Wednesday, 2015-03-04

*** catherineD has joined #openstack-qa00:01
*** annashen has joined #openstack-qa00:03
*** melwitt has quit IRC00:04
*** hemna is now known as hemnafk00:06
*** rbak has quit IRC00:06
*** zz_dimtruck is now known as dimtruck00:11
*** patrickeast_ has joined #openstack-qa00:19
*** patrickeast__ has joined #openstack-qa00:20
*** patri728 has joined #openstack-qa00:20
*** patri586 has joined #openstack-qa00:20
openstackgerritMerged openstack-dev/devstack: Revert change to remove lbaas from devstack; it breaks grenade.  https://review.openstack.org/16091300:21
*** dboik_ has joined #openstack-qa00:25
*** mfalatic has quit IRC00:27
naggappanhi i keep getting error "http://paste.openstack.org/show/187006/". when i run the tempest against my openstack environment00:30
naggappanhttp://paste.openstack.org/show/187006/00:31
naggappanis this " tempest.api.network.test_ports.PortsIpV6TestJSON.test_update_port_with_two_security_groups_and_extra_attributes" a valid test case?00:31
*** yamamoto has quit IRC00:31
naggappanmtreinish: could you please give me some clue like why i get the above error .00:32
*** carl_baldwin has quit IRC00:33
mtreinishnaggappan: that means the create security group call that is being sent to neutron is recieving a 500 response00:35
mtreinishyou'll have to look at the neutron logs to figure out what is going wrong00:36
mtreinisha 500 error normally means the service (in this case neutron) stack traced00:37
*** annashen has quit IRC00:38
mtreinishandreaf: the other thing I was just thinking about is sometime in the future we might want a tool which counts the peak cred requests from a test class so we can give a better estimate on how many accounts to put in the yaml00:39
naggappanmtreinish: yes i see error in neutron server log as this : http://paste.openstack.org/show/187008/    .  (but its a default configuration in neutron i did not change anything. Just when the tempest case run i get this error for security group" . May be if i know exactly what the tempest case is trying to do i can check why it fails by manually00:39
mtreinishnaggappan: well take a look at the test code, its normally pretty easy to read what the test is trying to do00:41
*** catohornet has quit IRC00:41
adam_gjogo, you okay with me re-proposing my revision of that spec under a new change, to get ownership?00:41
*** catohornet has joined #openstack-qa00:41
jogoadam_g sure, just abandon my patch or something00:42
naggappanmtreinish: ok i will check that out. It look similar to this bug :https://bugs.launchpad.net/neutron/+bug/1364166 .00:42
openstackLaunchpad bug 1364166 in neutron "Netaddr can't check IPv6 subnets for overlap" [Medium,Fix released] - Assigned to Sergey Shnaidman (sshnaidm)00:42
naggappanthanks00:42
adam_gjogo, i cant, only you can00:42
jogoadam_g: what happens if you push a new patch and take over author and committer?00:44
jogois that enough?00:44
*** mahito has joined #openstack-qa00:44
mtreinishnaggappan: yeah, that looks on point00:44
jogoas it would be nice to keep the review comments history00:44
adam_gjogo, maybe, lets see00:44
*** dmellado has quit IRC00:44
adam_gi dont think therse a way for me to get author, only comitter00:44
jogoadam_g: git can change the author not sure if gerrit lets you though00:44
adam_gjogo, yeah, gerrit wont00:45
mtreinishadam_g: it will, the patch owner in gerrit will still be jogo though00:45
mtreinishbut the git commit author will change00:45
naggappanmtreinish: I am wondering the fix is submitted for juno. But somehow tempest cases is passing for ice house also.00:45
adam_gjogo, you wanna just abandon yours with a link to https://review.openstack.org/#/c/161047/ ?00:45
naggappanmtreinish: May be i have to check that fix and apply in icehouse and try to run the test. will do that00:46
*** tsekiyam_ has joined #openstack-qa00:46
mtreinishnaggappan: we probably aren't running it on icehouse in the gate. On icehouse we only run a subset of tests00:46
jogoadam_g: done00:47
*** dmellado has joined #openstack-qa00:48
naggappani used the exact tempest config which is used for icehouse. Except that i have created a public network and changed that ID. And created a role Member and changed the IP address00:48
mtreinishnaggappan: yes, but we also run with this testr regex:  (?!.*\[.*\bslow\b.*\])((smoke)|(^tempest\.scenario))00:49
adam_gjogo, word.  having thought about it more while revving that, it makes most sense to create a .gate file during the bot proposed requirements sync, thats effectively a snapshot of requirements at that time. the other way would have required installing all of global-requirements.txt on slaves everytime00:49
mtreinishnaggappan: and also run it serially instead of in parallel00:49
mtreinishnaggappan: on icehouse neutron had issues so it wasn't capable of running everything reliably00:50
*** mtanino has quit IRC00:50
*** tsekiyama has quit IRC00:50
naggappanmtreinish: means ? could you point me an example how you run this? I am using ./run_tempest.sh00:50
*** tsekiyam_ has quit IRC00:51
openstackgerritJamie Lennox proposed openstack/tempest: Remove identity CLI tests  https://review.openstack.org/16105000:51
jamielennoxmtreinish, morganfainberg: ^00:51
mtreinishnaggappan: something like: ./run_tempest.sh -t -- '(?!.*\[.*\bslow\b.*\])((smoke)|(^tempest\.scenario))'00:51
mtreinishjamielennox: cool thanks00:52
*** topshare has joined #openstack-qa00:53
*** madhu has quit IRC00:56
*** topshare_ has joined #openstack-qa00:57
*** topshare has quit IRC01:00
openstackgerritMatthew Treinish proposed openstack/tempest-lib: Handle missing launchpadlib gracefully  https://review.openstack.org/15914701:02
*** marrusl has quit IRC01:04
*** mmedvede_ has quit IRC01:04
*** apevec has quit IRC01:05
naggappanmtreinish:  what does bslow regx pattern used for.. I am not seeing any difference in the test which is been run . will that be present in console log? to check how the run this test01:07
*** xgerman has quit IRC01:09
*** jaypipes has quit IRC01:11
*** dimtruck is now known as zz_dimtruck01:13
*** yamamoto has joined #openstack-qa01:14
*** dims_ has joined #openstack-qa01:17
*** dims has quit IRC01:19
*** salv-orlando has quit IRC01:20
mtreinishnaggappan: the slow part just excludes tests tagged as slow01:20
mtreinishthe part that matters is the smoke and the scenario01:20
mtreinishit means it'll only run tests that are tagged as smoke or in the scenario directory01:20
*** yamamoto has quit IRC01:20
*** madhu has joined #openstack-qa01:26
mtreinishblogan: I have thought about there isn't anything really stopping the switch to use the lib instead of the incubator code01:30
mtreinishblogan: my only concern at this point is the config options, because tempest still autogenerates the sample config files and runs check_uptodate. So switching to the lib will mean we risk an oslo.log release breaking that in tempest01:31
openstackgerritMerged openstack/tempest: Testcase to create domain with disable status  https://review.openstack.org/15768701:32
bloganmtreinish: ah okay, i'm just wondering because we are doing tempest tests in tree now (at least bootstrapping it) and they use the tempest oslo incubator, so if we just used oslo_log I don't see that causing an issue but I could be missing something01:32
*** achanda has quit IRC01:35
mtreinishblogan: yeah I doubt that would be an issue. For the most part in tempest it just treats it like normal python logging01:35
mtreinishthe only place where we really start to leverage anything in oslo logging was the base rest client. But that's been migrated to tempest-lib (which uses oslo_log now)01:36
*** achanda has joined #openstack-qa01:41
*** mlavalle has quit IRC01:42
*** zz_dimtruck is now known as dimtruck01:42
mtreinishstevebaker: could you give me a hand and take care of ripping the heatclient cli tests out of tempest? It shouldn't be to hard :)01:42
