volume size by created volume size https://review.openstack.org/163038 | 17:40 |
| openstackgerrit | Merged openstack/tempest: Copy the default params dict to avoid race https://review.openstack.org/163895 | 17:41 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 17:42 | |
| marun | mtreinish: do you have time today to revisit the issue of freezing neutron api test development? | 17:42 |
| *** jhenner has joined #openstack-qa | 17:43 | |
| notmyname | Noice: correct. the biggest "rules" are that it cannot be changed and that they must be secret | 17:43 |
| *** bkopilov has quit IRC | 17:43 | |
| *** dustins has joined #openstack-qa | 17:44 | |
| *** bkopilov has joined #openstack-qa | 17:44 | |
| notmyname | hmm...I used scare quotes and shouldn't have. those are the rules | 17:44 |
| *** ChuckC has quit IRC | 17:45 | |
| openstackgerrit | Zengfa Gao proposed openstack/tempest: Add API tests for Neutron subnetpool https://review.openstack.org/159644 | 17:46 |
| *** bkopilov has quit IRC | 17:50 | |
| *** bkopilov has joined #openstack-qa | 17:51 | |
| *** annashen has quit IRC | 17:51 | |
| *** rushiagr_away is now known as rushiagr | 17:51 | |
| openstackgerrit | Zengfa Gao proposed openstack/tempest: Add API tests for Neutron subnetpool https://review.openstack.org/159644 | 17:51 |
| *** Sukhdev has quit IRC | 17:52 | |
| *** Longgeek has joined #openstack-qa | 17:53 | |
| *** omnibus7 is now known as davidlenwell | 17:57 | |
| *** bkopilov has quit IRC | 17:58 | |
| *** bkopilov has joined #openstack-qa | 17:59 | |
| *** ajmiller has joined #openstack-qa | 18:06 | |
| *** idokungfu has joined #openstack-qa | 18:07 | |
| dkranz | mtreinish: This test fails test5 only if you have a chance to take a look https://review.openstack.org/#/c/164225/1/tempest/tests/common/test_admin_available.py,unified | 18:08 |
| *** mpaolino has quit IRC | 18:12 | |
| mriedem | mtreinish: so now i'm all paranoid about this change https://review.openstack.org/#/c/157054/ | 18:13 |
| mriedem | live migrating a paused instance doesn't work in juno or icehouse | 18:13 |
| mriedem | there are 2 config flags already to not run that test...but you could support both pause and live_migrate, and fail with that in juno... | 18:13 |
| mriedem | we won't hit any failures in the gate since multi-node is required and that's only on experimental atm | 18:14 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for detach_volume NoneType bug 1327218 https://review.openstack.org/164235 | 18:19 |
| openstack | bug 1327218 in OpenStack Compute (nova) "Volume detach failure because of invalid bdm.connection_info" [High,Confirmed] https://launchpad.net/bugs/1327218 | 18:19 |
| openstackgerrit | Morgan Fainberg proposed openstack-dev/devstack: Add response time to keystone access log https://review.openstack.org/164042 | 18:20 |
| *** woodster_ has quit IRC | 18:20 | |
| *** fandi has joined #openstack-qa | 18:20 | |
| *** fandi has quit IRC | 18:21 | |
| flashgordon | mtreinish: https://review.openstack.org/#/c/151056/ should I just abandon that patch? | 18:22 |
| *** Longgeek has quit IRC | 18:22 | |
| *** nikil has joined #openstack-qa | 18:23 | |
| *** shardy has quit IRC | 18:24 | |
| *** bkopilov has quit IRC | 18:25 | |
| *** bkopilov has joined #openstack-qa | 18:25 | |
| *** nikil has quit IRC | 18:28 | |
| *** nikil has joined #openstack-qa | 18:28 | |
| *** pal_ has quit IRC | 18:29 | |
| *** openstackrecheck has quit IRC | 18:31 | |
| *** openstackrecheck has joined #openstack-qa | 18:31 | |
| *** bkopilov has quit IRC | 18:32 | |
| *** bkopilov has joined #openstack-qa | 18:32 | |
| *** mlavalle has joined #openstack-qa | 18:33 | |
| *** e0ne has joined #openstack-qa | 18:34 | |
| *** Sukhdev has joined #openstack-qa | 18:35 | |
| dkranz | mriedem: about https://review.openstack.org/#/c/161768/ | 18:43 |
| mriedem | dkranz: yeah | 18:43 |
| dkranz | mriedem: Is this an atttempt to generate failures so logstash can be used to track them down? | 18:43 |
| mriedem | dkranz: well it was, but i haven't seen those tests fail after like 9 rechecks | 18:44 |
| mriedem | dkranz: so i basically changed it to just 'unskip' | 18:44 |
| mriedem | however, i'm all paranoid about stable branches now | 18:44 |
| dkranz | mriedem: that was not clear from the commit message | 18:44 |
| mriedem | yeah... | 18:45 |
| mriedem | it was originally dependent on a logging patch in nova which has since merged | 18:45 |
| dkranz | mriedem: I still think there should be a way to skip tests on stable branches | 18:45 |
| mriedem | i could update the commit message to just say 'we aren't seeing this anymore on master, unskip' | 18:45 |
| mriedem | yeah that would be nice | 18:46 |
| dkranz | mriedem: sure but I am also going to +A | 18:46 |
| dkranz | mriedem: and I am also pushing a patch that will run that test under stress as WIP | 18:46 |
| flashgordon | devstack is back down to about 15 minutes again | 18:46 |
| mriedem | dkranz: i checked the juno and icehouse neutron compat jobs and they are running the tests, so that makes me feel better | 18:47 |
| flashgordon | dtroyer: we make about 170 or so calls to OSC in devstack | 18:47 |
| flashgordon | and a lot of those take at least two seconds | 18:48 |
| openstackgerrit | David Kranz proposed openstack/tempest: DO NOT MERGE - Run network connectivity start/stop test as stress https://review.openstack.org/164300 | 18:48 |
| *** andreykurilin_ has joined #openstack-qa | 18:49 | |
| flashgordon | mtreinish: and tempest is 20 minutes or so | 18:50 |
| dtroyer | flashgordon: a lot of those calls are unnecessary… | 18:51 |
| openstackgerrit | Dean Troyer proposed openstack-dev/devstack: Batch up calls to OSC https://review.openstack.org/164302 | 18:51 |
| dtroyer | flashgordon: ^^^ is a simplistic way to reduce the number. it cut about 30 seconds off my slow VM tests | 18:52 |
| *** annashen has joined #openstack-qa | 18:52 | |
| flashgordon | dtroyer: nice | 18:53 |
| *** idokungfu has quit IRC | 18:53 | |
| flashgordon | very slick way to get around the load times | 18:54 |
| dtroyer | I'm playing with running osc as a long-running process instead and talking to it over pipes. the issue is where we need to create or find an ID, the simple batch mode doesn't do that nicely | 18:54 |
| flashgordon | dtroyer: what about the service create stuff | 18:54 |
| dtroyer | that's only one call and it is inside an if block | 18:55 |
| flashgordon | also are we re-using the same token for all of that ? | 18:55 |
| mtreinish | flashgordon: no, we can figure that patch out, we still want to do it | 18:55 |
| mtreinish | flashgordon: 20 min isn't too bad considering we're running ~1500 tests | 18:55 |
| flashgordon | mtreinish: no its not, are we ready to stop running say keystone API tests on nova changes? | 18:56 |
| dtroyer | flashgordon: yes, it's all the service token at this point, no auth available to set up keystone | 18:56 |
| flashgordon | dtroyer: ohh right doh | 18:56 |
