Wednesday, 2013-08-28

*** rgrober has quit IRC00:00
*** rgrober has joined #openstack-qa00:00
*** adalbas has quit IRC00:04
*** rgrober has quit IRC00:06
*** cyeoh__ has joined #openstack-qa00:59
*** cyeoh_ has quit IRC01:01
*** jasondotstar has joined #openstack-qa01:16
*** sthaha has joined #openstack-qa01:21
*** sthaha has joined #openstack-qa01:21
*** cyeoh__ is now known as cyeoh01:51
*** jasondotstar has quit IRC02:50
*** prekarat has joined #openstack-qa03:37
*** prekarat1 has joined #openstack-qa03:40
*** prekarat1 has quit IRC03:40
*** prekarat1 has joined #openstack-qa03:40
*** prekarat has quit IRC03:41
*** raies has joined #openstack-qa05:11
raiesis there someone from neutron team ?05:11
openstackgerritafazekas proposed a change to openstack/tempest: Remove wait_for_image_resp_code  https://review.openstack.org/3899505:16
*** SergeyLukjanov has joined #openstack-qa05:16
*** prekarat1 has quit IRC05:36
*** yfried has joined #openstack-qa05:47
*** prekarat has joined #openstack-qa05:48
mkodererraies: I think better ask in #openstack-dev05:50
raiesmkoderer: thank I will :)05:57
*** raies has quit IRC05:57
*** prekarat1 has joined #openstack-qa05:58
*** prekarat has quit IRC05:58
*** jhenner has joined #openstack-qa06:11
openstackgerritMarc Koderer proposed a change to openstack/tempest: Create discovery option for stress tests  https://review.openstack.org/4344406:12
openstackgerritKen'ichi Ohmichi proposed a change to openstack/tempest: Add a scenario test for verifying user scope  https://review.openstack.org/4399806:12
*** thomasbiege has joined #openstack-qa06:18
*** thomasbiege has quit IRC06:22
openstackgerritKen'ichi Ohmichi proposed a change to openstack/tempest: Add a scenario test for verifying user scope  https://review.openstack.org/4399806:22
openstackgerritKen'ichi Ohmichi proposed a change to openstack/tempest: Add a scenario test for verifying user scope  https://review.openstack.org/4399806:25
*** dkranz has quit IRC06:39
*** dkranz has joined #openstack-qa06:41
*** tkammer has joined #openstack-qa06:53
*** afazekas has joined #openstack-qa06:58
*** yassine has joined #openstack-qa07:53
*** yassine_ has joined #openstack-qa07:54
*** yassine_ has quit IRC07:55
*** jhenner1 has joined #openstack-qa08:03
*** jhenner has quit IRC08:03
*** afazekas has quit IRC08:03
*** giulivo has joined #openstack-qa08:05
*** thomasbiege has joined #openstack-qa08:16
*** thomasbiege has quit IRC08:23
*** SergeyLukjanov has quit IRC08:24
openstackgerritKen'ichi Ohmichi proposed a change to openstack/tempest: Add a scenario test for verifying user scope  https://review.openstack.org/4399808:37
openstackgerritA change was merged to openstack/tempest: Add unittest framework + tests for wrapper scripts  https://review.openstack.org/4177008:39
*** jhenner1 has quit IRC08:57
*** jhenner has joined #openstack-qa08:58
*** SergeyLukjanov has joined #openstack-qa09:04
*** afazekas has joined #openstack-qa09:06
openstackgerritPavel Sedlák proposed a change to openstack/tempest: Alway log stdout and stderr of CLI commands  https://review.openstack.org/4392309:24
Anjuafazekas : is there any differnce in quntum and neutron?09:26
*** SergeyLukjanov has quit IRC09:54
*** SergeyLukjanov has joined #openstack-qa09:59
openstackgerritChangBo Guo proposed a change to openstack/tempest: show details when nova-manage command fails  https://review.openstack.org/4402210:13
openstackgerritChangBo Guo proposed a change to openstack/tempest: show details when nova-manage command fails  https://review.openstack.org/4402210:15
openstackgerritLucas Alvares Gomes proposed a change to openstack-dev/devstack: Missing pxelinux.0 on RedHat systems  https://review.openstack.org/4402410:33
