Thursday, 2014-09-11

openstackgerritEmilien Macchi proposed a change to openstack/tempest: javelin: add network and secgroup resources  https://review.openstack.org/10016900:02
openstackgerritChris Dent proposed a change to openstack-dev/devstack: Replace screen_it() with run_process() throughout  https://review.openstack.org/12007700:07
cdentjogo, dtroyer ^ I have to give up for the night00:11
*** dmellado has quit IRC00:11
*** aimon has quit IRC00:13
*** dmellado has joined #openstack-qa00:13
*** dmorita has joined #openstack-qa00:24
*** harlowja_ has quit IRC00:25
*** cdent has quit IRC00:25
*** harlowja has joined #openstack-qa00:28
*** weshay has quit IRC00:30
*** aimon has joined #openstack-qa00:37
openstackgerritMiguel Lavalle proposed a change to openstack/tempest: Fix tempest.scenario.manager.ScenarioTest  https://review.openstack.org/11965900:38
*** suneelb has quit IRC00:46
*** dimtruck is now known as zz_dimtruck00:49
*** jpich has quit IRC00:51
*** yamahata has joined #openstack-qa00:51
*** jpich has joined #openstack-qa01:07
*** zz_dimtruck is now known as dimtruck01:24
*** dims_ has joined #openstack-qa01:32
*** amotoki has joined #openstack-qa01:34
*** mlavalle has quit IRC01:35
*** oomichi has joined #openstack-qa01:37
*** Sukhdev has joined #openstack-qa01:40
*** Sukhdev has quit IRC01:42
*** Sukhdev has joined #openstack-qa01:43
*** yjiang5 has quit IRC01:44
*** nosnos has joined #openstack-qa01:47
*** Sukhdev has quit IRC01:47
*** jaypipes has quit IRC01:58
*** rodrigods_ has joined #openstack-qa01:59
*** markmcclain has quit IRC02:05
*** jpich has quit IRC02:06
*** jyuso has joined #openstack-qa02:09
*** amotoki_ has joined #openstack-qa02:10
*** amotoki has quit IRC02:11
*** amotoki_ is now known as amotoki02:11
*** rodrigods_ has quit IRC02:14
*** dkranz has joined #openstack-qa02:16
*** jpich has joined #openstack-qa02:19
*** markmcclain has joined #openstack-qa02:26
*** sileht has quit IRC02:46
*** dims_ has quit IRC02:49
*** dims_ has joined #openstack-qa02:49
*** dims_ has quit IRC02:54
*** rodrigods_ has joined #openstack-qa03:00
*** rodrigods_ has quit IRC03:00
*** arnaud has quit IRC03:01
*** adalbas has quit IRC03:05
*** rfolco has joined #openstack-qa03:29
*** Poornima has joined #openstack-qa03:29
*** Longgeek has joined #openstack-qa03:29
*** Longgeek_ has joined #openstack-qa03:32
*** Longgeek has quit IRC03:36
*** rodrigods_ has joined #openstack-qa03:37
*** dimtruck is now known as zz_dimtruck03:42
*** rodrigods_ has quit IRC03:45
*** jpich has quit IRC03:58
*** zz_dimtruck is now known as dimtruck04:02
*** harlowja is now known as harlowja_away04:05
*** jpich has joined #openstack-qa04:10
*** krtaylor has joined #openstack-qa04:11
*** dimtruck is now known as zz_dimtruck04:11
*** chandankumar has joined #openstack-qa04:19
openstackgerritA change was merged to openstack/tempest: Allow out of quota failure status code to be 413 or 403  https://review.openstack.org/11809904:28
*** Poornima has quit IRC04:31
*** markmcclain has quit IRC04:34
*** HenryG is now known as HenryG_zzzz04:40
*** Sukhdev_ has joined #openstack-qa04:43
*** yamahata has quit IRC04:46
*** yamahata has joined #openstack-qa04:47
*** jaypipes has joined #openstack-qa04:54
*** ChanServ sets mode: +o jaypipes04:54
*** Madan has joined #openstack-qa04:57
*** Poornima has joined #openstack-qa04:57
*** torandu has quit IRC05:05
*** pcrews has quit IRC05:07
*** sabeen has quit IRC05:09
*** Madan has quit IRC05:18
*** neeti has joined #openstack-qa05:28
*** Madan has joined #openstack-qa05:33
*** nmagnezi has joined #openstack-qa05:44
*** yfried has joined #openstack-qa05:44
*** rdekel has joined #openstack-qa05:49
*** flaper87|afk is now known as flaper8705:58
*** nmagnezi has quit IRC05:59
*** vigneshvar has joined #openstack-qa05:59
*** vigneshvar_ has joined #openstack-qa06:01
*** vigneshvar_ has quit IRC06:03
*** oomichi_ has joined #openstack-qa06:03
*** oomichi has quit IRC06:05
*** zz_dimtruck is now known as dimtruck06:10
*** mpaolino has joined #openstack-qa06:12
*** andreaf has joined #openstack-qa06:13
*** SridharG has joined #openstack-qa06:17
*** Longgeek_ has quit IRC06:17
*** Longgeek has joined #openstack-qa06:18
*** dimtruck is now known as zz_dimtruck06:20
*** Longgeek has quit IRC06:20
*** mauricioliima has quit IRC06:23
*** Sukhdev_ has quit IRC06:23
*** dkranz has quit IRC06:28
*** dkranz has joined #openstack-qa06:28
*** dkranz has quit IRC06:33
*** dkranz has joined #openstack-qa06:34
*** Madan1 has joined #openstack-qa06:34
*** k4n0 has joined #openstack-qa06:34
*** afazekas is now known as __afazekas06:34
*** Madan1 has quit IRC06:35
*** Madan1 has joined #openstack-qa06:35
*** Madan has quit IRC06:36
*** gszasz has joined #openstack-qa06:37
openstackgerritDaisuke Morita proposed a change to openstack/tempest: Replace confusing member name  https://review.openstack.org/12070606:37
*** rhsu has quit IRC06:49
*** cmyster_away is now known as cmyster06:49
*** rfolco has quit IRC06:51
*** Longgeek has joined #openstack-qa06:52
*** mrodden has quit IRC06:56
*** mrodden has joined #openstack-qa06:58
*** jlibosva has joined #openstack-qa06:59
*** Qarekhani has quit IRC07:03
*** sileht has joined #openstack-qa07:03
openstackgerritDaisuke Morita proposed a change to openstack/tempest: Add a test for deleting multiple objects by POST method  https://review.openstack.org/11654707:06
*** chandankumar has quit IRC07:09
*** ildikov has joined #openstack-qa07:09
*** dkranz has quit IRC07:18
*** jpich has quit IRC07:20
*** dkranz has joined #openstack-qa07:20
openstackgerritA change was merged to openstack/tempest: Migrate security_groups_basic to tempest clients  https://review.openstack.org/11219907:21
yfriedmasayukig: andreaf: could you please take a look at this. it's ready for merge. https://review.openstack.org/#/c/113124/07:22
openstackgerritBenny Kopilov proposed a change to openstack/tempest: glance v1 add image with duplicate id  https://review.openstack.org/11961007:23
*** sileht has quit IRC07:29
*** sileht has joined #openstack-qa07:29
*** e0ne has joined #openstack-qa07:32
*** dkranz has quit IRC07:34
*** dkranz has joined #openstack-qa07:34
*** ashishg has joined #openstack-qa07:35
*** chandankumar has joined #openstack-qa07:39
*** jlanoux has joined #openstack-qa07:40
yfriedmasayukig: ping07:44
*** oomichi_ has quit IRC07:45
*** qba73 has joined #openstack-qa07:46
yfrieddoes anyone know why we expose the clients in specifically in scenario.manager.py? why can't we use "self.clients.network_clients" instead of "self.network_clients"?07:46
*** pkoniszewski has joined #openstack-qa07:47
*** dkranz has quit IRC07:50
*** dkranz has joined #openstack-qa07:52
ashishgdoes anybody know why i am hitting conflict issue with https://review.openstack.org/#/c/115527/07:54
*** cnesa9 has joined #openstack-qa08:00
openstackgerritMarc Koderer proposed a change to openstack/tempest: Add test to validate negative testing schema's  https://review.openstack.org/12003308:00
*** jordanP has joined #openstack-qa08:05
*** afazekas_ has joined #openstack-qa08:05
*** vigneshvar has quit IRC08:11
*** ianw has quit IRC08:18
*** ianw has joined #openstack-qa08:19
*** vigneshvar has joined #openstack-qa08:24
*** dkranz has quit IRC08:25
*** dkranz has joined #openstack-qa08:26
*** mpavlase has joined #openstack-qa08:30
*** gszasz has quit IRC08:33
*** zz_johnthetubagu is now known as johnthetubaguy08:34
*** dkranz has quit IRC08:36
*** dkranz has joined #openstack-qa08:36
*** Ch00k has joined #openstack-qa08:38
*** gszasz has joined #openstack-qa08:40
*** ianw has quit IRC08:41
*** ianw has joined #openstack-qa08:42
*** andreaf has quit IRC08:47
*** oomichi has joined #openstack-qa08:52
*** pkoniszewski has quit IRC08:54
*** mkollaro has joined #openstack-qa08:54
gmannashishg: may be this will help - fetch your patch - (git fetch https://review.openstack.org/openstack/tempest refs/changes/27/115527/1 && git checkout FETCH_HEAD)08:55
