| *** tkammer has quit IRC | 00:03 | |
| *** mahito has joined #openstack-qa | 00:05 | |
| *** annoyingspore has quit IRC | 00:18 | |
| *** dimsum__ has joined #openstack-qa | 00:34 | |
| *** dmorita has joined #openstack-qa | 00:34 | |
| *** dims_ has joined #openstack-qa | 00:35 | |
| *** dims_ is now known as dims | 00:36 | |
| *** dimsum__ has quit IRC | 00:39 | |
| *** achanda has joined #openstack-qa | 00:46 | |
| *** achanda has quit IRC | 00:51 | |
| *** dimtruck is now known as zz_dimtruck | 00:58 | |
| *** mriedem has quit IRC | 01:18 | |
| *** mlavalle has quit IRC | 01:21 | |
| *** wuhg has joined #openstack-qa | 01:34 | |
| *** cody-somerville has joined #openstack-qa | 01:37 | |
| *** yamamoto_ has joined #openstack-qa | 01:59 | |
| *** diegows has quit IRC | 02:07 | |
| *** mwagner_lap has joined #openstack-qa | 02:35 | |
| *** patri962 has joined #openstack-qa | 02:38 | |
| *** patri674 has joined #openstack-qa | 02:38 | |
| *** patri824 has joined #openstack-qa | 02:38 | |
| *** patri659 has joined #openstack-qa | 02:38 | |
| *** patrickeast___ has quit IRC | 02:38 | |
| *** patrickeast has quit IRC | 02:38 | |
| *** patrickeast__ has quit IRC | 02:38 | |
| *** patrickeast_ has quit IRC | 02:39 | |
| *** patri28 has quit IRC | 02:39 | |
| *** patrickeast has joined #openstack-qa | 02:41 | |
| openstackgerrit | Merged openstack/tempest: Fix response schema for compute aggregate API https://review.openstack.org/157194 | 02:58 |
|---|---|---|
| openstackgerrit | Ken Chen proposed openstack-dev/devstack: Change use_floating_ips to true when neutron is not enabled https://review.openstack.org/159728 | 03:26 |
| openstackgerrit | Jamie Lennox proposed openstack/tempest: Make V3 authentication more strict https://review.openstack.org/159342 | 03:30 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/149927 failed because of: gate-grenade-dsvm-partial-ncpu: https://bugs.launchpad.net/bugs/1313168 | 03:33 |
| openstack | Launchpad bug 1313168 in OpenStack-Gate "paramiko ValueError: Unable to compute factors p and q from exponent d" [Medium,Triaged] | 03:33 |
| *** dims has quit IRC | 03:35 | |
| *** amotoki has joined #openstack-qa | 03:44 | |
| *** achanda has joined #openstack-qa | 03:49 | |
| *** flwang1 has quit IRC | 03:52 | |
| *** achanda has quit IRC | 03:54 | |
| *** david-lyle_afk has quit IRC | 04:02 | |
| *** Sukhdev has joined #openstack-qa | 04:26 | |
| *** patrickeast_ has joined #openstack-qa | 04:27 | |
| *** patrickeast has quit IRC | 04:27 | |
| *** patrickeast_ is now known as patrickeast | 04:27 | |
| *** patri659 has quit IRC | 04:28 | |
| *** patri962 has quit IRC | 04:29 | |
| *** patri824 has quit IRC | 04:29 | |
| *** patri674 has quit IRC | 04:29 | |
| *** patrickeast_ has joined #openstack-qa | 04:30 | |
| *** patrickeast__ has joined #openstack-qa | 04:30 | |
| *** patri484 has joined #openstack-qa | 04:30 | |
| *** patri533 has joined #openstack-qa | 04:30 | |
| *** BharatK has joined #openstack-qa | 04:34 | |
| *** dims has joined #openstack-qa | 04:35 | |
| *** Madan has joined #openstack-qa | 04:39 | |
| *** chandankumar has joined #openstack-qa | 04:40 | |
| *** dims has quit IRC | 04:40 | |
| *** carl_baldwin has joined #openstack-qa | 04:41 | |
| *** coolsvap_ is now known as coolsvap | 04:43 | |
| *** achanda has joined #openstack-qa | 04:49 | |
| *** gamado has joined #openstack-qa | 05:02 | |
| *** chandankumar has quit IRC | 05:06 | |
| *** carl_baldwin has quit IRC | 05:18 | |
| *** yfried|afk has joined #openstack-qa | 05:19 | |
| openstackgerrit | Aishwarya Thangappa proposed openstack-dev/devstack: Added devstack-with-lbaas-v2 installation documentation https://review.openstack.org/158512 | 05:21 |
| *** patri338 has joined #openstack-qa | 05:23 | |
| *** patri781 has joined #openstack-qa | 05:23 | |
| *** patri996 has joined #openstack-qa | 05:23 | |
| *** patri952 has joined #openstack-qa | 05:23 | |
| *** patrickeast has quit IRC | 05:23 | |
| *** patrickeast_ has quit IRC | 05:24 | |
| *** patri484 has quit IRC | 05:24 | |
| *** patri533 has quit IRC | 05:24 | |
| *** patrickeast__ has quit IRC | 05:25 | |
| *** patrickeast has joined #openstack-qa | 05:27 | |
| *** BharatK has quit IRC | 05:39 | |
| *** yfried|afk has quit IRC | 05:46 | |
| *** yfried|afk has joined #openstack-qa | 05:46 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/qa-specs: Add rearrange-nova-response-schemas spec https://review.openstack.org/160228 | 05:48 |
| *** chandankumar has joined #openstack-qa | 05:48 | |
| openstackgerrit | Aishwarya Thangappa proposed openstack-dev/devstack: Added devstack-with-lbaas-v2 installation documentation https://review.openstack.org/158512 | 05:54 |
| *** yfried|afk has quit IRC | 05:55 | |
| openstackgerrit | Jamie Lennox proposed openstack/tempest: Allow full v3 authentication https://review.openstack.org/159342 | 06:00 |
| *** BharatK has joined #openstack-qa | 06:01 | |
| *** achanda has quit IRC | 06:03 | |
| *** deepakcs has joined #openstack-qa | 06:12 | |
| deepakcs | Hi all, i am looking for some info on the tempest volume snapshot in_use testcase, anybody here who can help ? | 06:12 |
| openstackgerrit | Ghanshyam Mann proposed openstack/qa-specs: Add rearrange-nova-response-schemas spec https://review.openstack.org/160228 | 06:16 |
| *** Sukhdev has quit IRC | 06:26 | |
| *** anshul has joined #openstack-qa | 06:27 | |
| *** achanda has joined #openstack-qa | 06:27 | |
| openstackgerrit | Ian Wienand proposed openstack-dev/devstack: Add a simple memory/process tracking service https://review.openstack.org/160232 | 06:30 |
| *** k4n0 has joined #openstack-qa | 06:34 | |
| *** SridharG has joined #openstack-qa | 06:42 | |
| *** pal has joined #openstack-qa | 06:47 | |
| pal | Hi, Can anyone tell me why my nova-api is not able to start? | 06:48 |
| openstackgerrit | Tong Liu proposed openstack/tempest: Fix ICMP code/type in security group rule test https://review.openstack.org/158928 | 06:48 |
| openstackgerrit | Tong Liu proposed openstack/tempest: Fix ICMP code/type in security group rule test https://review.openstack.org/158928 | 06:49 |
| *** yfried|afk has joined #openstack-qa | 06:52 | |
| *** hogepodge has quit IRC | 06:55 | |
| *** yfried|afk has quit IRC | 06:56 | |
| *** yfried has joined #openstack-qa | 06:56 | |
| *** yfried is now known as yfried|afk | 06:56 | |
| *** achanda has quit IRC | 07:13 | |
| *** achanda has joined #openstack-qa | 07:15 | |
| *** gamado has quit IRC | 07:16 | |
| *** gamado has joined #openstack-qa | 07:18 | |
| *** rdekel has joined #openstack-qa | 07:19 | |
| *** achanda has quit IRC | 07:23 | |
| *** steveg has quit IRC | 07:27 | |
| *** steveg has joined #openstack-qa | 07:28 | |
| *** steveg has quit IRC | 07:28 | |
| *** mahito has quit IRC | 07:29 | |
| *** steveg has joined #openstack-qa | 07:29 | |
| *** steveg has quit IRC | 07:29 | |
| *** achanda has joined #openstack-qa | 07:29 | |
| *** steveg has joined #openstack-qa | 07:30 | |
| *** neeti has joined #openstack-qa | 07:31 | |
| openstackgerrit | Ken Chen proposed openstack-dev/devstack: Change use_floating_ips to true when neutron is not enabled https://review.openstack.org/159728 | 07:32 |
| *** yfried|afk is now known as yfried | 07:34 | |
| *** SridharGaddam has joined #openstack-qa | 07:36 | |
| *** itzikb has joined #openstack-qa | 07:39 | |
| openstackgerrit | Sridhar Gaddam proposed openstack-dev/devstack: Fix curl issue when SERVICE_HOST is defined to IPv6 address https://review.openstack.org/158594 | 07:43 |
| *** steveg has quit IRC | 07:48 | |
| *** steveg has joined #openstack-qa | 07:48 | |
| *** hogepodge has joined #openstack-qa | 07:49 | |
| *** gszasz has joined #openstack-qa | 07:49 | |
| *** woodster_ has quit IRC | 07:50 | |
| *** ibiris_away is now known as ibiris | 07:50 | |
| *** pkoniszewski has joined #openstack-qa | 07:50 | |
| *** Sukhdev has joined #openstack-qa | 07:52 | |
| *** achanda has quit IRC | 07:52 | |
| *** achanda has joined #openstack-qa | 07:57 | |
| *** ihrachyshka has joined #openstack-qa | 07:58 | |
| *** ibiris is now known as ibiris_away | 08:00 | |
| *** steveg has quit IRC | 08:05 | |
| *** markus_z has joined #openstack-qa | 08:06 | |
| *** mpaolino has joined #openstack-qa | 08:06 | |
| *** mpaolino has quit IRC | 08:07 | |
| *** mpaolino has joined #openstack-qa | 08:08 | |
| *** ibiris_away is now known as ibiris | 08:08 | |
| *** mpaolino has quit IRC | 08:09 | |
| *** mpaolino has joined #openstack-qa | 08:09 | |
| *** mpaolino has quit IRC | 08:11 | |
| *** mpaolino has joined #openstack-qa | 08:12 | |
| *** mpaolino has quit IRC | 08:14 | |
| *** mpaolino has joined #openstack-qa | 08:15 | |
| *** achanda has quit IRC | 08:16 | |
| *** afazekas has joined #openstack-qa | 08:18 | |
| *** flwang1 has joined #openstack-qa | 08:20 | |
| *** Sukhdev has quit IRC | 08:23 | |
| *** sergsh has joined #openstack-qa | 08:24 | |
| *** sergsh has quit IRC | 08:26 | |
| *** sergsh has joined #openstack-qa | 08:27 | |
| *** mpaolino has quit IRC | 08:38 | |
| *** mpaolino has joined #openstack-qa | 08:38 | |
| *** mpaolino has quit IRC | 08:40 | |
| *** mpaolino has joined #openstack-qa | 08:41 | |
| *** mpaolino has quit IRC | 08:43 | |
| *** mpaolino has joined #openstack-qa | 08:44 | |
| *** salv-orlando has joined #openstack-qa | 08:44 | |
| *** jlanoux has joined #openstack-qa | 08:47 | |
| *** Poornima has joined #openstack-qa | 08:49 | |
| *** gfidente has joined #openstack-qa | 08:50 | |
| *** gfidente has quit IRC | 08:50 | |
| *** gfidente has joined #openstack-qa | 08:50 | |
| *** safchain has joined #openstack-qa | 08:54 | |
| *** yfried is now known as yfried|afk | 08:55 | |
| *** yfried|afk is now known as yfried | 08:57 | |
| *** bkopilov has quit IRC | 08:59 | |
| *** bkopilov has joined #openstack-qa | 08:59 | |
| openstackgerrit | Merged openstack/tempest: Prepare token clients for migration to tempest-lib https://review.openstack.org/159267 | 08:59 |
| *** pkoniszewski has quit IRC | 09:03 | |
| *** pkoniszewski has joined #openstack-qa | 09:03 | |
| openstackgerrit | Markus Zoeller proposed openstack/tempest: Replaces hardcoded volume size by configured value https://review.openstack.org/159878 | 09:03 |
| *** ihrachyshka has quit IRC | 09:07 | |
| *** mpaolino has quit IRC | 09:08 | |
| *** mpaolino has joined #openstack-qa | 09:09 | |
| *** tkammer has joined #openstack-qa | 09:11 | |
| *** mpaolino has quit IRC | 09:13 | |
| *** salv-orlando has quit IRC | 09:13 | |
| *** salv-orlando has joined #openstack-qa | 09:14 | |
| *** mpaolino has joined #openstack-qa | 09:14 | |
| *** dmorita has quit IRC | 09:31 | |
| *** pkoniszewski has quit IRC | 09:35 | |
| *** pkoniszewski has joined #openstack-qa | 09:36 | |
| *** mpaolino has quit IRC | 09:36 | |
| *** mpaolino has joined #openstack-qa | 09:37 | |
