Monday, 2014-04-14

*** matsuhashi has joined #openstack-qa00:01
*** Guest41065 is now known as StevenK00:06
openstackgerritA change was merged to openstack-dev/devstack: Fix typos in HACKING.rst  https://review.openstack.org/8711400:25
*** adalbas has quit IRC00:26
*** adalbas has joined #openstack-qa00:42
*** mestery has joined #openstack-qa00:48
*** mriedem1 has quit IRC01:01
*** mriedem has joined #openstack-qa01:05
*** nati_ueno has joined #openstack-qa01:12
*** ianw has quit IRC01:16
*** ianw has joined #openstack-qa01:16
*** mkollaro has quit IRC01:34
*** nosnos has joined #openstack-qa01:41
openstackgerritA change was merged to openstack-dev/devstack: Avoid pm-is-supported spam in logs  https://review.openstack.org/8427601:42
*** mriedem has quit IRC02:04
*** nati_ueno has quit IRC02:21
*** GMann has joined #openstack-qa02:23
*** mlavalle has quit IRC02:36
openstackgerritGhanshyam Mann proposed a change to openstack/tempest: Verify create/get flavor attributes of Nova APIs  https://review.openstack.org/8054302:37
*** amotoki has quit IRC02:38
*** adalbas has quit IRC02:51
*** Abhi has joined #openstack-qa03:06
*** GMann has quit IRC03:07
*** nati_ueno has joined #openstack-qa03:23
*** GMann has joined #openstack-qa03:24
*** matsuhashi has quit IRC03:31
*** nati_ueno has quit IRC03:35
openstackrecheckopenstack/heat change: https://review.openstack.org/75798 failed gate-heat-python26, gate-heat-python27, gate-grenade-dsvm-neutron in the gate queue with an unrecognized error03:47
*** nosnos has quit IRC03:49
*** nati_ueno has joined #openstack-qa03:50
*** nati_ueno has quit IRC04:05
*** matsuhashi has joined #openstack-qa04:38
*** nosnos has joined #openstack-qa04:48
*** chandan_kumar has joined #openstack-qa04:52
openstackgerritMasayuki Igawa proposed a change to openstack-dev/devstack: Add fping utility path to nova.conf  https://review.openstack.org/8680605:26
*** om has joined #openstack-qa05:35
*** nosnos_ has joined #openstack-qa05:36
*** nosnos has quit IRC05:36
*** nati_ueno has joined #openstack-qa05:42
*** zul has quit IRC05:46
*** nati_ueno has quit IRC05:56
*** zul has joined #openstack-qa06:00
*** matsuhas_ has joined #openstack-qa06:00
*** matsuhashi has quit IRC06:03
*** Abhi has quit IRC06:06
*** akamyshnikova has quit IRC06:11
*** afazekas has joined #openstack-qa06:12
*** psharma has joined #openstack-qa06:21
*** amotoki has joined #openstack-qa06:24
*** Abhi has joined #openstack-qa06:26
*** akamyshnikova has joined #openstack-qa06:26
*** nosnos has joined #openstack-qa06:32
*** nosnos_ has quit IRC06:32
*** matsuhas_ has quit IRC06:37
*** matsuhashi has joined #openstack-qa06:37
*** matsuhashi has quit IRC06:37
*** matsuhashi has joined #openstack-qa06:38
*** _sweston_ has joined #openstack-qa06:45
*** _sweston_ has quit IRC06:49
*** sweston has joined #openstack-qa06:49
openstackgerritZhi Kun Liu proposed a change to openstack/tempest: fix import for default clients in clients.py  https://review.openstack.org/8717306:52
openstackgerritxu-haiwei proposed a change to openstack/tempest: Verify the response attributes of 'create-aggregate' API  https://review.openstack.org/8369207:03
*** andreaf has joined #openstack-qa07:08
*** vponomaryov has joined #openstack-qa07:13
*** jamielennox is now known as jamielennox|away07:18
openstackgerritMh Raies proposed a change to openstack/tempest: Implementing XML client for VPNaaS  https://review.openstack.org/8525507:21
openstackgerritChristian Schwede proposed a change to openstack/tempest: Swift formpost cleanup  https://review.openstack.org/8624407:23
*** jordanP has joined #openstack-qa07:39
openstackgerritZhi Kun Liu proposed a change to openstack/tempest: fix import for default clients in clients.py  https://review.openstack.org/8717307:42
*** qba73 has joined #openstack-qa07:47
openstackrecheckConsole logs not available after 13:31s for gate-ceilometer-pep8 77612,34,70eb9ed07:56
*** morganfainberg is now known as morganfainberg_Z08:00
*** yassine has joined #openstack-qa08:04
*** giulivo has joined #openstack-qa08:08
*** jpich has joined #openstack-qa08:10
openstackrecheckConsole logs not available after 14:11s for gate-ceilometer-pep8 77845,30,4bd6a5608:11
*** jhenner has joined #openstack-qa08:20
openstackgerritGhanshyam Mann proposed a change to openstack/tempest: Adds test for volume attachments Nova V2 APIs  https://review.openstack.org/8719108:28
*** jlibosva has joined #openstack-qa08:58
openstackrecheckConsole logs not available after 13:36s for gate-heat-pep8 86526,6,bd7b54a09:05
*** fbo_away is now known as fbo09:18
*** mkollaro has joined #openstack-qa09:25
*** GMann has quit IRC09:30
*** StevenK has quit IRC09:44
openstackgerritAla Rezmerita proposed a change to openstack/tempest: Cleanup neutron default security groups on tenant deletion  https://review.openstack.org/7927509:47
*** yassine has quit IRC09:48
*** tosky has joined #openstack-qa10:00
akamyshnikovaafazekas, mkoderer, sdague Hello! I've got 3 changes with API test for lbaas and floating ips. They have already had +2s. Could you please find some time to have a look at them one more time? LBaaS - https://review.openstack.org/66541 Floating ips - https://review.openstack.org/63723 https://review.openstack.org/7125110:05
*** fbo is now known as fbo_away10:10
*** sahumada has quit IRC10:12
*** sahumada has joined #openstack-qa10:12
*** ianw has quit IRC10:17
*** mkollaro has quit IRC10:25
*** mkollaro has joined #openstack-qa10:32
openstackgerritSteven Hardy proposed a change to openstack-dev/grenade: Add initial support for Heat  https://review.openstack.org/8697810:40
*** matsuhashi has quit IRC10:40
*** matsuhashi has joined #openstack-qa10:41
*** matsuhashi has quit IRC10:50
*** matsuhas_ has joined #openstack-qa10:52
openstackgerritSean Dague proposed a change to openstack-infra/devstack-gate: normalize the services for a post grizzly world  https://review.openstack.org/8428710:54