*** e0ne has quit IRC08:55
gmannashishg: create a patch from this  - git log -p -n1 > <new patch file path>08:56
gmannashishg: checkout the master, update it and the apply your patch on that new branch from master by - patch -p1 < <created patch file>08:56
gmannashishg: if there is any conflict those you need to resolve manually08:57
*** e0ne has joined #openstack-qa09:01
*** gfidente has quit IRC09:04
*** gfidente has joined #openstack-qa09:05
*** oomichi has quit IRC09:10
*** dkranz has quit IRC09:11
*** dkranz has joined #openstack-qa09:11
*** rdekel has quit IRC09:13
afazekas_andreaf: https://review.openstack.org/#/c/120386/09:14
*** mkollaro has quit IRC09:14
yfriedandreaf_: a quick question about https://blueprints.launchpad.net/tempest/+spec/resource-cleanup09:15
yfriedandreaf_: to you mean that we want be able to override setUpClass, but instead we have our own "setup_class()" we can override, that will be always called form ancestor's setUpClass?09:18
ashishg<gmann>: apply your patch on that new branch from master by - patch -p1 < <created patch file>  not able to get it ?09:19
gmannashishg: after creating the patch file. checkout the master and then creat new branch from master (git checkout -b <new branhc name>)09:23
*** rfolco has joined #openstack-qa09:23
gmannashishg: then on that new branch run  patch -p1 < <created patch file> . it will apply the patch to new branch09:24
*** mkollaro has joined #openstack-qa09:26
*** rdekel has joined #openstack-qa09:26
*** aix has joined #openstack-qa09:26
openstackgerritafazekas proposed a change to openstack-infra/devstack-gate: multi-node dg setup  https://review.openstack.org/10604309:27
*** dkranz has quit IRC09:27
*** dkranz has joined #openstack-qa09:30
openstackgerritA change was merged to openstack/tempest: Move response code checking to client for v3 identity tests  https://review.openstack.org/11777809:31
openstackgerritNir Magnezi proposed a change to openstack/tempest: Adds status check for router Port in scenarios  https://review.openstack.org/11745409:31
openstackgerritA change was merged to openstack/tempest: Add baremetal API tests about console  https://review.openstack.org/11771509:31
*** mpaolino has quit IRC09:34
*** dkranz has quit IRC09:45
*** dkranz has joined #openstack-qa09:46
*** dims_ has joined #openstack-qa09:51
*** dims_ has quit IRC10:00
*** dims_ has joined #openstack-qa10:01
*** mpaolino has joined #openstack-qa10:03
ashishg<gmann>I was able to resolve the conflicts but in the commit message when i perform git commit -a --amend i am seeing some other commits message(Merge "Migrate security_groups_basic to tempest clients") should i edit the message or i am doing something wrong10:03
gmannashishg: that great. for commit msg as this is new branch you need to write the commit msg again. just copy paste from gerrit10:05
gmanndo not runt git commit --amend. just run 'git commit'10:05
*** dims_ has quit IRC10:06
ashishg<gmann> performing tsocks git review getting error : tsocks git review Errors running git rebase -i remotes/gerrit/master error: could not apply 47100bb... Port  API Tests Enhancement When you have resolved this problem run "git rebase --continue". If you would prefer to skip this patch, instead run "git rebase --skip". To check out the original branch and stop rebasing run "git rebase --abort". Could not apply 47100bb... Port10:14
ashishg<gmann > Switching to some no branch10:15
gmannashishg: how you created your new branch from master only right?10:16
ashishg<gmann> yes10:17
openstackgerritRohan Kanade proposed a change to openstack/tempest: Add subnet tests for extra attributes  https://review.openstack.org/10632610:20
ashishg<gmann> git checkout master git checkout -b Test-Port-User-Defined-Mac110:20
gmannashishg: what patch file shows? only your changes right10:21
*** rdekel has quit IRC10:23
ashishg<gmann>git checkout -b Test-Port-User-Defined-Mac1 patch -p1 < ashish.patch patching file tempest/api/network/test_ports.py Hunk #1 succeeded at 233 (offset -5 lines). git status # On branch Test-Port-User-Defined-Mac1 # Changes not staged for commit: #   (use "git add <file>..." to update what will be committed) #   (use "git checkout -- <file>..." to discard changes in working directory) # #       modified:   tempest/api/netw10:24
*** marun_afk has joined #openstack-qa10:25
gmannashishg: git status? only your 1 file change right10:27
ashishg<gmann> yes10:27
*** yamahata has quit IRC10:31
gmannashishg: just check manually in that file for any conflict10:32
ashishg<gmann> yes it has <<<<<<< HEAD  >>>>>>> 47100bb... Port  API Tests Enhancement10:36
*** e0ne has quit IRC10:37
gmannashishg: :). just fix those. then it will work10:37
*** jyuso has quit IRC10:49
*** marun_afk is now known as marun10:56
*** rodrigods_ has joined #openstack-qa11:07
*** e0ne has joined #openstack-qa11:10
*** johnthetubaguy is now known as zz_johnthetubagu11:12
*** marun has quit IRC11:15
*** dims_ has joined #openstack-qa11:17
*** rodrigods_ has quit IRC11:21
mkodererdkranz: could you have a look to https://review.openstack.org/#/c/120033/11:24
mkodererdkranz: some of the existing schemas are already broken :(11:24
*** dims_ has quit IRC11:30
*** dims_ has joined #openstack-qa11:31
*** ashishg has quit IRC11:32
*** dkranz has quit IRC11:32
*** dims__ has joined #openstack-qa11:34
*** dims_ has quit IRC11:35
openstackgerritTomas Sedovic proposed a change to openstack/tempest: Use the AWS instance type in the cfn_init scenario  https://review.openstack.org/12074911:38
*** cnesa9 has quit IRC11:43
afazekas_mkoderer, dkranz: https://review.openstack.org/#/c/120386/11:43
*** enikanorov_ has joined #openstack-qa11:54
*** enikanorov__ has quit IRC11:54
*** andreaf has joined #openstack-qa11:54
*** adalbas has joined #openstack-qa11:57
*** mwagner_lap has quit IRC11:59
*** bdperkin has quit IRC12:02
*** bdperkin has joined #openstack-qa12:02
*** weshay has joined #openstack-qa12:08
*** markmcclain has joined #openstack-qa12:09
*** dmorita has quit IRC12:11
openstackgerritTomas Sedovic proposed a change to openstack/tempest: Use the AWS instance type in the cfn_init scenario  https://review.openstack.org/12074912:12
*** mwagner_lap has joined #openstack-qa12:15
*** e0ne has quit IRC12:16
*** aysyd has joined #openstack-qa12:20
*** zz_jgrimm is now known as jgrimm12:20
*** nosnos has quit IRC12:22
*** zz_dimtruck is now known as dimtruck12:22
*** nosnos has joined #openstack-qa12:22
*** masayukig has quit IRC12:24
openstackgerritAndrea Frittoli  proposed a change to openstack/qa-specs: add bp:resource-cleanup  https://review.openstack.org/11835212:25
*** nosnos has quit IRC12:27
*** cnesa9 has joined #openstack-qa12:28
*** masayukig has joined #openstack-qa12:30
*** mpaolino has quit IRC12:31
*** dkranz has joined #openstack-qa12:32
*** zz_johnthetubagu is now known as johnthetubaguy12:37
*** Madan1 has quit IRC12:47
*** Poornima has quit IRC12:48
*** dims__ has quit IRC12:49
*** dims_ has joined #openstack-qa12:50
*** dims_ is now known as dims12:51
*** marun has joined #openstack-qa12:53
*** HenryG_zzzz is now known as HenryG12:55
*** jgrimm is now known as zz_jgrimm13:03
*** neeti has quit IRC13:10
andreafdkranz: ping - regarding  https://review.openstack.org/#/c/118352 - resource cleanup spec, I fixed two typos and added an etherpad link, if you have time for a quick re-review13:11
*** dustins has joined #openstack-qa13:13
dkranzandreaf: ok, done13:13
andreafdkranz: thank you!13:13