| *** sergsh has quit IRC | 09:37 | |
| *** mpaolino has quit IRC | 09:39 | |
| *** mpaolino has joined #openstack-qa | 09:40 | |
| *** mpaolino has quit IRC | 09:41 | |
| *** mpaolino has joined #openstack-qa | 09:42 | |
| *** mpaolino has quit IRC | 09:43 | |
| *** mpaolino has joined #openstack-qa | 09:43 | |
| *** ajo has joined #openstack-qa | 09:44 | |
| *** e0ne has joined #openstack-qa | 09:49 | |
| *** sergsh has joined #openstack-qa | 09:49 | |
| *** mpaolino has quit IRC | 09:51 | |
| *** mpaolino has joined #openstack-qa | 09:51 | |
| *** mpaolino has quit IRC | 09:55 | |
| openstackgerrit | Thalabathy Venkatesan proposed openstack/tempest: test_in_tenant_no_securitygroup https://review.openstack.org/158293 | 09:55 |
| *** mpaolino has joined #openstack-qa | 09:55 | |
| *** nmagnezi has joined #openstack-qa | 09:59 | |
| *** nmagnezi has quit IRC | 10:00 | |
| *** jordanP has joined #openstack-qa | 10:05 | |
| *** chandankumar_ has joined #openstack-qa | 10:06 | |
| *** mpaolino has quit IRC | 10:06 | |
| *** mpaolino has joined #openstack-qa | 10:07 | |
| *** chandankumar has quit IRC | 10:08 | |
| *** salv-orlando has quit IRC | 10:11 | |
| *** bkopilov has quit IRC | 10:11 | |
| *** mpaolino has quit IRC | 10:13 | |
| afazekas | sdague, https://review.openstack.org/#/c/148679/ | 10:13 |
| openstackrecheck | openstack/cinder change: https://review.openstack.org/153932 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1311066 https://bugs.launchpad.net/bugs/1425310 | 10:13 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 10:13 |
| openstack | Launchpad bug 1425310 in Cinder "tgtadm: this access control rule does not exist - results in timeout waiting for volume to attach (stuck in detach) or detach" [High,Confirmed] | 10:13 |
| *** mpaolino has joined #openstack-qa | 10:13 | |
| *** bkopilov has joined #openstack-qa | 10:16 | |
| *** mpaolino has quit IRC | 10:16 | |
| *** mpaolino has joined #openstack-qa | 10:16 | |
| *** rvcehimanshu has joined #openstack-qa | 10:17 | |
| rvcehimanshu | hello... How do pass/load a set of test while running testr | 10:18 |
| rvcehimanshu | in a file format | 10:18 |
| rvcehimanshu | using --load-test | 10:18 |
| deepakcs | afazekas, I am looking for someone to help respond to this : http://lists.openstack.org/pipermail/openstack-dev/2015-February/057948.html | 10:21 |
| *** tosky has joined #openstack-qa | 10:22 | |
| *** rvcehimanshu has quit IRC | 10:27 | |
| afazekas | deepakcs: at the first looks likes you need to configure the nova admin/service user in the cinder.conf (or create an another service user, but it usually we just using the same service user) | 10:28 |
| afazekas | Sit is looks like option 1 | 10:28 |
| deepakcs | afazekas, yeah, but putting nova admin tenant, user and passwd in cinder.conf isn't very good, .conf gets bundled in logs and shared with people so i was hoping that elevating the user to admin role, running the test, de-elevating back to normal user would be better approach ? | 10:29 |
| *** bkopilov has quit IRC | 10:29 | |
| *** bkopilov has joined #openstack-qa | 10:30 | |
| afazekas | deepakcs: no, this op expected to work with regular user account | 10:30 |
| afazekas | we does the same thing with neutron-nova notifications | 10:30 |
| deepakcs | afazekas, no, nova's policy.json says it needs admin_api, so we need a user with admin role iiuc | 10:30 |
| afazekas | the service users are admins | 10:32 |
| afazekas | or at least they have lot of admin capability | 10:32 |
| afazekas | If the service user is not good enough please change the policy | 10:32 |
| deepakcs | afazekas, ok, so when tempest run the full suite, do we have a service user w/ admin role ? | 10:33 |
| deepakcs | afazekas, can u point me to the testcase of neutron where similar stuff is done ? | 10:33 |
| afazekas | the service user created by devstack | 10:33 |
| afazekas | tempest `never` touches them | 10:33 |
| afazekas | at least not explicitly | 10:33 |
| deepakcs | afazekas, i don't have a devstack setup handy, but i thought devstack created admin and demo users. When you say service user, you mean user with name `service` ? | 10:34 |
| afazekas | create_service_user "nova" "admin" | 10:34 |
| afazekas | It also created a service user called 'nova' | 10:35 |
| *** mpaolino has quit IRC | 10:35 | |
| afazekas | probably you need to configure the nova user's credentials | 10:35 |
| *** mpaolino has joined #openstack-qa | 10:35 | |
| deepakcs | afazekas, thanks, will check it out. | 10:36 |
| *** pkoniszewski has quit IRC | 10:36 | |
| afazekas | In theory we could create multiple service user in nova , one for nova-cinder and one for nova-neutron (also one for neutron-nova) , but we alread have too many users .. | 10:36 |
| *** mpaolino has quit IRC | 10:37 | |
| *** mpaolino has joined #openstack-qa | 10:37 | |
| deepakcs | afazekas, " create_service_user "nova" "admin"" <-- where did you find this ? | 10:38 |
| deepakcs | afazekas, i just did a recursive grep in devstack and i dont' see it, unless this was added very recently | 10:39 |
| afazekas | devstack lib/nova | 10:39 |
| afazekas | pls wait I try to lookup the neutron-nova related code parts on github | 10:40 |
| deepakcs | afazekas, ok | 10:40 |
| *** aix has joined #openstack-qa | 10:41 | |
| *** yfried is now known as yfried|afk | 10:41 | |
| *** rvcehimanshu has joined #openstack-qa | 10:41 | |
| afazekas | https://github.com/openstack-dev/devstack/blob/master/lib/neutron#L1050 | 10:41 |
| afazekas | https://github.com/openstack-dev/devstack/blob/master/lib/nova#L361 | 10:43 |
| deepakcs | afazekas, thanks, 'guess we can use the nova service user as ti has admin role, for the volume In_use snapshot tempest test | 10:43 |
| deepakcs | afazekas, before i was grepping in an older devstack code | 10:44 |
| openstackgerrit | Ian Wienand proposed openstack-dev/devstack: Add a simple memory/process tracking service https://review.openstack.org/160232 | 10:45 |
| *** bkopilov has quit IRC | 10:46 | |
| *** ihrachyshka has joined #openstack-qa | 10:46 | |
| afazekas | deepakcs: We should not change anything on tempest side, the cinder service needs to be configured. The parameters mentioned here https://review.openstack.org/#/c/156940/1 needs to be set in the cinder.conf , as we set nova_admin_username in the neutron.conf | 10:49 |
| deepakcs | afazekas, yes I work with BharatK (author of that patch) , but the last time we tried we put in the admin user creds, which isn't good, since we are exposing the passwd. Didn't knew about the service user having admin role capability | 10:50 |
| afazekas | I'll tell you secret, the cinder.conf already contains an admin service user credentials which would be accepted by all service as an admin account :) | 10:52 |
| *** EmilienM is now known as EmilienM|afk | 10:54 | |
| *** Poornima has quit IRC | 10:54 | |
| deepakcs | afazekas, are u referring to "project_name = service" in cinder.conf ? | 10:55 |
| deepakcs | afazekas, thats havign service role, not admin | 10:56 |
| afazekas | looks like the policies a little stricter then before .., the `service` role was handled as the `admin` role in many places | 10:58 |
| deepakcs | afazekas, no, if u look at "lib/keystone +425" it default to service role, unless 2nd arg is passed as admin (which only few like nova, heat, ironic does). lib/cinder doesn't pass admin as 2nd arg | 11:00 |
| *** yfried|afk is now known as yfried | 11:02 | |
| *** rvcehimanshu has quit IRC | 11:02 | |
| *** rvcehimanshu has joined #openstack-qa | 11:03 | |
| *** tochi has quit IRC | 11:04 | |
| *** apevec has joined #openstack-qa | 11:04 | |
| amotoki | hi, I am now trying to change the default fields displayed in neutron xxxx-list command. tempest has strict test for a list of fields | 11:05 |
| amotoki | an example is http://git.openstack.org/cgit/openstack/tempest/tree/tempest/cli/simple_read_only/network/test_neutron.py#n158 | 11:05 |
| amotoki | what is our practice when we change the default fields in xxx-list command? | 11:06 |
| afazekas | deepakcs, yes, in the past it was handled differently https://github.com/openstack-dev/devstack/blob/stable/juno/lib/cinder#L332 | 11:06 |
| *** jhenner has joined #openstack-qa | 11:07 | |
| *** mpaolino has quit IRC | 11:09 | |
| *** mpaolino has joined #openstack-qa | 11:09 | |
| *** mpaolino has quit IRC | 11:11 | |
| *** mpaolino has joined #openstack-qa | 11:11 | |
| *** yfried is now known as yfried|afk | 11:12 | |
| deepakcs | afazekas, good 2 know, its not that old too, its stable/juno, so this change seems to be recent | 11:13 |
| openstackgerrit | Julien Danjou proposed openstack-dev/hacking: Add a check for staticmethod https://review.openstack.org/151952 | 11:14 |
| openstackgerrit | Julien Danjou proposed openstack-dev/hacking: tests: allow to run the same test for different code https://review.openstack.org/152108 | 11:14 |
| deepakcs | afazekas, if u can reply to that thread, it would help to close it | 11:15 |
| *** mpaolino has quit IRC | 11:17 | |
| *** achanda has joined #openstack-qa | 11:17 | |
| *** mpaolino has joined #openstack-qa | 11:17 | |
| *** rvcehimanshu has quit IRC | 11:19 | |
| *** achanda has quit IRC | 11:21 | |
| *** yamamoto_ has quit IRC | 11:22 | |
| *** yfried|afk is now known as yfried | 11:24 | |
| afazekas | deepakcs, Can you confirm the assisted snapshot also expected to work with nfs ? | 11:26 |
| *** pc_m has joined #openstack-qa | 11:31 | |
| *** mpaolino has quit IRC | 11:34 | |
| *** mpaolino has joined #openstack-qa | 11:34 | |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Support v3 in credential providers and subclasses https://review.openstack.org/133029 | 11:35 |
| *** mpaolino has quit IRC | 11:36 | |
| *** mpaolino has joined #openstack-qa | 11:36 | |
| *** rvcehimanshu has joined #openstack-qa | 11:37 | |
| *** mpaolino has quit IRC | 11:39 | |
| *** mpaolino has joined #openstack-qa | 11:40 | |
| deepakcs | afazekas, i remember eharney working on it. I am not sure if its ready to use. Haven't followed nfs in recent times | 11:46 |
| afazekas | deepakcs, Do you know is anybody working on a feature like this: https://blueprints.launchpad.net/nova/+spec/glusterfs-image-backend | 11:48 |
| *** acfleury has joined #openstack-qa | 11:49 | |
| afazekas | The fuse driver for ephemeral is not fast enough.., the qemu/libvirt ... driver just implemented for cinder volumes | 11:49 |
| *** mpaolino has quit IRC | 11:51 | |
| *** mpaolino has joined #openstack-qa | 11:51 | |
| deepakcs | afazekas, so this is a patch to use gluster's libgfapi to store ephermal images ? | 11:52 |
| afazekas | deepakcs: yes | 11:52 |
| *** mpaolino has quit IRC | 11:53 | |
| *** mpaolino has joined #openstack-qa | 11:53 | |
| *** yamamoto has joined #openstack-qa | 11:53 | |
| *** rvcehimanshu has quit IRC | 11:54 | |