| mtreinish | mriedem: hmm so for the live migrate thing, do you want to add another flag, pause_live_migrate? | 18:56 |
| *** liusheng has quit IRC | 18:56 | |
| *** annashen has quit IRC | 18:56 | |
| *** liusheng has joined #openstack-qa | 18:57 | |
| * morganfainberg perks up. | 18:57 | |
| mtreinish | flashgordon: keystone's api tests would be easy to segregate, but what are you trying to save? if it's runtime that won't do much | 18:57 |
| dtroyer | IIRC we stick with the service token for all of the account creation so auth isn't part of the times there | 18:58 |
| flashgordon | dtroyer: not too bad get_or_create_service takes 4 second start to end | 18:58 |
| mriedem | mtreinish: is that the only option? i kind of hate having to create a new flag for every new test we want just because we dont have the code back on stable | 18:58 |
| flashgordon | mtreinish: run time, I know its not a lot. but its an easy start | 18:58 |
| flashgordon | and the inverse. | 18:58 |
| flashgordon | don't run nova API tests on keystone patches | 18:58 |
| mtreinish | flashgordon: I'm sure morganfainberg doesn't care :) | 18:58 |
| mtreinish | mriedem: well it's that or backport | 18:58 |
| dtroyer | flashgordon: look at the time for the whole create_keystone_accounts(). in the logs, create_nova_accounts() immediately follows it so subtracting those two times is simple to compare the whole thing | 18:59 |
| morganfainberg | wait, i don't care about what? :P | 18:59 |
| mriedem | yeah, the backport isn't going to happen | 18:59 |
| mriedem | curse you, branchless tempest | 18:59 |
| morganfainberg | brrrraaaaaaaaaaaaanncchhhheessssss /zombie | 18:59 |
| morganfainberg | :P | 18:59 |
| mtreinish | mriedem: it's your own damn fault for breaking api consistency by fixing bugs :) | 18:59 |
| mriedem | fair point | 19:00 |
| mriedem | well let me wip that one too now | 19:00 |
| mriedem | remember that scene from wayne's world where wayne exclaims that he's being shit on? that's what i feel like today :) | 19:00 |
| openstackgerrit | Joe Gordon proposed openstack/tempest-lib: Deprecate the nova_manage client https://review.openstack.org/151056 | 19:01 |
| mtreinish | mriedem: heh, I could see that | 19:02 |
| mtreinish | flashgordon: fwiw, ironic already in the process of splitting there gating jobs off | 19:02 |
| mriedem | my green dashboard is all turning red now | 19:02 |
| flashgordon | mtreinish:oh nice what is the process? | 19:03 |
| mtreinish | flashgordon: they're just changing there gating regex, instead of running full | 19:03 |
| flashgordon | mtreinish: it may be easier to start running less of tempest on say keystone before nova actually | 19:03 |
| flashgordon | mtreinish: link | 19:03 |
| *** jordanP has joined #openstack-qa | 19:03 | |
| flashgordon | morganfainberg: would you mind if I make keystone gate faster? | 19:03 |
| *** jordanP has quit IRC | 19:03 | |
| morganfainberg | flashgordon, uhm... how would you do that? | 19:03 |
| *** cody-somerville has quit IRC | 19:03 | |
| mtreinish | yeah, but I would be -1 on doing that for nova | 19:04 |
| mtreinish | ironic and keystone are diff | 19:04 |
| morganfainberg | flashgordon, i wont say no. but i want to know what you're proposing. | 19:04 |
| flashgordon | morganfainberg: stop running compute tempest tests on keystone changes | 19:04 |
| *** cody-somerville has joined #openstack-qa | 19:04 | |
| flashgordon | mtreinish: what about nova? | 19:04 |
| mtreinish | flashgordon: https://review.openstack.org/#/c/161420/ | 19:04 |
| mtreinish | flashgordon: nova is the glue, I'm much more hesitant to change that regex | 19:05 |
| morganfainberg | flashgordon, i don't think this is a bad idea. though, do you see any risk in a keystone change breaking nova [in most cases]? we've been good, but just want to see what you think. | 19:05 |
| mtreinish | morganfainberg: well you've introduced a noticeable race at least once this cycle... :) | 19:06 |
| flashgordon | mtreinish: so your saying don't change the set of tests that are run against nova? or don't remove nova from say keystone? | 19:06 |
| morganfainberg | mtreinish, right. | 19:06 |
| mtreinish | flashgordon: don't change the set that runs against nova | 19:06 |
| *** mpopow has joined #openstack-qa | 19:06 | |
| flashgordon | mtreinish: ack, most of tempest is nova right now anyway | 19:06 |
| morganfainberg | mtreinish, but would that have been missed w/o nova cross-gating on the keystone changes | 19:06 |
| mtreinish | morganfainberg: actually I think it would, just because the keystone tests in tempest are limited | 19:07 |
| *** gfidente has quit IRC | 19:07 | |
| mtreinish | because it was a race on initial creation, but I'm not sure the keystone tempest tests would tickle that | 19:07 |
| morganfainberg | mtreinish, this is why i am asking before | 19:07 |
| flashgordon | but when did we catch the race? | 19:08 |
| dkranz | The tempest scenario tests that use nova would still be running | 19:08 |
| *** tosky has quit IRC | 19:08 | |
| mtreinish | flashgordon: like a month or 2 ago | 19:08 |
| dkranz | Even if we stopped running tempest api nova tests on keystone | 19:08 |
| morganfainberg | mtreinish, flashgordon, maybe we want to wait until keystone spins up our functional testing suite? | 19:08 |
| mtreinish | dkranz: no the proposal would be to set the regex on keystone jobs to be 'identity' | 19:08 |
| mtreinish | so unless the scenario tests set that service tag they wouldn't be run either | 19:09 |
| morganfainberg | mtreinish, flashgordon, but i'm not opoosed to this change now. i just want to make sure we're not opening ourselves up for scary scary stuff too quickly | 19:09 |
| morganfainberg | mtreinish, flashgordon, i'll defer to your judgement on the risk - we (keystone) will continue to try and avoid causing issues regardless. | 19:09 |
| dkranz | mtreinish: why not identity + scenario? | 19:09 |
| morganfainberg | and keep working towards functional testing | 19:10 |
| flashgordon | mtreinish: as in how did we catch the race | 19:10 |
| flashgordon | mtreinish: did it land or did we catch it before it landed? | 19:10 |
| *** dustins has quit IRC | 19:10 | |
| mtreinish | flashgordon: by running 2 compute tests with tenant isolation at the same time | 19:10 |
| flashgordon | mtreinish: not how to reproduce the bug, but at what point did we realize we had the bug? | 19:11 |
| mtreinish | flashgordon: I don't remember, it was pretty obvious when looking at the code. I pushed the fix as soon as we noticed it | 19:12 |
| openstackgerrit | Merged openstack/tempest: Skip test_connectivity_between_vms_on_different_networks for baremetal https://review.openstack.org/163934 | 19:12 |
| mtreinish | there should have been an e-r query if you wanted to dig it up | 19:12 |
| flashgordon | mtreinish: right, my point is it got into the code even with our tests | 19:12 |
| *** dustins has joined #openstack-qa | 19:12 | |