openstackgerritA change was merged to openstack/tempest: removes self.fail as suggested by HACKING.rst  https://review.openstack.org/4372310:40
*** weshay has joined #openstack-qa11:01
Anjumetrinish : a confusion  :-(11:01
openstackgerritA change was merged to openstack-dev/devstack: Add support for heat enviroments  https://review.openstack.org/4338711:12
*** prekarat1 has quit IRC11:13
mkodererAnju: https://wiki.openstack.org/wiki/Network/neutron-renaming11:16
Anjumkoderer : thanks  :)11:25
*** adalbas has joined #openstack-qa11:28
openstackgerritPavel Sedlák proposed a change to openstack/tempest: Alway log stdout and stderr of CLI commands  https://review.openstack.org/4392311:36
*** Anju has quit IRC11:41
*** jhenner has quit IRC11:48
*** Anju has joined #openstack-qa11:58
*** afazekas has quit IRC12:02
*** afazekas has joined #openstack-qa12:14
*** Anju has quit IRC12:21
*** Anju has joined #openstack-qa12:22
*** Anju has quit IRC12:29
mkoderergiulivo: ping12:30
giulivopong mkoderer12:39
mkoderergiulivo: I had a really contradicting discussion about tempest scenario tests ;)12:39
mkoderermaybe you can give your opinion on that12:40
giulivodoh12:40
mkoderer:)12:40
*** anteaya has joined #openstack-qa12:40
mkoderergiulivo: In my opinion a scenario test should cover a real-life use case12:41
mkodererwould you agree to that?12:41
giulivoyep12:41
mkoderergiulivo: ok, because in the README is a bit different12:41
mkodererhere it sounds like scenario test are to test multiple openstack services12:42
mkodererwhich is also ok but I don't see that this is the main focus ;)12:43
giulivook you mean it looks like an enforcement12:46
giulivoI agree it shouldn't why that would probably happen in every real case scenario12:46
*** afazekas has quit IRC12:46
*** woodspa has joined #openstack-qa12:50
*** SergeyLukjanov has quit IRC12:56
openstackgerritDavanum Srinivas (dims) proposed a change to openstack-dev/devstack: Skip DHCP variables for Nova Network FlatManager  https://review.openstack.org/4346812:58
mkoderergiulivo: yes. I see we have the same understanding :)12:58
mkoderergiulivo: I will ask sdague when he back from vacation12:58
*** jasondotstar has joined #openstack-qa12:59
giulivomkoderer, in my previous line s/why/while/12:59
*** maurosr has quit IRC13:00
*** SergeyLukjanov has joined #openstack-qa13:02
*** afazekas has joined #openstack-qa13:07
*** weshay has quit IRC13:10
*** mriedem has joined #openstack-qa13:13
openstackgerritPavel Sedlák proposed a change to openstack/tempest: Always log stdout and stderr of CLI commands  https://review.openstack.org/4392313:16
*** dkranz has quit IRC13:19
*** weshay has joined #openstack-qa13:24
*** timello_ is now known as timello13:24
*** Anju_ has joined #openstack-qa13:29
mkoderergiulivo: and I am not sure about this one https://review.openstack.org/#/c/43998/13:36
mkoderernever heard about negative scenario tests13:37
giulivoyeah I see what you mean13:37
mkodererI think the test itself is valid13:38
mkodererbut it shouldn't be in scenario tests13:38
Anju_I aggree with u mkoderer13:39
mkodererAnju_: could you describe what you want to change in test_server_basic_ops?13:41
Anju_yeah : giulivo , mkoderer : can we add start/stop test before terminating an instancce13:42
Anju_https://github.com/openstack/tempest/blob/master/tempest/scenario/test_server_basic_ops.py13:42
mkodererIMHO this wouldn't add a real benefit, since it's already covered13:44
Anju_mkoderer: how?13:45
mkoderergiulivo: for me the scope of "basic_ops" is really fuzzy13:46
*** burt has joined #openstack-qa13:46
Anju_afazekas : can u help please13:49
Anju_afazekas :^^^13:50