*** chandan_kumar has quit IRC10:55
*** dkorn has joined #openstack-qa11:00
*** matsuhas_ has quit IRC11:00
*** yassine has joined #openstack-qa11:01
*** matsuhas_ has joined #openstack-qa11:04
*** Abhi has quit IRC11:07
*** chandan_kumar has joined #openstack-qa11:08
*** matsuhas_ has quit IRC11:15
*** chandan_kumar has quit IRC11:16
*** matsuhashi has joined #openstack-qa11:16
*** fbo_away is now known as fbo11:26
*** amotoki has quit IRC11:28
*** weshay has joined #openstack-qa11:29
*** chandan_kumar has joined #openstack-qa11:29
*** weshay has quit IRC11:32
*** sthaha has joined #openstack-qa11:36
*** sthaha has quit IRC11:36
*** sthaha has joined #openstack-qa11:36
*** sthaha has quit IRC11:36
*** sthaha has joined #openstack-qa11:37
*** sthaha has quit IRC11:37
*** sthaha has joined #openstack-qa11:37
*** matsuhashi has quit IRC11:46
*** StevenK has joined #openstack-qa11:48
*** om has quit IRC11:50
*** om has joined #openstack-qa11:51
*** matsuhashi has joined #openstack-qa11:51
openstackrecheckopenstack/horizon change: https://review.openstack.org/87053 failed gate-grenade-dsvm-neutron, gate-horizon-python27-django14 in the gate queue with an unrecognized error11:58
*** nosnos has quit IRC12:01
openstackgerritnayna-patel proposed a change to openstack/tempest: Adds more testcases to test_telemetry_alarming_api  https://review.openstack.org/8249112:02
openstackgerritA change was merged to openstack-dev/devstack: Prevent the use of 'swift' in ENABLED_SERVICES  https://review.openstack.org/8714012:04
*** enikanorov has quit IRC12:04
openstackgerritA change was merged to openstack-dev/devstack: Delete unused neutron port  https://review.openstack.org/8501312:04
*** enikanorov has joined #openstack-qa12:04
shardyHi all, can anyone tell me what OS check-grenade-dsvm uses, or more specifically, what OS should I use for local VM testing of grenade?12:14
shardyI'm working on a new grenade test for heat, and a lot of stuff seems broken using a F20 image12:14
*** dkranz has joined #openstack-qa12:15
*** rfolco has joined #openstack-qa12:16
openstackgerritA change was merged to openstack-dev/devstack: Remove keystone_ec2_uri flag from lib/heat  https://review.openstack.org/8377712:18
*** aysyd has joined #openstack-qa12:20
maurosrshardy: try it with ubuntu 12.0412:25
*** bada has quit IRC12:25
shardymaurosr: Ok, will do, thanks12:26
maurosryw12:26
*** matsuhashi has quit IRC12:29
*** matsuhashi has joined #openstack-qa12:30
*** om has quit IRC12:31
*** weshay has joined #openstack-qa12:33
*** BobBall_ has quit IRC12:33
*** adalbas has joined #openstack-qa12:33
*** matsuhashi has quit IRC12:34
*** matsuhashi has joined #openstack-qa12:42
*** om has joined #openstack-qa12:45
*** zul has quit IRC12:48
sdagueshardy: it's running on 12.04 in the gate, fedora patches would be welcomed, though my guess is most of those would be patches against devstack12:48
shardysdague: Ok, I'd like to debug the heat test issues first, then I'll try to work out the Fedora problems12:49
sdaguecool12:49
*** gszasz has joined #openstack-qa12:49
*** zul has joined #openstack-qa12:51
openstackgerritYves-Gwenael Bourhis proposed a change to openstack-dev/devstack: Fixed pip with http(s)_proxy  https://review.openstack.org/8726112:52
*** qba73_ has joined #openstack-qa12:55
*** qba73 has quit IRC12:57
*** devvesa has joined #openstack-qa13:00
openstackgerritA change was merged to openstack-dev/devstack: Use nova bindir of /usr/bin on SUSE  https://review.openstack.org/8693313:04
*** matsuhashi has quit IRC13:06
*** matsuhas_ has joined #openstack-qa13:10
openstackrecheckConsole logs not available after 14:35s for check-tempest-dsvm-postgres-full 86966,4,392b95f13:17
*** mriedem has joined #openstack-qa13:18
*** HenryG has quit IRC13:21
*** gszasz has quit IRC13:23
*** om has quit IRC13:28
*** om has joined #openstack-qa13:29
*** jgrimm has quit IRC13:30
*** kgriffs|afk is now known as kgriffs13:30
openstackrecheckConsole logs not available after 13:40s for check-tempest-dsvm-full 83371,6,f29cbaa13:35
*** yfried__ has quit IRC13:39
*** mwagner_lap has quit IRC13:39
SergeyLukjanovsdague, morning13:40
SergeyLukjanovsdague, I have a question re adding sahara client to tempest13:40
SergeyLukjanovhttps://review.openstack.org/#/c/83317/6/tempest/clients.py13:40
sdagueSergeyLukjanov: sure13:41
SergeyLukjanovsdague, thx!13:41
SergeyLukjanovsdague, currently, our client searches for the endpoint in constructor and it means that it'll fail if no sahara available13:41
SergeyLukjanovsdague, we're initializing all stuff in client constructor13:42
*** matsuhas_ has quit IRC13:42
SergeyLukjanovsdague, is it incorrect?13:43
sdagueno, that seems fine13:43
*** salv-orlando has joined #openstack-qa13:43
SergeyLukjanovsdague, in this case it requires check that sahara is available before creating client itself13:43
sdaguesure13:44
SergeyLukjanovsdague, but it sounds fine for me that client could fail if service isn't available13:44
sdagueyep, that seems fine. It will throw an exception right? And we'll just fail from there13:45
SergeyLukjanovsdague, yup13:46
SergeyLukjanovsdague, do you remember that I asked for your help several weeks ago re this patch?13:46
SergeyLukjanovsdague, it was the issue that error from sahara client was magically wrapped with ImportError and failed like error in running testr13:47
openstackrecheckConsole logs not available after 13:27s for check-tempest-dsvm-postgres-full 83520,11,073ae8613:49
sdagueSergeyLukjanov: ah...13:50
sdagueso can we wrap the import and throw a skip exception explictly?13:51
*** matsuhashi has joined #openstack-qa13:52
*** yfried__ has joined #openstack-qa13:54
*** matsuhashi has quit IRC13:54
*** Aimon_ is now known as Aimon13:54
*** psharma has quit IRC13:56
*** enikanorov_ has joined #openstack-qa13:57
*** mestery_ has joined #openstack-qa13:58
SergeyLukjanovsdague, it's an ImportError itself13:58