| *** mpaolino has quit IRC | 11:55 | |
| *** mpaolino has joined #openstack-qa | 11:55 | |
| *** rvcehimanshu has joined #openstack-qa | 11:56 | |
| *** mpaolino has quit IRC | 11:58 | |
| *** mpaolino has joined #openstack-qa | 11:58 | |
| *** mpaolino has quit IRC | 12:01 | |
| *** mpaolino has joined #openstack-qa | 12:01 | |
| *** e0ne is now known as e0ne_ | 12:03 | |
| *** yamamoto has quit IRC | 12:04 | |
| *** pal has quit IRC | 12:06 | |
| *** coolsvap is now known as coolsvap_ | 12:06 | |
| *** dims has joined #openstack-qa | 12:09 | |
| *** mpaolino has quit IRC | 12:09 | |
| *** mpaolino has joined #openstack-qa | 12:09 | |
| *** pal has joined #openstack-qa | 12:10 | |
| *** mpaolino has quit IRC | 12:11 | |
| *** mpaolino has joined #openstack-qa | 12:11 | |
| *** e0ne_ has quit IRC | 12:13 | |
| *** coolsvap_ is now known as coolsvap | 12:15 | |
| *** pal has quit IRC | 12:23 | |
| *** BharatK has quit IRC | 12:24 | |
| openstackgerrit | Filip HubÃk proposed openstack/tempest: Verify the capabilities of the domain administrator https://review.openstack.org/159915 | 12:25 |
| *** e0ne has joined #openstack-qa | 12:27 | |
| *** yamamoto has joined #openstack-qa | 12:29 | |
| *** aysyd has joined #openstack-qa | 12:29 | |
| *** diegows has joined #openstack-qa | 12:41 | |
| *** bkopilov has joined #openstack-qa | 12:43 | |
| *** dkranz has joined #openstack-qa | 12:43 | |
| *** rvcehimanshu has quit IRC | 12:46 | |
| *** k4n0 has quit IRC | 12:46 | |
| afazekas | sdague, jogo : looks like we have a situation again (stable/ice, swift-functional): http://logs.openstack.org/95/159695/2/check/check-swift-dsvm-functional/c821d8a/console.html | 12:47 |
| *** rvcehimanshu has joined #openstack-qa | 12:47 | |
| *** yamamoto has quit IRC | 12:50 | |
| *** gamado has quit IRC | 12:52 | |
| *** deepakcs has quit IRC | 12:57 | |
| *** woodster_ has joined #openstack-qa | 12:59 | |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/151381 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1373513 https://bugs.launchpad.net/bugs/1376941 https://bugs.launchpad.net/bugs/1311066 | 13:01 |
| openstack | Launchpad bug 1373513 in Cinder "Lvm hang during tempest tests" [Critical,In progress] - Assigned to John Griffith (john-griffith) | 13:01 |
| openstack | Launchpad bug 1352498 in Cinder "duplicate for #1376941 Tempest fails spectacularly with "VolumeType c39f18bf-2408-4377-98f6-ef2839d6f70e deletion failed, VolumeType in use."" [Undecided,New] | 13:01 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 13:01 |
| *** dkranz has quit IRC | 13:04 | |
| HenryG | mtreinish: ping | 13:04 |
| *** weshay has joined #openstack-qa | 13:09 | |
| *** gamado has joined #openstack-qa | 13:11 | |
| *** dkranz has joined #openstack-qa | 13:14 | |
| *** mpaolino has quit IRC | 13:16 | |
| *** rvcehimanshu has quit IRC | 13:16 | |
| *** mpaolino has joined #openstack-qa | 13:16 | |
| *** rfolco has joined #openstack-qa | 13:17 | |
| *** rvcehimanshu has joined #openstack-qa | 13:17 | |
| *** yfried has quit IRC | 13:18 | |
| *** e0ne is now known as e0ne_ | 13:20 | |
| *** nithyag__ has joined #openstack-qa | 13:22 | |
| *** gamado has quit IRC | 13:23 | |
| *** nithyag_ has quit IRC | 13:24 | |
| *** yamamoto has joined #openstack-qa | 13:24 | |
| *** e0ne_ is now known as e0ne | 13:24 | |
| andreaf | mtreinish: hi - I fixed https://review.openstack.org/#/c/133029/ so it passes now v3 jobs as well, and I restored "name" in Accounts __init__ so that it does not conflict with your change - would you mind having a look? | 13:35 |
| *** marrusl has joined #openstack-qa | 13:39 | |
| *** salv-orlando has joined #openstack-qa | 13:43 | |
| *** pal has joined #openstack-qa | 13:44 | |
| *** bknudson has quit IRC | 13:46 | |
| *** chandankumar_ has quit IRC | 13:46 | |
| *** mpaolino has quit IRC | 13:46 | |
| *** rvcehimanshu has quit IRC | 13:46 | |
| *** mpaolino has joined #openstack-qa | 13:47 | |
| *** rvcehimanshu has joined #openstack-qa | 13:47 | |
| *** mpaolino has quit IRC | 13:48 | |
| *** mpaolino has joined #openstack-qa | 13:49 | |
| *** dboik has joined #openstack-qa | 13:54 | |
| *** amotoki has quit IRC | 13:55 | |
| *** yamamoto has quit IRC | 13:59 | |
| *** tkammer has quit IRC | 14:03 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/devstack: WIP: Try oslo.log from trunk to see if it works fine https://review.openstack.org/160149 | 14:03 |
| *** neeti has quit IRC | 14:04 | |
| openstackgerrit | Emily Hugenbruch proposed openstack/tempest: Split resource_setup for thirdparty tests https://review.openstack.org/159894 | 14:06 |
| *** cdent has joined #openstack-qa | 14:06 | |
| *** afazekas has quit IRC | 14:06 | |
| openstackgerrit | afazekas proposed openstack/tempest: Revert "Turn off sitepackages" https://review.openstack.org/160352 | 14:06 |
| *** rvcehimanshu has quit IRC | 14:08 | |
| *** dustins has joined #openstack-qa | 14:09 | |
| *** bknudson has joined #openstack-qa | 14:10 | |
| *** yfried has joined #openstack-qa | 14:10 | |
| *** mriedem has joined #openstack-qa | 14:10 | |
| *** mriedem has quit IRC | 14:11 | |
| *** mriedem has joined #openstack-qa | 14:11 | |
| *** tkammer has joined #openstack-qa | 14:15 | |
| *** yamamoto has joined #openstack-qa | 14:17 | |
| *** mpaolino has quit IRC | 14:19 | |
| *** mpaolino has joined #openstack-qa | 14:19 | |
| *** Madan has quit IRC | 14:21 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/devstack: WIP: Try oslo.log from trunk to see if it works fine https://review.openstack.org/160149 | 14:21 |
| *** afazekas has joined #openstack-qa | 14:23 | |
| *** dims has quit IRC | 14:25 | |
| *** dims has joined #openstack-qa | 14:26 | |
| *** mpaolino has quit IRC | 14:28 | |
| *** mpaolino has joined #openstack-qa | 14:28 | |
| *** mattt has joined #openstack-qa | 14:30 | |
| *** mpaolino has quit IRC | 14:30 | |
| *** mpaolino has joined #openstack-qa | 14:31 | |
| *** rbak_ has quit IRC | 14:32 | |
| *** mpaolino has quit IRC | 14:35 | |
| *** mpaolino has joined #openstack-qa | 14:35 | |
| openstackgerrit | Emily Hugenbruch proposed openstack/tempest: Split resource_setup for scenario tests https://review.openstack.org/159256 | 14:36 |
| *** mpaolino has quit IRC | 14:38 | |
| *** mpaolino has joined #openstack-qa | 14:38 | |
| *** mpaolino has quit IRC | 14:40 | |
| *** mpaolino has joined #openstack-qa | 14:41 | |
| *** afazekas has quit IRC | 14:41 | |
| *** SridharG has left #openstack-qa | 14:42 | |
| *** mpaolino has quit IRC | 14:43 | |
| *** mpaolino has joined #openstack-qa | 14:43 | |
| openstackgerrit | Merged openstack/tempest: Remove python-novaclient from requirements.txt https://review.openstack.org/157953 | 14:45 |
| *** mpaolino has quit IRC | 14:46 | |
| *** mpaolino has joined #openstack-qa | 14:47 | |
| *** dboik has quit IRC | 14:48 | |
| *** mpaolino has quit IRC | 14:48 | |
| *** dboik has joined #openstack-qa | 14:48 | |
| openstackgerrit | Andrea Frittoli proposed openstack-infra/devstack-gate: Get default branch from feature.yaml https://review.openstack.org/160366 | 14:48 |
| *** mpaolino has joined #openstack-qa | 14:49 | |
| *** jecarey has joined #openstack-qa | 14:49 | |
| *** mpaolino has quit IRC | 14:52 | |
| *** mpaolino has joined #openstack-qa | 14:52 | |
| *** mpaolino has quit IRC | 14:54 | |
| *** mpaolino has joined #openstack-qa | 14:54 | |
| *** tkammer has quit IRC | 14:54 | |
| *** mpaolino has joined #openstack-qa | 14:55 | |
| openstackrecheck | Console logs not available after 13:23s for gate-nova-tox-functional 159824,7,456e87b | 14:57 |
| *** yamamoto has quit IRC | 14:59 | |
| *** toure has joined #openstack-qa | 14:59 | |
| ihrachyshka | sdague, hey. a question on recent move to py34 | 15:00 |
| ihrachyshka | sdague, seems like icehouse oslo-incubator is broken due to it | 15:00 |
| ihrachyshka | no py34 interpreter found | 15:00 |
| ihrachyshka | I guess because jobs do run on smth older than trusty | 15:00 |
| ihrachyshka | right, it's bare-precise | 15:00 |
| ihrachyshka | do you know how to fix it correctly? | 15:01 |
| *** _dpaterson has joined #openstack-qa | 15:04 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/158355 failed because of: gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1251521 https://bugs.launchpad.net/bugs/1311066 | 15:04 |
| openstack | Launchpad bug 1251521 in OpenStack Compute (nova) "Volume detach in tempest fails because libvirt refuses connections" [Undecided,Invalid] | 15:04 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 15:04 |
| *** malini has joined #openstack-qa | 15:06 | |
| *** yamamoto has joined #openstack-qa | 15:07 | |
| *** tkammer has joined #openstack-qa | 15:07 | |
| *** yfried has quit IRC | 15:11 | |
| mtreinish | HenryG: pong | 15:12 |
| mtreinish | andreaf: sure I'll tkae a look right now | 15:13 |
| *** malini has left #openstack-qa | 15:13 | |
| andreaf | mtreinish: thanks | 15:13 |
| HenryG | mtreinish: https://review.openstack.org/159994 | 15:13 |
| mtreinish | HenryG: heh, oops looks like we forgot about the group for the CONF var :) | 15:14 |
| HenryG | mtreinish: yes, my bad, I should have checked the usage | 15:14 |
| *** mtanino has joined #openstack-qa | 15:14 | |
| *** rbak has joined #openstack-qa | 15:16 | |
| *** Ananci has joined #openstack-qa | 15:17 | |
| *** afazekas has joined #openstack-qa | 15:19 | |
| openstackgerrit | Kirill Shileev proposed openstack/tempest: Scenario test to check IPv6 mupltiprefix feature Starts 2 VMs on network with 2 IPv6 subnets and checks cross connectivity. https://review.openstack.org/157816 | 15:21 |
| *** chandankumar_ has joined #openstack-qa | 15:25 | |
| *** e0ne is now known as e0ne_ | 15:28 | |
| *** e0ne_ is now known as e0ne | 15:28 | |
| *** pal has quit IRC | 15:28 | |
| *** chandankumar_ has quit IRC | 15:31 | |
| *** yamamoto has quit IRC | 15:32 | |
| *** beekneemech is now known as bnemec | 15:32 | |
| *** achanda has joined #openstack-qa | 15:33 | |
| *** yamahata has joined #openstack-qa | 15:35 | |
| *** zz_jgrimm is now known as jgrimm | 15:35 | |
| *** jlvillal|afk has quit IRC | 15:37 | |
| *** zz_dimtruck is now known as dimtruck | 15:38 | |
| *** achanda has quit IRC | 15:44 | |
| mtreinish | andreaf: reviewed LGTM, just a few comments | 15:44 |
| *** chandankumar_ has joined #openstack-qa | 15:48 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/devstack: WIP: Try oslo.log from trunk to see if it works fine https://review.openstack.org/160149 | 15:49 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add microsecond columns to the test_runs table https://review.openstack.org/160057 | 15:53 |
| *** dustins_ has joined #openstack-qa | 15:54 | |
| *** salv-orlando has quit IRC | 15:55 | |
| *** salv-orlando has joined #openstack-qa | 15:55 | |
| *** dustins has quit IRC | 15:57 | |