stevebakermtreinish: sure, I can do that01:43
mtreinishjamielennox: oh, you know what, can you respin your removal patch to drop the keystonclient line from the requirements file too? I completely forgot about that, I'll +2 right after01:43
mtreinishstevebaker: cool, let me dig up jogo's ml post on how to do it01:43
mtreinishit's pretty simple01:43
mtreinishstevebaker: lists.openstack.org/pipermail/openstack-dev/2015-February/057458.html01:44
stevebakerthanks01:44
stevebakerbbl01:44
bloganmtreinish: would you recommend import tempest_lib or tempest for in-tree tempest tests?01:47
*** jhenner has quit IRC01:47
mtreinishblogan: it depends on what you're trying to do, tempest_lib only has a subset of functionality (we're working on migrating more)01:48
mtreinishblogan: but that being said tempest doesn't provide a stable interface for anything01:48
mtreinishtempest also consumes tempest-lib where appropriate01:48
*** beisner has quit IRC01:48
*** Sukhdev has quit IRC01:49
bloganmtreinish: well i know we need clients, smoke tests, and scenario tests01:49
bloganmtreinish: i assume the runner is just nosetests and that should carry over okay, but not sure about all the plumbing that may go on01:49
mtreinishblogan: heh, well the runner should never be nosetests, because it doesn't work for most things in tempest :)01:50
mtreinishblogan: what exactly are you trying to do?01:50
*** beisner has joined #openstack-qa01:51
bloganlol well im out of the loop then01:51
mtreinishs/most/some01:51
*** yamamoto has joined #openstack-qa01:52
bloganmtreinish: i'm working on neutron-lbaas, which is a new repository, but still in openstack.  we're trying to get as much things in our tree as we can, and tempest tests seem like they were going that way01:52
bloganmtreinish: so whatever we need to get in tree that gives us the ability to have full tempest test coverage is what we have to do, be it through tempest_lib or importing tempest for now and fixing as tempest_lib gets more features01:53
mtreinishblogan: so you're working on moving the existing lbaas tests from tempest into the lbaas repo01:53
bloganmtreinish: that would be the end goal, however we're keeping the v1 lbaas tests in tempest until we get a good path set by the v2 tests (we have a v1 lbaas api and v2 lbaas api)01:53
*** openstackrecheck has joined #openstack-qa01:54
mtreinishblogan: so have you connected with dougwig and marun on this, because that is slightly different from what we've been discussing01:54
bloganmtreinish: ive bene talking with dougwig, and it may be that they have plans for v1 to get in sooner but that has been my assumption01:54
mtreinishespecially because after talking to dougwig during the initial repo split the plan was to remove all adv services tests from inside the tempest repo01:54
bloganmtreinish: yeah i think thats the end goal, but right now we're focusing on the v2 tests right now, whcih are being written.  I woudl like the v1 tests to get in as well asap01:55
marunblogan: I'd like to see the tests removed from tempest asap01:56
mtreinishblogan: so I believe marun has a patch up on neutron which shows how neutron is running tempest-like tests from the neutron repo, you probably want to use that same mechanism for the lbaas tests01:56
marunblogan: so that neutron is no longer gating on them01:56
bloganokay so if we moved the v1 tests in tree, would tempest_lib work or would we need to import tempest?01:56
mtreinishblogan: most of the needed bits are still missing from tempest-lib01:57
marunblogan: you'd have to import wholesale for now...01:57
marunblogan: not all the required bits have been moved yet01:57
bloganokay so the question arose today, as to why the tests are not using tempest_lib and are using tempest, that answers the question01:57
marunblogan: https://review.openstack.org/#/c/158021/01:58
bloganthanks marun01:58
bloganoh good using a tempest directory too01:59
bloganalright i believe that answers my questions, thanks for the answers mtreinish and marun01:59
*** jamielennox is now known as jamielennox|away01:59
marunblogan: the idea is to remove/replace the in-tree tempest stuff as it gets moved to tempest-;lib01:59
bloganthats what i thought and i am perfectly fine with, similar to replacing the oslo incubator imports with the ones that graduate02:00
*** ChuckC has quit IRC02:04
dougwigmtreinish, blogan - the only difference i'd add from blogan is that the plan for v1 is separate from v2 (lbaas).  the existing adv.services tempest tests, we're waiting for marun's work to move tempest in-tree in neutron, and then we're further splitting into the service repos.  that's v1.   for lbaas v2, we just started it in-tree directly..02:05
dougwiglbaas v1, vpn, and fwaas splitting out of tempest will all happen at the same time.02:06
bloganah okay02:06
*** dimtruck is now known as zz_dimtruck02:11
*** jhenner has joined #openstack-qa02:11
*** patrickeast has quit IRC02:19
*** patrickeast__ has quit IRC02:20
*** patri728 has quit IRC02:20
*** patri586 has quit IRC02:20
*** patrickeast_ has quit IRC02:20
*** salv-orlando has joined #openstack-qa02:20
*** achanda has quit IRC02:25
*** zz_dimtruck is now known as dimtruck02:33
*** achanda has joined #openstack-qa02:33
*** jamielennox|away is now known as jamielennox02:37
*** weshay has quit IRC02:39
*** Longgeek has joined #openstack-qa02:47
*** marun has quit IRC02:49
*** ChuckC has joined #openstack-qa02:49
*** jamielennox is now known as jamielennox|away02:52
*** chandankumar has joined #openstack-qa02:54
*** ChuckC has quit IRC02:54
*** ChuckC has joined #openstack-qa02:54
*** jamielennox|away is now known as jamielennox02:55
*** catohornet has quit IRC02:55
*** jamielennox is now known as jamielennox|away02:56
*** jamielennox|away is now known as jamielennox02:57
*** achanda has quit IRC03:05
*** ChuckC has quit IRC03:06
*** ChuckC has joined #openstack-qa03:06
*** BharatK has joined #openstack-qa03:09
openstackrecheckConsole logs not available after 13:24s for gate-cinder-pylint 160417,4,6d7ef8f03:10
*** achanda has joined #openstack-qa03:10
openstackrecheckopenstack/nova change: https://review.openstack.org/159329 failed because of: gate-tempest-dsvm-nova-v21-full: https://bugs.launchpad.net/bugs/131106603:10
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]03:10
*** dimtruck is now known as zz_dimtruck03:13
*** phd2 has quit IRC03:13
*** dims_ has quit IRC03:17
*** naggappan has quit IRC03:17
*** FatDarrel has joined #openstack-qa03:17
*** FatDarrel has quit IRC03:19
*** harlowja_ is now known as harlowja_away03:21
*** jecarey has quit IRC03:21
*** salv-orlando has quit IRC03:21
*** Poornima has joined #openstack-qa03:29
*** yamahata has quit IRC03:29
*** achanda has quit IRC03:30
*** patrickeast has joined #openstack-qa03:31
*** patrickeast has quit IRC03:33
*** xgerman has joined #openstack-qa03:47
*** jhenner has quit IRC03:50
*** zz_dimtruck is now known as dimtruck03:52
openstackgerritZengfa Gao proposed openstack/tempest: Add API tests for Neutron subnetpool  https://review.openstack.org/15964403:53
*** achanda has joined #openstack-qa03:55
*** flwang1 has quit IRC03:56
xgermanHi04:13
xgermanwhat's the difference between tempest-lib and tempest and what should we use for in-tree tempest tests?04:14
*** madhu has quit IRC04:16
*** adam_g has quit IRC04:20
*** ygoto has joined #openstack-qa04:21
*** stevebaker has quit IRC04:21
*** hughsaunders has quit IRC04:24
*** adam_g has joined #openstack-qa04:28
*** adam_g has quit IRC04:28
*** adam_g has joined #openstack-qa04:28
*** Madan has joined #openstack-qa04:30
*** stevebaker has joined #openstack-qa04:34
gmannxgerman: some common framework has been split out from tempest into a lib (tempest-lib)04:35