| mtreinish | dkranz: that would probably be fine too | 19:12 |
| dkranz | mtreinish: that is just "virtualizing" what will happen when api tests are moved to projects | 19:13 |
| mtreinish | flashgordon: sure, but I just blame morganfainberg for the recheck grind to land it (even though it's not his fault) :) | 19:13 |
| flashgordon | mtreinish: can I set the regex to" tempest.api.identity|tempest.scenario | 19:14 |
| flashgordon | to run both? | 19:14 |
| mtreinish | flashgordon: yes that'll will do what you intend, but I would honestly do 'identity|scenario' | 19:15 |
| flashgordon | mtreinish: that works for me | 19:15 |
| mtreinish | in case something else is tagged as saying it needs keystone | 19:15 |
| flashgordon | ahh so tags match too nice | 19:15 |
| morganfainberg | flashgordon, mtreinish so just ping me if you're going to change how we gate and point me at the patch. | 19:16 |
| mtreinish | flashgordon: but I feel before we start doing this, I think we need to have some people sit down and improve the test tagging. Which I view as a prereq before we can do any of this | 19:16 |
| mtreinish | although everything would be tagged as using keystone, because every test does | 19:17 |
| flashgordon | mtreinish: how about this, I'll post a patch up for morganfainberg to see | 19:18 |
| flashgordon | mark it as WIP | 19:18 |
| flashgordon | for the idea | 19:18 |
| flashgordon | and go from there | 19:18 |
| morganfainberg | flashgordon, ++ let me know if you need anything else | 19:18 |
| flashgordon | and then we talk tagging details | 19:18 |
| mtreinish | flashgordon: sure | 19:19 |
| mtreinish | dkranz: yeah it does, but honestly I'm much more comfortable with this approach vs the complete split out | 19:20 |
| dkranz | mtreinish: I would like it better too, if we could solve the problem of a horizontal tempest core team | 19:21 |
| *** annashen has joined #openstack-qa | 19:21 | |
| dkranz | mtreinish: with big tent, I don't see how saying all the project api tests (and more deep functional) are in tempest | 19:22 |
| mtreinish | dkranz: I'm not sure I see that as a problem, if we just limit the scope of tempest to something like what I had in that WIP (which I need to revisit) | 19:22 |
| flashgordon | FYI running just keystone tests takes about a minute | 19:23 |
| mtreinish | flashgordon: yeah with ~20 min of setup time | 19:23 |
| flashgordon | mtreinish: lol yeah ... | 19:23 |
| dkranz | mtreinish: I guess I don't see the value of having blackbox api/functional tests divided between tempest and project repos, but there are many different views about that | 19:23 |
| flashgordon | 15 minutes now ! | 19:23 |
| mtreinish | flashgordon: oh, you know trove is already doing this. I just remembered | 19:24 |
| dkranz | mtreinish: we should make sure it is resolved for real in Vancouver | 19:24 |
| mtreinish | flashgordon: because trove has 6 tests which are even quicker | 19:26 |
| mtreinish | dkranz: I've always viewed it as just a matter of tiering, that there are classes of things which belong in projects (like negative testing) and others that I view as belonging in tempest | 19:26 |
| mtreinish | so it becomes hard to make blanket statements about dirs of tests which I don't think are as clear cut | 19:27 |
| mtreinish | dkranz: but at the same time that also makes it difficult to make rules that are handled other than at the individual patch level about things we agree should be tested by tempest proper | 19:30 |
| flashgordon | mtreinish: yeah I thought I saw trove do this | 19:32 |
| flashgordon | https://review.openstack.org/#/c/164314/1 | 19:32 |
| flashgordon | morganfainberg mtreinish: that is the idea ^ | 19:32 |
| flashgordon | now lets talk about tagging | 19:33 |
| flashgordon | mtreinish morganfainberg: what are your thoughts? | 19:33 |
| mtreinish | flashgordon: have you done a run with that regex before? | 19:34 |
| flashgordon | running it now | 19:34 |
| mtreinish | I'm curious how much time it actually saves | 19:35 |
| flashgordon | me too | 19:35 |
| flashgordon | well ident only is about a minute or 2 | 19:35 |
| mtreinish | flashgordon: but scenario are normally the slowest tests | 19:36 |
| flashgordon | mtreinish: right ... I ma seeing that now | 19:36 |
| flashgordon | maybe worth dropping them and just tagging a few with identity ? | 19:36 |
| flashgordon | test_encrypted_cinder_volumes etc | 19:36 |
| flashgordon | that doesn't need to run against keystone I hope | 19:37 |
| mtreinish | flashgordon: but encryption that sounds like something keystoney :) | 19:37 |
| flashgordon | lol | 19:37 |
| mtreinish | morganfainberg: ^^^ you're with me right | 19:37 |
| flashgordon | there is another project that does that | 19:37 |
| flashgordon | in keystone org | 19:38 |
| morganfainberg | mtreinish, encryption is uhhh | 19:38 |
| morganfainberg | i mean somewhat | 19:38 |
| morganfainberg | keystone, but i mean we don't "own" encryption stuff :P | 19:38 |
| morganfainberg | we have encryption things. | 19:39 |
| *** dustins has quit IRC | 19:39 | |
| mtreinish | heh, I was giving you an opportunity to rule it all, and you turned it down :) | 19:39 |
| morganfainberg | mtreinish, i am scared to rule it all | 19:39 |
| morganfainberg | mtreinish, honestly, i don't want to own all encryption :P | 19:39 |
| flashgordon | mtreinish: my scenario times will be off, this is a fakevirt driver env | 19:40 |
| flashgordon | but its still going | 19:40 |
| mtreinish | flashgordon: it will timeout for everything then | 19:40 |
| mtreinish | because scenario tests normally login to the guest | 19:40 |
| flashgordon | ahh 1101.3275 | 19:40 |
| flashgordon | - Worker 0 (48 tests) => 0:06:23.703093s | 19:41 |
| flashgordon | so 6 minutes totatl | 19:41 |
| flashgordon | not bad actually | 19:41 |
| mtreinish | flashgordon: you really have to use a real backend to get a feel for it. The cheap way to do it is just push a tempest patch which changes the regex in tox.ini for -efull | 19:42 |
| mtreinish | that will give you real run times, like 20 of them | 19:42 |
| openstackgerrit | Joe Gordon proposed openstack/tempest: DO NOT MERGE: testing regex https://review.openstack.org/164322 | 19:44 |
| flashgordon | like that ^ ? | 19:44 |
| mtreinish | flashgordon: yep that should do it | 19:45 |
| flashgordon | cool | 19:45 |
| flashgordon | so what about tags? | 19:45 |
| *** bkopilov has quit IRC | 19:45 | |
| flashgordon | you said that was something to be done before this? | 19:45 |
| *** ajo has joined #openstack-qa | 19:46 | |
| mtreinish | flashgordon: so it's probably not an issue with keystone, but we're pretty poor about applying service tags everywhere | 19:46 |
| flashgordon | so when should they be applied? | 19:47 |
| mtreinish | so I just think we need some people to go through everything to make sure things are sane | 19:47 |
| flashgordon | on say a nova test that uses volumes? | 19:47 |