| openstackgerrit | gordon chung proposed openstack-dev/devstack: elasticsearch event support in ceilometer https://review.openstack.org/155104 | 15:59 |
| *** anshul has quit IRC | 15:59 | |
| *** salv-orlando has quit IRC | 16:00 | |
| *** FatDarrel has joined #openstack-qa | 16:01 | |
| *** FatDarrel is now known as phd2 | 16:02 | |
| *** coolsvap is now known as coolsvap_ | 16:05 | |
| *** hemnafk is now known as hemna | 16:06 | |
| *** itzikb has quit IRC | 16:06 | |
| andreaf | mtreinish: it should be enough to pass name into Accounts, or shall I still pass it to NotLockingAccounts as well? | 16:08 |
| mtreinish | andreaf: well the name isn't used for logging with the NotLockingAccounts so it shouldn't matter whether you pass it in there or not | 16:10 |
| mtreinish | andreaf: but I think passing it in to NotLockingAccounts might be useful in the future if we want to improve logging | 16:11 |
| mtreinish | I'll leave it up to you | 16:12 |
| openstackgerrit | Andrea Frittoli proposed openstack/tempest: Support v3 in credential providers and subclasses https://review.openstack.org/133029 | 16:12 |
| andreaf | mtreinish: ok - I'll pass it there as well | 16:12 |
| andreaf | mtreinish: regarding the microseconds in subunit2sql, do you have an idea what the impact might be on a DB the size of infra? if you had a chance to test the migration and the impact is small, I think it's not worth engineering a solution to avoid re-writing microseconds during the migration | 16:14 |
| mtreinish | andreaf: I made a local backup of the infra db over the weekend to test that :) | 16:15 |
| mtreinish | I was going to try the migration today and see what happens | 16:15 |
| *** chandankumar_ has quit IRC | 16:15 | |
| *** carl_baldwin has joined #openstack-qa | 16:19 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/devstack: WIP: Try oslo.log from trunk to see if it works fine https://review.openstack.org/160149 | 16:21 |
| openstackgerrit | Martin Pavlásek proposed openstack/tempest: Added test of default domain in Keystone https://review.openstack.org/31782 | 16:21 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/devstack: WIP: Try oslo.log from trunk to see if it works fine https://review.openstack.org/160149 | 16:22 |
| openstackgerrit | Andrea Frittoli proposed openstack-infra/devstack-gate: Get default branch from feature.yaml https://review.openstack.org/160366 | 16:22 |
| *** tsekiyama has joined #openstack-qa | 16:23 | |
| *** carl_baldwin_ has joined #openstack-qa | 16:24 | |
| *** rvcehimanshu has joined #openstack-qa | 16:24 | |
| *** patrickeast has quit IRC | 16:25 | |
| *** patri952 has quit IRC | 16:25 | |
| *** patri781 has quit IRC | 16:25 | |
| *** patri338 has quit IRC | 16:25 | |
| *** patri996 has quit IRC | 16:25 | |
| *** carl_baldwin has quit IRC | 16:26 | |
| *** carl_baldwin_ is now known as carl_baldwin | 16:26 | |
| *** carl_baldwin has quit IRC | 16:32 | |
| *** boris-42 has quit IRC | 16:32 | |
| *** mrodden has quit IRC | 16:33 | |
| *** mrodden has joined #openstack-qa | 16:35 | |
| *** Noice has joined #openstack-qa | 16:35 | |
| *** phd2 has quit IRC | 16:36 | |
| Noice | Hey, Why does the install guide have you configure nova on the controller if neutron is being used? | 16:36 |
| Noice | and does nova ahve to be installed or do I just have to make sure the config file is present? | 16:37 |
| mrodden | you need nova-api and nova-conductor, etc on the controller | 16:38 |
| mrodden | Noice, ^ | 16:38 |
| *** carl_baldwin has joined #openstack-qa | 16:38 | |
| mrodden | for neutron, you'll be swapping nova-network with neutron | 16:38 |
| mrodden | all the other nova-* services are still important | 16:39 |
| *** wuhg has quit IRC | 16:39 | |
| Noice | why are they considered legacy if I sitll need them for compute? | 16:41 |
| *** melwitt has joined #openstack-qa | 16:41 | |
| *** apevec has quit IRC | 16:42 | |
| *** rdekel has quit IRC | 16:42 | |
| mrodden | nova-network would be | 16:42 |
| mrodden | would be legacy* that is | 16:42 |
| mrodden | i dont think the others are... | 16:42 |
| *** ymorkovn has quit IRC | 16:44 | |
| Noice | ah ok | 16:49 |
| *** diegows has quit IRC | 16:51 | |
| openstackgerrit | Aishwarya Thangappa proposed openstack-dev/devstack: Added devstack-with-lbaas-v2 installation documentation https://review.openstack.org/158512 | 16:52 |
| *** yfried has joined #openstack-qa | 16:53 | |
| *** markus_z has quit IRC | 16:54 | |
| *** dustins_ has quit IRC | 16:55 | |
| *** e0ne is now known as e0ne_ | 16:55 | |
| *** e0ne_ is now known as e0ne | 16:56 | |
| Noice | so is there part of the instlal guide that documents what nova compenents need to be installed and on what node? The install guide covers what services to restart but I want to make sure I'm not missing something before I proceed | 16:56 |
| *** jaypipes has joined #openstack-qa | 17:00 | |
| *** ChanServ sets mode: +o jaypipes | 17:00 | |
| *** jordanP has quit IRC | 17:02 | |
| *** mlavalle has joined #openstack-qa | 17:03 | |
| *** jordanP has joined #openstack-qa | 17:03 | |
| *** diegows has joined #openstack-qa | 17:04 | |
| *** Sukhdev has joined #openstack-qa | 17:05 | |
| *** yfried has quit IRC | 17:05 | |
| *** jlanoux has quit IRC | 17:06 | |
| *** rvcehimanshu has quit IRC | 17:06 | |
| *** sergsh has quit IRC | 17:09 | |
| *** rwsu has joined #openstack-qa | 17:09 | |
| *** e0ne has quit IRC | 17:14 | |
| *** yamahata has quit IRC | 17:16 | |
| *** salv-orlando has joined #openstack-qa | 17:16 | |
| jogo | afazekas: not too surprised I didn't do the full pinning of stable/icehouse | 17:18 |
| afazekas | jogo: looks like the venv does not have python-keystoneclient | 17:21 |
| afazekas | maybe the tox.ini needs to ba changed to use site packages , or the test-requirements should be extended by python-keystoneclient | 17:22 |
| afazekas | The swift functional job on ice is re-enabled recently, but I just see one failing change .. | 17:22 |
| *** safchain has quit IRC | 17:24 | |
| afazekas | jogo: BTW, sitepackages=True expected to use the system packages when it is within the requirements version range, otherwise it installs a different version into the venv | 17:25 |
| *** vigneshvar has joined #openstack-qa | 17:25 | |
| *** patrickeast has joined #openstack-qa | 17:26 | |
| *** carl_baldwin has quit IRC | 17:26 | |
| *** adam_g_out is now known as adam_g | 17:27 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add microsecond columns to the test_runs table https://review.openstack.org/160057 | 17:29 |
| *** dboik_ has joined #openstack-qa | 17:29 | |
| jogo | afazekas: that isn't sufficient, the risk is other things breaking etc, | 17:31 |
| *** annashen has joined #openstack-qa | 17:31 | |
| jogo | afazekas: err, rather. the version range isn't always accurate | 17:32 |
| jogo | afazekas: and full isolation is a really nice property | 17:32 |
| jogo | afazekas: and why revert | 17:32 |
| afazekas | jogo: in this case we just need to change the version range in the master tempest only | 17:32 |
| jogo | afazekas: so maybe, but put this in the commit message | 17:32 |
| *** dboik has quit IRC | 17:33 | |
| openstackgerrit | Matthew Treinish proposed openstack/tempest-lib: Add api docs to the rest client https://review.openstack.org/158017 | 17:35 |
| *** vigneshvar has quit IRC | 17:37 | |
| * afazekas end of day.. | 17:37 | |
| *** afazekas is now known as _afezekas|pub | 17:37 | |
| *** patrickeast_ has joined #openstack-qa | 17:44 | |
| *** patrickeast__ has joined #openstack-qa | 17:44 | |
| *** patri136 has joined #openstack-qa | 17:44 | |
| *** patri505 has joined #openstack-qa | 17:44 | |
| *** yamahata has joined #openstack-qa | 17:44 | |
| *** ChuckC has quit IRC | 17:45 | |
| *** vigneshvar has joined #openstack-qa | 17:46 | |
| *** tkammer has quit IRC | 17:50 | |
| *** ihrachyshka has quit IRC | 17:50 | |
| *** carl_baldwin has joined #openstack-qa | 17:53 | |
| *** yfried has joined #openstack-qa | 17:54 | |
| openstackgerrit | Zengfa Gao proposed openstack/tempest: Add API tests for Neutron subnetpool https://review.openstack.org/159644 | 18:03 |
| *** boris-42 has joined #openstack-qa | 18:05 | |
| *** gszasz has quit IRC | 18:06 | |
| openstackgerrit | Emily Hugenbruch proposed openstack/tempest: Split resource_setup for scenario tests https://review.openstack.org/159256 | 18:09 |
| *** dustins has joined #openstack-qa | 18:09 | |
| *** rbak has quit IRC | 18:10 | |
| *** _afezekas|pub has quit IRC | 18:10 | |
| *** rbak has joined #openstack-qa | 18:11 | |
| *** dboik_ has quit IRC | 18:12 | |
| *** dboik has joined #openstack-qa | 18:12 | |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Revert "Revert "Make neutron the default in devstack"" https://review.openstack.org/159216 | 18:13 |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/160233 failed because of: gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1311066 | 18:14 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 18:14 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Revert "Revert "Make neutron the default in devstack"" https://review.openstack.org/159216 | 18:15 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Support for single interface Neutron networking https://review.openstack.org/158424 | 18:15 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Set a default for OVS_PHYSICAL_BRIDGE https://review.openstack.org/159215 | 18:15 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Revert "Revert "Make neutron the default in devstack"" https://review.openstack.org/159216 | 18:17 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Support for single interface Neutron networking https://review.openstack.org/158424 | 18:17 |
| openstackgerrit | Sean M. Collins proposed openstack-dev/devstack: Set a default for OVS_PHYSICAL_BRIDGE https://review.openstack.org/159215 | 18:17 |
| openstackgerrit | Kirill Shileev proposed openstack/tempest: Scenario test to check IPv6 mupltiprefix feature Starts 2 VMs on network with 2 IPv6 subnets and checks cross connectivity. https://review.openstack.org/157816 | 18:18 |
| *** naggappan has joined #openstack-qa | 18:20 | |
| *** harlowja_away is now known as harlowja_ | 18:21 | |
| *** jordanP has quit IRC | 18:25 | |
| sc68cal | ugh. we have provider networking stuff sprinkled in all kinds of bad places in lib/neutron of devstack | 18:25 |
| *** Nic has joined #openstack-qa | 18:26 | |
| *** yfried has quit IRC | 18:28 | |
| *** achanda has joined #openstack-qa | 18:31 | |
| *** tkammer has joined #openstack-qa | 18:37 | |
| *** aix has quit IRC | 18:41 | |
| *** achanda has quit IRC | 18:46 | |
| *** achanda has joined #openstack-qa | 18:46 | |
| *** mpaolino_ has joined #openstack-qa | 18:47 | |
| *** marun has joined #openstack-qa | 18:49 | |