*** madhu has joined #openstack-qa04:35
gmannxgerman: so that other projects can use the same to develop their functional tests04:35
*** hughsaunders has joined #openstack-qa04:35
xgermanthanks gmann - so if we put tempest tests in tree we should use the temepst-lib or tempest or both? I am from neutron-lbaas04:36
gmannxgerman: actually there many items to be move to lib to use tempest-lib, till then i think you need to use tempest04:37
gmannxgerman: service clients are not yet moved ti tempest-lib which you use for tests.04:38
xgermanwe mostly do api tests...04:38
gmannxgerman: please check today discussion between mtreinish, blogan and marun. Same was discussed there about lbaas testing04:41
xgermanok, thanks04:41
gmannxgerman: around 3 hr back on this channel only04:41
*** dimtruck is now known as zz_dimtruck04:44
*** rvcehimanshu has joined #openstack-qa04:45
*** openstackgerrit has quit IRC04:46
*** openstackgerrit has joined #openstack-qa04:52
*** comstud has quit IRC05:00
*** rvcehimanshu has quit IRC05:04
*** bkopilov has joined #openstack-qa05:05
*** rvcehimanshu has joined #openstack-qa05:06
*** chandankumar has quit IRC05:09
*** chandankumar has joined #openstack-qa05:10
*** rvcehimanshu has quit IRC05:13
*** rvcehimanshu has joined #openstack-qa05:14
*** yamamoto_ has joined #openstack-qa05:20
*** yamamoto has quit IRC05:20
openstackgerritJamie Lennox proposed openstack/tempest: Remove identity CLI tests  https://review.openstack.org/16105005:27
*** achanda has quit IRC05:28
*** Longgeek has quit IRC05:28
*** BharatK has quit IRC05:28
*** woodster_ has quit IRC05:30
*** jamielennox is now known as jamielennox|away05:32
*** openstack has joined #openstack-qa05:36
-sendak.freenode.net- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp05:36
*** salv-orlando has joined #openstack-qa05:38
*** amotoki has joined #openstack-qa05:43
*** rvcehimanshu has quit IRC05:44
*** BharatK has joined #openstack-qa05:45
*** rvcehimanshu has joined #openstack-qa05:45
*** xgerman has quit IRC05:49
*** k4n0 has joined #openstack-qa05:49
*** k4n0 has quit IRC05:50
*** k4n0 has joined #openstack-qa05:54
openstackgerrityalei wang proposed openstack/tempest: Add port-security extension test for Neutron  https://review.openstack.org/16072705:56
*** chandankumar has quit IRC06:06
*** Longgeek has joined #openstack-qa06:08
*** lbragstad has quit IRC06:08
*** lbragstad has joined #openstack-qa06:10
*** woodster_ has joined #openstack-qa06:10
*** ymorkovn has joined #openstack-qa06:12
*** rvcehimanshu has quit IRC06:14
*** ymorkovn1 has joined #openstack-qa06:15
*** achanda has joined #openstack-qa06:16
*** ymorkovn has quit IRC06:17
*** JayF has quit IRC06:17
*** rvcehimanshu has joined #openstack-qa06:18
*** pcrews has quit IRC06:19
*** yfried has quit IRC06:19
*** lbragstad has quit IRC06:19
*** lbragstad has joined #openstack-qa06:22
*** topshare_ has quit IRC06:28
*** chandankumar has joined #openstack-qa06:28
*** salv-orlando has quit IRC06:29
*** openstackrecheck has quit IRC06:33
*** topshare has joined #openstack-qa06:33
*** topshare_ has joined #openstack-qa06:35
*** itzikb has joined #openstack-qa06:37
*** topshare has quit IRC06:38
*** gamado has quit IRC06:40
*** rvcehimanshu has quit IRC06:47
*** rvcehimanshu has joined #openstack-qa06:48
*** k4n0 has quit IRC06:51
*** jogo has quit IRC06:53
*** yfried has joined #openstack-qa06:55
openstackgerritSergey Reshetnyak proposed openstack/tempest: Replacing data_processing with data-processing  https://review.openstack.org/13548907:00
*** xgerman has joined #openstack-qa07:01
*** rvcehimanshu has quit IRC07:02
BharatKhttps://review.openstack.org/#/c/160664/07:03
BharatKI got  Code-Review+2 Workflow+1 to this patch07:03
BharatKWaiting for Jenkins approval07:03
BharatKBut I waiting since 3 hours, and I am not able to find this integration job in http://status.openstack.org/zuul/07:03
*** flwang has quit IRC07:04
*** rvcehimanshu has joined #openstack-qa07:07
*** flwang has joined #openstack-qa07:07
*** lbragstad has quit IRC07:08
*** pcrews has joined #openstack-qa07:09
*** xgerman_ has joined #openstack-qa07:17
*** gfidente has joined #openstack-qa07:20
*** xgerman has quit IRC07:21
*** Longgeek has quit IRC07:21
*** xgerman_ has quit IRC07:22
*** yfried is now known as yfried|afk07:23
*** yfried|afk is now known as yfried07:26
*** SridharG has joined #openstack-qa07:29
*** neeti has joined #openstack-qa07:30
*** rvcehimanshu has quit IRC07:31
*** david-lyle_afk has joined #openstack-qa07:33
*** Longgeek has joined #openstack-qa07:34
*** vigneshvar has joined #openstack-qa07:34
*** gszasz has joined #openstack-qa07:37
*** mpaolino has joined #openstack-qa07:38
*** topshare_ has quit IRC07:39
openstackgerritVincent Hou proposed openstack-dev/devstack: Set my_ip in cinder.conf with value of HOST_IP  https://review.openstack.org/16111607:41
*** topshare has joined #openstack-qa07:43
openstackgerritMerged openstack/tempest: Move to hacking 0.10  https://review.openstack.org/15924607:43
openstackgerritMerged openstack/tempest: Fix and re-enable H105  https://review.openstack.org/15924707:44
*** mpaolino has quit IRC07:46
*** nmagnezi has joined #openstack-qa07:46
*** mpaolino has joined #openstack-qa07:47
*** ygoto has quit IRC07:47
*** afazekas has quit IRC07:48
*** openstackgerrit has quit IRC07:49
*** openstackgerrit has joined #openstack-qa07:49
*** flwang1 has joined #openstack-qa07:50
*** anshul has joined #openstack-qa07:50
*** anshul has quit IRC07:50
*** anshul has joined #openstack-qa07:51
*** amotoki has quit IRC07:57
*** amotoki has joined #openstack-qa07:57
*** rdekel has joined #openstack-qa07:58
*** madhu has quit IRC07:58
*** k4n0 has joined #openstack-qa08:02
openstackgerritRuijing proposed openstack-dev/devstack: Closes-Bug: #1427086 Author: Ruijing Guo <ruijing.guo@intel.com>    export OS_* creds variables even if keystone is not installed in    network/computer node  https://review.openstack.org/16112108:02
openstackbug 1427086 in devstack "network node without keystone service is unsuccefully installed" [Undecided,New] https://launchpad.net/bugs/1427086 - Assigned to Ruijing (ruijing-guo)08:02
*** ajo has quit IRC08:02
*** markus_z has joined #openstack-qa08:03
*** psedlak has quit IRC08:06
*** shardy has joined #openstack-qa08:12
*** achanda has quit IRC08:12
*** sergsh has joined #openstack-qa08:12
*** openstackrecheck has joined #openstack-qa08:14
*** mpaolino has quit IRC08:15
*** salv-orlando has joined #openstack-qa08:15
*** mpaolino has joined #openstack-qa08:15
*** sergsh has quit IRC08:17
*** mpaolino has quit IRC08:18
*** mpaolino has joined #openstack-qa08:19
*** woodster_ has quit IRC08:20
*** e0ne has joined #openstack-qa08:20
*** mpaolino has quit IRC08:21
*** mpaolino has joined #openstack-qa08:21
openstackgerritRuijing proposed openstack-dev/devstack: Closes-Bug: #1427086 Author: Ruijing Guo <ruijing.guo@intel.com>  https://review.openstack.org/16112508:22
openstackbug 1427086 in devstack "network node without keystone service is unsuccefully installed" [Undecided,In progress] https://launchpad.net/bugs/1427086 - Assigned to Ruijing (ruijing-guo)08:22
*** mpaolino has quit IRC08:23
*** mpaolino has joined #openstack-qa08:23
*** yamahata has joined #openstack-qa08:26
openstackgerritMerged openstack/tempest: Remove CONF reference from servers client  https://review.openstack.org/16058108:28
*** mpaolino has quit IRC08:28
*** mpaolino has joined #openstack-qa08:29
*** topshare has quit IRC08:29
*** vigneshvar has quit IRC08:30
*** topshare has joined #openstack-qa08:30
-openstackstatus- NOTICE: Zuul check queue stuck due to reboot maintenance window at one of our cloud providers - no need to recheck changes at the moment, they won't move forward.08:31