*** mestery has quit IRC13:58
SergeyLukjanovsdague, "service not found" exception from sahara client was wrapped somehow13:59
*** enikanorov has quit IRC13:59
*** mlavalle has joined #openstack-qa13:59
*** mlavalle has quit IRC14:00
SergeyLukjanovsdague, personally, I prefer to rely on service_available.sahara because if endpoint isn't available than something goes wrong14:00
*** mlavalle has joined #openstack-qa14:01
openstackrecheckConsole logs not available after 13:47s for check-tempest-dsvm-postgres-full 83370,6,3af957814:03
*** malini_afk is now known as malini14:03
openstackrecheckopenstack/tempest change: https://review.openstack.org/87013 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1295876, gate-grenade-dsvm-neutron: unrecognized error14:03
*** jgrimm has joined #openstack-qa14:06
*** rwsu has joined #openstack-qa14:09
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: added query for neutron migrations issue  https://review.openstack.org/8727414:12
sdaguemriedem: got a quick sec for that?14:12
mriedemsdague: yeah if my system doesn't lock up again this morning for the 5th time14:14
openstackgerritSergey Lukjanov proposed a change to openstack/tempest: Add Sahara client for scenarios tests  https://review.openstack.org/8331714:14
mriedemand if these corporate IT security guys get off my back about updating openssl14:14
SergeyLukjanovsdague, could you please restore your +2? (it's just a rebase)14:14
sdaguemriedem: are you using windows?14:14
mriedemsdague: yeah14:14
mriedemsdague: firefox decided to shit the bed14:15
*** kgriffs is now known as kgriffs|afk14:15
openstackrecheckConsole logs not available after 13:23s for check-tempest-dsvm-full 84454,4,35a06cd14:16
mriedemsdague: looks good, only comment would be you could filter on the build_name also14:17
*** om has quit IRC14:17
mriedemAND (build_name:"check_grenade_dvsm_neutron" OR build_name:"gate_grenade_dvsm_neutron")14:18
*** beekneemech is now known as bnemec14:18
mriedemoops, s/_/-/14:19
sdaguesure14:20
mriedems/dvsm/dsvm/14:20
mriedemdamn14:20
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: added query for neutron migrations issue  https://review.openstack.org/8727414:20
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: added query for neutron migrations issue  https://review.openstack.org/8727414:21
sdagueyeh, I make that mistake all the time14:21
mriedemi need more coffee14:21
mriedem+A14:21
sdagueI'm done on more coffee for this morning14:21
*** otherwiseguy has joined #openstack-qa14:22
mriedemnow to just get these heartbleed guys off my back14:22
mriedem"you have to update!" "no i don't your scan tool sux"14:22
mtreinishmriedem: just hack the scan tool to say your in compliance14:23
*** kgriffs|afk is now known as kgriffs14:23
mriedemmtreinish: meh14:23
mtreinishmriedem: oh I missed the windows part sorry14:23
mriedemi already did the run around with 2 lab guys in POK14:23
mriedemmtreinish: unrelated14:23
mriedemmy laptop is windows, my lab servers are RHEL 6.414:23
openstackgerritA change was merged to openstack-infra/elastic-recheck: added query for neutron migrations issue  https://review.openstack.org/8727414:28
*** openstackrecheck has quit IRC14:30
*** openstackrecheck has joined #openstack-qa14:30
*** mestery_ is now known as mestery14:32
*** dims has quit IRC14:35
*** kgriffs is now known as kgriffs_66614:39
*** sthaha has quit IRC14:39
*** kgriffs_666 is now known as kgriffs14:40
*** dkorn has quit IRC14:41
*** dims has joined #openstack-qa14:42
*** otherwiseguy has quit IRC14:43
malinidkranz: ping14:43
dkranzmalini: Hi14:44
malinidkranz: how are you?14:44
openstackgerritA change was merged to openstack-dev/devstack: Fix typo in nova_admin_username  https://review.openstack.org/8548414:44
dkranzmalini: pretty well. What's up?14:45
malinidkranz: I see you have a session suggestion in https://etherpad.openstack.org/p/Juno-QA-design-summit-topics14:45
dkranzmalini: Yes14:45
maliniI created a summit session a while back http://summit.openstack.org/cfp/details/134 ..Its along the same lines14:45
dkranzmalini: Yes, that is the same though I did not see it.14:46
malinidkranz: I missed adding it to etherpad14:46
maliniI was wondering if its ok to add this one in the session link to your etherpad entry14:47
dkranzmalini: Sure14:47
*** amotoki has joined #openstack-qa14:47
dkranzmalini: I think there was a little mixup between the etherpad and the "official" submission place14:47
*** HenryG has joined #openstack-qa14:48
malinidkranz: awesome.I just updated etherpad with the session link14:48
maliniWe can update the session details as our thoughts evolve14:48
openstackgerritA change was merged to openstack/tempest: Add list roles api to identity v3  https://review.openstack.org/7614314:51
yfried__anyone knows how to get test logs for successful runs? I want to compare failed tests to succesfull ones running on the same setup, but tempest.log isn't devided by test, and test logs are only printed for failed tests14:53
*** Guest38269 is now known as jeblair14:57
*** jeblair is now known as Guest3464214:57
*** mwagner_lap has joined #openstack-qa14:58
*** Guest34642 is now known as corvus14:58
openstackrecheckConsole logs not available after 13:30s for gate-trove-python27 63789,22,be2ca7515:02
openstackrecheckopenstack/nova change: https://review.openstack.org/85567 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1253896, gate-grenade-dsvm-neutron: unrecognized error15:03
*** chandan_kumar has quit IRC15:11
jordanPdtroyer, hi/ https://review.openstack.org/#/c/87140/ was commited recently "Prevent the use of 'swift' in ENABLED_SERVICES"15:12
jordanPdtroyer, but now I can't use swift3 middleware...15:12
*** jpich has quit IRC15:12
*** dims has quit IRC15:13
*** otherwiseguy has joined #openstack-qa15:16
*** jpich has joined #openstack-qa15:19
*** om has joined #openstack-qa15:21
*** kmartin has joined #openstack-qa15:24
*** zul has quit IRC15:24
*** devvesa has quit IRC15:25
*** zul has joined #openstack-qa15:25
*** enikanorov has joined #openstack-qa15:27
*** chuck_ has joined #openstack-qa15:31
openstackrecheckConsole logs not available after 13:35s for gate-python-cinderclient-pypy 69898,3,286180b15:31