| dkranz | marun: Do you have a rough time line for when you think you would start the procedure defined here https://wiki.openstack.org/wiki/QA/Tempest-test-removal | 18:52 |
| *** adalba has joined #openstack-qa | 18:52 | |
| marun | dkranz: I've already started on (1) | 18:52 |
| *** mpaolino_ has quit IRC | 18:52 | |
| *** acfleury has quit IRC | 18:52 | |
| marun | kevinbenton: https://review.openstack.org/#/c/158021/ | 18:52 |
| marun | oop | 18:52 |
| marun | dkranz: https://review.openstack.org/#/c/158021/ | 18:52 |
| *** bkopilov has quit IRC | 18:53 | |
| marun | sorry kevinbenton that was in error | 18:53 |
| dkranz | marun: yeah, I guess meant when you thought you might complete it :) | 18:53 |
| *** bkopilov has joined #openstack-qa | 18:53 | |
| *** flwang1 has quit IRC | 18:54 | |
| *** Rockyg has joined #openstack-qa | 18:54 | |
| marun | dkranz: I guess it depends on how much cleanup I want to see done before removal from tempest. | 18:54 |
| dkranz | marun: fair enough | 18:54 |
| openstackgerrit | Roey Chen proposed openstack-dev/devstack: vmware-nsx: Vendor code split https://review.openstack.org/144312 | 18:54 |
| marun | dkranz: As is, (3) is the only issue that we haven't really resolved. | 18:54 |
| dkranz | marun: The zaqar tests in tempest don't even run in the gate, and there was interest in moving them in-project. I said you were paving the path of how-to with neutron. | 18:55 |
| marun | dkranz: It should be easy enough to ensure that the tests in the neutron tree can run agains tstable branches | 18:55 |
| marun | tstable -> stable | 18:55 |
| dkranz | marun: cool | 18:55 |
| marun | dkranz: It's less clear how the in-tree tests should be consumable for refstack/defcore purposes. | 18:56 |
| marun | dkranz: do you have any thoughts on that? | 18:56 |
| mtreinish | marun: heh, I wouldn't be so confident about #2 for some of the tests :) | 18:56 |
| dkranz | mtreinish: :) | 18:56 |
| marun | mtreinish: well, I'd be hard-pressed to see how they could do worse than when they are running from the tempest repo | 18:56 |
| dkranz | mtreinish: At the summit (I think) some of us talked about having a kind of pluggable way for tempest to discover tests | 18:57 |
| dkranz | marun: sorry, I meant to address you | 18:57 |
| mtreinish | dkranz: well honestly that's not hard since tempest doesn't do discovery, you just need to write a module that will use load_tests to pull from different places | 18:57 |
| marun | dkranz: https://review.openstack.org/#/c/158021/4/neutron/tests/api/__init__.py | 18:58 |
| marun | dkranz: I'm assuming we'd need to do something like this but use module paths instead of file paths | 18:58 |
| mtreinish | dkranz: or: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/test_discover/test_discover.py | 18:58 |
| dkranz | mtreinish: It is not technically hard, but we need to make it a requirement IMO which means spelling out exactly how to do it as part of the removal procedure. | 18:59 |
| mtreinish | marun: yeah you can just construct a big suite and manually add the tests instead of relying on discovery | 18:59 |
| Rockyg | Defcore should have full pathnames of all test in a file | 18:59 |
| dkranz | mtreinish: but I'm not sure there was universal agreement about it being required | 18:59 |
| mtreinish | dkranz: well the thing I'm not sure about is whether it's worth the effort to even bother trying with that. It's not a burden at all to keep the set of tests for interop in tempest | 18:59 |
| mtreinish | which was the intent of 3 on the wiki really, someone from defcore says they need test X so we don't remove it | 19:00 |
| Rockyg | Dkranz, that's also an interesting idea | 19:00 |
| marun | mtreinish: Wouldn't that imply duplication? Or are you tinking that | 19:00 |
| dkranz | mtreinish: no, but any one making use of full tempest now is going to want to replace it with a similar function | 19:00 |
| mtreinish | marun: yes there would probably be some duplication which really isn't a big deal | 19:00 |
| dkranz | mtreinish: and it would be nice if we did that once and not make every one do it | 19:00 |
| dkranz | mtreinish: and the refstack tests include all the api tests, right? | 19:01 |
| Rockyg | Refstack meeting is starting about now in #openstack-meeting-alt Wanna discuss this over there? | 19:01 |
| marun | mtreinish: What justifies the cost of duplication? | 19:01 |
| dkranz | mtreinish: so what exactly are you proposing once neutron api tests are removed from tempest? | 19:01 |
| marun | Is being able to reuse the tests out of the project tree perceived as too expensive? | 19:01 |
| Rockyg | dkranz: not all the API tests. Only those for interop | 19:01 |
| dkranz | Rockyg: what does that mean? | 19:01 |
| mtreinish | dkranz: they only use a subset | 19:01 |
| mtreinish | they have a big list somewhere | 19:02 |
| Rockyg | So right now, no admin apis | 19:02 |
| dkranz | Rockyg: is that all that is missing? | 19:02 |
| dkranz | Rockyg: is so, I would say that is essentially the same as full tempest for the purpose of the issue we are discussing | 19:02 |
| Rockyg | Defcore lags by a half year to a year on what is in it. Some things might never get in it if it's an API that few people use | 19:03 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Fix check for FloatingIP status before check connection https://review.openstack.org/160135 | 19:03 |
| mtreinish | marun: well, if we want to make tempest really an integration suite, I think a set of tests useful for interop probably would be fine to keep in tempest | 19:03 |
| marun | mtreinish: If you mean cross-project, sure. | 19:03 |
| dkranz | marun: I don't think that's what he means | 19:03 |
| marun | mtreinish: but if a test is against a single project's api, I'm not sure i follow. | 19:03 |
| Rockyg | Also consider that integrated API tests might be more useful for interop than ones that don't require a full cloud setup | 19:04 |
| dkranz | marun: but I'm not sure exactly what mtreinish does mean by interop | 19:04 |
| dkranz | marun: what Rockyg said sounds more like just exclude admin and some tests they have not gotten around to including | 19:04 |
| *** Sukhdev has quit IRC | 19:04 | |
| mtreinish | dkranz: well mostly because it's not a well defined term, with the mess that defcore is. But the theory is there is a subset of tests they care about for verifying different clouds operate the same | 19:05 |
| dkranz | Rockyg: the bottom line is, do all neutron api tests get moved, and we have a way to run them with tempest? | 19:05 |
| mtreinish | dkranz: hogepodge has a proposal for tagging them specifically | 19:05 |
| dkranz | Rockyg: Or do some neutron tests remain in tempest. | 19:05 |
| dkranz | mtreinish: But doesn't that include most apis, and hence most api tests? | 19:06 |
| Rockyg | Good question. If tempest actually becomes the tempest lib plus test server, would it matter? | 19:06 |
| mtreinish | dkranz: and given that it'll be a subset of functionality I personally have no issue with keeping those tests in tempest even given neutron has a local copy in tree | 19:06 |
| dkranz | Rockyg: Yes,, once neutron api tests appear in the neutron tree | 19:06 |
| hogepodge | dkranz: my feeling is that if a test uses a very limited set of credentials (non-admin) and not much else it should be tagged as interop | 19:06 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 19:07 | |
| Rockyg | Also, I would think no drivers except default would be in defcore | 19:07 |
| mtreinish | dkranz: I have no idea how many things will end up getting a tag | 19:07 |
| marun | mtreinish: who maintains them though? | 19:07 |
| dkranz | hogepodge: in that case almost all of tempest is in that set | 19:07 |
| mtreinish | marun: which ones? | 19:07 |
| hogepodge | dkranz: the problem with a bunch of api tests is they assume admin creds and need to heavily touch the environment to set themselves up. | 19:07 |
| marun | mtreinish: I don't think tempest should be maintaining any tests that validate the stability of Neutron's api. | 19:07 |
| marun | mtreinish: if you want to keep copies, sure. | 19:07 |
| Rockyg | interop tests should be maintained by the community. | 19:07 |
| hogepodge | dkranz: the assumption isn't true in practive | 19:07 |
| dkranz | hogepodge: a bunch, yes. BUt a very small fraction | 19:07 |
| hogepodge | s/practive/practice | 19:07 |
| marun | mtreinish: but we've seen pretty definitively that developers need incentive to care about API stability, and giving responsibility to a different team/repo is not an effective way to achieve that goal. | 19:08 |
| mtreinish | marun: yeah that's what I'm proposing. The api stability will be primarily maintained by neutron, but there might be copies of things that remain in tempest | 19:09 |
| mtreinish | but the primary burden of api stability will shift to neutron | 19:09 |
| dkranz | mtreinish: for what purpose? | 19:09 |
| mtreinish | dkranz: for the purpose we've been discussing | 19:09 |
| hogepodge | dkranz: I disagree, maybe I'm wrong. In defcore we have 380 capabilities listed, but only 160 don't require admin. That's not a small fraction. | 19:09 |
| dkranz | hogepodge: Can I see this list? | 19:10 |
| hogepodge | dkranz: My proposal is that [interop] can only be applied if a strict requirement is met. | 19:10 |
| hogepodge | dkranz: https://github.com/openstack/defcore/blob/master/havanacore.json | 19:10 |
| marun | mtreinish: So long as it's a read-only copy, I'm ok with that. I think it should be possible to consume tests directly from neutron, but I'm ok with that being a long-term goal. | 19:10 |
| dkranz | hogepodge: So is the goal to minimize the number of tests, or to make sure some set is covered? | 19:10 |
| hogepodge | dkranz: the goal is to arrive at a set of tests that cover the api, but any end user can point at any cloud they have access to and check api compatibility | 19:11 |
| Rockyg | goal is to get really good coverage on any capability that is designated interop | 19:11 |
| *** ChuckC has joined #openstack-qa | 19:11 | |
| hogepodge | dkranz: with coverage being encouraged because nobody wants to look the fool | 19:11 |
| mtreinish | marun: sure, I'm fine with that being a long term goal. Right now the only thing I'm thinking is as an external tester it would be kinda weird to depend on having all of the project code installed | 19:11 |
| mtreinish | but I'm sure we can figure a way around that too | 19:12 |
| dkranz | hogepodge: I think the disconnect here is that tempest has nearly 1700 tests now | 19:12 |
| marun | mtreinish: I'm not sure why that would be a problem. It's not like having the project code is going to affect the targeting of a deployed service. | 19:12 |
| marun | mtreinish: that said... | 19:13 |
| dkranz | hogepodge: some of them are keystone, which requires admin, but most do not require admin | 19:13 |