andreafmtreinish: ping - regarding resource cleanup spec (https://review.openstack.org/#/c/118352) - do you think we got enough reviews? I'd like to start working on that pretty soon if possible13:15
*** vrovachev has joined #openstack-qa13:22
*** dimtruck is now known as zz_dimtruck13:24
*** mriedem has joined #openstack-qa13:24
mtreinishandreaf: ok, I'll take a look at it now13:27
andreafmtreinish, thank you13:28
*** pcrews has joined #openstack-qa13:31
*** jecarey has joined #openstack-qa13:32
*** ylobankov has joined #openstack-qa13:32
*** torandu has joined #openstack-qa13:33
*** sabeen has joined #openstack-qa13:34
*** weshay has quit IRC13:36
*** weshay has joined #openstack-qa13:39
*** toure has joined #openstack-qa13:39
*** Longgeek has quit IRC13:41
*** e0ne has joined #openstack-qa13:42
*** chandankumar has quit IRC13:47
*** bobj has joined #openstack-qa13:50
bobjHello everybody. I'm trying run stress test on object storage module using tempest, I using @test.stresstest but these approach doesn't work. Somebody can help me?13:50
*** balles has joined #openstack-qa13:50
mtreinishmkoderer: ^^^13:53
*** samuelbercovici has joined #openstack-qa13:58
*** markmcclain has quit IRC13:58
*** arnaud has joined #openstack-qa14:00
mtreinishandreaf: reviewed, if you respin it quickly we can fast path it through14:02
mtreinishbobj: you need to provide a bit more info on why it doesn't work14:03
*** mtreinish changes topic to "All Things QA..."14:03
*** Longgeek has joined #openstack-qa14:05
*** mpaolino has joined #openstack-qa14:14
*** zz_dimtruck is now known as dimtruck14:14
*** zz_jgrimm is now known as jgrimm14:15
*** rbak has joined #openstack-qa14:15
*** mpavlase has quit IRC14:17
*** mpaolino has quit IRC14:17
*** amotoki has quit IRC14:21
*** yamahata has joined #openstack-qa14:23
*** dmellado_ has joined #openstack-qa14:26
*** vigneshvar has quit IRC14:31
*** dmellado_ has quit IRC14:32
openstackgerritgit-harry proposed a change to openstack/tempest: VolumeMultiBackendTest: delete error volumes  https://review.openstack.org/12079314:33
*** gordc has joined #openstack-qa14:36
*** amotoki has joined #openstack-qa14:38
openstackgerritSean Dague proposed a change to openstack/tempest: always check servers are active before rescuing  https://review.openstack.org/12079614:40
*** arnaud has quit IRC14:40
sdaguedkranz: so that's based on a bug you triaged, where I think the tempest side is actually part of the problem (or at least not helping in the solution)14:41
hockeynutGreetings all - trying to figure out where our Barbican tests need to live (in our Barbican repo or in tempest repo).  Was under the impression that they live in Barbican repo, so that's where our devstack gated tests live for now.  Do they need to move?14:42
*** Ch00k has quit IRC14:42
*** Ch00k_ has joined #openstack-qa14:42
*** pkoniszewski has joined #openstack-qa14:45
mtreinishhockeynut: it depends on what the tests are doing14:45
mtreinishhockeynut: that being said, having some tempest coverage is a graduation requirement though14:47
mtreinishthe tests which would be best to have in tempest are ones that tickle integration between barbican and another project14:47
*** markmcclain has joined #openstack-qa14:47
hockeynutmtreinish - these are our functional tests - only integration is with keystone for auth14:48
*** Ch00k_ has quit IRC14:48
mtreinishhockeynut: do you have a link?14:48
*** markmcclain1 has joined #openstack-qa14:49
*** markmcclain1 has quit IRC14:49
hockeynutmtreinish https://github.com/openstack/barbican/tree/master/functionaltests14:49
*** balles has quit IRC14:49
*** markmcclain1 has joined #openstack-qa14:50
*** samuelbercovici has quit IRC14:51
hockeynutmtreinish that is the beginning - we are adding more, but want to be sure we're on the right track14:51
mtreinishhockeynut: ok so that kinda of testing is something we're pushing to have back in the projects (right now they're in tempest)14:52
*** dmorita has joined #openstack-qa14:52
mtreinishbut, the way you're using tempest inside barbican like that probably isn't the best idea14:52
*** markmcclain has quit IRC14:52
mtreinishbecause tempest doesn't provide any stability on it's interfaces14:52
hockeynutmtreinish I'm open to suggestions...14:52
mtreinishhockeynut: what you really want is: http://specs.openstack.org/openstack/qa-specs/specs/tempest-library.html14:53
mtreinishbut that's still a WIP14:53
hockeynutmtreinish - for our info, what kind of tests are supposed to go into tempest repo?  Anything we need for Barbican?14:53
*** balles has joined #openstack-qa14:54
hockeynutmtreinish just integration?  or other?14:54
*** cdent has joined #openstack-qa14:55
mtreinishhockeynut: ideally long term it'll be just integration. But for right now I'm thinking that those tests maybe should be in tempest (which I'm reluctant to say especially if we're going to ask you to move them back in a cycle or 2)14:55
*** gszasz has quit IRC14:55
mtreinishbut it really is a project level decision to a certain degree, what goes in tempest is waht you want to be co-gating when barbican is integrated14:55
mtreinishand you still need to setup a devstack-gate job that runs tempest with barbican enabled14:56
hockeynutmtreinish we do have that job and its working today14:56
mtreinishwell, if it runs tempest then it should be an experimental job on tempest, and any other projects which will start using barbican then14:57
hockeynutalso I have no problem with the WIP aspect - that's business as usual :-)14:57
hockeynutright now we don't do anything w/experimental.14:57
mtreinishhockeynut: http://git.openstack.org/cgit/openstack/tempest-lib/ well it hasn't even existed for a week yet :)14:57
*** cbader has joined #openstack-qa14:57
mtreinishit's only for cli testing right now, and I haven't pushed it to pypi yet14:58
*** atiwari has joined #openstack-qa14:58
hockeynuta week is better than nothing :-)  Howsabout we stick with our current setup and when you're happy with the tempest-lib then we migrate?14:59
dkranzsdague: I agree with the comments Matt R made in the ticket and that was the reason I pushed it to nova. I could be wrong of course.14:59
sdaguedkranz: did you see my proposed tempest side enhancement for it?14:59
*** marun has quit IRC15:00
dkranzsdague: yes15:00
bobjmtreinish I'm using @test.stresstest on test_create_object, but when I ran the tempest and checked results I get test_create_object[gate,stress] OK  0.96. However I think these time is very low to a stress test execution15:00
bobjmtreinish: I'm using @test.stresstest on test_create_object, but when I ran the tempest and checked results I get test_create_object[gate,stress] OK  0.96. However I think these time is very low to a stress test execution15:00
*** dimtruck is now known as zz_dimtruck15:01
mtreinishhockeynut: that's fine, but you'll still need to have something in tempest. If there are tests from your functional test suite that you think makes sense to eventually be co-gating you should put that in tempest15:01
*** k4n0 has quit IRC15:01
hockeynuthow do you determine if a test needs to be co-gating?15:02
dkranzsdague: It seemed to me that setUpClass and all the addCleanUps mean that the test is always ACTIVE when it starts15:02
*** afazekas_ has quit IRC15:03
hockeynutwould that be other components who have a dependency on Barbican?15:03
dkranzsdague: So I don't think your change is wrong but I don't see how it fixes the problem is there was one.15:03
*** zz_dimtruck is now known as dimtruck15:04
mtreinishhockeynut: yeah, that's one way. Another is to have a test case which involves more than just barbican to work15:04
*** balles has quit IRC15:04
hockeynutok - that makes sense15:04
*** SridharG has left #openstack-qa15:05
hockeynutmtreinish thanks for the info - greatly appreciated!!15:05