| dkranz | mtreinish: your last comment is what pushed me over the line towards thinking that moving them out of tempest was the way to go. | 19:47 |
| flashgordon | should get a cinder tag | 19:47 |
| mtreinish | flashgordon: yeah exactly (although the tag is volume) | 19:48 |
| flashgordon | err right | 19:48 |
| flashgordon | so for the special case of keystone what needs to be done | 19:48 |
| flashgordon | and for the rest of the tags, sounds like a great sprint session | 19:48 |
| *** bkopilov has joined #openstack-qa | 19:48 | |
| mtreinish | flashgordon: well keystone should be good, unless we get pedantic and start tagging everything | 19:49 |
| mtreinish | flashgordon: unless there is a test which explicity calls keystone's api somewhere (which ther emight be) | 19:50 |
| *** Longgeek has joined #openstack-qa | 19:50 | |
| flashgordon | any easy way (grep) way to find those | 19:50 |
| mtreinish | flashgordon: err, I'm not sure, it all eventually goes to tempest.services.identity, but things normally go through an object initialized in tempest.clients and are aliased in local test classes | 19:51 |
| mtreinish | dkranz: which statement? | 19:52 |
| mtreinish | I had a network blip and my client dropped the history | 19:52 |
| flashgordon | mtreinish: http://paste.openstack.org/show/192178 | 19:52 |
| flashgordon | git grep identity_client tempest/api | grep -v 'api/identity/' | 19:53 |
| *** sabeen has quit IRC | 19:53 | |
| dkranz | mtreinish: back at 15:30 | 19:53 |
| flashgordon | quotas tests sound reasonable to tag | 19:54 |
| dkranz | mtreinish: IMO it is arbitrary and the current set of tempest api tests are all over the map | 19:54 |
| flashgordon | anyway lunch time bbiab | 19:54 |
| mtreinish | dkranz: ok, yeah but that's because we've mostly just allowed everything. I think if we came up with stricter guidelines and then applied them to all new patches and used them as an additional criteria for removal that's a much better approach | 19:56 |
| *** ajmiller_ has joined #openstack-qa | 19:56 | |
| *** rushiagr is now known as rushiagr_away | 19:57 | |
| dkranz | mtreinish: I would be fine with that and we have discussed it in the abstract last three summits | 19:57 |
| mtreinish | dkranz: well the real problem before was there wasn't anywhere else for anything to live. That's not true today (at least for some projects) | 19:58 |
| mtreinish | dkranz: and we've also started this process, like for cli testing | 19:58 |
| dkranz | mtreinish: yes. Maybe there can be some discussion about this during the NY thing (which I will not be there physically but will be around Wed/Thur) | 19:58 |
| *** ajmiller has quit IRC | 20:00 | |
| mtreinish | dkranz: well, to be honest I was trying to avoid discussions like that in NY, because it's the appearance of being behind closed. (also because of limited space) I was mostly trying to save that time to be high bandwidth get some bps and other todos done | 20:01 |
| mtreinish | but I'm sure the topic will at least come up | 20:01 |
| dkranz | mtreinish: your call :) | 20:02 |
| *** bkopilov has quit IRC | 20:02 | |
| dkranz | mtreinish: I am actually not a subscriber to the theory that "closed-door" discussions are bad as long as any theory is then published | 20:02 |
| dkranz | mtreinish: not everyone can travel to the summits either | 20:03 |
| mtreinish | dkranz: neither am I, I'm just cognizant of that opinion. Also the sprint has much more limited attendance | 20:04 |
| *** bkopilov has joined #openstack-qa | 20:04 | |
| *** achanda has quit IRC | 20:06 | |
| openstackgerrit | Zengfa Gao proposed openstack/tempest: Add API tests for Neutron subnetpool https://review.openstack.org/159644 | 20:08 |
| *** yamahata has joined #openstack-qa | 20:09 | |
| mriedem | mtreinish: this is the culmination of my day with that detach_volume thing https://review.openstack.org/#/c/164330/ | 20:09 |
| *** mpopow has quit IRC | 20:09 | |
| *** achanda has joined #openstack-qa | 20:11 | |
| mtreinish | mriedem: cool, I'll tkae a look | 20:13 |
| mriedem | it's really addressing the symptom, which sucks | 20:13 |
| *** Longgeek has quit IRC | 20:18 | |
| mtreinish | mriedem: I wonder if we should look at doing a callback mechanism with cinder, it might help with things like this | 20:18 |
| mriedem | callback how? | 20:18 |
| marun | mtreinish: do you have time to revisit the issue of neutron api tests? | 20:19 |
| mtreinish | mriedem: similar to the nova api that neutron uses to inform about state of things | 20:19 |
| mtreinish | mriedem: so we can better track the state of volumes between the services | 20:20 |
| mriedem | mtreinish: you mean the events stuff? | 20:20 |
| mtreinish | yeah | 20:21 |
| mriedem | ndipanov was working on that at one point i thought | 20:21 |
| mtreinish | I'm just thinking out loud | 20:21 |
| mriedem | basically, attach a volume and wait for cinder to send an event back | 20:21 |
| marun | mriedem: oh, nova doesn't wait for that to complete? | 20:21 |
| mtreinish | mriedem: yeah something like that | 20:21 |
| * marun wonders why people keep forgetting this is supposed to be a distributed system | 20:22 | |
| mtreinish | marun: heh, it's surprising isn't it | 20:22 |
| mtreinish | marun: yeah we can probably talk about it now | 20:22 |
| openstackgerrit | Merged openstack/tempest: remove redundant negative flavor admin tests https://review.openstack.org/158852 | 20:23 |
| openstackgerrit | Merged openstack/tempest: remove redundant negative flavor admin tests https://review.openstack.org/158852 | 20:23 |
| mriedem | marun: well, i don't think cinder does it | 20:23 |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/volume.py#n394 | 20:23 |
| marun | mtreinish: As I said yesterday, I'm hoping we can devise a strategy for freezing. | 20:23 |
| mriedem | nova does everything locally, plugs vifs, attaches volumes, etc etc | 20:24 |
| marun | mriedem: ouch. good luck testing something like that comprehensively. | 20:24 |
| mriedem | it's like nova-volume and the network stuff never moved out of nova :) | 20:24 |
| *** achanda has quit IRC | 20:25 | |
| marun | mriedem: ah, so when the responsibility was removed, no rearchitecture was done. | 20:25 |
| mriedem | and now we have multiple services keeping track asynchronously | 20:25 |
| marun | mriedem: :/ | 20:25 |
| mriedem | i guess? | 20:25 |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/162266 failed because of: gate-keystone-pep8: unrecognized error, gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1311066 | 20:25 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 20:25 |
| mriedem | nova-volume was before my time | 20:25 |
| marun | mriedem: that certainly was the case with neutron | 20:25 |
| *** e0ne has quit IRC | 20:25 | |
| mtreinish | mriedem: was it? I remember having to work with it. Although that probably was from before you started on openstack stuff | 20:25 |
| marun | mriedem: it took me looking through the source to figure out that nova was racing vm boot against neutron port wiring | 20:25 |