| hogepodge | dkranz: I'd like for us to identify the [interop] criteria, then nail down the api tests that meet the criteria, and modify tests that don't make it or write tests that hit coverage. | 19:13 |
| marun | mtreinish: there is an ongoing discussion around splitting out code that is common between neutron, plugins, and services. | 19:13 |
| marun | mtreinish: it may be that the api testing ends up in the neutron-lib repo and that will aid in making the tests consumable. | 19:13 |
| mtreinish | marun: sure that would alleviate some of my concerns around that | 19:14 |
| marun | hogepodge: where is the criteria for 'interop' defined? | 19:14 |
| dkranz | hogepodge: ok, all I am saying is that I believe a large majority of the current 1700 tests are expected to behave interoperably | 19:14 |
| hogepodge | marun: It isn't defined yet. I'd like it to be a test that only requires a guest account, an enpoint, and possibly a network id and image id. | 19:15 |
| hogepodge | marun: zero admin | 19:15 |
| marun | hogepodge: that seems like a pretty important thing to define | 19:15 |
| marun | hogepodge: not sure why work would be even contemplated without knowing what is needed | 19:15 |
| dkranz | hogepodge: We are close to being able to run tempest in a way where it will just skip any tests that require admin | 19:16 |
| hogepodge | marun: saying it isn't defined doesn't mean that I don't have a minimum idea of what the definition should be. In whatever proposal I write up, I expect criticism and possible expansion of the maximum allowable configuration | 19:16 |
| dkranz | hogepodge: If there were such a list of skipped tests, would you want to include the rest? | 19:16 |
| hogepodge | dkranz: what do you mean? I think that list of skipped tests would be valuable. | 19:17 |
| dkranz | hogepodge: I am trying to figure out which of the following two is your goal: | 19:17 |
| hogepodge | dkranz: marun: I'd just like to have a well-defined criteria that makes it possible to simply configure tempest and point it any any cloud. | 19:17 |
| hogepodge | dkranz: marun: in an ideal world that would be credentials and endpoint. | 19:17 |
| dkranz | hogepodge: 1) Have a minimum set of tempest tests that cover your criteria | 19:17 |
| dkranz | hogepodge: 2) Use all tempest tests that meet your criteria | 19:18 |
| dkranz | hogepodge: The first is easy to do and easy to maintain. The second is work for both. | 19:18 |
| openstackrecheck | openstack/ceilometer change: https://review.openstack.org/159907 failed gate-ceilometer-python27 in the gate queue with an unrecognized error | 19:18 |
| dkranz | hogepodge: I had always assumed it was the first. | 19:18 |
| dkranz | hogepodge: and so assumed it was important to continue to be able to collect these tests in a run even if they were moved out of tempest (along with their uuids) | 19:19 |
| Rockyg | dkranz: I think it's more (2), but with a delay for new capabilities | 19:19 |
| dkranz | Rockyg, hogepodge : sorry, I meant I assumed (2) and similar for the next comment | 19:20 |
| *** tosky has quit IRC | 19:20 | |
| *** cdent has quit IRC | 19:20 | |
| Rockyg | dkranz. Yes. but new tests for already selected interop capabilites would want to be rolled into the next defcore release to happen | 19:21 |
| Rockyg | Encouraging adding coverage for interop flagged capabilities | 19:21 |
| dkranz | Rockyg: ok, I am just arguing that the test selection and ability to run tempest should not be impacted by whether a test happens to have been moved frolm the tempest repo to a project repo | 19:21 |
| Rockyg | dkranz: agreed | 19:22 |
| dkranz | marun, mtreinish : Do you agree with that ^^^ and we can work on the implementation? I can't quite tell. | 19:22 |
| *** ibiris is now known as ibiris_away | 19:22 | |
| hogepodge | dkranz: the promise of smoke and api was that we would have criteria like I mentioned. They've failed in that regard. By explicitly setting criteria and admitting (automatically or manually, if automatic is possible then +2 to that) only those tests as 'interop' tests we create a clear means for easily identifying the publicly testable features of a deployment | 19:23 |
| dkranz | hogepodge: I believe we are close to automatically being able to tell which tests require admin by trying to run them. | 19:24 |
| hogepodge | dkranz: what do you mean? | 19:25 |
| dkranz | hogepodge: In any event, I now understand what is meant by interop and think we are on the same page | 19:25 |
| dkranz | hogepodge: I just meant you could tag a test as not needing admin but the test would fail if it actually did require it. | 19:25 |
| dkranz | hogepodge: ANother alternative is to run all of tempest, configured with no admin, and just run all the tests. | 19:26 |
| dkranz | hogepodge: But it sounds like that is not where you are going. | 19:26 |
| hogepodge | dkranz: if it were expected to be a test that didn't need admin I would consider that a bug. | 19:26 |
| Rockyg | right, but then you file the bug and remove the tag | 19:27 |
| dkranz | hogepodge: yes, it would be a bug. As would be the reverse. | 19:27 |
| marun | dkranz: I'm not entirely sure what is required. | 19:28 |
| hogepodge | dkranz: no, I'd like a way to just run tests that are known to not need admin, and only rely on public api | 19:28 |
| dkranz | marun: I think just the discovery thing | 19:28 |
| dkranz | hogepodge: I understand, but another way to do that is to run tempest in a way that all tests requiring admin are skipped | 19:28 |
| marun | dkranz: I'd like to see that happen if it means we can have only one copy of the tests. | 19:29 |
| hogepodge | I'm willing to admit some additional infomation though, like ids of existing resources that we can reasonably expect a user to have access to | 19:29 |
| mtreinish | andreaf: did you see: http://logs.openstack.org/11/158511/15/check/check-tempest-dsvm-f21/a051107/console.html.gz#_2015-02-27_23_08_18_898 | 19:29 |
| dkranz | marun: I think the best way is to just make it so you can do that, and worry about removing all tests from tempest once it is proved to have no benefit and just be a headache. | 19:29 |
| hogepodge | dkranz: not just admin, but anything else that makes exceptional assumptions. But yes, admin is the mose exceptional. | 19:29 |
| mtreinish | looks like some cases things get held for too long with roles | 19:29 |
| dkranz | hogepodge: agreed. My only point is that it is a lot easier to stay current if you mark tests that should *not* be run than if you have to mark tests that *are* to be run. | 19:30 |
| *** yfried has joined #openstack-qa | 19:31 | |
| hogepodge | dkranz: yes, rot is always a threat | 19:31 |
| dkranz | hogepodge: that is similar to the point of branchless tempest | 19:32 |
| *** dimtruck is now known as zz_dimtruck | 19:32 | |
| Rockyg | dkranz, unfortunately, defcore isn't about current, it's about mature. But yes. any test not marked admin or interop would be a candidate undfer consideration during the defcore discussions | 19:32 |
| *** e0ne has joined #openstack-qa | 19:32 | |
| *** e0ne is now known as e0ne_ | 19:32 | |
| *** e0ne_ is now known as e0ne | 19:32 | |
| dkranz | hogepodge: It's not just rot, but missing a new test that was added, and is valuable, but has not been catalogued yet | 19:32 |
| Rockyg | defcore discussions about whether the capabilitiy is mature but not included yet. | 19:32 |
| hogepodge | dkranz: One of my goals is to get those involved with efforts like defcore (and anyone can be involved with defcore) more involved with the testing process, to truly become the stake holders in having a strong interoperability test suite as part of the overall gating process | 19:33 |
| dkranz | Rockyg: I am not talking about mature vs immature apis | 19:33 |
| dkranz | Rockyg: Just the number of tests that are related to a mature api. More are always being added. | 19:33 |
| dkranz | Rockyg: And this will be even more so when they are moved to projects because projects will have more of a budget to run them. | 19:33 |
| hogepodge | dkranz: if a test is valuable I would have a personal interest in making sure it's captured. | 19:33 |
| Rockyg | Yes, new tests in already marked categories would still need to be identified and added to the list by the defcore process | 19:34 |
| dkranz | hogepodge: So my proposal, for what it's worth, is to start by assuming tempest tests (whether in tempest or moved to a project) are valuable, and exclude the ones that are not with some tag or by allowing it to be skipped as in the admin case. | 19:35 |
| dkranz | hogepodge: rather than assuming that none are valuable until marked as such | 19:35 |
| hogepodge | dkranz: that's what we're doing right now | 19:35 |
| Rockyg | good proposal | 19:35 |
| Rockyg | ttfn | 19:36 |
| hogepodge | dkranz: (the assuming all api tests are valuable) | 19:36 |
| hogepodge | dkranz: it sank efforts like tcup, which were an attempt to give a reusable and minimally configurable tool to tests api compatibility | 19:37 |
| dkranz | hogepodge: Another point I am trying to make is that true interop testing has to be very deep. | 19:37 |
| marun | dkranz: ? | 19:37 |
| dkranz | hogepodge: and is not adequately covered in tempest, never mind a smaller subset | 19:37 |
| hogepodge | dkranz: deep in what way? | 19:37 |
| dkranz | hogepodge: covering the dictionary formats, headers, etc. | 19:38 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/156957 failed because of: gate-grenade-dsvm-ironic-sideways: https://bugs.launchpad.net/bugs/1425258 | 19:38 |
| openstack | Launchpad bug 1425258 in OpenStack Compute (nova) "test_list_baremetal_nodes race fails with a node not found 404" [Medium,Confirmed] | 19:38 |
| dkranz | hogepodge: many areas of tempest do not cover that and current thinking is that projects should do deep testing of the api themselves which is part of the motivation for moving that testing to projects | 19:39 |
| hogepodge | dkranz: How about I send up a blueprint before the next meeting, and publicize it on the mailing list. I think that there's a lot of good discussion here and I'd like to capture it more formally. | 19:39 |
| hogepodge | dkranz: agreed. | 19:39 |
| dkranz | hogepodge: sounds great | 19:39 |
| hogepodge | dkranz: Part of what I'm asking for is a criteria that gives test writers a platonic target for what it means to test an api. | 19:39 |
| hogepodge | (platonic ideal of a target) | 19:40 |
| dkranz | hogepodge: there is a lot of discussion going on about that in various groups. It includes both detailed testing of the REST API and schemas in use, as well as some testing of actual function. | 19:41 |
| dkranz | hogepodge: for example, if I call 'nova boot' and it accepts and returns all the right stuff, it should still fail if a vm is not actually created. | 19:41 |
| hogepodge | dkranz: It's a hard problem, no doubt. | 19:42 |