*** zul has quit IRC15:31
*** chuck_ is now known as zul15:32
mriedemsdague: bash8 doesn't support # noqa?15:33
openstackrecheckopenstack/neutron change: https://review.openstack.org/85585 failed because of: gate-grenade-dsvm-neutron: unrecognized error, gate-tempest-dsvm-full: https://bugs.launchpad.net/bugs/130277415:34
*** atiwari has joined #openstack-qa15:36
*** dims has joined #openstack-qa15:39
*** enikanorov_ has quit IRC15:45
*** mmaglana has joined #openstack-qa15:48
*** asselin has joined #openstack-qa15:49
dkranzmtreinish: In your last comment about non-admin with regard to configuration, are you talking about essentially having a service tag for admin?15:51
*** dims has quit IRC15:52
mtreinishdkranz: we already basically have it because the tests are named admin15:53
mtreinishbut yes if that doesn't cover everything15:53
mtreinishwe add an attr for it15:53
dkranzmtreinish: ok, it does not cover everything. I think it would be cleaner to just put the attr on the class even if there was Admin in the name.15:54
dkranzmtreinish: I will fix up that spec and resubmit.15:55
*** zul has quit IRC16:02
*** zul has joined #openstack-qa16:04
*** zul has quit IRC16:04
sdaguemriedem: no16:04
sdagueit's pretty basic16:04
*** dims has joined #openstack-qa16:05
mriedemyeah :) you provide something that's better than nothing and now i want it to do everything!16:05
sdagueheh16:05
mriedem"patches welcome"16:05
mtreinishmriedem: nothing is stopping to from doing that16:05
mriedemyeah yeah16:05
*** qba73_ has quit IRC16:06
sdagueyeh, honestly, if I get a few moments pre summit I'll finally split it off16:06
mriedemfirst i have to see if we have a regression in nova resizes16:06
openstackgerritMark Maglana proposed a change to openstack/tempest: Make sure to release the floating ip afterwards  https://review.openstack.org/7915316:06
*** mmaglana has quit IRC16:07
*** mmaglana has joined #openstack-qa16:07
*** zul has joined #openstack-qa16:08
*** mmaglana has quit IRC16:12
*** dims has quit IRC16:13
*** jaypipes has joined #openstack-qa16:16
*** ChanServ sets mode: +o jaypipes16:16
*** Aimon has quit IRC16:16
mriedemmtreinish: we don't have any api tests that test server resize from shutoff state do we? i'm not seeing any16:24
*** jpich has quit IRC16:24
mriedemnope, i need to add one16:24
*** dims has joined #openstack-qa16:27
jaypipesmkoderer: hey! want to chat about https://bugs.launchpad.net/nova/+bug/130660216:28
jaypipesmkoderer: I'm a bit confused as to why this is not expected behaviour...16:28
mkodererjaypipes: hi16:29
mkodererjaypipes: yep, so if you specify the default AZ in the cli it behaves different if you don't16:30
jaypipesmkoderer: that is correct behaviour, no?16:30
jaypipesAZ=None != AZ=default16:31
mkodererjaypipes: is it? why.. ;)16:31
mkodererjaypipes: so meaning AZ=None matches all nodes is correct?16:31
jaypipesmkoderer: well, I'm interpreting AZ=None as "any AZ", and interpreting AZ=default as "choose the AZ that is the default for the endpoint that the CLI is connecting to"16:32
jaypipesmkoderer: but this is one of those areas that is very poorly tested to date :( So, perhaps it is worth a ML post about it.16:32
*** chandan_kumar has joined #openstack-qa16:33
mkodererjaypipes: I interpret it as "all hosts are in the default AZ as long there are not assigned to another one".16:33
mkodererjaypipes: question if there is a case where you want to boot a vm and you don't care about the AZ16:34
mkodererjaypipes: but you're right.. let's take it to the ML16:34
jaypipesmkoderer: yes, agree with you on the above.16:34
jaypipesre: host is in the default AZ (i.e. the availablity_zone option in nova.conf) if it is not specified to be in a host aggregate that is tagged with an availability zone16:35
jaypipes(this is, BTW, one of the reasons why I think hacking host aggregates with the concept of an availability zone was A Bad Idea ;)16:36
jaypipesmkoderer: regardless of result of ML discussion, looks like there's a couple unit test issues to fix on https://review.openstack.org/#/c/87247/16:37
mkodererjaypipes: yep I think the design is not really clear in that area16:37
mkodererjaypipes: Yep.. I will fix them tomorrow... ;)16:37
jaypipescoolio.16:37
jaypipesmkoderer: I'll write a ML post, ccing you.16:38
mkodererjaypipes: thx!16:38
jaypipesnp16:38
*** om has quit IRC16:38
openstackgerritMatt Riedemann proposed a change to openstack/tempest: WIP: Add API test for resizing a stopped server  https://review.openstack.org/8731216:39
*** yassine has quit IRC16:40
*** Aimon_ has joined #openstack-qa16:42
*** mmaglana has joined #openstack-qa16:43
openstackrecheckConsole logs not available after 13:29s for check-grenade-dsvm 57715,13,dff8aaa16:43
openstackrecheckopenstack/horizon change: https://review.openstack.org/86505 failed gate-tempest-dsvm-neutron-heat-slow, gate-grenade-dsvm-neutron in the gate queue with an unrecognized error16:43
*** jordanP has quit IRC16:44
*** akamyshnikova has quit IRC16:46
openstackgerritMatt Riedemann proposed a change to openstack/tempest: WIP: Add API test for resizing a stopped server  https://review.openstack.org/8731216:46
*** harlowja_away is now known as harlowja16:54
*** afazekas has quit IRC16:55
*** fbo is now known as fbo_away16:56
openstackgerritA change was merged to openstack-infra/devstack-gate: normalize the services for a post grizzly world  https://review.openstack.org/8428716:57
*** akamyshnikova has joined #openstack-qa16:59
openstackrecheckopenstack/nova change: https://review.openstack.org/76029 failed because of: gate-grenade-dsvm-neutron: unrecognized error, gate-tempest-dsvm-neutron-heat-slow: https://bugs.launchpad.net/bugs/129587617:03
