Tuesday, 2014-11-25

*** dims has quit IRC00:03
*** dims has joined #openstack-qa00:03
*** gordc has quit IRC00:05
*** dims has quit IRC00:08
*** toure has quit IRC00:08
*** david-lyle is now known as david-lyle_afk00:17
*** rushiagr_away is now known as rushiagr00:17
*** mlavalle has quit IRC00:17
*** dims has joined #openstack-qa00:21
*** markmcclain has quit IRC00:23
JayFI have https://review.openstack.org/#/c/136867/2/lib/ironic,cm tested working locally, would love to get it in to help troubleshoot any agent_ssh failures in devstack00:24
*** jhenner has quit IRC00:24
JayFOn another note; has there ever been investigation into making the ironic-bm-logs/baremetalbrbm_*_console.log files not clear on reboot?00:24
JayFbecause if there was an issue imaging an instance and it got far enough to reboot; it'd completely kill any logs from the Ironic Python Agent00:25
JayFadam_g: ^00:25
adam_gJayF, yeah actually i started workig on a patch set to do that00:26
JayFadam_g: you see my change? (136867 linked above)00:26
JayFadam_g: makes anything from inside the ramdisk log to console; I verified locally it worked and it looked great00:26
JayFadam_g: not the easiest logs to parse because it basically funnels everything through kmsg, but it's better than no logs and is a very small change00:27
adam_gJayF, cool00:27
adam_gJayF, so i started this a while ago, i need to dust it off https://review.openstack.org/#/c/129099/00:27
*** rushiagr is now known as rushiagr_away00:28
JayFdtroyer: if you have a chance, https://review.openstack.org/#/c/136867 is awaiting a +2A and you were reccomended to me as a friendly person who is kind and reviews with ferocity :P00:28
*** markmcclain has joined #openstack-qa00:28
adam_gJayF, works fine if youre using screen (new screen tabs to monitor each log) but not so much when screen is disabled. need a good way to launch that procescs in the bg00:28
JayFwait, those logs are in screen somewhere?00:29
* JayF looks00:29
JayFadam_g: not /yet/ you mean, just in that change?00:29
adam_gJayF, no, with that patch--it'll spawn processes in screen, to monitor the console log and preserve stuff that'd be truncated00:29
JayFwon't that fix the gate issue then?00:30
adam_gJayF, but we dont use screen in the gate jobs, so its kinda useless00:30
JayFadam_g: there are logs *labelled* screen in teh gate?00:30
*** dims has quit IRC00:31
adam_gJayF, yeah--i dont know why those are labeled like that, they run USE_SCREEN=False00:31
*** dims has joined #openstack-qa00:31
JayFWould be nice to see if clarkb or those guys had any advice on that patch then00:32
clarkbbecause they were run with screen for like 2 years00:32
clarkbI would not rename them to rename them00:32
JayFbecause that's going to pretty much required to figure out when IPA starts breaking jobs00:32
JayFwhich will happen, eventually, in the IPA pipeline only (hopefully)00:32
clarkbit would be something to do as part of the larger break the world service renaming00:32
adam_gclarkb, makes sense00:33
JayFany advice for me or adam_g on persisting those ironic-bm-logs then?00:33
*** melwitt has quit IRC00:33
clarkbcan't you do it like all the other?00:34
adam_gJayF, we just need to spawn a background process to 'wait for console log to exist && start tailing console log to a new persistent log file'  i need to play with that patch more, spawning that with the facilities offered by devstack is tricky00:34
adam_gclarkb, we setup libvirt to log VM console output to a file, but that file is truncated every time theres a power cycle, so we lose lots context unless the last test to run failed00:35