mtreinishbobj: if you've decoratored the test you'll need to use the run_stress command to run it as a stress test. Doing a normal test run doesn't do that15:05
mtreinishbobj: see the stress tox job in tox.ini for an example. I also think there is something in the tempest docs about using the command15:05
sdaguedkranz: well, it would at least make it more clear to debug15:05
*** balles has joined #openstack-qa15:06
mtreinishhockeynut: sure np, if you need anymore help feel free to ping me15:06
hockeynutyou can bet on it :-)15:07
*** Longgeek has quit IRC15:07
bobjmtreinish: I'm looking a little bit more in the tempest docs and I'll try to run using tox.ini too. Thank!15:11
andreafmtreinish: ok thanks for your comments. I'm fine with the approach of you proposed of first making the setupclass safe and then do the staged approach15:11
andreafmtreinish: for the names, is something like safe_class_setup and safe_class_teardown or did you have anything specific in mind15:13
mtreinishandreaf: no I didn't have anything in mind, I just thought that calling it setup_class wasn't really different enough15:15
*** yamahata has quit IRC15:15
*** marun has joined #openstack-qa15:17
andreafmtreinish: well I don't know it's what they are - it could be pre_tests_hook - once we go to the staged approach it will be split in several methods anyways15:18
mtreinishandreaf: well, it was a suggestion. I was just thinking about trying to talk about it in person, how would you know which one we were talking about15:19
mtreinishI'll leave it up to you though15:19
*** markmcclain1 has quit IRC15:20
*** markmcclain has joined #openstack-qa15:20
*** balles has quit IRC15:24
*** balles has joined #openstack-qa15:26
*** rhsu has joined #openstack-qa15:34
openstackgerritJim Rollenhagen proposed a change to openstack/tempest: Support lack of ephemeral volumes for baremetal  https://review.openstack.org/12081415:34
*** atiwari has quit IRC15:35
*** jgrimm is now known as zz_jgrimm15:37
openstackgerritJim Rollenhagen proposed a change to openstack-infra/devstack-gate: Don't use ephemeral volumes for Ironic agent driver  https://review.openstack.org/12081715:39
openstackgerritJim Rollenhagen proposed a change to openstack/tempest: Support lack of ephemeral volumes for baremetal  https://review.openstack.org/12081415:43
openstackgerritTomas Sedovic proposed a change to openstack/tempest: Use the AWS instance type in the cfn_init scenario  https://review.openstack.org/12074915:43
*** cbader_ has joined #openstack-qa15:43
openstackgerritJohn Davidge proposed a change to openstack-dev/devstack: Add IPv6 support for tenant data network  https://review.openstack.org/8798715:44
openstackgerritJim Rollenhagen proposed a change to openstack/tempest: Support lack of ephemeral volumes for baremetal  https://review.openstack.org/12081415:45
*** zz_jgrimm is now known as jgrimm15:48
*** balles has quit IRC15:50
*** ZZelle has quit IRC15:52
*** ZZelle has joined #openstack-qa15:53
*** atiwari has joined #openstack-qa15:53
*** balles has joined #openstack-qa15:54
*** Ch00k has joined #openstack-qa15:57
*** bkopilov has quit IRC15:58
openstackgerritMauro Rodrigues proposed a change to openstack/tempest: Change LOG.exception to LOG.info for skipped tests  https://review.openstack.org/12082416:01
openstackgerritMauro Rodrigues proposed a change to openstack/tempest: Fix safe_setup decorator to not call tearDownClass for skipped tests  https://review.openstack.org/12082516:01
openstackgerritMauro Rodrigues proposed a change to openstack/tempest: Add unittests for safe setup decorator  https://review.openstack.org/12082616:01
cdentdtroyer, jogo, sdague: This change https://review.openstack.org/#/c/120077/ is now passing its tests but only because nova-compute has gone back to using screen_it to start up. When using run_process, the cells job fails.16:02
cdentI fought with it last night to try to figure out why but the late hour and my ignorance got the better of me. Any ideas/help?16:02
sdaguecdent: it works for the non cells jobs?16:02
cdentYup16:02
sdagueso... honestly, I'd say propose deleting the cells job at the same time16:02
sdagueit doesn't do anything useful16:02
*** jlanoux has quit IRC16:03
*** mlavalle has joined #openstack-qa16:03
cdentthis is the failure from before I switched it back to screen it: http://logs.openstack.org/77/120077/7/check/check-devstack-dsvm-cells/f5054ad/16:04
cdentI'd be curious to know what's going on because clearly there is something different between how the process is started and that difference is meaningful16:05
cdentand potentially a clue to other broken crap16:05
cdentthat is: we have a situation now where a minor change is making a significant difference in behavior, if we can determine what that is we'll have super powers16:05
cdentbut at the moment, I'm stuck on how to dig effectively16:05
cdentbrb16:06
cdent(note the failure happens in a local devstack too)16:07
*** e0ne has quit IRC16:09
*** bkopilov has joined #openstack-qa16:12
*** qba73 has quit IRC16:13
sdaguecdent: actually, what's the version that fails cells?16:14
*** qba73 has joined #openstack-qa16:14
*** yjiang5 has joined #openstack-qa16:16
cdentsorry sdague had to run something to the neighbors, what you after?16:17
sdagueI wanted to see the patch version that worked on everything except cells16:17
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Remove query for (fixed) bug 1338844  https://review.openstack.org/12083116:18
cdenthttps://review.openstack.org/#/c/120077/716:18
*** jordanP has quit IRC16:18
*** markmcclain has quit IRC16:19
*** qba73 has quit IRC16:20
cdenthowever you can see the crux of the biscuit at line 675 here: https://review.openstack.org/#/c/120077/9/lib/nova,cm16:20
*** yjiang5 has quit IRC16:20
cdentit may be as simple as the directory change16:20
*** bkopilov has quit IRC16:20
sdaguecdent: we're still missing a set of quotes there16:21
*** danfinch has quit IRC16:21
cdentthe implication being that nova-compute is not getting the proper config?16:22
* cdent looks in logs from local tests16:22
openstackgerritTomas Sedovic proposed a change to openstack/tempest: Use the AWS instance type in the cfn_init scenario  https://review.openstack.org/12074916:23
*** rwsu has quit IRC16:24
*** mkollaro has quit IRC16:26
*** rwsu has joined #openstack-qa16:29
*** yjiang5 has joined #openstack-qa16:31
*** pkoniszewski has quit IRC16:32
*** rwsu has quit IRC16:33
*** yfried has quit IRC16:33
* cdent recalls now16:34
*** balles has quit IRC16:35
cdentsdague: simple quotes at that section do not work: they get out of balance as the args are passed through run_process and _run_process16:35
*** rwsu has joined #openstack-qa16:35
cdentyou end up with a syntax error when the command is actually called16:35
*** danfinch has joined #openstack-qa16:36
*** balles has joined #openstack-qa16:38
openstackgerritJohn Davidge proposed a change to openstack-dev/devstack: Add IPv6 support for tenant data network  https://review.openstack.org/8798716:38
*** suneelb has joined #openstack-qa16:41
*** jgrimm is now known as zz_jgrimm16:42
openstackgerritA change was merged to openstack-infra/elastic-recheck: Remove query for (fixed) bug 1338844  https://review.openstack.org/12083116:42
*** ildikov has quit IRC16:45
*** salv-orlando has joined #openstack-qa16:45
*** jlanoux has joined #openstack-qa16:55
*** zz_jgrimm is now known as jgrimm16:55
*** aix has quit IRC16:55
*** jlibosva has quit IRC16:58
openstackgerritAndrea Frittoli  proposed a change to openstack/qa-specs: add bp:resource-cleanup  https://review.openstack.org/11835217:00
andreafmtreinish, ^^^^17:01