*** nati_ueno has joined #openstack-qa17:04
mtreinishmriedem: nope I don't think that we do17:10
mriedemmtreinish: we will if this passes https://review.openstack.org/#/c/87312/17:10
mriedemthen i'll add the same for the v3 api17:11
*** corvus is now known as jeblair17:11
*** mrodden has quit IRC17:12
mkodererjaypipes: what came in my mind while driving home ;) Why should you need a "default AZ" if this is not default in the CLI17:20
jaypipesmkoderer: because the default AZ in the CLI is whatever Nova endpoint you are hitting, whereas the default AZ for a host is whatever is in nova.conf on that host, yes?17:22
*** mrodden has joined #openstack-qa17:24
mkodererjaypipes: isn't the default AZ in nova.conf deprecated.. not sure about that, need to check the code17:26
jaypipesmkoderer: I didn't think so...17:26
mkodererjaypipes: ah no this was cinder17:31
mkodererjaypipes: anyway.. so yes the default AZ is the one defined in nova.conf in the API endpoint17:33
jaypipesyeah17:33
mkodererjaypipes: for me if a default is set I don't have to specify it on the cli17:34
jaypipesmkoderer: right, but what if you specify availability-zone=None? :)17:36
mkodererjaypipes: better the CLI wouldn't set availability-zone at all17:38
jaypipesagreed.17:38
mkodererjaypipes: so then it should take "default" AZ right? ;)17:39
jaypipesyeah, i think so.17:39
mkodererjaypipes: cool, let me fix it tomorrow in nova cli and nova17:40
dkranzchmouel: The swift client set up in OfficialClientManager needs admin creds but it doesn't seem like the swift scenario test does. Is it possible to run test_swift_basic_ops with admin creds?17:40
dkranzchmouel: without admin creds I mean17:40
dkranznotmyname: ^^^17:41
jaypipesmkoderer: sounds good.17:41
mkodererjaypipes: but anyway we can take it to the ML.. maybe there are some other thoughts around17:42
jaypipesmkoderer: frankly, I think the discussion may devolve into yet another debate about what is messed up with host aggregates and availability zones... not sure it would be all that useful at this point -- or add anything more than what has already been said about the topic :(17:44
notmynamedkranz: I have no idea (because I don't know what OfficialClientManager is or what admin means)17:44
dkranznotmyname: ok, who is the best person to ask about tempest/swift issues?17:44
shardyCan anyone advise the best way to run just one upgrade test in grenade?17:45
mkodererjaypipes: lol all right ;) pls just help me to get it merged :D17:45
shardyRunning grenade.sh sets up /opt/stack/havana, but then running upgrade-* individually looks for /opt/stack/icehouse which doesn't exist?17:45
mtreinishdkranz: it probably just needs permissions to make a container17:46
mtreinishwhich I believe we have a role config option for17:46
jaypipesmkoderer: will do my best :)17:46
mtreinishthe admin perms are probably just for tenant isolation17:46
dkranzmtreinish: Yeah, but the role needs to be added which is why it is making a swiftclient admin17:46
dkranzmtreinish: I will just turn it off for now so I can see what fails with no admin creds17:47
jaypipesshardy: dtroyer may be able to assist you on that.17:48
notmynamedkranz: my understanding is that "admin" from a keystone perspective means that you can modify keystone. is that still correct?17:48
dkranzmtreinish: Unfortunately I have not found a way to blank the admin creds, fix this client issue, and turn off tenant isolation with a change to a single project17:48
shardyjaypipes: thanks, the docs I've found are a bit light on details, I'm probably missing something obvious ;)17:48
dkranznotmyname: Yes, and each service has some apis that are admin-only17:48
dkranznotmyname: In this case it is calling keystone to add an operator role17:49
notmynamedkranz: if my understanding is correct, I don't think you should need keystone admin rights to do anything in swift17:49
dkranznotmyname: I will find out soon :)17:50
*** otherwiseguy has quit IRC17:50
notmynamedkranz: "it is calling keystone" <-- what is "it" referring to?17:50
dkranznotmyname: The code that is creating the swiftclient connection17:50
notmynamedkranz: in swiftclient or tempest code?17:50
dkranznotmyname: tempest17:51
notmynamedkranz: what is the operator role?17:51
dkranznotmyname: It would be easiest to point you at the code. Just a sec17:51
notmynamedkranz: line 574 of clients.py?17:51
dkranznotmyname: yes17:52
dkranznotmyname: 54717:52
notmynamewell, that says it's being added, but not what it is :-)17:52
*** mkollaro has quit IRC17:52
*** enikanorov_ has joined #openstack-qa17:52
dkranznotmyname: Right, and the comment says it is necassry so the test user can operate swift.17:53
dkranznotmyname: But I'm not sure why that is.17:53
dkranznotmyname: In any event I am trying to run without it and will see what happens. That will give more concrete info.17:53
mtreinishnotmyname: the operator role is just a user role in keystone that has permissions to create containers17:55
mtreinishnotmyname: it's set via a config option and it should match the keystoneauth middleware setting17:55
*** mkollaro has joined #openstack-qa17:57
notmynamemtreinish: looking at sample configs in the respective codebases, tempest has operator_role set to "Member", and swift has keystone middleware configured as "operator_roles= = admin, swiftoperator"17:59
openstackrecheckConsole logs not available after 13:41s for gate-nova-pep8 85804,6,3b38de918:00
*** marun has joined #openstack-qa18:00
mtreinishnotmyname: well that just comes from the default in tempest.config, I think Member is what devstack sets so that's probably where the default value came from18:01
mtreinishbut we can change it to be more consistent with the swift default18:01
notmynamemtreinish: is that a string match that has to be equal?18:02
mtreinishnotmyname: is what a string match? the sample config with the value in the config.py file?18:02
notmynamemtreinish: sorry. is "Member" a string that is compared to the keystone middleware settings or is is a reference to something else?18:03