*** maurosr has joined #openstack-qa13:52
afazekastempest.api.compute.servers.test_server_actions.test_stop_start_server13:55
Anju_afazekas : please a bit more clear13:56
mkodererafazekas: yes but Anju_ want to add it in the scenario test13:56
afazekashttps://github.com/openstack/tempest/blob/master/tempest/api/compute/servers/test_server_actions.py#L28314:03
afazekasmkoderer: an additional question, do we need the test_server_basic_ops.py file at all ?14:04
Anju_afazekas : mkoderer  is no testst agreee with theses14:06
Anju_mkoderer is not agree with these tests14:06
afazekas?14:08
Anju_lest_server_basic_ops14:09
afazekasAnju_: which operation not covered by another test cases ?14:10
Anju_afazekas : my question was why we cant add start /stop tests  here14:11
Anju_before the termination of an instance14:11
Anju_afazekas : https://github.com/openstack/tempest/blob/master/tempest/scenario/test_server_basic_ops.py14:11
Anju_here14:11
afazekasevery instance is terminated14:13
*** sthaha has quit IRC14:13
*** thomasbiege has joined #openstack-qa14:13
Anju_so afazekas :  https://github.com/openstack/tempest/blob/master/tempest/scenario/test_server_basic_ops.py14:14
Anju_here14:14
Anju_can we add self.start_server()14:14
*** annegentle has quit IRC14:15
psedlakafazekas: https://bugs.launchpad.net/openstack-ci/+bug/121779514:16
psedlakafazekas: http://logs.openstack.org/23/43923/4/check/gate-tempest-devstack-vm-postgres-full/aa95e8c/logs/screen-g-api.txt.gz14:23
*** annegentle_ has joined #openstack-qa14:25
mtreinishpsedlak: yeah, I've seen that a couple of times before too14:26
psedlakmtreinish: hm i wonder what can be behind it ...14:26
*** yassine has quit IRC14:31
*** yassine_ has joined #openstack-qa14:32
openstackgerritgongysh proposed a change to openstack-dev/devstack: Use the varialbe to export right keystone api version  https://review.openstack.org/4405914:32
afazekaspsedlka: scheduler: 2013-08-28 12:07:56.744 DEBUG nova.filters [req-90748ae1-6d1d-4e01-be96-f80bf1fae424 demo demo] Filter ComputeFilter returned 0 host(s) get_filtered_objects /opt/stack/new/nova/nova/filters.py:8514:35
afazekasafter 11:44:49.130  no logging in the n-cpu14:36
*** tkammer has quit IRC14:37
mtreinishafazekas: https://review.openstack.org/#/c/43923/14:42
*** dkranz has joined #openstack-qa14:44
*** prekarat has joined #openstack-qa14:45
*** mlavalle has joined #openstack-qa14:45
afazekasmtreinish: +A14:48
mtreinishafazekas: great thanks14:50
mtreinishnow hopefully we can debug the keystone client cli test failures we're hitting14:50
*** yfried has quit IRC14:51
mkoderermtreinish: fixed your finding https://review.openstack.org/#/c/43444/14:55
mtreinishmkoderer: ok, I'll look at it in a bit. But it's probably a +A :)14:57
*** thomasbiege has quit IRC14:59
*** thomasbiege has joined #openstack-qa15:02
*** thomasbiege has quit IRC15:03
*** thomasbiege has joined #openstack-qa15:03
*** thomasbiege has quit IRC15:11
*** dkranz has quit IRC15:29
*** prekarat has quit IRC15:34
*** Anju has joined #openstack-qa15:35
mtreinishmkoderer: ok approved15:36
mtreinishthanks for adding seconds to the comments15:36
mkoderermtreinish: np :)15:38
mkodererafazekas: IMHO we can delete  test_server_basic_ops.py15:41
*** dkranz has joined #openstack-qa15:43
mkodererafazekas: it seems this file is a leftover? maybe we could delete all basic ops tests....?15:44
mtreinishgiulivo: I just saw your meeting topic on the agenda. Just triage all the time :), we need to get things more stable by the end of the week.15:46
mtreinishit should be top priority15:46
giulivomtreinish, yeah I thought a new bugday could have been helpful15:49
giulivoand congrats for the parallel testr15:49
mtreinishgiulivo: it probably would I'm just wondering if we should make it bug rest of the week?15:49