*** ChanServ changes topic to "Zuul check queue stuck due to reboot maintenance window at one of our cloud providers - no need to recheck changes at the moment, they won't move forward."08:31
*** sergsh has joined #openstack-qa08:31
*** mahito has quit IRC08:31
*** mpaolino has quit IRC08:32
*** mpaolino has joined #openstack-qa08:32
*** psedlak has joined #openstack-qa08:34
openstackgerritMerged openstack/tempest: Remove CONF values from images clients  https://review.openstack.org/16061308:34
*** yfried is now known as yfried|afk08:36
*** yfried|afk is now known as yfried08:36
*** ajo has joined #openstack-qa08:38
*** yamamoto_ has quit IRC08:38
*** e0ne has quit IRC08:42
*** _afezekas|pub has joined #openstack-qa08:42
*** jlanoux has joined #openstack-qa08:42
*** vigneshvar has joined #openstack-qa08:45
*** salv-orlando has quit IRC08:45
*** salv-orlando has joined #openstack-qa08:45
*** jordanP has joined #openstack-qa08:48
*** mpaolino has quit IRC08:50
*** mpaolino has joined #openstack-qa08:50
*** achanda has joined #openstack-qa08:52
*** mpaolino has quit IRC09:00
*** mpaolino has joined #openstack-qa09:01
*** yfried is now known as yfried|afk09:01
*** yfried|afk is now known as yfried09:03
*** mpaolino has quit IRC09:05
*** mpaolino has joined #openstack-qa09:06
openstackgerritgit-harry proposed openstack/tempest: Start instances using fixed network when possible  https://review.openstack.org/15977609:07
*** openstack has joined #openstack-qa15:28
*** zz_dimtruck is now known as dimtruck15:34
andreafmtreinish: I think we should still catch the InvalidConfiguration exception - when tenant isolation is true, if admin creds are missing or misconfigured, you'll get an InvalidConfiguration15:34
mtreinishandreaf: but, that's a failure isn't it? Shouldn't we just explode/fail15:35
andreafmtreinish: oh, actually yes, because everything else will also fail15:36
*** Poornima has joined #openstack-qa15:37
andreafmtreinish: +215:38
*** marun has joined #openstack-qa15:38
mtreinishhooray \o/15:39
mtreinishmy OCD would really like it if rev. 30 was the one that landed :)15:41
*** yfried has joined #openstack-qa15:41
*** marun has quit IRC15:43
*** mattoliverau has quit IRC15:43
*** hemnafk is now known as hemna15:44
*** Sukhdev has joined #openstack-qa15:46
*** dustins_ has quit IRC15:50
andreafmtreinish: :D15:51
*** dkranz has quit IRC15:51
*** ChuckC_ has joined #openstack-qa15:54
*** lbragstad has joined #openstack-qa15:56
*** dustins has joined #openstack-qa15:57
*** ihrachyshka has quit IRC15:58
*** Bsony has quit IRC16:03
*** dustins has quit IRC16:04
*** dkranz has joined #openstack-qa16:05
*** mattoliverau has joined #openstack-qa16:05
*** tsekiyama has joined #openstack-qa16:05
mtreinishandreaf: on https://review.openstack.org/#/c/157935/ was there a reason you didn't +A it?16:12
*** ihrachyshka has joined #openstack-qa16:13
*** ihrachyshka has quit IRC16:15
*** yjiang5 has quit IRC16:15
*** ihrachyshka has joined #openstack-qa16:15
*** rdekel has quit IRC16:24
openstackgerritMerged openstack/tempest: Editing comment in test "test_trust_expire"  https://review.openstack.org/15903316:25
*** jaypipes has joined #openstack-qa16:26
*** ChanServ sets mode: +o jaypipes16:26
openstackgerritMerged openstack/tempest: Skip test_subnet_details for baremetal  https://review.openstack.org/16094816:27
openstackgerritMerged openstack/tempest: Allow tenant isolation by default  https://review.openstack.org/15705216:27
*** neeti has joined #openstack-qa16:30
*** nmagnezi has quit IRC16:31
*** ymorkovn has quit IRC16:33
openstackrecheckopenstack/tempest change: https://review.openstack.org/158533 failed because of: gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1311066 https://bugs.launchpad.net/bugs/1425310, gate-tempest-dsvm-nova-v21-full: https://bugs.launchpad.net/bugs/1311066 https://bugs.launchpad.net/bugs/134451316:34
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]16:34
openstackLaunchpad 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]16:34
openstackLaunchpad bug 1344513 in Cinder "test_cinder_encryption_type_list fails" [Undecided,New]16:34
*** neeti has quit IRC16:37
*** k4n0 has quit IRC16:37
*** Poornima has quit IRC16:38
openstackgerritMatthew Treinish proposed openstack/tempest: Switch swift tests to use cred_provider with roles  https://review.openstack.org/15894116:39
openstackgerritMatthew Treinish proposed openstack-dev/devstack: Do Not Merge: Test test-accounts with roles  https://review.openstack.org/15851116:40
*** rwsu-afk is now known as rwsu16:42
*** mrodden has quit IRC16:42
*** mrda has quit IRC16:43
*** mrodden has joined #openstack-qa16:44
*** cbader has joined #openstack-qa16:45
*** chandankumar has quit IRC16:49
mriedemsdague: semi-interesting discussion in https://review.openstack.org/#/c/156166/ regarding how long sqlalchemy-migrate should support sqlalchemy>=0.7.816:50
mriedemjuno required sqla 0.8.4, kilo requires 0.9.716:50
mriedembut it's a library, so i guess we could bend over backwards as long as it works16:50
mriedem0.7.10 and 0.9.8 are actually tested by ci too, 0.8.x isn't16:50
mriedemi'd have to talk to zzzeek to see if there are any things we're keeping around as cruft to support 0.716:51
sdaguemriedem: honestly, as it's a life support library, I think it would be fine to do a 0.9 release which drops support of sqla < 0.9.816:52
sdaguedrop the jobs, and stop worring about it otherwise16:53
*** anshul has quit IRC16:54
mriedemcut the cord?!16:54
mriedemi need that old spice my boy has become a man commercial right about now16:55
mriedemit is due for a 1.0.0 release16:55
openstackgerritnithya-ganesan proposed openstack/tempest: WIP - Define validation_resources function for ssh validation  https://review.openstack.org/14922916:55
mriedemi'd be cool with dropping sqla < 0.9.8 at that point16:55
*** Noice has joined #openstack-qa16:56
*** ihrachyshka has quit IRC16:59
*** patrickeast has quit IRC16:59
*** safchain has quit IRC17:00
*** tonyb has quit IRC17:03
*** mikal has quit IRC17:03
andreafmtreinish: only I wasn't sure if for docs there was a special policy (like specs)17:03
mtreinishandreaf: nope, just a regular patch :)17:03
andreafmtreinish: ok +A17:03
*** dustins has joined #openstack-qa17:03
*** dustins has quit IRC17:04
*** amotoki has quit IRC17:04
*** jlanoux has quit IRC17:05
*** rdekel has joined #openstack-qa17:05
*** SridharG has joined #openstack-qa17:05
*** pkoniszewski has joined #openstack-qa17:05
*** mrodden has quit IRC17:06
*** mrodden has joined #openstack-qa17:06
*** rdekel has quit IRC17:09
mtreinishandreaf: cool, thanks17:10
bloganmtreinish, dtroyer: got a question about the grenade job and devstack if either of you can answer17:11
bloganor anyone who can answer grenade and devstack questions17:11
openstackgerritMerged openstack-dev/devstack: Add bash completion for OSC  https://review.openstack.org/14867917:11
*** e0ne is now known as e0ne_17:12
*** e0ne_ is now known as e0ne17:12
-openstackstatus- NOTICE: Issue solved, gate slowly digesting accumulated changes17:16
*** pkoniszewski has quit IRC17:16
*** Nic has joined #openstack-qa17:17
*** Sukhdev has quit IRC17:20
openstackgerritMerged openstack-dev/devstack: Add support for using IPA with iSCSI  https://review.openstack.org/15917817:22
*** e0ne is now known as e0ne_17:23
openstackgerritMerged openstack-dev/devstack: Complete the support of MatchMakerRedis driver  https://review.openstack.org/14353317:23
openstackgerritMerged openstack-dev/devstack: Add guestagent to defined Trove conf variables  https://review.openstack.org/16063217:23
openstackrecheckConsole logs not available after 13:29s for check-tempest-dsvm-neutron-full 160727,2,abeeaf517:24
openstackrecheckopenstack-dev/devstack change: https://review.openstack.org/160863 failed because of: gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1384373 https://bugs.launchpad.net/bugs/131106617:24