*** catohornet has joined #openstack-qa18:04
notmynamemtreinish: I'm slowly understanding how the tempest/keystone parts are put together with swift18:04
mtreinishMember is just the default value we set for the tempest config option. The burden is really on the user to verify that it matches what's in the keystone middleware setting18:05
mtreinishif it's incorrect the swift tests that use it will just fail18:05
mtreinishI think it's only really used in the tenant isolation case (where each test class makes a keystone tenant and user) so we add the role to the new user18:06
mtreinishto ensure that the user for the swift tests have permissions to do what's needed18:06
notmynamemtreinish: what you are calling "operator role" is, in swift, the "owner" of the swift account. an identity (or rather, an auth token) with that role has the ability to anything they want inside of a swift account18:06
notmyname(including creating containers)18:07
*** nati_ueno has quit IRC18:08
mtreinishnotmyname: ok, that makes sense. Do you think that there is a better way we could be handling things here then?18:08
notmynamemtreinish: in which case, yes that operator role will be required for swift tests.18:08
mtreinishbecause I added the operator stuff in when I was trying to make the tests run with tenant isolation enabled18:09
*** nati_ueno has joined #openstack-qa18:09
mtreinishand I didn't dive to deeply into the swift-keystone user permission model18:09
notmynamemtreinish: from my (limited) understanding of keystone and tempest, it seems like the right thing is being done, but the wording or description isn't all that clear. also, I don't know if it's the best or right way to accomplish that goal in swift18:10
notmynamemtreinish: I too don't fully grok the swift-keystone permission model. it's actually set completely on the keystone side of things (swift has an auth interface usable by many auth systems, but doesn't provide semantic meaning to the values stored in the designated fields--thus my lack of understanding of keystone's particular semantics)18:12
notmynameoops..earlier statement should have said "...I don't know if it's the best or right way to accomplish that goal in keystone"18:12
openstackgerritSteven Hardy proposed a change to openstack-dev/grenade: Add initial support for Heat  https://review.openstack.org/8697818:13
mtreinishok, well we can definitely improving the description of the config option to make it clearer what it is18:13
mtreinishother than that I guess we can just leave it as good enough for now, if we need to make further refinements down the road we can tackle that as the need arises18:14
mtreinishdkranz: for your non-admin work non-tenant isolated stuff, you might just want to check that the user has the role set up front and error out if it doesn't18:15
*** morganfainberg_Z is now known as morganfainberg18:15
dkranzmtreinish: Yes. By coincidence the offending code was just proposed as stable/havana backport.18:15
dkranzmtreinish: I am going to see what happens without it18:16
mtreinishheh, yeah I think that I just approved that through recently18:16
dkranzmtreinish: IMO, tempest should not be doing this18:16
openstackgerritAdam Gandelman proposed a change to openstack/tempest: Consider state in firewall create/delete test  https://review.openstack.org/8562218:16
mtreinishdoing what?18:16
dkranzmtreinish: Creating some role that is needed to run a set of tests.18:17
dkranzmtreinish: It should be part of tempest setup/config18:17
mtreinishit is, tempest doesn't create the role18:17
mtreinishit just assigns create users the role18:17
mtreinishso they have permissions to create containers18:17
dkranzmtreinish: I misspoke but it is really the same thing if we want to be able to run as non-admin18:17
mtreinishdkranz: it's basically the same except that since it's limited to just swift we should error out if swift is enabled18:18
dkranzmtreinish: tempest should not try to do any admin thing that is not directly controlled by a flag or explicitly part of some test that needs admin18:18
mtreinishand the user doesn't have the proper permissions18:18
mtreinishdkranz: assigning the role should be limited to the tenant isolation code18:19
dkranzmtreinish: That is the fall back way but I will try to find something better.18:19
mtreinishwhich requires admin18:19
dkranzmtreinish: That would be fine with me.18:19
*** andreaf has quit IRC18:20
dkranzmtreinish: So if I comment out the role assignment, the only test that uses it still passes...18:24
*** lcostantino has joined #openstack-qa18:25
lcostantinohi, i was trying to deploy a cinder-volume host only with devstack, but apts/cinder seems to be missing python-mysqldb, etc as nova has. is that expected?18:26
*** jlibosva has quit IRC18:29
mtreinishdkranz: that's probably because devstack assigns the role the demo user when swift is enabled18:40
dkranzmtreinish: Another reason that code should not be there18:41
mtreinishwhich code are you talking about?18:41
dkranzmtreinish: The code in tempest that adds the role18:41
dkranzmtreinish: The heat tests require an admin role in the demo user but tempest does not try to add it.18:42
mtreinishdkranz: link?18:42
openstackgerritMatt Riedemann proposed a change to openstack/tempest: Add API test for resizing a stopped server  https://review.openstack.org/8731218:42
*** aysyd has quit IRC18:42
dkranzmtreinish: I will chase it down.18:43
dkranzmtreinish: from lib/keystone:  openstack role add --project $DEMO_TENANT --user $ADMIN_USER $ADMIN_ROLE18:47
mtreinishyeah that's the devstack part of it18:47
dkranzmtreinish: But I am just agreeing with what you said which was that such code should either be outside of tempest or in the isolated creds code.18:48
mtreinishdkranz: which I thought it was, where do we do it elsewhere?18:48
mtreinishthat's what I was asking18:48
dkranzmtreinish: In the OfficialClientManager code18:48