*** dimtruck is now known as zz_dimtruck17:02
*** markmcclain has joined #openstack-qa17:02
*** suneelb has quit IRC17:03
*** suneelb has joined #openstack-qa17:03
*** Sukhdev has joined #openstack-qa17:04
*** suneelb has quit IRC17:04
*** suneelb has joined #openstack-qa17:04
bobjHello, I'm trying to run stress test in the object storage module using tempest, but when I ran "run-tempest-stress" an message is showed "run-tempest-stress command not found". What am I forgetting?17:07
*** harlowja_away is now known as harlowja17:09
*** Sukhdev has quit IRC17:09
*** chandankumar has joined #openstack-qa17:10
*** bkopilov has joined #openstack-qa17:17
openstackgerritOpenStack Proposal Bot proposed a change to openstack/tempest: Updated from global requirements  https://review.openstack.org/11011017:21
sdaguelifeless: so this smells very much like a fixtures bug itself - https://bugs.launchpad.net/nova/+bug/1330763 - at least I can't see how it's raisable in nova or oslo.db code17:24
sdaguealso, I can't assign it to python-fixtures, because launchpad has timed out on my twice trying to do so17:24
*** masayukig has quit IRC17:28
*** Ch00k has quit IRC17:30
*** masayukig has joined #openstack-qa17:34
*** Sukhdev has joined #openstack-qa17:35
*** balles has quit IRC17:35
*** marun has quit IRC17:36
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Update query for bug 1334142  https://review.openstack.org/12085117:37
*** balles has joined #openstack-qa17:38
*** johnthetubaguy is now known as zz_johnthetubagu17:39
*** chandankumar has quit IRC17:40
*** salv-orlando has quit IRC17:42
*** aimon has quit IRC17:46
*** openstackgerrit has quit IRC17:46
*** zz_dimtruck is now known as dimtruck17:47
*** openstackgerrit has joined #openstack-qa17:47
openstackgerritArmando Migliaccio proposed a change to openstack/tempest: Fix MismatchError for LB scenario test  https://review.openstack.org/12053317:51
andreafmtreinish, non-isolated jobs in experimental pipeline: https://review.openstack.org/#/c/120861/117:55
notmynamedkranz: also, (so as not to consume your meeting time)...I looked at swift's internal functional tests. we're testing for specific status codes. I want to know if we change a response code with a patch. but at the same time, eg a new possibility of 2xx-series response to a PUT request isn't something that should a priori be rejected18:01
openstackgerritChris Dent proposed a change to openstack-dev/devstack: Replace screen_it() with run_process() throughout  https://review.openstack.org/12007718:01
*** jlanoux has quit IRC18:01
*** dmorita has quit IRC18:01
mtreinishandreaf: cool, thanks18:02
mkoderermtreinish: about the tempest-lib.. wouldn't it a good idea to split tempest/test.py into logical pieces?18:02
dkranznotmyname: It is a philosophical question whether a particular 2xx value is specified as part of the contract18:02
dkranznotmyname: all I was saying is that swift takes one view and every other project takes the other view18:03
dkranznotmyname: I am not even sure which is right18:03
notmynamedkranz: yeah. or at least theres some tension there over what the server is expected to send vs what the client is expected to accept18:03
mtreinishmkoderer: yeah, there is going to be a decorators class, etc18:03
mkoderermtreinish: yep... and we should move all the negative testing classes out18:03
mtreinishright now I just migrated the history for everything (which was probably a mistake), but am only using the very core of the base test class18:03
mtreinishmkoderer: yeah, feel free to do that after we push the first release out18:04
mkoderermtreinish: yep, I will do that18:05
mtreinishmkoderer: the only thing is I'm trying to avoid using oslo.config for the library is that something that you can do as a refactor for the negative testing?18:06
mtreinishI don't remember how coupled it was to config18:06
mkoderermtreinish: negative testing has only one config parameter18:07
mkodererif I remember correctly18:07
mtreinishmkoderer: ok cool, could that be passed in as a param instead?18:07
mkoderermtreinish: it's just the generator... yep we can do that in a param18:08
*** dimtruck is now known as zz_dimtruck18:09
mtreinishok, that's all I was concerned about. I'm trying to avoid having a tempest-lib config file if we can18:09
mkoderermtreinish: +1 :)18:09
mkoderermtreinish: but we have more configs in the stress test area18:09
mtreinishmkoderer: yeah, I'm thinking that might become a separate tool at some point18:10
notmynamedkranz: so the thing I'm curious about is the comment about "every other project takes the other view". I'm certainly not trying to find gratuitous differences for swift. can you tell me more what you mean by that?18:10
mkodererI think we should discuss about the future of stress testing anyway though18:10
openstackgerritEmilien Macchi proposed a change to openstack/tempest: javelin: add network and secgroup resources  https://review.openstack.org/10016918:10
mtreinishmkoderer: yeah, that's a long term discussion :)18:10
mtreinishbobj: to use that as a cli binary you need to pip install tempest18:12
*** e0ne has joined #openstack-qa18:12
dkranznotmyname: Just that if you look at the docs for every other project, they specify for each api call 200 or 201, etc. One specific code.18:12
mtreinishbobj: so from the root of the tempest repo you can do 'pip install -e .' and then the run-tempest-stress command by itself18:12
cdentdoes swift say '2xx'?18:12
dkranznotmyname: Swift specifies 2xx18:12
* cdent is surprised18:12
dkranznotmyname: I think there are two ways of looking at these codes in rest apis18:13
notmynameas I said in gerrit, we're clearly not just returning random.choice(range(100)+200) for every response18:13
dkranz1. The 2xx value is like a return value from a normal api in any language18:13
notmynamethere are specific things we respond with, and there are specific meaning for specific codes18:13
dkranz2. The client should interpret any 2xx response based on the meaning (sort of) in the http spec and act accordingly18:14
dkranznotmyname: right, that is what I called (2)18:14
*** aimon has joined #openstack-qa18:15
dkranznotmyname: But the other projects are also committing to not changing the actual response except in a new api version18:15
dkranznotmyname: that is a difference in philosophy18:15
mtreinishbobj: you could also just call the script directly without installing it. From the tempest root dir it would be: tempest/cmd/run_stress.py18:15
*** zz_dimtruck is now known as dimtruck18:16
dkranznotmyname: so the other project's tempest tests look like swift's internal ones18:16
notmynamedkranz: yeah, so the question is testing as a well-behaved client should act or as a server should act (or has acted)18:20
dkranznotmyname: My guess is that this debate never happened early on. swift did it one way and nova the other, and then others copied nova18:21
dkranznotmyname: But I am not sure about that.18:21
notmynamethat is very similar to other things in openstack :-)18:21
dkranznotmyname: :)18:22
dkranznotmyname: Well, I hope we now have a shared understanding of the issue18:22
notmynamedkranz: what's the tempest view? what should tempest be testing? the server's expectations of the clients (as the current swift patch does) or the client's expectations of the server (as you say the other projects do)?18:23
notmynamedkranz: and how does that translate to the future in-project testing that -infra is pushing for?18:24
dkranznotmyname: with regard to the second, it should not change anything. The semantics should depend on a decision, not on where the repo is18:25
*** dimtruck is now known as zz_dimtruck18:25
dkranznotmyname: But tempest does not have a view about this. I think both arguments are valid.18:25
dkranznotmyname: And as mtreinish said, we have just gone by what the PTL of each project says18:25