JayFadam_g: or even if the last test failed; if the provision got far enough for ironic to reboot the node the data is gone00:36
clarkblibvirt doesn't have an append mode?00:36
*** dmorita has joined #openstack-qa00:36
adam_gJayF, yeah00:36
*** dims has quit IRC00:36
adam_gclarkb, not that i know of ?00:36
*** mpavlase has quit IRC00:40
*** david-lyle_afk has quit IRC00:45
gmanncoolsvap: ping00:55
*** toure has joined #openstack-qa01:02
*** salv-orlando has quit IRC01:04
*** marun has joined #openstack-qa01:08
*** ranger81 has quit IRC01:08
*** ranger81 has joined #openstack-qa01:09
*** catohornet has joined #openstack-qa01:10
*** topshare has joined #openstack-qa01:11
*** adalbas has quit IRC01:13
*** wuhg has joined #openstack-qa01:14
*** jecarey has joined #openstack-qa01:23
*** weshay has joined #openstack-qa01:34
*** yamamoto has quit IRC01:36
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Set up ssh_known_host based on hostname  https://review.openstack.org/13659601:38
jogomtreinish: I think you had the right idea, just used tr in the wrong place ^01:38
jogoI tried echo foo `remote_command chatty hostname | tr -d '\r'` locally and it looks better01:39
*** dims has joined #openstack-qa01:42
*** toure has quit IRC01:43
*** toure has joined #openstack-qa01:46
*** achanda has quit IRC01:50
*** ranger81 has quit IRC01:51
*** weshay has quit IRC01:54
*** ranger81_ has joined #openstack-qa01:55
*** catohornet has quit IRC01:59
*** dims has quit IRC02:00
*** dims has joined #openstack-qa02:00
*** nosnos has joined #openstack-qa02:01
*** dims has quit IRC02:05
*** jecarey has quit IRC02:08
openstackgerritAdam Gandelman proposed openstack-dev/devstack: Spawn console log monitors for Ironic VM nodes  https://review.openstack.org/12909902:09
adam_gJayF, that might work ^02:09
*** yamamoto has joined #openstack-qa02:09
*** shashankhegde has quit IRC02:13
openstackgerrithuangtianhua proposed openstack/tempest: Take InstanceType for launchconfig res creation  https://review.openstack.org/13551902:17
*** catohornet has joined #openstack-qa02:18
*** dkranz has joined #openstack-qa02:28
*** chandankumar has joined #openstack-qa02:32
*** patrickeast has quit IRC02:40
*** dkranz has quit IRC02:40
*** esmute has quit IRC02:44
*** catohornet has quit IRC02:45
*** mrodden has quit IRC02:46
*** jyuso has quit IRC02:53
*** mrodden has joined #openstack-qa02:56
*** yamahata has joined #openstack-qa02:57
*** chandankumar has quit IRC03:02
*** chandankumar has joined #openstack-qa03:03
*** ranger81_ has quit IRC03:09
*** JayJ has joined #openstack-qa03:28
*** chandankumar has quit IRC03:28
*** nosnos has quit IRC03:29
*** JayJ has quit IRC03:33
openstackrecheckConsole logs not available after 14:23s for check-tempest-dsvm-neutron-full 136456,4,393906203:33
*** bkopilov has quit IRC03:38
*** dpaterson has quit IRC03:38
*** markmcclain has quit IRC03:42
*** tiantian has joined #openstack-qa03:43
tiantiancyeoh: hi03:43
*** tiantian is now known as huangtianhua03:46
huangtianhuacyeoh: hi, would you please to review the patch https://review.openstack.org/#/c/135519/03:47
*** nosnos has joined #openstack-qa04:03
openstackgerritJamie Lennox proposed openstack-dev/devstack: Use new cafile option in nova cinder options  https://review.openstack.org/13696004:06
openstackgerritJamie Lennox proposed openstack-dev/devstack: Use the service catalog when talking to cinder  https://review.openstack.org/13696104:06
openstackrecheckConsole logs not available after 13:33s for check-grenade-dsvm 134347,4,5bd0bac04:10