dkranzmtreinish: Sorry, you probably missed the initial start of this irc thread18:49
dkranzmtreinish: _get_object_storage_client18:49
mtreinishdkranz: yeah propose a patch to remove that try block it was just a hack because of neutron18:51
mtreinishwe may still need it on stable/havanna though because of tenant isolation issues with neutron18:51
dkranzmtreinish: I was objecting to the code inside the try block too.18:52
*** tosky has quit IRC18:52
dkranzmtreinish: How do you propose that we can run the swift scenario test as non-admin with that code there?18:52
dkranzmtreinish: You should not have to be an admin to use swift, right?18:53
mtreinishdkranz: no, the user just needs to have that role assigned18:53
mtreinishwhich you could either explicitly check for in the client18:53
mtreinishand raise an exception if it doesnt18:53
mtreinishor just let it fail down the line18:53
dkranzmtreinish: Yes, and devstack already does that so we could just move that code to isolated tenant18:53
dkranzmtreinish: Why should it fail if you are non-isolated and non-admin?18:54
mtreinishdkranz: that code is already in tenant isolation18:54
dkranzmtreinish: Ah, I didn't realize that. So we can get rid of the whole thing then (role assignment in that method)18:54
*** aysyd has joined #openstack-qa18:55
mtreinisherr, maybe. I think we actually do it in the swift baseclass if tenant isolation is enabled18:55
mtreinishbecause we didn't want it assigned everywhere18:55
mtreinishdkranz: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/api/object_storage/base.py#n3918:56
mtreinishwe need something like that for the scenario test18:56
*** catohornet has quit IRC18:56
mtreinishbut only if tenant_isolation18:56
dkranzmtreinish: I see. But that is not used by scenario.18:56
dkranzmtreinish: So I guess we just need to expand the try to include the call to get keystone admin creds18:57
mtreinishno it's not, and that's the crux of the confusion here I think. We need an equiv method using python-keystoneclient for the swift scenario test18:58
dkranzmtreinish: I think we have it, just not conditioned on non-tenant-isolation18:58
dkranzmtreinish: So that is the right fix.18:58
dkranzmtreinish: Only call that method if no isolation18:59
openstackgerritAdam Gandelman proposed a change to openstack-dev/devstack: Remove sshd reconfig from lib/ironic  https://review.openstack.org/8735518:59
mtreinishdkranz: I think you have that backwards. We should only assign the role to the user if tenant isolation is enabled18:59
dkranzmtreinish: Yes, that is what I meant.19:00
dkranzmtreinish: I'll push a fix after I understand why the get_network_client is *not* failing :)19:00
*** nati_ueno has quit IRC19:01
*** jhenner has quit IRC19:02
*** jhenner has joined #openstack-qa19:02
*** jhenner has quit IRC19:02
*** nati_uen_ has joined #openstack-qa19:02
openstackgerritMatthew Treinish proposed a change to openstack/tempest: Add support for updating the config file  https://review.openstack.org/8358519:05
*** aysyd has quit IRC19:10
mtreinishsdague: have you seen this before: http://logs.openstack.org/85/85085/2/gate/gate-tempest-dsvm-neutron/f828d37/console.html.gz19:12
mtreinishwhere devstack takes >20min to finish19:12
mtreinishI looked in the devstack log and I didn't see anything obvious19:12
mtreinishthere were some errors but logstash showed them mostly on successful runs19:13
*** aysyd has joined #openstack-qa19:16
openstackgerritA change was merged to openstack-dev/devstack: increase heat guests  https://review.openstack.org/8507819:21
*** chandan_kumar has quit IRC19:22
*** enikanorov_ has quit IRC19:26
*** enikanorov_ has joined #openstack-qa19:27
openstackrecheckConsole logs not available after 14:25s for gate-horizon-python27 76133,5,d24f61f19:27
sdaguemtreinish: looks like we're building kernel modules - http://logs.openstack.org/85/85085/2/gate/gate-tempest-dsvm-neutron/f828d37/logs/devstacklog.txt.gz#_2014-04-11_23_09_16_10719:31
*** lcostantino has quit IRC19:31
sdaguealso, there are just some super slow nodes in rax19:32
sdaguewe probably hit one19:32
mtreinishahh ok just recheck no bug then?19:32
mtreinishor do we have a generic timeout one19:32
openstackrecheckopenstack/nova change: https://review.openstack.org/76029 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1273908, gate-grenade-dsvm-neutron: unrecognized error19:32
mtreinishthe kernel module build only took a little over a min though19:33
*** malini is now known as malini_afk19:38
dkranzmtreinish: One more thing. I am going to change the proposal to say that admin-required tests should be tagged as such and attempts to do admin things without admin creds will just fail.19:39
dkranzmtreinish: That is, no skipping of admin tests if there are no admin creds. Do you have a problem with that?19:40
dkranzmtreinish: So you run non-admin tests by setting isolated to false, admin creds to none, and selecting the regexp to not include admin tests19:40
dkranzmtreinish: I don't see the value in implementing skipping everywhere19:41
mtreinishdkranz: honestly the best way to do it is probably the helper method/decorator approach19:41
mtreinishlike what I did with the service tags19:41
mtreinishthe helper will check that admin creds are set to None19:42
openstackgerritDirk Mueller proposed a change to openstack/tempest: Fix occurrences of F812  https://review.openstack.org/8736619:42
mtreinishthen you can skip based on that19:42
*** lcostantino has joined #openstack-qa19:42
mtreinishthe decorator will also add the attr besides skipping19:42
mtreinishbecause if admin creds aren't set we really shouldn't try to run the admin tests19:42
dkranzmtreinish: ok19:43
openstackgerritRob Crittenden proposed a change to openstack/tempest: botoclient does not honor disable_ssl_certificate_validation  https://review.openstack.org/8736719:43