openstackLaunchpad bug 1384373 in OpenStack-Gate "fatal: Could not read from remote repository" [Medium,Triaged] - Assigned to Clark Boylan (cboylan)17:24
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]17:24
openstackgerritMatt Riedemann proposed openstack/tempest: Add scenario 'test_preserve_preexisting_port'  https://review.openstack.org/16092417:24
*** cdent has quit IRC17:33
*** sergsh has quit IRC17:40
*** e0ne has joined #openstack-qa17:41
*** e0ne has quit IRC17:41
*** chandankumar has joined #openstack-qa17:43
*** nmagnezi has joined #openstack-qa17:44
*** afazekas has quit IRC17:44
*** mrodden has quit IRC17:45
*** mrodden has joined #openstack-qa17:45
*** patrickeast has joined #openstack-qa17:47
*** jogo has joined #openstack-qa17:47
*** mwagner_lap has quit IRC17:49
*** dkranz has quit IRC17:51
*** anshul has joined #openstack-qa17:53
*** achanda has joined #openstack-qa17:54
hogepodgeclarkb For some reason this patch has fallen out of the gate. https://review.openstack.org/#/c/139499/17:57
*** afazekas has joined #openstack-qa17:57
mtreinishblogan: you should just ask the question people will respond when they're around that way17:57
*** patrickeast_ has joined #openstack-qa17:58
clarkbhogepodge: not sure I have logs that go back to december to figure out why17:58
*** patrickeast__ has joined #openstack-qa17:58
clarkbhogepodge: try reapproving it17:58
*** patri251 has joined #openstack-qa17:58
hogepodgeclarkb yeah, I tried to recheck and it was missed17:58
*** patri498 has joined #openstack-qa17:58
hogepodgeI don't have the juice to reapprove.17:58
clarkbrecheck won't help because gerrit17:58
clarkbhogepodge: ok there is a thing I can do, one moment17:58
clarkbits queued (I had to delete the jenkins +1 and recheck because gerrit events don't emit values if they don't change beause gerrit is high)17:59
*** dkranz has joined #openstack-qa18:00
*** erlon has joined #openstack-qa18:00
*** patri461 has joined #openstack-qa18:01
*** patri582 has joined #openstack-qa18:01
*** SridharG has left #openstack-qa18:02
*** markus_z has quit IRC18:03
*** annashen has joined #openstack-qa18:03
*** marun has joined #openstack-qa18:06
*** gordc has left #openstack-qa18:07
openstackgerritMatthew Treinish proposed openstack/tempest: Switch swift tests to use cred_provider with roles  https://review.openstack.org/15894118:09
openstackgerritMatthew Treinish proposed openstack-dev/devstack: Do Not Merge: Test test-accounts with roles  https://review.openstack.org/15851118:10
*** flwang has quit IRC18:18
*** luqas__ has quit IRC18:19
*** harlowja_away is now known as harlowja_18:19
*** jordanP has quit IRC18:20
*** e0ne has joined #openstack-qa18:22
*** gszasz has quit IRC18:22
*** afazekas has quit IRC18:31
*** flwang1 has quit IRC18:35
openstackgerritRichard Winters proposed openstack/tempest: Update test to fail instead of skip if floating pool is already allocated  https://review.openstack.org/16135218:36
*** FatDarrel has joined #openstack-qa18:36
*** ChuckC_ has quit IRC18:46
*** jhesketh has quit IRC18:47
*** chandankumar has quit IRC18:52
*** jhesketh has joined #openstack-qa18:53
jogodid all of rax just reboot?18:55
clarkbjogo: not all of it18:56
jogoclarkb: my IRC box went down last night18:56
clarkbdfw was last night18:56
clarkbord and iad are different days I forget which18:56
jogoahh18:56
jogosome sort of security fix?18:56
*** chandankumar has joined #openstack-qa18:56
jogomtreinish: FYI your thoughts on https://review.openstack.org/#/c/160914/ would be useful18:57
clarkbjogo: yes they announced the reboots last week18:57
hogepodgemtreinish did I mess up on the uuids?18:57
clarkband the Xen vulnerability is still under embargo while the various clouds reboot18:57
jogomtreinish: add grenade+neutron to devstack18:57
jogoclarkb: ahh18:57
jogoclarkb: if only they could do live migration :)18:57
*** dustins_ has joined #openstack-qa18:57
jogoso does this mean AWS has a rolling reboot as well?18:58
clarkbI think AWS said not all of their regions are affected but for those that are reboots are happening18:58
sdaguejogo: they put something out there today that said they had a way around that, I believe18:58
jogosdague: ahh18:58
*** dustins_ is now known as dustins18:58
*** mwagner_lap has joined #openstack-qa18:59
*** mrodden has quit IRC19:02
*** e0ne is now known as e0ne_19:05
*** e0ne_ has quit IRC19:05
*** yfried has quit IRC19:06
*** mrodden has joined #openstack-qa19:09
*** Sukhdev has joined #openstack-qa19:10
*** e0ne has joined #openstack-qa19:16
*** ChanServ changes topic to "All Things QA..."19:16
*** tkammer has quit IRC19:16
*** e0ne is now known as e0ne_19:16
marundkranz: hi19:18
marundkranz: regarding your comment on https://review.openstack.org/#/c/158021/19:18
marundkranz: I think it would be great if we could pin on tempest-lib, but would that be compatible with global requirements?19:18
*** Ananci has quit IRC19:19
*** chandankumar has quit IRC19:19
marunmtreinish: ping19:19
*** e0ne_ is now known as e0ne19:20
marunmtreinish: copying tempest code into neutron results in some pep8 failures because the 2 projects have differential enforcement19:20
marunhttps://review.openstack.org/#/c/158021/4//COMMIT_MSG19:20
marunmtreinish: do you have any interest in resolving any of the 3 outstanding issues I'm seeing in the tempest repo (and I'm happy to help) or should it be done entirely on the neutron side?19:21
marunmtreinish: make that 2 issues - N321 and H10519:21
*** Bsony has joined #openstack-qa19:24
dkranzmarun: I guess I was getting ahead of myself in assuming that of course there could not be any problem with pinning a version19:24
*** BharatK has quit IRC19:24
*** aix has quit IRC19:24
dkranzmarun: but in your gate job, wouldn't neutron be the only thing using tempest-lib?19:25
marundkranz: yes19:25
dkranzmarun: That job should not be installing tempest19:25
marundkranz: my conception of global requirements is that it is more about packaging than anything else19:25
marundkranz: so that distros don't have to ship multiple copies of the same library for different applications19:26
marundkranz: it would seem to have the same problem as co-gating, though19:26
marundkranz: in that we always have to be on the same page19:26
dkranzmarun: I think it is more that pip is global and you cannot have more than one version installed. Same for rpm/deb, etc.19:27
*** Nic has quit IRC19:27
dkranzmarun: but jogo and others are dismantling the global requirements I think. He could answer this.19:27
marundkranz: pip doesn't have to be global.  venvs ftw19:28
openstackgerritLuigi Toscano proposed openstack/tempest: Decouple Sahara templates test from vanilla plugin  https://review.openstack.org/16137019:28
marundkranz: but what I'm suggesting is that maybe we need to be allowed to converge on common versions of libraries rather than forcing it all the time19:28
dkranzmarun: yes, but I think all components are now installed in the same venv19:28
dkranzmarun: and that is what is being changed19:28
jogodkranz: ?19:29
marundkranz: I'm not sure that gets us away from global requirements though19:29
marunjogo: what was the intent of installing in venvs?19:29
dkranzjogo: point (2) on the last comment in https://review.openstack.org/#/c/158021/19:29
marunmy memory is rusty19:29
jogomarun: not sure about the per service venvs19:30
jogomarun: but we CLIs need to be in venvs19:30
marunjogo: ah, ok19:30
*** tosky has quit IRC19:30
dkranzjogo: the question was can neutron running with tempest lib pin the version of tempest lib so it does not need to co-gate with it19:30
jogosince we want to test latest CLI with stable branch servers19:30
marunjogo: we were just discussing the possibility of tempest-lib not gating on neutron's api job, but rather pinning the version the api job uses and using periodic jobs to evaluate newer revisions19:30