| hogepodge | dkranz: a complete mock library should be admitted as iteroperable. :-D | 19:42 |
| dkranz | hogepodge: right, and in my thinking, what refstack really ought to be doing is similar to what you would do to really validate a cloud | 19:42 |
| dkranz | hogepodge: but limited to the apis that are declared to be interoperable | 19:43 |
| dkranz | hogepodge: that limitation is a concept that is not really in tempest right now | 19:43 |
| dkranz | hogepodge: so really what I am suggesting is that "refstack" tests are a discoverable subset of tempest + api tests in projects | 19:44 |
| dkranz | hogepodge: rather than a manually generated list. But I understand it is not that way. | 19:44 |
| hogepodge | dkranz: what would that discovery process look like? | 19:44 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check Floating ip status before load balancer check https://review.openstack.org/160471 | 19:45 |
| hogepodge | dkranz: I can imagine one test being using an interop criteria for configuration, then flagging tests that don't pass as possible non-interop compatible. Is there a better way to check to see which tests use which config options? I would assume no. | 19:45 |
| dkranz | hogepodge: assuming appropriate tagging of tests that are not interop, just an expression that uses the structure of tempest directories and tags on tests | 19:46 |
| hogepodge | dkranz: the problem with that being tempest defaults may create false positives | 19:46 |
| hogepodge | (or false negatives depending on your point of view) | 19:46 |
| dkranz | hogepodge: Tempest is supposed to run against any cloud | 19:47 |
| dkranz | hogepodge: THis obviously requires more thought | 19:47 |
| dkranz | hogepodge: And I am not trying to derail anything. What I am arguing for is something I want anyway, even if refstack does not do it that way. | 19:48 |
| dkranz | hogepodge: I mean with regard to pulling in tempest tests from whatever repo they live in and running them | 19:48 |
| hogepodge | dkranz: I want to be clear that I want tempest to run in a way that makes it possible to identify capabilities with any tool, refstack or rally for example, and for any initiative, defcore being one of them, but one can imagine the existence of other cross-platforms standards, or even tooling for automated cloud bursting | 19:50 |
| hogepodge | dkranz: so it's not a refstack or defcore issue, I see this primarily as a tempest issue | 19:50 |
| dkranz | hogepodge: agreed | 19:51 |
| hogepodge | (or more broadly, tempest issue) | 19:51 |
| hogepodge | s/tempest/testing/ | 19:51 |
| dkranz | hogepodge: and I think there are really only two issues for tempest | 19:51 |
| dkranz | hogepodge: 1. Make sure it can be run even with api tests in a project | 19:51 |
| *** zz_dimtruck is now known as dimtruck | 19:51 | |
| dkranz | 2. Make sure there is a way to tag tests as being interop or not. | 19:51 |
| dkranz | hogepodge: more likely tagging as not-interop since most will be. | 19:52 |
| hogepodge | dkranz: I like that idea. | 19:53 |
| *** carl_baldwin has quit IRC | 19:56 | |
| *** e0ne is now known as e0ne_ | 19:59 | |
| *** e0ne has joined #openstack-qa | 20:00 | |
| *** flwang1 has joined #openstack-qa | 20:02 | |
| *** tkammer has quit IRC | 20:04 | |
| *** patrickeast__ has quit IRC | 20:04 | |
| *** patri505 has quit IRC | 20:04 | |
| *** patri136 has quit IRC | 20:04 | |
| *** patrickeast_ has quit IRC | 20:04 | |
| *** patrickeast has quit IRC | 20:04 | |
| mtreinish | adam_g: qq, there was a reason we don't run the ironic tempest job with tenant isolation right? | 20:05 |
| mtreinish | I couldn't remember | 20:05 |
| mtreinish | it caught a bug in my patch, but I was surprised it wasn't running with it | 20:05 |
| *** patrickeast has joined #openstack-qa | 20:07 | |
| adam_g | mtreinish, oh, that goes back a while. at the time, tempest's tenant isolation didn't have required special casing for baremetal, specifically around network isolation | 20:07 |
| *** achanda has quit IRC | 20:07 | |
| *** patrickeast_ has joined #openstack-qa | 20:08 | |
| *** patrickeast__ has joined #openstack-qa | 20:08 | |
| *** patri875 has joined #openstack-qa | 20:08 | |
| *** patri928 has joined #openstack-qa | 20:08 | |
| *** achanda has joined #openstack-qa | 20:08 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate: apt-get needs access to tty so give it access https://review.openstack.org/160480 | 20:08 |
| adam_g | mtreinish, that was all fixed for the parallel job, but im not sure the future of that test. i was going to propose switching the old job to isolation, but its caught a few other bugs in the past so maybe its better to keep it isolated so we have that code running somewhere? | 20:09 |
| mtreinish | adam_g: ok that's what I thought, but was unsure because when I checked the isolation code there was special casing for baremetal :) | 20:09 |
| *** EmilienM|afk is now known as EmilienM | 20:10 | |
| mtreinish | adam_g: I'll defer to what you think is best. If it's running serially there isn't a reason to make it use tenant isolation | 20:10 |
| naggappan | hi in setup class it is trying to setup EC2 also and each time it fails as i have not configured ect. Do some one know how to dissable ec2 in tempest.conf. I tried the flag "ec2_api = False" but it did not work out | 20:10 |
| *** Rockyg has quit IRC | 20:11 | |
| *** _afezekas|pub has joined #openstack-qa | 20:14 | |
| *** Sukhdev has joined #openstack-qa | 20:15 | |
| adam_g | mtreinish, yeah, we'll probably end up disabling the parallel job. the motivation with that one was to allow the entire tempest suite to run against baremetal, but attitudes have changed since toward testing only the subset that is relevant.. which is exactly what the original job is doing | 20:16 |
| mtreinish | adam_g: well fwiw, having things run in parallel normally just shakes loose bugs that would be missed normally | 20:17 |
| mtreinish | so having the job configured and running somewhere might be useful even if it's just on ironic | 20:18 |
| adam_g | mtreinish, ya, tho then we miss bugs around non-isolation :) im not sure the status of it, but i know there is a non-isolated periodic job running somewhere | 20:19 |
| mtreinish | adam_g: yeah there is a nightly job which runs tempest without tenant isolation in serial | 20:21 |
| openstackgerrit | Matthew Treinish proposed openstack/tempest: Switch swift tests to use cred_provider with roles https://review.openstack.org/158941 | 20:23 |
| openstackgerrit | Matthew Treinish proposed openstack-dev/devstack: Do Not Merge: Test test-accounts with roles https://review.openstack.org/158511 | 20:23 |
| *** _afezekas|pub has quit IRC | 20:29 | |
| openstackgerrit | Merged openstack/tempest: Make tempest's python 2.6 policy clear https://review.openstack.org/158355 | 20:31 |
| *** vigneshvar has quit IRC | 20:33 | |
| naggappan | mtreinish: I tried to provide flag "ec2_api = False" but then it skips the whole test "api.network.test_port" . So is this the only flag to skip ec2 | 20:38 |
| mtreinish | naggappan: yes that flag is only used in one place: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/thirdparty/boto/test.py#n198 | 20:39 |
| mtreinish | that will skip all the ec2 and s3 tests if the flag is set to false | 20:39 |
| naggappan | Mtrinish: if i set that flag it skips the whole test like "api.network.test_port, api.network.test_network" etc. It skips where ever the setup class is required to configure ec2 | 20:40 |
| naggappan | Mtrinish: so i guess it would be no point in testing those things right | 20:41 |
| mtreinish | naggappan: api.network.test_port is not an ec2 test that's a test that uses the neutron rest api | 20:41 |
| openstackgerrit | Merged openstack-dev/devstack: Fix curl issue when SERVICE_HOST is defined to IPv6 address https://review.openstack.org/158594 | 20:43 |
| openstackgerrit | Merged openstack-dev/devstack: Remove LBaaS support from devstack https://review.openstack.org/153079 | 20:43 |
| naggappan | mtreinish: but when i run most of the test i get the following error repeatedly . "http://paste.openstack.org/show/185441/" | 20:44 |
| openstackgerrit | Henry Gessau proposed openstack/tempest: Allow time for ipv6 auto addresses to get assigned https://review.openstack.org/159994 | 20:45 |
| HenryG | mtreinish: ^^ again please :) | 20:46 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check FloatingIP status before SSH in network_v6 https://review.openstack.org/160502 | 20:46 |
| mtreinish | HenryG: heh, death of a thousand nits :) | 20:47 |
| HenryG | mtreinish: Sometimes I give nits, so I have to learn to receive them. :) | 20:49 |
| *** ibiris_away is now known as ibiris | 20:50 | |
| *** achanda has quit IRC | 20:52 | |
| *** yamamoto has joined #openstack-qa | 20:56 | |
| Noice | should the default nova.conf have a fair bit if information in it like the image and identity services or is 12 lines normal? | 21:02 |
| flwang | mtreinish: ping | 21:05 |
| *** andreaf_ has joined #openstack-qa | 21:08 | |
| flwang | mtreinish: i'm still trying to figure out how to safely cleanup the tenants created by tempest | 21:08 |
| *** andreaf_ has quit IRC | 21:08 | |
| flwang | because we're trying to use it in prod | 21:09 |
| *** diegows has quit IRC | 21:13 | |
| mtreinish | flwang: is it leaving tenants lying around from tenant isolation? | 21:14 |
| mtreinish | flwang: I guess I'm curious what exactly is being left hanging around after a run | 21:14 |
| flwang | mtreinish: as i mentioned before, we're planning to leverage scenario test to get some preactive view in prod | 21:15 |
| flwang | but in our test, after run scenario test, there are always some tenants left | 21:15 |
| *** mfalatic has joined #openstack-qa | 21:16 | |
| mtreinish | flwang: hmm that seems like a bug, the cleanup in tenant isolation should be handling deleting any tenants that get created | 21:16 |
| mtreinish | flwang: do you have a pastebin or an example I can look at? | 21:16 |
| flwang | mtreinish: not for now, but I can do that later and let you know | 21:19 |
| flwang | so did you mean the scenario test should handle the temp tenants very well, right? | 21:20 |
| flwang | if it can't, there must be a bug, is it? :) | 21:20 |
| mtreinish | flwang: sure, I'm just looking for a bit more specific context so I can give you better pointers | 21:20 |
| mtreinish | flwang: yes, so if it's creating tenants as part of tenant isolation those should be deleted after the class is done everytime | 21:20 |
| mtreinish | if they're not that is 100% a bug :) | 21:20 |
| flwang | mtreinish: i see, but most of the time, there is no error log, what we can see is the scenario test run completely, but there are some tenants left in openstack | 21:21 |