mtreinishgiulivo: thanks!15:49
afazekasmkoderer: ok, I wonder how difficult to implement a stress test case for "random" action order.  The stressAction needs to know when the state transition is allowed.15:52
afazekasmtreinish: do you know about any stability issue which might be related to tempest itself ?15:53
mtreinishafazekas: https://bugs.launchpad.net/tempest/+bug/1217265 this one might be a race between tests but I need to fix the logging from nova-manage before I can be sure15:54
mkodererafazekas: as far all the actions are independent it's quite easy15:55
*** burt has quit IRC15:56
mkodererafazekas: btw I raised a bug https://bugs.launchpad.net/tempest/+bug/121627415:56
afazekasmkoderer: some action might throws 409 or 42215:56
mkodererproblem is that you can't see which action is executed if you use your UnitTest class15:57
mkodererif I have time tomorrow I will fix it15:57
mkoderershouldn't be hard15:57
afazekaspsedlak: https://bugs.launchpad.net/tempest/+bug/121726515:58
psedlakyeah, i'm not sure how 'simple-read-only' the nova-manage calls are (in general) ...16:02
psedlakafazekas: ^^16:02
psedlakesp. something like "archive_deleted_rows" does not seems as read-only for me (or is it expected to fail?)16:03
mtreinishpsedlak: yeah they're not read only16:03
psedlakso if nothing else they are in wrong folder ;)16:04
psedlakhm my cli loggin is still in the gate pipe as it got canceled in the meantime :| after that we can try recheck on that nova-manage as it's master16:06
mtreinishpsedlak: actually a different flaky test just bumped it from the gate queue16:09
* psedlak regrets saying anything16:10
*** Anju_ has quit IRC16:10
*** mlavalle has quit IRC16:12
psedlakafazekas: and guess which one ... postgres, server in error :D16:13
psedlaklibvir: QEMU error : Domain not found: no domain with matching name 'instance-00000055'16:14
psedlakhttp://logs.openstack.org/23/43923/5/gate/gate-tempest-devstack-vm-postgres-full/c38a1fd/logs/screen-ceilometer-acompute.txt.gz16:14
psedlakand it's just 160 lines compared to the one we were looking at before (~560)16:15
openstackgerritYassine Lamgarchal proposed a change to openstack/tempest: Add client and tests for Ceilometer compute metering  https://review.openstack.org/4348116:16
*** hemnafk is now known as hemna16:17
afazekaspsedlak: does the n-cpu does not stopped logging,  so it is a  different issue16:18
psedlakafazekas: seems so16:18
afazekas'./safe-devstack-vm-gate-wrap.sh: line 422: [: missing `]''16:18
psedlakyeah i noticed that :]16:19
afazekashttps://bugs.launchpad.net/heat/+bug/1217329 here some git fetch is missing16:19
*** yassine_ has quit IRC16:20
psedlakafazekas: but i'm not sure if the n-cpu not logging is acceptable evidence that it can be something else16:21
afazekaspsedlak: I used '==' with '['16:21
psedlakafazekas: that's why always [[ ;)16:22
psedlakafazekas: and those ceilo libvirt errors/warnings are 'ok', they appear also in green run16:22
afazekaspsedlak: the services has periodic task, so if they are not logging , they are waiting for something or they are crashed16:23
afazekasin the first case the libvirt lock the is a suspicious thing16:24
afazekasWhat can hold it ? The "dead" n-cpu ? ..16:25
*** SergeyLukjanov has quit IRC16:40
* afazekas EOF16:42