*** shashankhegde has joined #openstack-qa04:13
*** tochi has joined #openstack-qa04:15
openstackrecheckopenstack/nova change: https://review.openstack.org/131490 failed because of: gate-grenade-dsvm-partial-ncpu: https://bugs.launchpad.net/bugs/137351304:16
jogomtreinish: woot it worked!04:19
jogomtreinish: tr FTW04:20
*** OmPrakashPandey has joined #openstack-qa04:35
*** oomichi has quit IRC04:38
*** toure has quit IRC04:38
*** Madan has joined #openstack-qa04:40
*** subscope_ has joined #openstack-qa04:55
openstackgerritMerged openstack/tempest: Add Conflict exception in test_floating_ips_actions case  https://review.openstack.org/13136404:58
openstackgerritMerged openstack/tempest: Update documentation for clarity  https://review.openstack.org/13573704:59
*** esmute has joined #openstack-qa05:04
openstackgerritMerged openstack/qa-specs: Add spec for clients returning one value  https://review.openstack.org/12600405:11
*** jamielennox has quit IRC05:11
*** jamielennox has joined #openstack-qa05:11
*** anshul has joined #openstack-qa05:13
*** bkopilov has joined #openstack-qa05:14
*** ranger81 has joined #openstack-qa05:18
*** chandankumar has joined #openstack-qa05:25
*** subscope_ has quit IRC05:25
openstackgerritMerged openstack-infra/subunit2sql: Flush out the db api documenation  https://review.openstack.org/12940505:27
openstackgerritMerged openstack-infra/subunit2sql: Reorganize the python api documentation  https://review.openstack.org/12940605:27
openstackgerritMerged openstack-infra/subunit2sql: Add section on the basic data model  https://review.openstack.org/12940705:29
*** nosnos has quit IRC05:30
*** nosnos_ has joined #openstack-qa05:30
*** oomichi has joined #openstack-qa05:32
*** ranger81 has quit IRC05:41
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: pulling in unmerged devstack patch  https://review.openstack.org/13650405:53
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Set up ssh_known_host based on hostname  https://review.openstack.org/13659605:53
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Setup /etc/hosts for aiopcpu  https://review.openstack.org/13615805:53
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT COMMIT disable config drive  https://review.openstack.org/13583105:53
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13463905:53
openstackrecheckopenstack/tempest change: https://review.openstack.org/134889 failed gate-grenade-dsvm-partial-ncpu in the gate queue with an unrecognized error05:56
chandankumargmann, good morning :)05:57
*** shashankhegde has quit IRC05:58
gmannchandankumar: gm06:01
*** shashankhegde has joined #openstack-qa06:02
chandankumargmann, http://logs.openstack.org/89/134889/4/gate/gate-grenade-dsvm-partial-ncpu/d5a710d06:03
chandankumargmann, the jenkin gate is failing, i have added a recheck.06:03
gmannchandankumar: i did not find any failure related to patch changes.06:07
gmannlet's see recheck results06:07
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13650406:10
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Set up ssh_known_host based on hostname  https://review.openstack.org/13659606:10
*** shashankhegde has quit IRC06:10
*** shashankhegde has joined #openstack-qa06:14
*** rushiagr_away is now known as rushiagr06:19
*** jyuso has joined #openstack-qa06:23
*** k4n0 has joined #openstack-qa06:24
*** vigneshvar has joined #openstack-qa06:25
*** vigneshvar has quit IRC06:33
*** nosnos_ has quit IRC06:34
*** nosnos has joined #openstack-qa06:34
openstackrecheckopenstack/glance change: https://review.openstack.org/132113 failed because of: gate-tempest-dsvm-full: https://bugs.launchpad.net/bugs/1373513 https://bugs.launchpad.net/bugs/1348204 https://bugs.launchpad.net/bugs/135249806:35