| mtreinish | flwang: so the tenant delete is being issued and it returns with a success code but it doesn't ever go away? | 21:21 |
| flwang | mtreinish: and for now, we still can't use the cleanup script, right? because I was told there is a bug like the new tenant created after the snapshot will be deleted as well | 21:22 |
| flwang | mtreinish: i think so | 21:22 |
| mtreinish | flwang: well if the delete returns a success and nothing gets deleted then that's a keystone bug :) | 21:23 |
| *** DinaBelova has quit IRC | 21:24 | |
| *** SergeyLukjanov has quit IRC | 21:24 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check FloatingIP status before connecting to access-points https://review.openstack.org/160517 | 21:24 |
| mtreinish | flwang: yeah the cleanup script doesn't work right now if you have additional tenants being created outside of tempest after you generate the whitelist before the run | 21:25 |
| mtreinish | which is a known limitation right now, I don't think anyone is working on fixing that right now, but it was something to be addressed long term | 21:25 |
| mtreinish | naggappan: you're telling me 2 different things. ec2_api in the compute-feature-enabled group is used to tell tempest whether the ec2 api in nova is available or not (so it either skips or runs those tests) | 21:27 |
| flwang | mtreinish: do you think it's a good idea if we can add a prefix(like a UUID, generating each time for the tempest run), and then after running, just looking for the tenant with the specific prefix | 21:27 |
| flwang | to delete them | 21:28 |
| mtreinish | naggappan: it has nothing to do with the tests in tempest.api.network. That pastebin is from trying to run the ec2 tests which will be skipped if you set ec2_api to false | 21:28 |
| *** carl_baldwin has joined #openstack-qa | 21:29 | |
| mtreinish | flwang: sure that'll be one way to get around it, although we do something like that now. The testclass name is used as the prefix of the random name | 21:29 |
| mtreinish | the issue with that is running parallel won't have a shared prefix if you're using a uuid (or some other generation mechanism) because things run in separate processes | 21:30 |
| flwang | mtreinish: right | 21:30 |
| mtreinish | flwang: we could change the tenant isolation code to add 'tempest-' in front of the test class name if you think that would help | 21:31 |
| flwang | and I just have another question, if everything works fine and we assume there should not be a bug for all the teardown method, then why we need a cleanup script? :) | 21:31 |
| flwang | am I missing anything? | 21:31 |
| flwang | mtreinish: it would be a simple but very good idea | 21:32 |
| mtreinish | flwang: heh, well it's for 2 things. 1. when tenant isolation isn't used and we want to ensure the pre-existing tenants don't leak resources. (we also can potentially use this for debugging resource leaks) | 21:33 |
| mtreinish | 2. things rarely are bug free :) (both in tempest and in the services) | 21:34 |
| mtreinish | so having the script help provide some automation around when things inevitably go bad | 21:34 |
| mtreinish | flwang: well if adding the tempest- prefix works for you can you file a bug and push a patch? :) | 21:35 |
| naggappan | mtreinish: I have figured it out. under [boto] all should be commented out except ece_api = False option | 21:37 |
| flwang | mtreinish: yep, I will hack it to see if it works and will upstream it back | 21:37 |
| flwang | mtreinish: thanks a lot for the suggestion | 21:37 |
| mtreinish | andreaf: so it looks like the teardown bug is in the cli tests. They don't give up there creds back to the pool which consumes most of the available admin creds | 21:38 |
| mtreinish | flwang: sure np | 21:38 |
| mtreinish | naggappan: ok good | 21:38 |
| andreaf | mtreinish: ok, cool, your debug patch is really useful for that | 21:39 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 21:39 | |
| mtreinish | andreaf: yeah that's how I was able to figure out cli was always the thing left holding creds :) | 21:40 |
| *** FatDarrel has joined #openstack-qa | 21:40 | |
| mtreinish | dkranz: if you get a sec: https://review.openstack.org/#/c/159917/ should be a quick review | 21:40 |
| *** FatDarrel is now known as phd2 | 21:41 | |
| mtreinish | andreaf: I still need to figure out why it's not giving them back, but it's at least a start to know what is holding on to them | 21:41 |
| *** gfidente has quit IRC | 21:42 | |
| *** ChuckC has quit IRC | 21:44 | |
| *** e0ne is now known as e0ne_ | 21:46 | |
| *** achanda has joined #openstack-qa | 21:46 | |
| mtreinish | andreaf: hah, figured it out the cli tests call there isolated_creds class cred_prov but the base test class only checks for the existence of isolated_creds | 21:48 |
| mtreinish | flwang: you're not running the cli tests in your runs are you? because this bug would affect you too and leave tenants lying around | 21:48 |
| *** e0ne_ is now known as e0ne | 21:48 | |
| flwang | mtreinish: no, I just run scenario test | 21:49 |
| andreaf | mtreinish: so it's in tempest lib to be fixed? | 21:49 |
| mtreinish | andreaf: no we don't use tempest-lib's base class for the cli framework for other reasons | 21:50 |
| mtreinish | we just named the class var the wrong thing so the cleanup_isolated_creds hasattr check failed and cleanup was never run | 21:50 |
| andreaf | ...upz :D | 21:51 |
| andreaf | good catch | 21:51 |
| *** achanda has quit IRC | 21:52 | |
| openstackgerrit | Matthew Treinish proposed openstack/tempest: Fix isolated cred cleanup for cli tests https://review.openstack.org/160525 | 21:54 |
| mtreinish | andreaf: ^^^ | 21:55 |
| *** aix has joined #openstack-qa | 21:56 | |
| *** toure has quit IRC | 21:56 | |
| openstackgerrit | Matthew Treinish proposed openstack-dev/devstack: Do Not Merge: Test test-accounts with roles https://review.openstack.org/158511 | 21:56 |
| *** _afezekas|pub has joined #openstack-qa | 21:56 | |
| *** yamamoto has quit IRC | 22:00 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Fix check for FloatingIP status before check connection https://review.openstack.org/160135 | 22:01 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check Floating ip status before load balancer check https://review.openstack.org/160471 | 22:01 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check FloatingIP status before SSH in network_v6 https://review.openstack.org/160502 | 22:01 |
| andreaf | mtreinish: +2 | 22:02 |
| mtreinish | andreaf: cool, thanks | 22:02 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check FloatingIP status before connecting to access-points https://review.openstack.org/160517 | 22:03 |
| mtreinish | andreaf: https://jenkins07.openstack.org/job/check-tempest-dsvm-f21/153/console includes that fix, let's see if we pass everywhere | 22:04 |
| *** dustins has quit IRC | 22:04 | |
| *** jecarey has quit IRC | 22:08 | |
| *** _afezekas|pub has quit IRC | 22:10 | |
| *** Nic has quit IRC | 22:12 | |
| *** Nic has joined #openstack-qa | 22:13 | |
| *** Sukhdev has quit IRC | 22:15 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/tempest: Check FloatingIP status before SSH in server ops https://review.openstack.org/160533 | 22:16 |
| *** andreykurilin_ has joined #openstack-qa | 22:18 | |
| *** e0ne has quit IRC | 22:23 | |
| *** Nic has quit IRC | 22:24 | |
| *** pc_m has quit IRC | 22:27 | |
| *** e0ne has joined #openstack-qa | 22:27 | |
| *** ibiris is now known as ibiris_away | 22:29 | |
| openstackgerrit | Jamie Lennox proposed openstack/tempest: Allow full v3 authentication https://review.openstack.org/159342 | 22:29 |
| *** jecarey has joined #openstack-qa | 22:32 | |
| *** andreykurilin_ has quit IRC | 22:33 | |
| *** andreykurilin_ has joined #openstack-qa | 22:34 | |
| *** e0ne has quit IRC | 22:34 | |
| openstackgerrit | Jamie Lennox proposed openstack/tempest: Move identity v2 tests to their own folders https://review.openstack.org/158518 | 22:34 |
| *** DinaBelova has joined #openstack-qa | 22:37 | |
| *** SergeyLukjanov has joined #openstack-qa | 22:37 | |
| *** e0ne has joined #openstack-qa | 22:38 | |
| *** mtanino has quit IRC | 22:39 | |
| *** marun has quit IRC | 22:39 | |
| *** patrickeast has quit IRC | 22:39 | |
| *** achanda has joined #openstack-qa | 22:42 | |
| *** patrickeast__ has quit IRC | 22:44 | |
| *** patri875 has quit IRC | 22:44 | |
| *** patri928 has quit IRC | 22:44 | |
| *** patrickeast_ has quit IRC | 22:44 | |
| *** tkammer has joined #openstack-qa | 22:44 | |
| *** Ananci has quit IRC | 22:46 | |
| *** weshay has quit IRC | 22:46 | |
| *** mtanino has joined #openstack-qa | 22:47 | |
| *** e0ne has quit IRC | 22:49 | |
| *** jhenner has quit IRC | 22:50 | |
| openstackgerrit | Jamie Lennox proposed openstack-dev/devstack: Create _member_ role in keystone bootstrap https://review.openstack.org/159690 | 22:51 |
| *** wuhg has joined #openstack-qa | 22:52 | |
| *** e0ne has joined #openstack-qa | 22:53 | |
| *** andreykurilin_ has quit IRC | 23:00 | |
| mtreinish | dkranz: https://review.openstack.org/#/c/160525/ is another simple review if you get a sec | 23:04 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/157269 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1311066 https://bugs.launchpad.net/bugs/1425310 | 23:08 |
| openstack | Launchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed] | 23:08 |
| openstack | Launchpad bug 1425310 in Cinder "tgtadm: this access control rule does not exist - results in timeout waiting for volume to attach (stuck in detach) or detach" [High,Confirmed] | 23:08 |
| *** bknudson has quit IRC | 23:08 | |
| *** dboik_ has joined #openstack-qa | 23:09 | |
| *** dboik has quit IRC | 23:13 | |
| *** dboik_ has quit IRC | 23:14 | |
| *** dimtruck is now known as zz_dimtruck | 23:18 | |
| *** salv-orlando has quit IRC | 23:20 | |
| *** salv-orlando has joined #openstack-qa | 23:21 | |
| *** marun has joined #openstack-qa | 23:24 | |
| *** jgrimm is now known as zz_jgrimm | 23:26 | |
| *** yamamoto has joined #openstack-qa | 23:26 | |
| *** toure has joined #openstack-qa | 23:26 | |
| *** tkammer has quit IRC | 23:30 | |
| *** marun has quit IRC | 23:30 | |
| *** yamamoto has quit IRC | 23:34 | |
| *** rfolco has quit IRC | 23:35 | |
| *** melwitt has quit IRC | 23:37 | |
| *** openstackgerrit has quit IRC | 23:38 | |
| *** openstackgerrit has joined #openstack-qa | 23:39 | |
| phd2 | hey guys | 23:39 |
| phd2 | any tempest people here? | 23:40 |
| *** tochi has joined #openstack-qa | 23:41 | |
| *** mriedem is now known as mriedem_away | 23:46 | |
| *** erlon has quit IRC | 23:51 | |
| *** salv-orlando has quit IRC | 23:51 | |
| *** Sukhdev has joined #openstack-qa | 23:52 | |
| *** hemna is now known as hemnafk | 23:53 | |
| *** EmilienM is now known as EmilienM|afk | 23:58 | |
| *** aysyd has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!