*** thomasbiege has joined #openstack-qa16:47
*** afazekas has quit IRC16:49
*** SergeyLukjanov has joined #openstack-qa17:05
*** burt has joined #openstack-qa17:05
*** thomasbiege has quit IRC17:13
*** fbo is now known as fbo_away17:18
*** thomasbiege has joined #openstack-qa17:43
*** mlavalle has joined #openstack-qa17:47
*** thomasbiege has quit IRC17:52
*** yfried has joined #openstack-qa18:01
*** mlavalle has quit IRC18:05
*** adalbas has quit IRC18:16
*** mlavalle has joined #openstack-qa18:22
*** openstack has joined #openstack-qa18:46
*** Anju has left #openstack-qa18:49
*** SergeyLukjanov has quit IRC18:58
*** SergeyLukjanov has joined #openstack-qa18:59
openstackgerritA change was merged to openstack-dev/devstack: use keystone service port instead of admin port  https://review.openstack.org/4360519:13
openstackgerritDean Troyer proposed a change to openstack-dev/devstack: Add support for Docker as Nova hypervisor  https://review.openstack.org/4075919:20
openstackgerritDolph Mathews proposed a change to openstack/tempest: dump keystoneclient output if table label is None  https://review.openstack.org/4413519:26
*** adalbas has joined #openstack-qa19:27
*** yfried has quit IRC19:38
*** mlavalle has quit IRC19:40
*** mlavalle has joined #openstack-qa19:41
*** dkranz has quit IRC19:41
*** mlavalle has quit IRC19:41
*** dkranz has joined #openstack-qa19:42
*** mlavalle has joined #openstack-qa19:48
*** SergeyLukjanov has quit IRC20:05
*** thomasbiege has joined #openstack-qa20:10
*** thomasbiege has quit IRC20:11
*** fbo_away is now known as fbo20:13
*** fbo is now known as fbo_away20:13
dkranzmtreinish: What do you think of my review of https://review.openstack.org/#/c/41038/ ?20:23
mtreinishdkranz: yeah I tend to agree with what you said20:24
mtreinishwe don't want to merge code that isn't run in the gate20:24
mtreinishbut I'm wondering if we should stop accepting neutron fixes until the full run starts working again.20:25
dkranzmtreinish: I had a talk with Jim Blair about this. He sent an email to the neutron ptl asking what was up with this.20:26
dkranzmtreinish: I was afraid that moving full neutron to experimental would lead to out-of-site, out-of-mind but he was right to do it.20:26
dkranzmtreinish: Let's see what mlavalle has to say at the meeting tomorrow.20:26
mtreinishdkranz: yeah it was just wasted resources eating a devstack node that can be better used for another tempest job20:27
mtreinishdkranz: yeah we should bring this all up during the meeting tomorrow20:27
dkranzmtreinish: And training people to ignore failing jobs.20:27
dkranzmtreinish: I think accepting tests but only as smoke is a reasonable temporizing measure20:27
mtreinishdkranz: yeah at least they'll be run. I'm just worried about increasing the runtime too much.20:28
mtreinishright now it's faster than the parallel jobs because it's such a small set20:28
mtreinishalthough I doubt we'll add so many tests to it that will make go above ~25min20:29
dkranzmtreinish: Oh yeah. Hopefully that parallel neutron bug will be fixed soon.20:29
dkranzmtreinish: mlavalle seemed to have a devstack change that would fix it.20:29
dkranzmtreinish: Or so it says in the ticket.20:29
dkranzmtreinish: ANd that would be a good problem to have :)20:29
mtreinishyeah I talked to mlavelle and afazekas about it yesterday. We worked out a plan to get around it and mlavalle said he take care of implementing it20:30
dkranzmtreinish: Great. I should add a heat update to the agenda, which is getting long.20:31
mtreinishdkranz: yeah it is, but we'll get to what we can I'll do a priority reorder before the meeting20:32
*** adalbas has quit IRC20:34
openstackgerritA change was merged to openstack/tempest: Always log stdout and stderr of CLI commands  https://review.openstack.org/4392320:57
mtreinishpsedlak: ^^^ now lets find a failure :)20:58
*** thomasbiege has joined #openstack-qa20:59
mlavallemtreinish, dkranz: Yeah, I am implementing the plan we talked about yesterday. I will have some code to push for review over the weekend21:01