*** torgomatic has joined #openstack-qa18:26
notmynameok, cool18:26
mtreinishdkranz: heh, I said that :)18:26
notmynamejust had an interesting in-office conversation about it :-)18:27
dkranznotmyname: I would guess though, that there is a lot of client code that would break if the server started returning 204 instead of 20018:27
dkranznotmyname: In other words, if a server starts acting asynchronously all of a sudden I might consider that to be an api change.18:27
* cdent pages postel18:28
notmynameso the http spec also says that headers are case-insensitive and can be in any order. but we aren't testing that. eg we could strip out timestamp headers and then take the md5 of the response to see that _nothing_ has changed18:28
*** vigneshvar has joined #openstack-qa18:28
notmynamebut we clearly don't do that, even though I've come across java swift clients that do in face rely on header name case sensitivity (ie Etag != ETag, therefore the client broke)18:28
notmynameand the http spec says that clients shoudld treat 2xx as success18:29
*** zz_dimtruck is now known as dimtruck18:29
notmynamedkranz: I'd expect 204 to be returned from a GET if in fact it was a zero-byte object ;-)  /pedant18:29
dkranznotmyname: I was thinking of a post that creates a resource18:30
dkranznotmyname: It really matters if the resource is ready to use when the call returns or not18:30
dkranznotmyname: even if the client treats both cases as success18:30
notmynameright. does "success" mean that the request was accepted or that it was completed successfully18:31
*** amotoki has quit IRC18:31
notmyname(there's a code for that. 202)18:31
dkranznotmyname: Sorry, that is what I meant when I said 20418:31
notmynameah ok18:31
dkranznotmyname: my http is a little rusty :)18:31
notmynameheh18:32
*** amotoki has joined #openstack-qa18:32
* notmyname needs to start memorizing the new sections of the new rfcs that replaced 261618:32
*** suneelb has quit IRC18:32
*** suneelb has joined #openstack-qa18:32
dkranznotmyname: so I do fall more on the way other projects are doing it only in the sense that these kinds of changes might warrant a new api version anyway due to the probability of breaking clients if you went from 200 to 202 or something like that.18:33
notmynamedkranz: then why not go all the way and look for case sensitivity and order and new headers?  (/me playing devil's advocated)18:33
dkranznotmyname: I would not be opposed to that necessarily. It  is tricky to define what is an api change.18:34
*** yfried has joined #openstack-qa18:34
dkranznotmyname: THis is why I don't like trying to slavishly map an http spec onto api type signatures18:35
notmynameright. which is where my point comes in that the rfc says that clients should tread any 2xx response as a success. and yes particular 2xx codes have particular meanings :-)18:35
dkranznotmyname: But that is kind of what we do18:35
cdentTests aren't for testing success are they, they are for testing the contract18:35
cdentthat's not the same thing18:35
cdent(I'm sorry to horn in on you guys, but this is fascinating and interesting)18:35
dkranzcdent: Yes, but it is a philosophical discussion about what the contract should actually be18:36
dkranzcdent: and we have two different examples in OpenStack18:36
notmynamecdent: yes, but the whole this is the questions of "what is the contract". that's just another word for "api change"18:36
cdentI don't really see it that way: A test should test whatever the current reality is18:36
notmynamecdent: and yes, I too think this is fascinating and interestion :-)18:36
cdentwhen that reality changes, it should fail18:36
cdentthat's _separate_ from "is this an api change"18:36
*** amotoki has quit IRC18:36
notmynamecdent: that I completely agree with18:37
dkranzcdent: I think the issue is that every code patch changes something, otherwise you would not do it. The question is what kind of changes should clients expect?18:37
notmynamecdent: as shown by how the in-tree functional tests for swift are written18:37
dkranznotmyname: IMO, an actual server change from 200 to 202 in anger would break a lot of client code, even if it should not have.18:38
*** chandankumar has joined #openstack-qa18:38
cdent200 to 202 _should_ break client code18:39
dkranzcdent: not if the code looks at the response and doesn't to use the resource until it becomes available in the 202 case18:39
dkranzcdent: if the contract is "returns 200" then you can just use it.18:40
cdentIf the server sent 200 in response to resource creation request it's busted18:40
dkranzcdent: If the contract is "returns 2xx" then you have to check and act accordingly.18:40
cdent201, 202 or 20418:40
cdentbut not 20018:40
cdentbad contract18:41
dkranzcdent: sorry I meant 20118:41
notmynamecdent: do you mean that in the sense that 200 shouldn't be returned for resource creation? (I agree)18:41
dkranznotmyname: that was my mistake (again)18:42
cdentyes notmyname18:42
notmynameheh18:42
notmynamedkranz: go lurn you some http!18:42
dkranznotmyname: I'll say sync vs async instead18:42
cdenthttps is hard, let's go rpc-ing18:42
notmynamehere's a different way for me to phrase things with relation to the current proposed patch18:43
notmynameI'm actually not opposed to tempest testing for specific codes. I think that means that some patch to swift in the future might be harder, but things that change codes are (intentionally) very rare. but, I prefer tempest to check for response code classes since that's what I expect clients to use. and I _never_ want a swift patch blocked because a tempest reviewer things swift should implement a new api version18:45
notmynamewhile it may be that a new api version is needed, that shoudl happen with swift contributors18:45
openstackgerritMatthew Treinish proposed a change to openstack/qa-specs: Move implemented bps to the implemented dir  https://review.openstack.org/12087918:46
notmynameon the other hand, I do think that docs should talk about response code classes and not specific codes (and there's a patch to docs to do that now)18:46
cdentIt sounds like you are trying to work around a problem in tempest's responsiveness...18:47
notmynamecdent: no, not at all18:47
notmynameI'm not wanting to come across that way18:47
cdent(which would be understandable)18:47
notmynameI think I actually very much agree with what you said that the tests of reality are separate from "is this an api change"18:48
notmynamecdent: my point is only that I think the swift devs are the ones to make the call on "is this an api change", and I think that tempest checking for the same constraints as the client is expected to have makes those conversations happen in the right place18:49
notmynamein no way am I complaining about tempest or tempest devs. I'm actually really happy to see how much for swift has been added into tempest recently :-)18:50
cdentI'm not familiar with the mechanics of those conversations (too new to the scene).18:50
dkranzcdent, notmyname : The bottom line is that tempest checks for what the spec (docs) says should be returned18:51
notmynamecdent: honestly, I'm not sure they've happened too often yet. the only example I can think of is the comment in gerrit on the current patch where a reviewer set -118:51
dkranzThat is determined by projects and that is not going to change18:52
notmynamedkranz: that, in fact, is a very clear standard :-)18:52
notmynamefun having these conversations in a text-based medium, huh? it would be much better in person :-)18:52
cdentJust out of curiosity, not quite on topic, how would you, notmyname, expect a client developer to know about and/or deal with a creation request that responded with 202?18:52