*** neeti has joined #openstack-qa06:35
*** nmagnezi has joined #openstack-qa06:39
*** shashankhegde has quit IRC06:40
*** Longgeek has joined #openstack-qa06:42
*** Longgeek has quit IRC06:42
*** Longgeek has joined #openstack-qa06:42
*** shashankhegde has joined #openstack-qa06:45
*** shashankhegde has quit IRC06:47
*** SridharG has joined #openstack-qa06:48
*** harlowja is now known as harlowja_away07:01
*** Longgeek has quit IRC07:01
*** Longgeek has joined #openstack-qa07:02
*** Longgeek has quit IRC07:06
*** achanda has joined #openstack-qa07:08
openstackgerritSergey Shnaidman proposed openstack/tempest: Test delete of SLAAC subnet without port delete  https://review.openstack.org/13690907:11
*** Longgeek has joined #openstack-qa07:14
*** yfried has joined #openstack-qa07:24
yfriedsergsh: about https://review.openstack.org/#/c/136789/07:28
chandankumargmann, it is still failing and 4 jobs are not able to run: https://review.openstack.org/13488907:28
chandankumarplease have a look07:28
yfriedsergsh: I'd like to help you run it, cause I'd like more practical eyes on that test07:28
*** achanda has quit IRC07:35
*** e0ne has joined #openstack-qa07:40
*** achanda has joined #openstack-qa07:43
openstackgerritYair Fried proposed openstack/tempest: Cleanup class resources for large-ops scenario  https://review.openstack.org/13698107:44
*** achanda has quit IRC07:46
*** achanda has joined #openstack-qa07:48
*** jamielennox is now known as jamielennox|away07:49
*** obondarev_ has joined #openstack-qa07:51
*** e0ne has quit IRC07:53
*** achanda has quit IRC07:53
*** pkoniszewski has joined #openstack-qa07:57
*** e0ne has joined #openstack-qa07:57
*** pkonisze has joined #openstack-qa07:58
*** achanda has joined #openstack-qa07:58
*** e0ne has quit IRC07:59
*** pkoniszewski has quit IRC08:01
*** e0ne has joined #openstack-qa08:03
*** salv-orlando has joined #openstack-qa08:08
*** e0ne has quit IRC08:08
*** achanda has quit IRC08:10
*** e0ne has joined #openstack-qa08:12
*** ci-testing has quit IRC08:14
*** btran has joined #openstack-qa08:14
*** qba73 has joined #openstack-qa08:16
*** markus_z has joined #openstack-qa08:16
*** yamamoto has quit IRC08:18
*** e0ne has quit IRC08:18
openstackgerritUdi Kalifon proposed openstack/tempest: Support creating users with a default_project_id  https://review.openstack.org/13592208:21
*** tkammer has joined #openstack-qa08:24
*** afazekas has joined #openstack-qa08:34
*** OmPrakashPandey has quit IRC08:39
*** eezhova has quit IRC08:40
*** OmPrakashPandey has joined #openstack-qa08:40
*** eezhova has joined #openstack-qa08:41
*** jlanoux has joined #openstack-qa08:42
*** ranger81 has joined #openstack-qa08:42
*** safchain has joined #openstack-qa08:43
*** OmPrakashPandey has quit IRC08:45
*** ranger81 has quit IRC08:46
*** matrohon has joined #openstack-qa08:52
*** jpich has joined #openstack-qa09:01
*** obondarev_ is now known as obondarev09:06
*** yamamoto has joined #openstack-qa09:11
*** yamahata has quit IRC09:11
*** svasheka has quit IRC09:11
*** adalbas has joined #openstack-qa09:14
*** topshare has quit IRC09:18
*** yamamoto has quit IRC09:18
*** adalbas has quit IRC09:21
*** andreykurilin_ has joined #openstack-qa09:22
*** avico has joined #openstack-qa09:22
*** jordanP has joined #openstack-qa09:23
*** yfried is now known as yfried|afk09:24
*** nmagnezi has quit IRC09:25
*** yfried|afk is now known as yfried09:25
*** e0ne has joined #openstack-qa09:25
*** OmPrakashPandey has joined #openstack-qa09:27
*** luqas has joined #openstack-qa09:27
*** oomichi has quit IRC09:33
*** nmagnezi has joined #openstack-qa09:38
sergshyfried, np, will run09:41