*** dkranz has quit IRC21:02
*** jasondotstar has quit IRC21:19
*** mlavalle has quit IRC21:21
*** thomasbiege has quit IRC21:31
mtreinishpsedlak: any ideas: http://logs.openstack.org/66/43966/1/gate/gate-tempest-devstack-vm-postgres-full/630986e/logs/tempest.txt.gz#_2013-08-28_21_19_32_46221:32
mtreinishthat one hit the failure: http://logs.openstack.org/66/43966/1/gate/gate-tempest-devstack-vm-postgres-full/630986e/testr_results.html.gz21:32
*** mriedem has quit IRC21:38
mtreinishjog0: ^^^ I'm looking at the logs now I'm not seing anything too weird looking21:39
*** thomasbiege has joined #openstack-qa21:41
jog0mtreinish: which one?21:48
*** mlavalle has joined #openstack-qa21:49
mtreinishjog0: http://logs.openstack.org/66/43966/1/gate/gate-tempest-devstack-vm-postgres-full/630986e/logs/tempest.txt.gz#_2013-08-28_21_19_32_462 that's the output when the keystoneclient cli test fails on catalog21:49
mtreinishit's up there for top flaky test21:50
jog0ahh21:50
jog0I need to head out but just a hunch 'Service: type--47301037521:56
jog0'21:56
jog0that may be  breaking the table parser21:56
mtreinishjog0: yeah I manually am running the parser and that's what it looks like21:56
mtreinishthe result for that table is messed up21:56
jog0yeah so  its the test itself sadly :/21:57
mtreinishit probably is just a random name usage bug21:57
mtreinishwe should probably fix the parser too21:57
jog0yeah anyway I need to head but looks like we have this one mostly sorted out21:58
jog0mtreinish: thanks for working on these bugs21:58
mtreinishjog0: I was the one who pushed them into the gate. I figure I should at least help fix them21:59
*** thomasbiege has quit IRC22:03
openstackgerritMatthew Treinish proposed a change to openstack/tempest: Remove '--' from rand_name calls  https://review.openstack.org/4416022:11
openstackgerritMatthew Treinish proposed a change to openstack/tempest: Add a hacking rule for dashes with rand_name  https://review.openstack.org/4416822:30
*** dkranz has joined #openstack-qa22:31
*** woodspa has quit IRC22:31
*** kashyap has quit IRC22:32
*** kashyap has joined #openstack-qa22:32
mtreinishjog0: actually I double checked the parser results its not the double dash that got picked up fine22:40
mtreinishhttp://paste.openstack.org/show/45326/22:40
mtreinishit's the 2 subtables for that same entry22:41
mtreinishalso a good indicator that my fix patch failed jenkins: http://logs.openstack.org/60/44160/1/check/gate-tempest-devstack-vm-full/9cc7665/testr_results.html.gz22:41
openstackgerritMatthew Treinish proposed a change to openstack/tempest: Remove '--' from rand_name calls and enforce it  https://review.openstack.org/4416822:48
*** mlavalle has left #openstack-qa23:05
openstackgerritA change was merged to openstack/tempest: Create discovery option for stress tests  https://review.openstack.org/4344423:09
openstackgerritMatthew Treinish proposed a change to openstack/tempest: Fix test_admin_catalog_list  https://review.openstack.org/4417823:13
clarkbmtreinish: ^ that fixes the keystone cli flaky test?23:15
mtreinishclarkb: It should23:15
*** cyeoh is now known as cyeoh_23:19
mtreinishclarkb: there also might be a bug in keystoneclient, but regardless that should get merged because it just makes it a better test that won't fail23:19
mtreinishfor no reason23:19
*** jaypipes has quit IRC23:21
*** marun has quit IRC23:22
openstackgerritA change was merged to openstack/tempest: Rename heat logical_resource_id to resource_name  https://review.openstack.org/4381923:24
openstackgerritgongysh proposed a change to openstack-dev/devstack: Use the varialbe to export right keystone api version  https://review.openstack.org/4405923:25

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!