cdent(given that you've already said client code would think in terms of 2xx)18:53
*** vigneshvar has quit IRC18:53
*** __afazekas is now known as afazekas18:53
notmynamecdent: well, to be honest, there are some creation requests in swift that _do_ respond with 202. but the difference is that in swift we don't provision things that you use. instead, we only return success when your data is durably persisted to disk18:53
afazekasmtreinish: https://review.openstack.org/#/c/120386/18:54
notmynamecdent: or more specifically, you could get 201 or 20218:54
cdentin what situations does a 202 happen?18:54
*** suneelb has quit IRC18:54
*** suneelb has joined #openstack-qa18:55
notmynamecdent: PUT request to a container that has already been created18:56
afazekascdent: after a 202 you usually need to do polling18:56
cdentafazekas: yes, I know, I'm just trying imagine how 202 syncs with "we only return success when your data is durably persisted" and "need to do polling"18:57
cdents/imagine/understand/18:57
*** jecarey has quit IRC18:58
notmynamewe return 2xx because the resource (the container) is available right then. ie you can still read a container18:58
afazekas202 usually used with the server actions18:58
notmynamedkranz: heh. temptest tests what the docs say. the docs patch (https://review.openstack.org/#/c/120622/) just got a -2 because the tempest code hasn't landed :-)18:59
cdentnotmyname: in the example you're thinking of what (if anything) is in the headers and body of the response that has the 202 status?19:00
cdent(not trying to find holes here, just trying to expand my understanding)19:00
dkranznotmyname: more evidence this discussion never happened...19:00
*** melwitt has joined #openstack-qa19:01
notmynamecdent: making a paste for you19:01
cdentcool, thanks very much19:01
andreafmtreinish, sdague have you ever seen this before: http://logs.openstack.org/85/107685/31/check/gate-tempest-dsvm-neutron-large-ops/8157620/logs/tempest.txt.gz?level=WARNING19:01
notmynamecdent: https://gist.github.com/notmyname/6254035176dd3deb17ae19:02
* cdent read19:02
cdents19:02
*** jecarey has joined #openstack-qa19:03
afazekasBTW: which  code is used for http://docs.openstack.org/developer/swift/overview_container_sync.html ?19:03
notmynameafazekas: implementing that feature or generating those docs?19:04
*** vigneshvar has joined #openstack-qa19:06
*** dtroyer has quit IRC19:06
*** adam_g has quit IRC19:06
*** StevenK has quit IRC19:06
*** lbragstad has quit IRC19:07
*** dtroyer has joined #openstack-qa19:08
*** fungi has quit IRC19:09
*** freyes has quit IRC19:09
*** freyes has joined #openstack-qa19:10
*** hockeynut has quit IRC19:10
*** lbragstad has joined #openstack-qa19:10
cdentnotmyname: is the second container put at line 47 effectively a no op?19:11
openstackgerritA change was merged to openstack-infra/elastic-recheck: Update query for bug 1334142  https://review.openstack.org/12085119:11
notmynamecdent: internally, I think it might update a timestamp19:12
*** StevenK_ has joined #openstack-qa19:12
*** hockeynut has joined #openstack-qa19:12
cdent202 feels wrong to me there. I'm not sure why.19:13
cdentThere is no future thing that you expect to happen.19:13
cdentWould 204 be appropriate?19:14
*** Guest75250 has quit IRC19:14
afazekasnotmyname: forget it, wrong example..19:15
mtreinishandreaf: personally I've never really noticed it before19:15
mtreinishandreaf: but according to logstash it happens all the time19:15
notmynamecdent: maybe. and we're not returning 202 for an object overwrite either. I think this is along the lines of not looking to change response codes :-)19:15
cdent:)19:15
notmynameand FWIW if you POSt to a container (eg with new metadata), you'll get a 20119:16
cdentI'd think that was a 204 too19:16
cdent(if it already exists)19:16
*** dtroyer has quit IRC19:17
notmynamecdent: from rackspace cloud files docs from 2008: "No content is returned. A status code of 201 (Created) indicates that the Container was created as requested. Container PUTs are idempotent and a code of 202 (Accepted) is returned when the Container already existed."19:17
notmyname;-)19:17
cdentbut yeah, once people start expecting specific codes things get hard to change19:17
*** StevenK_ has quit IRC19:17
*** rhsu has quit IRC19:17
*** StevenK_ has joined #openstack-qa19:17
andreafmtreinish, ok so probably a red herring19:18
andreafmtreinish, thanks19:18
notmynameright. which comes back to the discussion around which change means it's an api change or not :-)19:18
cdentindeed it does19:18
cdentthat's why I was trying to tease out the expectations on the 20219:18
cdentbecause if it was an async thing requring a followup poll, then 202 is part of the contract: it tells you want to do19:19
*** freyes has quit IRC19:19
*** lbragstad has quit IRC19:19
*** jroll has quit IRC19:19
cdentbut at the moment it pretty much just says "you didn't break it, carry on"19:19
notmynamethat may be the only place you can get 202 from swift. not sure. maybe also if you PUT to an account that exists (if you have those permissions)19:19
*** nsaje_ has joined #openstack-qa19:19
notmynamecdent: yeah. the async stuff could be interpreted that if you do a read and get a response it might not be the version that you just PUT19:19
notmynamecdent: in which case 202 makes more sense19:20
mtreinishandreaf: are you looking at a specific failure?19:20
notmynamecdent: whereas with a new container, if you make a read request and get data, you'll have the right version (thus the 201 response to the create)19:20
*** hockeynut has quit IRC19:21
* notmyname goes downstairs to pick up stuff at a food truck19:21
*** hockeynut has joined #openstack-qa19:21
andreafmtreinish, large opts tests seems to be failing all the time on the last patch of test accounts19:21
* cdent needs food too19:22
andreafmtreinish, https://review.openstack.org/#/c/107685/19:22
*** StevenK_ has quit IRC19:22
andreafmtreinish, but I fail to see any relationship between the change and the failures19:22
*** StevenK_ has joined #openstack-qa19:23
andreafmtreinish, I'll try yet another recheck19:23
*** lbragstad has joined #openstack-qa19:23
*** chandankumar has quit IRC19:23
mtreinishandreaf: yeah, it looks like it's going into an error state19:24
mtreinishjogo: when you figure out the broken migration stuff ^^^19:24
*** StevenK_ has quit IRC19:27
*** lbragstad has quit IRC19:28
*** vigneshvar has quit IRC19:28
*** StevenK_ has joined #openstack-qa19:28
jogomtreinish: ?19:28
*** markmcclain has quit IRC19:28
mtreinisha large ops job failure19:29
mtreinishjogo: http://logs.openstack.org/85/107685/31/check/gate-tempest-dsvm-neutron-large-ops/8157620/console.html#_2014-09-08_02_56_22_27319:30
*** hockeynut has quit IRC19:30
*** hockeynut has joined #openstack-qa19:30
jogomtreinish: http://logs.openstack.org/85/107685/31/check/gate-tempest-dsvm-neutron-large-ops/8157620/logs/screen-n-cpu.txt.gz?level=TRACE#_2014-09-08_02_55_52_89019:31
jogomtreinish: neutron is out of fixed IPs19:31
mtreinishoh, yeah that would do it19:31
*** StevenK_ has quit IRC19:33
jogonot sure how that happened19:33
*** fungi has joined #openstack-qa19:33
*** stevebaker has quit IRC19:33
*** StevenK_ has joined #openstack-qa19:33
*** lbragstad has joined #openstack-qa19:34
*** Sukhdev has quit IRC19:35
*** sc68cal_ has joined #openstack-qa19:36
*** lbragstad has quit IRC19:37
*** mlavalle_ has joined #openstack-qa19:37
*** jecarey has quit IRC19:38
andreafmtreinish, jogo thanks for your help, I'll continue digging into this tomorroq19:38
*** StevenK_ has quit IRC19:38
*** jecarey has joined #openstack-qa19:38
andreaftomorrow19:38