| mriedem | so now it makes sense to me why every time there is a new cinder driver or neutron plugin we have new volume and vif types to handle in the libvirt code in nova | 20:26 |
| mtreinish | mriedem: to be fair though it was basically identical to cinder at that point | 20:26 |
| mriedem | why can't nova just talk to black box volume / network service and that can hide all of the backend driver details from nova? | 20:27 |
| marun | mriedem: people like making work for themselves? | 20:27 |
| mriedem | network_service.attach_port(instance, port) | 20:27 |
| mriedem | go! | 20:27 |
| marun | mriedem: if things were architected as distributed from the get-go, that's where we'd be | 20:28 |
| marun | mriedem: but since it's not, we're stuck with a crap-tonne of code that assumes synchronous | 20:28 |
| marun | it would take a major rewrite to remove those assumptions, not just the calling points | 20:28 |
| mriedem | well, i'm not dumping on it too much, things move forward and it's hard to forklift | 20:28 |
| mriedem | but it certainly makes things more complex | 20:29 |
| marun | mriedem: still, it's kind of sad that a distributed system, going on 5 years old by now, isn't architected to be properly distributed in all kinds of ways. | 20:29 |
| marun | the joys of legacy | 20:29 |
| marun | mtreinish: neutron api tests? | 20:30 |
| mtreinish | marun: so do you have the numbers on in flight api.neutron tests? | 20:31 |
| marun | mtreinish: building a gerrit query for that as we speak | 20:31 |
| marun | mtreinish: I always struggle with the file searches | 20:31 |
| * marun goes to look at gerrit query docs | 20:31 | |
| marun | mtreinish: https://review.openstack.org/#/q/project:openstack/tempest+file:%255Etempest/api/network/.*+status:open,n,z | 20:32 |
| marun | mtreinish: doesn't look too bad, honestly | 20:33 |
| mtreinish | marun: tbh, I'm also still kinda hesitant to freeze dev here, because this really feels like a neutron decision. Especially when we don't have a clear direction yet on the scope of tempest with functional testing | 20:33 |
| mtreinish | marun: yeah that's not too many, although a bunch have been around for awhile | 20:33 |
| mtreinish | like I know armax's dvr thing is at least from the fall | 20:33 |
| mtreinish | marun: by neutron decision I meant a neutron policy decision to say all new api tests belong in neutron | 20:34 |
| marun | mtreinish: I can make that happen. | 20:34 |
| *** Sukhdev has quit IRC | 20:35 | |
| marun | mtreinish: I'll work with Kyle to make it official, and once we have it committed to our repo, you'll be willing to freeze? | 20:35 |
| marun | mestery: ^^ | 20:35 |
| *** BharatK has joined #openstack-qa | 20:35 | |
| mestery | marun: Yay! | 20:36 |
| mtreinish | marun: no that's what I'm saying you guys say that's your policy and watch the tempest queue to tell people | 20:36 |
| mtreinish | then your frozen without input from me :) | 20:36 |
| mestery | mtreinish: I see what you did there ;) | 20:36 |
| marun | mtreinish: ah, so we'll be responsible for policing it, at least in the near term? | 20:36 |
| mtreinish | because that's really what it comes down to | 20:36 |
| marun | mtreinish: and once the tests are removed, the policing becomes unnecessary since there won't be a place for them to go? | 20:37 |
| mtreinish | marun: yeah, I think that's probably best way to handle that, you guys self police :) (how could that go wrong) | 20:38 |
| marun | longer term, obviously, we still have to solidify how the in-tree tests are to be consumable and can run against stable | 20:38 |
| marun | mtreinish: ok, fair enough. | 20:38 |
| mtreinish | marun: yeah, I think the deletion is a a different discussion, but in the short term if you want to move dev that's how I think we should go about it | 20:38 |
| marun | mtreinish: sounds good | 20:38 |
| marun | mestery: thoughts? | 20:38 |
| mestery | marun: This all seems reasonable and puts us in control of our own destiny. +2 from me | 20:39 |
| marun | great! | 20:39 |
| * mtreinish hopes no one notices he indemnified himself from tests slipping freeze :) | 20:40 | |
| flashgordon | mtreinish: another possibly easy test thing to sort out | 20:41 |
| flashgordon | horizon | 20:41 |
| flashgordon | how many tests does tempest even have for horizon? | 20:41 |
| mtreinish | flashgordon: heh, 1 | 20:41 |
| flashgordon | we run 3 full tempest and a grenade job on it | 20:41 |
| flashgordon | does grenade even test horizon? | 20:41 |
| flashgordon | (I hope so) | 20:41 |
| mtreinish | flashgordon: heh, shh... | 20:41 |
| mtreinish | flashgordon: all it does is login with raw http calls | 20:42 |
| *** annashen has quit IRC | 20:42 | |
| mtreinish | flashgordon: we could easily move it to somewhere else | 20:42 |
| mtreinish | david-lyle: ^^^ | 20:42 |
| flashgordon | mtreinish: well for starters we don't need to run it 3x | 20:42 |
| mtreinish | flashgordon: sure we do, postgres breaks the horizon all the time :) | 20:43 |
| david-lyle | move it where? | 20:43 |
| flashgordon | mtreinish: I can't tell if your joking or not sadly | 20:43 |
| mtreinish | flashgordon: haha, I am joking | 20:43 |
| mtreinish | david-lyle: somewhere you think it makes sense | 20:43 |
| flashgordon | david-lyle: tempest/scenario/test_dashboard_basic_ops.py | 20:43 |
| david-lyle | we have no db so upgrade is of no concern | 20:44 |
| flashgordon | david-lyle: do you have a test like that one elsewhere? | 20:44 |
| mtreinish | david-lyle: what about the consistent config during an upgrade? | 20:44 |
| david-lyle | mtreinish: we don't test that | 20:44 |
| david-lyle | buyer beware | 20:44 |
| mtreinish | david-lyle: that's part of what grenade gives you, the config doesn't change between releases | 20:45 |
| mtreinish | unless you explicitly add it | 20:45 |
| mtreinish | david-lyle: but regardless the upgrade question is different from the question of having that test in tempest | 20:45 |
| david-lyle | our config is different than most of openstack and uses some django settings files | 20:45 |
| david-lyle | only a very small set a required to run | 20:45 |
| david-lyle | and the only time that base few change is when django changes how it works | 20:46 |
| mtreinish | david-lyle: sure, I was just saying that's part of what grenade is doing | 20:46 |
| david-lyle | we have an integration test suite in horizon that runs approximately the same test as is in tempest | 20:46 |
| mtreinish | david-lyle: link? | 20:47 |
| david-lyle | mtreinish: sure, just saying I believe were not testing anything like that | 20:47 |
| david-lyle | mtreinish: looking | 20:47 |
| mtreinish | david-lyle: then do you want to go through: https://wiki.openstack.org/wiki/QA/Tempest-test-removal and propose removing the test during the next removal meeting? | 20:47 |
| mtreinish | then you can safetly kill all your tempest jobs | 20:48 |
| flashgordon | mtreinish: oh that is even better | 20:48 |