yfriedsergsh: I should put this in comments, but you need to disable nova and neutron quotas, as well as set a large enough subnet CIDR in tempest.conf/network section09:42
*** zz_johnthetubagu is now known as johnthetubaguy09:44
sergshyfried, the default cidr is not good?09:46
openstackgerritJordan Pittier proposed openstack/tempest: Scenario manager: catch Exception in get_remote_client  https://review.openstack.org/13491409:46
chandankumarafazekas, hello09:46
openstackgerritSergey Shnaidman proposed openstack/tempest: Test delete of SLAAC subnet without port delete  https://review.openstack.org/13690909:47
afazekaschandankumar, hi09:47
chandankumarafazekas, pleae have a look why jenkins are failing on this review https://review.openstack.org/#/c/134889/09:48
chandankumar*please09:48
*** nfedotov has joined #openstack-qa09:48
yfriedsergsh: it's using /28 mask. I think09:50
yfriedsergsh: not sure09:50
yfriedsergsh: you need the CIDR to allow for creation of large_ops number09:51
sergshok09:51
*** aix has quit IRC09:51
*** andreykurilin_ has quit IRC10:00
*** andreykurilin_ has joined #openstack-qa10:00
*** salv-orlando has quit IRC10:00
*** salv-orlando has joined #openstack-qa10:02
openstackgerritMerged openstack-dev/devstack: Enable admin mode by default  https://review.openstack.org/13680310:03
*** topshare has joined #openstack-qa10:05
*** topshare has quit IRC10:05
*** mpavlase has joined #openstack-qa10:05
*** andreykurilin_ has quit IRC10:05
*** topshare has joined #openstack-qa10:05
*** andreykurilin_ has joined #openstack-qa10:06
*** nosnos has quit IRC10:08
*** nosnos has joined #openstack-qa10:09
afazekaschandankumar,  cirros boot stopped at 'info: initramfs: up at 3.11' line10:10
openstackgerritMerged openstack-dev/devstack: Restore linuxbridge-agent compatibility  https://review.openstack.org/13649410:11
afazekaschandankumar, Are you asking about the 'NOT_REGISTERED' ?10:12
afazekaschandankumar: looks like other people also get this response from Jenkins. It is some kind of infra/ config issue.10:15
afazekasThe #openstack-infra channel is better for speaking about these kind of issues10:16
*** aix has joined #openstack-qa10:17
openstackgerritYuiko Takada proposed openstack-dev/devstack: add the kernel/ramdisk id to driver_info  https://review.openstack.org/13589710:19
*** yamamoto has joined #openstack-qa10:19
*** yfried is now known as yfried|afk10:22
*** dmorita has quit IRC10:23
*** yamamoto has quit IRC10:24
openstackgerritjun xie proposed openstack/tempest: add cinder v2 api tests for volume_services  https://review.openstack.org/13586010:29
*** jlanoux_ has joined #openstack-qa10:31
*** jlanoux has quit IRC10:34
*** mpaolino has joined #openstack-qa10:37
*** topshare has quit IRC10:37
*** jhenner has joined #openstack-qa10:37
*** vigneshvar has joined #openstack-qa10:41
kashyapafazekas, Hi, After a successful DevStack run, I always see most of the projects' requirements.txt/test-requirements.txt  files modified, e.g. bumping up eventlet version change, etc. Can this be avoided somehow I wonder10:42
*** yfried|afk is now known as yfried10:42
chandankumarafazekas, asking the same on infra.10:44
*** jlanoux has joined #openstack-qa10:45
*** jlanoux_ has quit IRC10:48
yfriedafazekas: do you know where AndreaF is?10:52