*** yamahata__ has joined #openstack-qa19:39
*** StevenK_ has joined #openstack-qa19:39
*** hockeynut has quit IRC19:39
*** hockeynut has joined #openstack-qa19:40
*** sc68cal has quit IRC19:40
*** mlavalle has quit IRC19:40
*** yamahata_ has quit IRC19:40
*** mriedem1 has joined #openstack-qa19:40
*** mriedem has quit IRC19:40
*** nsaje_ is now known as nsaje19:42
jogoandreaf: thanks19:42
jogocan you make sure a bug is filed for that19:42
jogoandreaf mtreinish: for e-r tracking19:42
mtreinishjogo: yeah I'll do that now19:43
jogomtreinish: thanks19:43
jogomtreinish: just dealt with https://bugs.launchpad.net/nova/+bug/136839119:43
*** StevenK_ has quit IRC19:44
*** StevenK_ has joined #openstack-qa19:44
*** dims has quit IRC19:44
*** vigneshvar has joined #openstack-qa19:45
*** Dafna has quit IRC19:45
openstackgerritAnderson Mesquita proposed a change to openstack/tempest: Refactor stack actions and add stack check  https://review.openstack.org/10645619:46
mtreinishjogo: hmm, actually maybe not. It's only shown up on 2 check queue changes, andreaf's cred provider patch and https://review.openstack.org/#/c/12007019:46
mtreinishso I'm not sure it's a real bug19:47
jogomtreinish: cool glad its no agate bug19:47
jogowe have enough of those today19:47
*** hockeynut has quit IRC19:48
*** StevenK_ has quit IRC19:49
*** StevenK_ has joined #openstack-qa19:50
*** mgagne has joined #openstack-qa19:51
*** jroll has joined #openstack-qa19:51
*** mestery has quit IRC19:54
*** mestery has joined #openstack-qa19:54
*** weshay has quit IRC20:01
*** dimtruck is now known as zz_dimtruck20:02
*** lbragstad has joined #openstack-qa20:02
*** dims_ has joined #openstack-qa20:03
*** rhsu has joined #openstack-qa20:03
*** markmcclain has joined #openstack-qa20:10
openstackgerritBenny Kopilov proposed a change to openstack/tempest: compute-image : create server from snapshot image  https://review.openstack.org/12089920:12
*** yfried has quit IRC20:13
*** weshay has joined #openstack-qa20:15
*** masayukig has quit IRC20:17
*** dims__ has joined #openstack-qa20:17
*** e0ne has quit IRC20:19
*** dims_ has quit IRC20:20
*** masayukig has joined #openstack-qa20:21
*** dims__ has quit IRC20:22
*** flaper87 is now known as flaper87|afk20:24
*** zz_dimtruck is now known as dimtruck20:25
bobjmtreinish: Thank for your support.20:25
mtreinishbobj: so you got it working?20:26
bobjmtreinish: Not yet, I'm researching how do it better.20:30
bobjmtreinish: I'll configure one json file to test just object storage module. Thank you again, your comments were very important.20:33
*** e0ne has joined #openstack-qa20:45
*** e0ne has quit IRC20:45
*** dimtruck is now known as zz_dimtruck20:48
openstackgerritBenny Kopilov proposed a change to openstack/tempest: compute-image : create server from saved image saved image can be reused by new servers  https://review.openstack.org/12089920:51
*** stevebaker has joined #openstack-qa20:55
*** cdent has quit IRC20:57
*** zz_dimtruck is now known as dimtruck20:59
*** markmcclain has quit IRC20:59
*** mestery has quit IRC21:00
*** enikanorov_ has quit IRC21:00
*** enikanorov_ has joined #openstack-qa21:00
*** mestery has joined #openstack-qa21:00
*** mwagner_lap has quit IRC21:04
*** toure has quit IRC21:06
*** bobj has quit IRC21:10
*** vigneshvar has quit IRC21:11
*** aysyd has quit IRC21:12
*** dustins has quit IRC21:13
lifelesssdague: looking21:14
*** torgomatic has left #openstack-qa21:16
*** adam_g has joined #openstack-qa21:26
*** rfolco has quit IRC21:26
*** adam_g has quit IRC21:27
*** adam_g has joined #openstack-qa21:27
*** nsaje has quit IRC21:27
*** cbader_ has quit IRC21:29
*** cbader has quit IRC21:29
*** dkranz has quit IRC21:30
*** catohornet1 has joined #openstack-qa21:32
openstackgerritJim Rollenhagen proposed a change to openstack/tempest: Support lack of ephemeral volumes for baremetal  https://review.openstack.org/12081421:37
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Remove query for bug 1347692  https://review.openstack.org/12091621:42
lifelesssdague: pushing a possible patch up now21:45
lifelesssdague: https://review.openstack.org/12091921:45
sdaguelifeless: ok, cool, appreciated21:46
lifelesssdague: added a little more commentary in the bug21:47
*** StevenK_ is now known as StevenK21:47
lifelesssdague: I need to context switch today sorry, but it should be clear, I hope :)21:47
sdaguesure, I'll early propose that to nova21:47
sdaguebecause I don't think oslo has any testing infrastructure to expose that21:48
lifelesssdague: the change is in oslo.incubator21:48
lifelesssdague: oh, you mean its not tested locally to oslo.incubator?21:48
*** jecarey has quit IRC21:48
lifelesssdague: so yeah, it will need some shenanigans to verify it works etc.21:48
*** atiwari has quit IRC21:50
sdagueyeh, honestly, I'm fine violating normal oslo rules here21:54
sdaguebecause I don't think that will get us to a sane fix21:54
*** qba73 has joined #openstack-qa21:55
lifelesssdague: in case its not obvious this is actually a secondary issue, its not the cause of the failure21:56
sdaguelifeless: sure21:56
lifelesssdague: the postgresql db is reporting available but isn't actually connectable to21:56
sdaguebut we can't see the base issue21:56
lifelessyup21:56
lifelessI'm going to concurrently fix that aspect in testtools21:56
lifelesswhich is fully tested :)21:56
sdagueappreciated21:56
lifelessbut first, context switching21:56
lifelesssince lynne just crashed the car21:56
sdagueoh, bummer21:57
sdaguetake care21:57
lifelessshes fine21:57
lifelessbut the light at the riccarton cinema parking lot, and the car bonnet... not so much21:57
*** mriedem1 has quit IRC21:59
*** weshay has quit IRC22:06
*** mwagner_lap has joined #openstack-qa22:09
*** dims_ has joined #openstack-qa22:10
*** jhenner has quit IRC22:11
*** dims_ has quit IRC22:14
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Remove query for bug 1357677  https://review.openstack.org/12092522:17
*** weshay has joined #openstack-qa22:19
*** gfidente has quit IRC22:20
*** cnesa9 has quit IRC22:20
*** andreaf has quit IRC22:23
*** weshay has quit IRC22:25
openstackgerritClark Boylan proposed a change to openstack-infra/devstack-gate: Install PyYaml for test-matrix.py  https://review.openstack.org/12092922:30
*** gordc has quit IRC22:32
*** jgrimm is now known as zz_jgrimm22:32
*** qba73 has quit IRC22:33
*** qba73 has joined #openstack-qa22:34
*** openstackgerrit has quit IRC22:38
*** openstackgerrit_ has joined #openstack-qa22:38
*** qba73 has quit IRC22:39
*** openstackgerrit_ is now known as openstackgerrit22:40
*** dims_ has joined #openstack-qa22:44
*** dimtruck is now known as zz_dimtruck22:54
*** jpich has joined #openstack-qa22:59
*** ianw has quit IRC23:04
*** ianw has joined #openstack-qa23:04
*** rbak has quit IRC23:06
openstackgerritClark Boylan proposed a change to openstack-infra/devstack-gate: Install PyYaml for test-matrix.py  https://review.openstack.org/12092923:28
*** dims_ has quit IRC23:34
*** oomichi has joined #openstack-qa23:37
openstackgerritDavid Lyle proposed a change to openstack/tempest: Skipping dashboard test to unblock gate  https://review.openstack.org/12093823:46
openstackgerritJim Rollenhagen proposed a change to openstack-dev/devstack: Fix Ironic agent ramdisk URL  https://review.openstack.org/12093923:46
jrollianw, quick bugfix if you have a sec, this is my fault: https://review.openstack.org/12093923:47
*** dims_ has joined #openstack-qa23:52
openstackgerritDavid Lyle proposed a change to openstack-dev/devstack: Moving horizon compression offline  https://review.openstack.org/12094023:53
*** chuckC has joined #openstack-qa23:53
*** dims__ has joined #openstack-qa23:55
*** suneelb has quit IRC23:55
*** suneelb has joined #openstack-qa23:55
*** aix has joined #openstack-qa23:56
*** dims_ has quit IRC23:58

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