| flashgordon | mtreinish: for now I have https://review.openstack.org/164344 | 20:48 |
| flashgordon | david-lyle: ^ | 20:48 |
| mtreinish | flashgordon: honestly I'd rather just wait until the 5th and kill all the jobs | 20:48 |
| flashgordon | mtreinish: I am fine with that too | 20:48 |
| flashgordon | the 5th? why so long? | 20:48 |
| david-lyle | mtreinish: so your option would pull us out of the integrated gate/ | 20:49 |
| mtreinish | david-lyle: yep | 20:49 |
| *** annashen has joined #openstack-qa | 20:49 | |
| mtreinish | david-lyle: to be fair I think the only reason horizon was part of the integrated gate was because of project policy for being integrated | 20:50 |
| david-lyle | mtreinish: the only thing being in the integrated gate really validates is that keystone didn't break, but our tests should cover that | 20:50 |
| flashgordon | mtreinish: if this was wait until next week then why not | 20:50 |
| david-lyle | we usually have to fix the issue locally anyway | 20:50 |
| david-lyle | I'd be ok with removing | 20:50 |
| mtreinish | flashgordon: we do the removal meetings once a month | 20:50 |
| flashgordon | mtreinish: ahh, so this is a quick win for now | 20:51 |
| david-lyle | we don't need to fail jobs on neutron/nova/cinder tests | 20:51 |
| flashgordon | and frees up a ton of test nodes | 20:51 |
| flashgordon | mtreinish: and what about the grenade testing? | 20:51 |
| flashgordon | mtreinish: we do basic grenade testing of horizon (not even sure what that means here though) | 20:51 |
| david-lyle | so this will run more like the client project tests? | 20:51 |
| clarkb | mtreinish: I think it was in there for two reasons. First its installed by devstack defaults and second there were plans to add tests it just never happened | 20:52 |
| mtreinish | david-lyle: sure, and if you do hit an issue with keystone we should be adding keystone tests to prevent that breakage in the future | 20:52 |
| david-lyle | yup | 20:52 |
| david-lyle | clarkb: my take is our tests are no longer wanted | 20:53 |
| clarkb | ya things may have changed, but there were reasons for it being there | 20:54 |
| clarkb | and hey we caught at least one largish bug that I can think of | 20:54 |
| clarkb | the template compile thing | 20:54 |
| mtreinish | david-lyle: heh s/tests/test :) | 20:54 |
| flashgordon | david-lyle: how would upgrade testing work | 20:54 |
| david-lyle | I meant the integration tests we've been creating | 20:54 |
| mtreinish | clarkb: sure, but if horizon had it's own equiv test now we would catch that | 20:54 |
| david-lyle | I think they are effective in our tree as well | 20:55 |
| mtreinish | david-lyle: oh, ok | 20:55 |
| clarkb | mtreinish: yup, we just never had that in the past | 20:55 |
| clarkb | mtreinish: but thats changing which is good | 20:55 |
| mtreinish | fwiw, I've always been a fan of that horizon tempest test. I thought it was cleverly done | 20:57 |
| flashgordon | mtreinish: hehe | 20:57 |
| david-lyle | log in is all you ever need | 20:57 |
| *** achanda has joined #openstack-qa | 20:57 | |
| mtreinish | david-lyle: heh, exactly :) | 20:58 |
| flashgordon | david-lyle: how would upgrade testing work if we pull the test out of tempest. What does horizon need for upgrade testing in genera;? | 20:58 |
| *** melwitt has quit IRC | 20:58 | |
| *** rfolco has quit IRC | 20:59 | |
| mtreinish | flashgordon: well right now all we do is: http://git.openstack.org/cgit/openstack-dev/grenade/tree/upgrade-horizon and verify you can login before and after | 20:59 |
| mtreinish | if that's still important we can move the login part to javelin pretty easily | 20:59 |
| *** melwitt has joined #openstack-qa | 21:00 | |
| *** harlowja has quit IRC | 21:00 | |
| flashgordon | mtreinish: that is true | 21:00 |
| *** melwitt has quit IRC | 21:00 | |
| david-lyle | not familiar with javelin | 21:01 |
| flashgordon | mtreinish: I still think we should drop the duplicate tempest-full jobs from horizon before the 5th | 21:01 |
| flashgordon | mtreinish: it will free up a non-trivial amount of nodes during crunch time | 21:01 |
| david-lyle | flashgordon: I fully agree with that | 21:01 |
| flashgordon | david-lyle: then take a look at https://review.openstack.org/#/c/164344/1 | 21:02 |
| mtreinish | flashgordon: is that what your patch did? I didn't actually read it. i'm fine with that, I just don't want to delete them until the test is out of tempest | 21:02 |
| mtreinish | flashgordon: ok, yeah I'm fine with dropping the job count | 21:02 |
| flashgordon | mtreinish: ohh yeah not delete no | 21:03 |
| *** dpaterson has quit IRC | 21:03 | |
| flashgordon | just prune | 21:03 |
| mtreinish | david-lyle: javelin is the tool we use for creating resources before an upgrade to verify they survive the upgrade process | 21:04 |
| mtreinish | david-lyle: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/cmd/javelin.py | 21:04 |
| *** harlowja has joined #openstack-qa | 21:06 | |
| *** melwitt has joined #openstack-qa | 21:07 | |
| *** ajo has quit IRC | 21:09 | |
| dkranz | mtreinish: I found why my unit test is failing and have two questions | 21:13 |
| *** marun has quit IRC | 21:14 | |
| mtreinish | dkranz: oh, sorry I forgot to look at it before. Sure shoot | 21:14 |
| *** diegows has quit IRC | 21:14 | |
| dkranz | mtreinish: the reason is because in ConfigFIxture.setUp we set all the identity confs to 'fake_...' | 21:14 |
| dkranz | mtreinish: My code was assuming these values had their defaults from config if I did not set them | 21:15 |
| dkranz | mtreinish: I can fix that but do we want that in the generic fixture class all the tests are using? | 21:15 |
| mtreinish | dkranz: nah, I'd say just override them locally in your tests | 21:15 |
| *** pc_m has quit IRC | 21:16 | |
| dkranz | mtreinish: The second pain was some random behavior that is different when tempest.conf is there. That seems like a bug. | 21:16 |
| mtreinish | because I think a bunch of the other tests are expecting those in there | 21:16 |
| morganfainberg | mtreinish, so.. | 21:16 |
| dkranz | mtreinish: ok, no big deal just painful to find | 21:16 |
| dkranz | mtreinish: the different behavior with tempest.conf there or not is really strange | 21:16 |
| mtreinish | morganfainberg: uh oh what'd I do? | 21:16 |
| dkranz | mtreinish: I looked at the code but did not find anything strange. I guess I will keep looking. With the file there I would get different behavior depending on how many tests I was running! | 21:17 |
| mtreinish | dkranz: yeah that sounds weird, after I migrate home, I'll try dig into that | 21:17 |
| mtreinish | dkranz: do you ahve stack traces or something as an example? | 21:17 |
| dkranz | mtreinish: not stacktraces, but the manifestation was that certain conf values that were set in the test came out different when read by the code | 21:18 |
| morganfainberg | mtreinish, is there a desire to cut back on the number of ddevstacks we use for keystone runs as well? | 21:18 |