marunjogo: I don't think that is compatible with global requirements the way it is today19:31
marunjogo: and I'm wondering if maybe we shouldn't revisit19:31
marunjogo: and allow convergence to common versions rather than mandating versions19:31
jogomarun: so the theory behind how we do it today, is we want master to roll with the punches to make sure we don't get stuck on ancient versions of libs19:31
marunjogo: it's a forcing function19:32
marunjogo: it's a mechanism of control exercised by infra on the projects19:32
jogomurun: yup19:32
marunjogo: and I'm not sure there aren't better ways to accomplish the same result19:32
jogomarun: I agree with you19:33
jogothat being said, tempest-lib should *not* gate on neutron's usage of it19:33
marunjogo: well, then we can't gate19:33
jogoif tempest-lib breaks you, then tempest-lib needs more testing around its API19:33
marunjogo: tempest-lib should not be allowed to break us, period19:33
jogomarun: you can, but risk breaking if mtreinish gets drunk :)19:33
marun:/19:33
marunjogo, dkranz: without pinning, i don't see much of a choice but to vendor indefinitely19:34
jogomarun: many things can break you, tempest-lib is not special in that way19:35
marunwe can't risk impacting x number of developers because of a lack of rigor on the tempest-lib side of thigns19:35
marunjogo: I don't care if something breaks me19:35
marunjogo: I care if it halts all neutron merges while we diagnose the problem19:35
jogomarun: so if that happens, we add a '!=1.1.1' to g-r and land it quickly and move on19:36
marunjogo: my 'stupid-o-meter' is pegging19:36
jogomarun: I think you are over estimating the amount of time this would break things for19:36
jogomarun: that being said, I agree with your sentiment about g-r as a forcing mechanism.19:36
marunjogo: I think you're underestimating the cost of random errors interrupting development :/19:36
jogomarun: but lets better deal with the general case, not this case19:37
marunjogo: fair enough19:37
jogomarun: I actually think the current approach on stable g-r may be most of the solution19:38
jogomarun: cap versions to currently installed. And then we have a periodic job proposing new caps to g-r19:38
jogoand if that passes tests, its a trivial approval19:38
marunjogo: that's pretty much what I want to see19:38
marunjogo: it's pinning19:39
marunjogo: on whatever we have some degree of confidence is going to work19:39
jogomarun: yeah, if you can get buy in my guess is its <1 week of work19:39
marunjogo: what do you think the stumbling block would be?19:39
jogomarun: getting buy in19:39
jogomarun: the basic steps would be:19:40
jogouse output of pip-freeze and pass it into my cap.py tool to generate caps19:40
*** pkoniszewski has joined #openstack-qa19:41
jogokeep a version of g-r without caps around, to use on periodic jobs that generate new pip-freeze files19:41
*** mlavalle has joined #openstack-qa19:42
openstackgerritSergey Reshetnyak proposed openstack/tempest: Replacing data_processing with data-processing  https://review.openstack.org/13548919:43
openstackgerritMerged openstack/tempest: Add tempest config guide doc  https://review.openstack.org/15793519:44
*** mikal has joined #openstack-qa19:46
*** Bsony has quit IRC19:48
marunjogo: that seems pretty straightforward19:48
marunjogo: what are the downsides of such an approach?19:49
*** Nic has joined #openstack-qa19:49
marunjogo: and what do you see potential resistence based on?19:49
openstackrecheckopenstack/nova change: https://review.openstack.org/161112 failed because of: gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/1251521 https://bugs.launchpad.net/bugs/131106619:50
openstackLaunchpad bug 1251521 in OpenStack Compute (nova) "Volume detach in tempest fails because libvirt refuses connections" [Undecided,Invalid]19:50
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]19:50
jogomtreinish: looks like we still need your wheel cache patch to land ^19:50
jogomtreinish: re: slow19:51
*** zz_jgrimm is now known as jgrimm19:51
*** anshul has quit IRC19:51
jogomarun: we need people to chase down failures and fix them in a reasonable fashion19:51
jogoturns out we haven't done a great job of that in the past19:51
*** Rockyg has joined #openstack-qa19:51
jogobut IMHO, at the very least this is worth a try19:51
*** dkranz has quit IRC19:54
*** e0ne is now known as e0ne_19:55
*** EmilienM is now known as EmilienM|afk19:56
*** harlowja_ has quit IRC19:57
mtreinishjogo: I'm getting drunk and breaking tempest-lib's api?19:59
jogomtreinish: its a risk :)20:00
*** harlowja has joined #openstack-qa20:01
*** e0ne_ has quit IRC20:01
mtreinishjogo: yep, well the patch is up there. I would start pestering in -infra if we need it to land soon :)20:01
*** jaypipes has quit IRC20:01
jogomtreinish: I was talking about a hypothetical here20:01
mtreinishhogepodge: nope, they seem to be working as intended20:02
mtreinishjogo: heh, I switched to the wheel cache patch :)20:02
*** jaypipes has joined #openstack-qa20:02
*** ChanServ sets mode: +o jaypipes20:02
mtreinishjogo: sure it's the same risk with any library20:02
jogomtreinish: oh that one20:02
*** naggappan has joined #openstack-qa20:02
mtreinishjogo: just my unordered response to the 5 pings from you when I was getting food20:03
mtreinishhogepodge: did my response mail not go through?20:04
*** e0ne has joined #openstack-qa20:05
*** dkranz has joined #openstack-qa20:06
*** anshul has joined #openstack-qa20:07
*** achanda has quit IRC20:07
bloganmtreinish: the grenade job doesn't install devstack plugins in the upgrade script, and neutron-lbaas has put its devstack installation in tree as a devstack plugin20:09
bloganmetrienish: the grenade upgrade script calls install_neutron devstack function, so would it be appropriate to patch teh grenade job to install the devstack plugins or to patch devstack's install_neutron to install any *aas repos devstack plugins20:10
*** yamahata has quit IRC20:11
*** pkoniszewski has quit IRC20:12
*** andreykurilin_ has joined #openstack-qa20:13
mtreinishblogan: so there isn't a good plugin story right now for grenade. That's something we need to address. The goal is to have grenade be pluggable in the same way as devstack20:13
*** dustins has quit IRC20:13
mtreinishblogan: that being said I don't have a specific answer right now for that. I can say that modifying devstack to point to the external plugin for the *aas stuff isn't the solution though :)20:14
bloganmtreinish: yeah i didnt like that either, but don't have the experience with devstack or grenade to know the best approach20:15
mtreinishblogan: I think working on making grenade understand devstack plugins would probably be a good start20:15
mtreinishthe only caveat is that it might conflict with a future plugin direction for grenade.20:16
mtreinishwhen that's decided on20:16
bloganmtreinish: ok thanks, i'll go with that approach20:16
*** annashen has quit IRC20:17
bloganmtreinish: oh thats fine, if it requires a refactor later, thats usually what happens with being one of the first attempters20:17
*** flwang has joined #openstack-qa20:17
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: If wheelhouse is pre-generated, set WHEELHOUSE in localrc  https://review.openstack.org/16093320:27
*** nmagnezi has quit IRC20:27
mtreinishblogan: I think your biggest issue is going to be on the old side (pre upgrade) because plugins don't exist on stable devstack IIRC20:27
dtroyeradding plugin support to grenade shouldn't be too hard, it'll basically need to do the setup stuff (sourcing files and defining vars) so the right functions are available.  long-term it might make sense for grenade to pull the upgrade scripts form the plugin dir too, but we hoestly haven't spent much time thinking about that20:28