mtreinishdkranz: for the tenant isolation stuff we probably should raise an Invalid configuration exception if tenant isolation is set without admin creds19:44
mtreinishor do you think failing through is better there? I really don't have a preference19:44
dkranzmtreinish: I was going to fail it19:44
dkranzmtreinish: Because it is going to fail anyway just in a less obvious way19:45
*** nati_uen_ has quit IRC19:47
*** otherwiseguy has joined #openstack-qa19:48
*** fbo_away is now known as fbo19:59
*** otherwiseguy has quit IRC19:59
*** lcostantino has left #openstack-qa20:03
openstackgerritDavid Kranz proposed a change to openstack/qa-specs: Spec for running tempest as non-admin  https://review.openstack.org/8696720:04
sdaguemtreinish: so the real issue is that job is still not parallel, and it landed on a slow node20:19
sdaguethe ec2 test took 4 minutes20:19
sdaguefor instance20:19
*** mwagner_lap has quit IRC20:22
openstackgerritAdam Gandelman proposed a change to openstack-dev/devstack: Insert required iptables rules for Ironic  https://review.openstack.org/8737720:24
openstackgerritAdam Gandelman proposed a change to openstack-dev/devstack: Insert required iptables rules for Ironic  https://review.openstack.org/8737720:25
*** otherwiseguy has joined #openstack-qa20:28
*** dkranz has quit IRC20:32
*** nati_ueno has joined #openstack-qa20:32
openstackrecheckopenstack/nova change: https://review.openstack.org/79618 failed because of: gate-grenade-dsvm-neutron: unrecognized error, gate-tempest-dsvm-neutron-heat-slow: https://bugs.launchpad.net/bugs/129756020:33
*** Aimon_ is now known as Aimon20:42
*** yamahata has joined #openstack-qa20:47
*** nati_ueno has quit IRC20:48
*** nati_uen_ has joined #openstack-qa20:52
*** nati_uen_ has quit IRC20:56
*** jlibosva has joined #openstack-qa20:56
*** Sukhdev has joined #openstack-qa20:57
mtreinishsdague: ok I'll just no bug it then, since it was just a slow node20:59
*** fbo is now known as fbo_away20:59
*** markmcclain has joined #openstack-qa21:00
*** nati_ueno has joined #openstack-qa21:00
*** nati_ueno has quit IRC21:01
*** nati_ueno has joined #openstack-qa21:02
sdaguemtreinish: honestly, we should probably open an issue to track it21:05
sdagueso we can see how often we're getting hit by it21:05
openstackrecheckConsole logs not available after 13:27s for gate-glance-pep8 82899,4,e89980621:16
openstackgerritSteven Hardy proposed a change to openstack-dev/grenade: Add initial support for Heat  https://review.openstack.org/8697821:19
mtreinishsdague: with what fingerprint just the timeout one, because that seems a bit too generic21:32
sdaguetimeout is fine21:33
sdaguewell, timeout + job21:33
sdaguethat will let us track jobs that are hitting a threshold regularly21:33
mtreinishok I'll add a bug and fingerprint for that later tonight21:34
sdaguecool, thanks21:36
*** giulivo has quit IRC21:44
openstackrecheckopenstack/tempest change: https://review.openstack.org/87173 failed because of: gate-grenade-dsvm-neutron: unrecognized error, gate-tempest-dsvm-neutron-heat-slow: https://bugs.launchpad.net/bugs/129756021:45
openstackrecheckopenstack/tempest change: https://review.openstack.org/87013 failed because of: gate-grenade-dsvm-neutron: unrecognized error, gate-tempest-dsvm-neutron-heat-slow: https://bugs.launchpad.net/bugs/129756021:49
*** chuck_ has joined #openstack-qa21:50
*** zul has quit IRC21:51
*** mriedem has quit IRC21:53
*** chuck_ has quit IRC21:53
*** zul has joined #openstack-qa21:53
*** Sukhdev_ has joined #openstack-qa21:55
*** zul has quit IRC21:57
*** Sukhdev__ has joined #openstack-qa21:57
*** Sukhdev_ has quit IRC21:57
*** dims has quit IRC21:57
*** Sukhdev has quit IRC21:58
*** markmcclain has quit IRC22:01
*** jlibosva has quit IRC22:02
*** om has joined #openstack-qa22:04
*** dims has joined #openstack-qa22:10
*** Sukhdev__ has quit IRC22:14
*** zul has joined #openstack-qa22:17
*** zul has quit IRC22:21
*** zul has joined #openstack-qa22:21
*** zul has quit IRC22:24
*** chuck_ has joined #openstack-qa22:25
*** om has quit IRC22:28
*** rbak has joined #openstack-qa22:28
*** mfisch has joined #openstack-qa22:30
openstackrecheckopenstack/nova change: https://review.openstack.org/79619 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1226412 https://bugs.launchpad.net/bugs/1254872, gate-grenade-dsvm-neutron: unrecognized error22:30
*** marun has quit IRC22:31
*** chuck_ is now known as zul22:34
*** otherwiseguy has quit IRC22:38
*** adalbas has quit IRC22:42
*** mrodden has quit IRC22:45
openstackgerritPaul Van Eck proposed a change to openstack/tempest: Add tests for nova floating IP Heat resources  https://review.openstack.org/8228222:46
*** nati_ueno has quit IRC22:49
*** nati_ueno has joined #openstack-qa22:49
*** dims has quit IRC22:57
*** adalbas has joined #openstack-qa22:57
*** rbak has quit IRC22:59
*** yamahata has quit IRC23:01
*** mmaglana has quit IRC23:03
*** jamielennox|away is now known as jamielennox23:06
*** mrodden1 has joined #openstack-qa23:06
*** mwagner_lap has joined #openstack-qa23:11
*** mkollaro has quit IRC23:12
openstackgerritAdam Gandelman proposed a change to openstack-dev/devstack: Add sudoers.d entry for Ironic conductor  https://review.openstack.org/8740823:15
openstackgerritAdam Gandelman proposed a change to openstack-dev/devstack: Insert required iptables rules for Ironic  https://review.openstack.org/8737723:15
*** nati_ueno has quit IRC23:21
*** dims has joined #openstack-qa23:22
*** nati_ueno has joined #openstack-qa23:22
*** mrodden has joined #openstack-qa23:24
*** atiwari has quit IRC23:26
*** mrodden1 has quit IRC23:27
*** nati_ueno has quit IRC23:40
openstackgerritGhanshyam Mann proposed a change to openstack/tempest: Adds test for volume attachments Nova V2 APIs  https://review.openstack.org/8719123:41
*** nati_ueno has joined #openstack-qa23:42
*** ianw has joined #openstack-qa23:47
*** yamahata has joined #openstack-qa23:47
*** yamahata has quit IRC23:56

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