| morganfainberg | mtreinish, not just eliminating the compute stuff flashgordon proposed | 21:19 |
| mtreinish | morganfainberg: well they're actually are used by keystone, especially since the postgres one overloads eventlet | 21:19 |
| morganfainberg | mtreinish, we can make any other one do eventlet, that is trivial | 21:19 |
| morganfainberg | mtreinish, it was just stuck there out of convienence | 21:20 |
| mtreinish | morganfainberg: for keystone I just see it as less clear cut, because the different configs are testing different things | 21:20 |
| mtreinish | in the horizon case there was no reason to be doing that | 21:20 |
| morganfainberg | mtreinish, well the only *really* difference is postgres = eventlet, otherwise it's the same config for us. | 21:20 |
| morganfainberg | for the rest. | 21:20 |
| mtreinish | morganfainberg: well except it uses postgres instead of mysql :) | 21:21 |
| mtreinish | which raises the larger question about the value of that | 21:21 |
| morganfainberg | mtreinish, so? :P | 21:21 |
| mtreinish | morganfainberg: heh, I tend to agree, but the last few times we tried to kill that people got there torches and pitchforks | 21:21 |
| *** annashen has quit IRC | 21:22 | |
| *** annashen has joined #openstack-qa | 21:22 | |
| flashgordon | mtreinish: yeah postgres is a third rail these days | 21:23 |
| *** andreykurilin_ has quit IRC | 21:23 | |
| *** andreykurilin_ has joined #openstack-qa | 21:23 | |
| *** ajo has joined #openstack-qa | 21:23 | |
| dkranz | mtreinish: I pushed a working version. I don't see the randomness with tempest.conf present for the code that passes all the tests. I don't think it is worth you looking at this more unless something else goes wrong. | 21:25 |
| dkranz | mtreinish: I may poke some more before quitting today. | 21:25 |
| mtreinish | dkranz: ok | 21:26 |
| *** achanda has quit IRC | 21:27 | |
| flashgordon | err morganfainberg ^^ | 21:27 |
| openstackgerrit | David Kranz proposed openstack/tempest: Add test for the various config cases of is_admin_available() https://review.openstack.org/164225 | 21:27 |
| david-lyle | we in horizon don't do anything, so you don't have to watch over us very closely ;) | 21:27 |
| david-lyle | it's all just an illusion | 21:28 |
| david-lyle | I'm head mirror guy, someone else is in charge of smoke | 21:28 |
| mtreinish | david-lyle: heh, I might hold you to that :) | 21:30 |
| openstackgerrit | Merged openstack/tempest: javelin: implement floating-ip support https://review.openstack.org/163622 | 21:30 |
| mtreinish | EmilienM: ^^^ | 21:30 |
| EmilienM | mtreinish: \o/ we can now use javelin in multi node | 21:31 |
| EmilienM | mtreinish: cool thx! | 21:31 |
| *** cbader has quit IRC | 21:31 | |
| morganfainberg | flashgordon, hehehe | 21:32 |
| dkranz | mtreinish: I have a reprodicible example and am going to file a bug. | 21:32 |
| mtreinish | dkranz: ok cool, if you've got a reproduce that should make it easy to track down | 21:33 |
| *** achanda has joined #openstack-qa | 21:33 | |
| dkranz | mtreinish: when I change the value in tempest.conf to match what the test is setting in the fake config, the test passes. | 21:33 |
| dkranz | mtreinish: but you should look at https://review.openstack.org/#/c/164225/2 when you get a chance to make sure it is not due to me doing something stupid | 21:34 |
| *** jgrimm_ has joined #openstack-qa | 21:34 | |
| *** timello_ has joined #openstack-qa | 21:35 | |
| *** maurosr has quit IRC | 21:35 | |
| *** bknudson has quit IRC | 21:35 | |
| *** timello has quit IRC | 21:36 | |
| *** ekhugen has quit IRC | 21:36 | |
| nikil | hi does any one have idea on parameter in "tenant_networks_reachable=true" tempest.conf. Does true means it do ssh/ping test ? Is it ok to enable this parameter if i run from other network | 21:36 |
| *** ekhugen has joined #openstack-qa | 21:36 | |
| *** jgrimm- has joined #openstack-qa | 21:36 | |
| *** jgrimm has quit IRC | 21:36 | |
| *** maurosr has joined #openstack-qa | 21:36 | |
| *** timello has joined #openstack-qa | 21:37 | |
| *** annashen has quit IRC | 21:39 | |
| mtreinish | nikil: it's how you tell tempest that the default network used fby tenants is not reachable by tempest. It means that certain tests which rely on that either create floating ips or are skipped | 21:39 |
| *** jgrimm_ has quit IRC | 21:39 | |
| *** timello_ has quit IRC | 21:39 | |
| mtreinish | nikil: but that entire section of config is confusing and is about to be refactored to be clearer and more explicit | 21:40 |
| nikil | mtreinish: thanks, basically i run the tempest from other machine and have some floating IP issue. So thanks i will skip this | 21:40 |
| *** bknudson has joined #openstack-qa | 21:41 | |
| *** annashen has joined #openstack-qa | 21:41 | |
| *** andreykurilin_ has quit IRC | 21:41 | |
| *** andreykurilin_ has joined #openstack-qa | 21:42 | |
| *** rushiagr_away is now known as rushiagr | 21:44 | |
| *** marun has joined #openstack-qa | 21:44 | |
| *** dboik has quit IRC | 21:46 | |
| *** annashen has quit IRC | 21:56 | |
| nikil | mtreinish: I try to run tempest like this "bash tools/pretty_tox_serial.sh (?!.*\[.*\bslow\b.*\])((smoke)|(^tempest\.scenario))" as they do in CI system. But it through me error "-bash: !.*\[.*\bslow\b.*\]: event not found" . Any idea on this . | 21:56 |
| *** mpopow has joined #openstack-qa | 21:58 | |
| *** annashen has joined #openstack-qa | 22:00 | |
| mtreinish | nikil: you need quotes around the regex otherwise your shell will try to interpret it | 22:00 |
| *** rwsu has quit IRC | 22:00 | |
| *** Longgeek has joined #openstack-qa | 22:03 | |
| nikil | mtreinish: Yes i tried even like this /bin/bash tools/pretty_tox_serial.sh "(?!.*\[.*\bslow\b.*\])((smoke)|(^tempest\.scenario))" -> this has quotes . Still it doesn't work. Also please let me knwo the blue print of the project because i have diged much of configuration so i can contribute few | 22:03 |
| *** harlowja has quit IRC | 22:04 | |
| *** jecarey has quit IRC | 22:04 | |
| mtreinish | nikil: with quotes you still get the error about event not found? | 22:05 |
| mtreinish | oh, you need to add -- between the script and the regex | 22:05 |
| nikil | mtreinish: I just took from community test log "http://logs.openstack.org/84/163784/4/check/check-tempest-dsvm-neutron/9fe5176/console.html" and this is how they run. I also tried with quots but it doesn't help | 22:05 |
| mtreinish | I think it looks for that (to be fair I don't often call pretty_tox directly, normally through tox | 22:06 |
| *** harlowja has joined #openstack-qa | 22:06 | |
| mtreinish | nikil: like do tox -esmoke-serial | 22:06 |
| *** harlowja has quit IRC | 22:06 | |
| *** harlowja has joined #openstack-qa | 22:06 | |
| mtreinish | nikil: do you mean to contribute to the config docs? I started an etherpad here: https://etherpad.openstack.org/p/tempest-configuration-doc just add a section with your name to that list if you want to add something | 22:07 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:07 |