dtroyerI think the old side is not a worry, as it runst stack.sh and should Just Work(TM).  it's the new side that picks bits out of stack.sh instead of running it20:28
bloganmtreinish: good point, we'll have to research that for sure, though I don't think we can get away with it not making it in for kilo20:28
openstackgerritMaru Newby proposed openstack-infra/devstack-gate: Revert "Add features.yaml entry for 'no services'"  https://review.openstack.org/16140220:28
*** dims has quit IRC20:29
blogani mean we can get away with it not making it for kilo20:29
*** dims has joined #openstack-qa20:29
*** andreykurilin_ has quit IRC20:30
mtreinishdtroyer: yeah, it's own plugin dir/options for the long term is what I was referring to before20:30
blogandtroyer: good to know, we'll just have to do our due diligence on it to make sure it all works, and we don't kill neutron grenade again20:31
*** achanda has joined #openstack-qa20:41
adam_gblogan, funny,i just spent some time moving all the ironic stuff out of devstack to an in-tree plugin and realized its going to hit the same problem /w grenade20:42
*** apevec has quit IRC20:53
*** annashen has joined #openstack-qa20:54
*** Bsony has joined #openstack-qa20:55
*** dustins has joined #openstack-qa20:59
*** Bsony has quit IRC20:59
adam_gmtreinish, devananda would love to get your thoughts here: https://review.openstack.org/16142021:06
adam_gdevananda, that would drop test runtime significantly of the 1hr+ grenade job, and mean we can start to run it in the ironic queue (we really need that for kilo)21:07
*** Sukhdev has quit IRC21:10
*** Longgeek has quit IRC21:10
mtreinishadam_g: you don't want to also run some of the compute tests too? Because, won't the baremetal tag just run the tests which touch the ironic api (so no guest booting)?21:10
* devananda looks21:11
mtreinishbut I'm generally a big fan of getting rid of that regex21:11
adam_gmtreinish, that'll also run the baremetal scenario tests21:11
adam_gmtreinish, which cover testing glance/keystone/nova integration21:11
devanandaadam_g: so this drops the nova coverage?21:12
devanandaie, where we configure nova w/ ironic, then run some subset of nova's tests21:12
*** ylobanko_ has joined #openstack-qa21:12
mtreinishadam_g: ok, well I don't have an issue with that. You guys can decide which subset is useful for you. :)21:13
adam_gdevananda, it skips the tempest.api.compute.* tests21:13
*** ylobanko_ has quit IRC21:13
openstackgerritMaru Newby proposed openstack-infra/devstack-gate: Remove configurable testr artifact processing  https://review.openstack.org/16142221:13
adam_gdevananda, we still end up stressing the nova api during the baremetal scenario tests21:13
devanandaadam_g: of which there's only one, right?21:14
*** e0ne has quit IRC21:14
devanandaso I believe there is/was value in running some subset of the tempest.api.compute tests against nova+ironic21:14
bloganadam_g: glad what we are doing is helping out, even when we make mistakes along the way21:15
devanandafolks put in a bunch of effort to make them pass (yourself included) and I think I'm missing some context as to why you'd wholesale drop all of them?21:15
devanandaadam_g: oh -- is the goal to drop the regex, and then apply the "baremteal" tag to some of the tempest.compute.api tests?21:15
*** aysyd has quit IRC21:16
adam_gdevananda, TBH i can't see what we are gaining by running those, other than verifying we can boot an instance--which is what the baremetal scenario test is doing.  with the API tests, that tends to happen in setUp() of some test like, create_keypair21:16
*** dkranz has quit IRC21:17
adam_gthese are tests tagged compute http://paste.ubuntu.com/10531286/21:17
devanandatempest.api.compute.admin.test_baremetal_nodes.BaremetalNodesAdminTestJSON.test_list_baremetal_nodes[gate,id-e475aa6e-416d-4fa4-b3af-28d5e84250fb,smoke]21:18
mtreinishdevananda: that'll get picked up by the baremetal regex filter21:18
devanandak21:18
adam_ghttp://paste.ubuntu.com/10531301/ is the baremetal tag21:18
mtreinishdevananda: testr just does a straight re.match() on whatever you pass in for each test_id in the full list21:19
devanandaoooh. I thought there was add'l metadata being added to tempest test definitions that was going to be used for this21:19
adam_ga big hunk of the compute tests would be skipped when run against ironic, and the ones that would run will be going through the trouble of booting the baremetal node to validate something that could more easily be validated against a kvm21:19
*** achanda has quit IRC21:19
devanandaadam_g: so for example, we do want to make sure that nothing we do breaks tempest.api.compute.admin.test_hypervisor21:20
devanandaand that needs to be run againts nova configured to use the ironic virt driver21:20
mtreinishdevananda: because of how test attrs are handled they get appended to the test_id (what's inside the [] are the attrs)21:20
*** rfolco has quit IRC21:20
mtreinishso when we add test tags with a decorator, they just get appended to the test id21:21
adam_gdevananda, ah, thats a good one21:21
devanandamtreinish: got it. so we could add 'baremetal' attr to test_hypervisor*, they'd get added to [..] and then match the regex21:21
mtreinishdevananda: yep21:21
adam_gmtreinish, using the services decorator also does a config check to make sure the service is enabled?21:21
mtreinishadam_g: yes, the service decorator does. But, you can just use the normal attr decorator21:22
mtreinishto tag but not skip on config21:22
adam_gmtreinish, i see, cool21:23
devanandaalso, tempest.scenario.test_baremetal_basic_ops.BaremetalBasicOps.test_baremetal_server_ops[baremetal,compute, should probably not have "compute" tag21:23
mtreinishwe'd just need to document that's what we're doing21:23
mtreinishdevananda: well if it makes nova calls that's the criteria for the service tag there21:23
mtreinishdevananda: because it depends on nova being present21:23
adam_gdevananda, it needs nova21:23
devanandaoh21:24
devanandaright. i mean, we dont want that to run against nova if ironic is -not- present21:24
devanandaare the tags AND'd ?21:24
*** ylobankov_ has joined #openstack-qa21:24
mtreinishdevananda: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/test.py#n10221:25
mtreinishdevananda: it'll be skipped if barmetal isn't enabled in config21:25
devanandaah, great21:25
devanandathen yea, both tags is correct21:25
*** Rockyg has quit IRC21:26
*** tonyb has joined #openstack-qa21:28
*** patrickeast has quit IRC21:29
*** tonyb has quit IRC21:29
devanandaadam_g: ok, after reading through that, I mostly agree. in addition to ^, there might be a few others that we should run, but overall -- yea, not helpful21:30
*** tonyb has joined #openstack-qa21:30
adam_gdevananda, cool. ill put up a patch to tag those hypervisor tests. if any others jump out at you, let me know21:30
devanandasome of tempest.scenario.test_security_groups_basic_ops.TestSecurityGroupsBasicOps could be interesting -- but I dont know if they're supported today, and it looks like they are currently bieng skipped21:31
*** patri582 has quit IRC21:31
*** ChuckC has joined #openstack-qa21:32
*** patrickeast__ has quit IRC21:32
*** patrickeast_ has quit IRC21:32
*** patri251 has quit IRC21:32
*** patri498 has quit IRC21:32
devanandaadam_g: some of test_server_actions.ServerActionsTestJSON too, maybe21:32
adam_gdevananda, tempest.api.compute.admin.test_hosts21:33
*** patri461 has quit IRC21:33
devanandaadam_g: yep21:33
devanandaadam_g: and the negative of that21:33
adam_gya21:34
*** patrickeast has joined #openstack-qa21:34
*** naggappan has quit IRC21:35
mtreinishmasayukig: https://review.openstack.org/#/c/159147/ i think you asked for that at some point21:35
*** patrickeast_ has joined #openstack-qa21:35
*** patrickeast__ has joined #openstack-qa21:35
*** patri255 has joined #openstack-qa21:35
*** patri910 has joined #openstack-qa21:35
*** patri169 has joined #openstack-qa21:35
*** patri733 has joined #openstack-qa21:35
*** ylobankov_ has quit IRC21:35
*** ylobanko_ has joined #openstack-qa21:37
openstackgerritAdam Gandelman proposed openstack/tempest: Tag compute hosts + hypervisor tests for baremetal  https://review.openstack.org/16142921:43
adam_gdevananda, ^21:43
*** EmilienM|afk is now known as EmilienM21:43
adam_gdevananda, i intentionally left off the test_server_actions for now. with the changes to IPA by default, we can't run /w lowmem/parallel anymore by default and that test will increase serial runtime significantly21:44
devanandaadam_g: ack21:45
*** melwitt has joined #openstack-qa21:45
devanandaadam_g: see the comment in def test_get_hypervisor_uptime21:47
*** dustins_ has joined #openstack-qa21:47
*** dustins has quit IRC21:48
devanandaadam_g: should we run it just to ensure that it is skipped? ...21:48
adam_gdevananda, we can probably just untag it21:49
*** dustins_ has quit IRC21:49
devanandathat's what I'm thinking21:49
adam_gbut good catch :)21:49
openstackgerritnithya-ganesan proposed openstack/tempest: WIP - Define validation_resources function for ssh validation  https://review.openstack.org/14922921:49
openstackgerritAdam Gandelman proposed openstack/tempest: Tag compute hosts + hypervisor tests for baremetal  https://review.openstack.org/16142921:49
*** ylobanko_ is now known as ylobankov_21:51
*** dhellmann has quit IRC21:52
openstackgerritAl Miller proposed openstack-dev/devstack: Remove LBaaS support from devstack  https://review.openstack.org/16143621:53
*** dhellmann has joined #openstack-qa21:55
openstackgerritMatthew Treinish proposed openstack-infra/subunit2sql: Add microsecond columns to the test_runs table  https://review.openstack.org/16005721:56
mtreinishandreaf: ^^^ I added a flag to allow bypassing the data migration if the daata was already lost21:56
*** dhellmann has quit IRC21:56
*** dhellmann has joined #openstack-qa21:57
mtreinishif that approach works for you I'll push up a patch to the puppet repo to add the flag to the config file for running migrations after it lands21:57
*** harlowja has quit IRC21:58
hogepodgemtreinish, no I saw it.21:58
hogepodgeI'll clarify with Catherine that auto-generated tests are going to show repeats.21:58
hogepodgemtreinish: I've been away from my computer this morning. I had this fear that the checking tool was broken and many functions had been tagged with repeated ids21:59
*** patrickeast has quit IRC22:06
*** pc_m has quit IRC22:07
*** patrickeast_ has quit IRC22:09
*** patrickeast__ has quit IRC22:09
*** patri169 has quit IRC22:10
*** patri910 has quit IRC22:10
*** patri255 has quit IRC22:10
*** patri733 has quit IRC22:10
*** patrickeast has joined #openstack-qa22:11
*** ChuckC has quit IRC22:11
*** patrickeast has quit IRC22:12
*** patrickeast has joined #openstack-qa22:12
*** patrickeast_ has joined #openstack-qa22:13
*** patrickeast__ has joined #openstack-qa22:13
*** patri582 has joined #openstack-qa22:13
*** patri455 has joined #openstack-qa22:13
*** patri696 has joined #openstack-qa22:13
*** jamielennox|away is now known as jamielennox22:14
openstackrecheckopenstack/horizon change: https://review.openstack.org/158867 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1384373 https://bugs.launchpad.net/bugs/1311066 https://bugs.launchpad.net/bugs/142531022:18
openstackLaunchpad bug 1384373 in OpenStack-Gate "fatal: Could not read from remote repository" [Medium,Triaged] - Assigned to Clark Boylan (cboylan)22:18
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]22:18
openstackLaunchpad 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]22:18
mtreinishhogepodge: well it's not autogenerated tests (those are different) this is just subclasses of test classes. It's how we handle a lot of the ipv6 and cinder v2 api testing22:19
mtreinishsince most of what has to change for that is a couple of params and the test code is the same22:19
*** anshul has quit IRC22:20
*** achanda has joined #openstack-qa22:20
mtreinishhogepodge: no worries, catherineD's email did make it sound like there was a bug in the ids. But, that was expected and desired behavior going into it22:20
*** e0ne has joined #openstack-qa22:22
*** jgrimm is now known as zz_jgrimm22:23
*** Noice has quit IRC22:25
*** ylobankov_ has quit IRC22:25
*** achanda has quit IRC22:26
*** dpaterson has joined #openstack-qa22:26
*** dimtruck is now known as zz_dimtruck22:26
*** yfried has joined #openstack-qa22:27
*** harlowja has joined #openstack-qa22:28
*** e0ne has quit IRC22:30
*** mriedem is now known as mriedem-away22:30
*** patrickeast___ has joined #openstack-qa22:31
mtreinishhogepodge: I'm not sure why you would need to generate a second id, presumably if the test is being run you have an opinion in defcore/refstack on the feature too.22:36
openstackgerritIan Wienand proposed openstack-dev/devstack: Remove packaged rpm python libraries  https://review.openstack.org/16145522:36
mtreinishlike is it ok to run with ipv6 disabled. Or cinder v2 disabled? If not then those would be non-optional config options in the tempest config for refstack's usage22:37
*** cbader has quit IRC22:40
*** erlon has quit IRC22:41
*** jogo has quit IRC22:41
*** jogo has joined #openstack-qa22:42
hogepodgemtreinish: we do, but we'd like to have a 1:1 mapping of features and ids. I completely understand that for code reuse there may not be that 1:1 mapping between function names and features.22:42
mtreinishhogepodge: but you'd still have that wouldn't you? It just depends on your definition of feature :)22:43
*** dboik has quit IRC22:46
hogepodgemtreinish: yeah, but say we require ipv4 but not ipv6 for some reason, could there be confusion on tests that pass or don't. Just thinking out loud. I think that Catherine would expect that ever named test has a unique identifier, and I don't think it's an unreasonable one.22:48
mtreinishhogepodge: well I think that's require v4 but not v6 is wrong. But ignoring that, that case wouldn't be a problem because the ipv6 tests will be skipped.22:49
mtreinishwhich is basically my point if feature X has multiple configs/ways to run it. Given what refstack says is required if they don't all pass then the feature doesn't work on the cloud22:50
mtreinishit only gets murky if you guys aren't opinionated about a feature that we subclass on22:50
*** bknudson has quit IRC22:50
openstackgerritAdam Gandelman proposed openstack-dev/devstack: Set rootfstype=ramfs for low memory Ironic nodes  https://review.openstack.org/16146122:52
*** jecarey_ has quit IRC22:54
*** gfidente has quit IRC22:59
*** Rockyg has joined #openstack-qa23:00
*** mrodden has quit IRC23:03
*** mrodden has joined #openstack-qa23:05
*** Nic has quit IRC23:07
*** jaypipes has quit IRC23:09
openstackgerritMerged openstack/tempest-lib: Handle missing launchpadlib gracefully  https://review.openstack.org/15914723:12
*** marun has quit IRC23:12
openstackrecheckopenstack/heat change: https://review.openstack.org/159090 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/131106623:16
openstackLaunchpad bug 1311066 in OpenStack-Gate "Some nodes allocated in node pool are very very slow" [High,Confirmed]23:16
*** dkranz has joined #openstack-qa23:21
*** dkranz has quit IRC23:38
*** achanda has joined #openstack-qa23:40
*** carl_baldwin has joined #openstack-qa23:43
*** tochi has joined #openstack-qa23:43
*** annashen has quit IRC23:46
*** bknudson has joined #openstack-qa23:52
*** EmilienM is now known as EmilienM|afk23:56
*** yamamoto has joined #openstack-qa23:57
*** yamahata has joined #openstack-qa23:57

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