| nikil | mtreinish: yes i will put what i can contribute | 22:08 |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/155974 failed gate-tempest-dsvm-neutron-full in the gate queue with an unrecognized error | 22:09 |
| mriedem | marun: mestery: do you know if this is a known issue? http://logs.openstack.org/78/163978/1/check/check-tempest-dsvm-neutron-full/792a4e4/logs/screen-q-svc.txt.gz?level=TRACE#_2015-03-13_16_07_36_406 | 22:09 |
| mtreinish | mriedem: ooh, that looks ugly | 22:09 |
| mriedem | fo sho | 22:10 |
| mriedem | mtreinish: of course, it's not 100% fail | 22:10 |
| nikil | mtreinish : so what is the difference between "tox -esmoke-serial" and /bin/bash tools/pretty_tox_serial.sh "(?!.*\[.*\bslow\b.*\])((smoke)|(^tempest\.scenario))" . Because now when i use tox -e is started to install dependencies etc.. I guess the test will run | 22:10 |
| mtreinish | mriedem: heh, of course because db explosions are par for the course | 22:10 |
| mriedem | http://goo.gl/WPuxqU | 22:11 |
| mriedem | basically | 22:11 |
| mriedem | https://bugs.launchpad.net/neutron/+bug/1432065 | 22:11 |
| openstack | Launchpad bug 1432065 in neutron "test_ports fails with "StaleDataError: UPDATE statement on table 'ports' expected to update 1 row(s); 0 were matched."" [Undecided,New] | 22:11 |
| mtreinish | nikil: tox is a command for setting up isolated venvs and running test commands: https://tox.readthedocs.org/en/latest/ | 22:11 |
| mriedem | mtreinish: hm, oslo.db 1.7.0 released yesterday | 22:12 |
| mriedem | this db explosion is new in the last 24 hours | 22:12 |
| mtreinish | nikil: we maintain the job definitions for the gate in the tox.ini in the root of the tempest repo, it automates setting up the env and running the tests | 22:12 |
| mtreinish | nikil: its normally what I use for running tempest | 22:12 |
| mtreinish | mriedem: oh, that's probably it | 22:12 |
| mriedem | mtreinish: or https://github.com/openstack/neutron/commit/5dbb34b56fc42d9c68bf6647910a437a2ad6b29e | 22:12 |
| mtreinish | mriedem: although looking at dhellmann's release notes I doubt it, the release looks pretty simple | 22:13 |
| mriedem | https://review.openstack.org/#/c/149261/ | 22:13 |
| *** cdent has quit IRC | 22:13 | |
| mtreinish | mriedem: yeah that looks more likely | 22:13 |
| mriedem | everytime i try reverting a thing in neutron they tell me it's not whatever i thought it was | 22:15 |
| mriedem | so i've given up | 22:15 |
| *** melwitt has quit IRC | 22:15 | |
| *** melwitt has joined #openstack-qa | 22:16 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:17 |
| mriedem | and there is this http://logs.openstack.org/78/163978/1/check/check-tempest-dsvm-neutron-full/792a4e4/logs/screen-q-svc.txt.gz?level=TRACE#_2015-03-13_16_14_02_595 | 22:17 |
| mriedem | and this http://logs.openstack.org/78/163978/1/check/check-tempest-dsvm-neutron-full/792a4e4/logs/screen-q-svc.txt.gz?level=TRACE#_2015-03-13_16_15_21_786 | 22:17 |
| mriedem | man | 22:17 |
| mtreinish | mestery: ^^^ why so many stack traces :( | 22:18 |
| mriedem | the last one is the killer | 22:18 |
| mriedem | tempest is cleaning up resources and hits the db deadlock | 22:19 |
| *** ihrachyshka has quit IRC | 22:19 | |
| *** melwitt has quit IRC | 22:19 | |
| *** melwitt has joined #openstack-qa | 22:19 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:21 |
| mtreinish | mriedem: yeah that would probably cause an issue, but honestly all 3 look like they should be fatal to me | 22:22 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for neutron db deadlock bug 1432065 https://review.openstack.org/164358 | 22:22 |
| openstack | bug 1432065 in neutron "DBDeadlock: (OperationalError) (1213, 'Deadlock found when trying to get lock; try restarting transaction') 'DELETE FROM ipallocationpools WHERE ipallocationpools.id" [Critical,Confirmed] https://launchpad.net/bugs/1432065 | 22:22 |
| mriedem | ^ | 22:22 |
| mriedem | e-r query | 22:22 |
| mriedem | yeah they all look terrible | 22:22 |
| mriedem | the 2nd one was related to that change i pointed out in neutron yesterday | 22:22 |
| mriedem | i think that was the retry thing they added | 22:22 |
| mriedem | so they trace but continue going | 22:22 |
| mtreinish | mriedem: yeah that's just ugly if it's expected and retries why trace... | 22:24 |
| *** Longgeek has quit IRC | 22:25 | |
| *** mtanino has quit IRC | 22:27 | |
| *** melwitt has quit IRC | 22:28 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:30 |
| *** dimtruck is now known as zz_dimtruck | 22:30 | |
| *** rwsu has joined #openstack-qa | 22:33 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:34 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack-infra/devstack-gate: Configure sub-nodes for Neutron DVR aiopcpu https://review.openstack.org/158525 | 22:35 |
| *** rbak has quit IRC | 22:35 | |
| *** ajo has quit IRC | 22:44 | |
| *** beisner has quit IRC | 22:49 | |
| *** apevec has joined #openstack-qa | 22:50 | |
| *** beisner has joined #openstack-qa | 22:52 | |
| *** tsekiyam_ has joined #openstack-qa | 22:52 | |
| *** mpopow has quit IRC | 22:52 | |
| *** mpopow has joined #openstack-qa | 22:53 | |
| *** tsekiyama has quit IRC | 22:56 | |
| *** tsekiyam_ has quit IRC | 22:56 | |
| *** mpopow has quit IRC | 22:57 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for neutron db deadlock bug 1432065 https://review.openstack.org/164358 | 23:05 |
| openstack | bug 1432065 in neutron "DBDeadlock: (OperationalError) (1213, 'Deadlock found when trying to get lock; try restarting transaction') 'DELETE FROM ipallocationpools WHERE ipallocationpools.id" [Critical,Confirmed] https://launchpad.net/bugs/1432065 | 23:05 |
| *** mpopow has joined #openstack-qa | 23:06 | |
| *** openstackrecheck has quit IRC | 23:13 | |
| *** openstackrecheck has joined #openstack-qa | 23:13 | |
| *** aysyd has quit IRC | 23:13 | |
| *** annashen has quit IRC | 23:18 | |
| *** annashen has joined #openstack-qa | 23:20 | |
| *** david-lyle is now known as david-lyle_afk | 23:29 | |
| *** annashen has quit IRC | 23:29 | |
| *** patrickeast has quit IRC | 23:32 | |
| *** zz_dimtruck is now known as dimtruck | 23:32 | |
| *** rwsu has quit IRC | 23:35 | |
| openstackrecheck | openstack/trove change: https://review.openstack.org/162354 failed gate-trove-python27 in the gate queue with an unrecognized error | 23:37 |
| *** ajmiller_ has quit IRC | 23:37 | |
| *** rwsu has joined #openstack-qa | 23:37 | |
| *** ChuckC has joined #openstack-qa | 23:41 | |
| *** dimtruck is now known as zz_dimtruck | 23:42 | |
| *** melwitt has joined #openstack-qa | 23:49 | |
| *** dimsum__ has quit IRC | 23:52 | |
| *** dims_ has joined #openstack-qa | 23:55 | |
| *** dims_ has quit IRC | 23:57 | |
| *** dims_ has joined #openstack-qa | 23:57 | |
| *** apevec has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!