afazekasyfried: Usually close to our TZ, I do not know where is he at the moment10:54
yfriedafazekas: he's been offline since Paris. I was wondering if he's on vacation and when he's coming back10:55
jlanouxyfried: AndreaF is on holiday until the 29th10:56
yfriedjlanoux: tnx10:57
*** salv-orlando has quit IRC10:57
yfriedjlanoux: guess I'll ping him next week10:57
jlanouxyfried: np10:58
*** avico has quit IRC11:00
*** salv-orlando has joined #openstack-qa11:02
yfriedjogo: ping11:02
*** luqas has quit IRC11:03
yfriedafazekas: what does NOT_REGISTERED means for Jenkins results in gate?11:06
*** X019 has joined #openstack-qa11:08
*** gszasz has joined #openstack-qa11:08
*** tosky has joined #openstack-qa11:09
*** salv-orlando has quit IRC11:11
*** salv-orlando has joined #openstack-qa11:11
openstackgerritAbhishek G M proposed openstack/tempest: Test to verify  shared attribute of network  https://review.openstack.org/12499811:12
*** andreykurilin_ has quit IRC11:12
*** avico has joined #openstack-qa11:13
afazekasyfried: maybe miss config in https://github.com/openstack-infra/project-config11:15
*** rfolco has joined #openstack-qa11:16
*** luqas has joined #openstack-qa11:17
*** aix has quit IRC11:18
*** k4n0 has quit IRC11:20
*** dims has joined #openstack-qa11:20
*** yfried has quit IRC11:20
*** rfolco has quit IRC11:20
*** mpavlase has left #openstack-qa11:20
*** yamamoto has joined #openstack-qa11:21
*** yfried has joined #openstack-qa11:22
*** avico has quit IRC11:22
*** avico has joined #openstack-qa11:25
afazekasyfried: try #openstack-infra I do not know what causes it11:26
*** vigneshvar has quit IRC11:27
openstackrecheckopenstack/nova change: https://review.openstack.org/134598 failed gate-tempest-dsvm-neutron-full in the gate queue with an unrecognized error11:28
*** vigneshvar has joined #openstack-qa11:30
*** aix has joined #openstack-qa11:31
*** yfried is now known as yfried|afk11:35
*** aysyd has joined #openstack-qa11:37
*** jhenner has quit IRC11:37
sergshwhat should I do if tested feature is only in Kilo, not in Juno?11:39
*** yfried|afk is now known as yfried11:40
yfriedsergsh: what do you mean? you found a bug in kilo?11:40
sergshyfried, no, I mean some bug was resolved in Kilo, I wrote test for it, but it fails on Jenkins jobs for juno11:41
sergshso it's impossible to push it to upstream11:41
yfriedsergsh: I had the same problem with icehouse last week11:42
*** pkonisze has quit IRC11:42
yfriedsergsh: the obvious, easy and long solution is to ask for a back port11:43
yfriedsergsh: the quick and dirty is to add a conf flag for a skip11:43
yfriedsergsh: and then edit devstack-gate11:43
*** yamamoto has quit IRC11:44
sergshyfried, yeah, it seems the same problem as yours, maybe version flag is good idea anyway11:44
yfriedsergsh: well, afazekas said (and I agree) that a version flag goes against the standard practice in tempest11:45
yfriedsergsh: a feature-specific flag is better11:45
yfriedsergsh: but a backport request is easiest11:45
sergshand how is backporting going? I don't know about it much)11:46
yfriedsergsh: where is the bug?11:48
sergshneutron11:48
afazekassergsh: do you have to the fixing change ?11:48
*** vigneshvar has quit IRC11:49
afazekashttps://wiki.openstack.org/wiki/StableBranch11:49
afazekasBTW: If you are lucky the backport is clickable on gerrit11:50
*** Madan has quit IRC11:53
SridharGsergsh: may i know the bug you are referring to in Neutron?11:54
sergshafazekas, SridharG , the change was committed already to Kilo by Dane Leblanc: https://review.openstack.org/#/c/134978/11:57
sergshand my test for it fails in Juno jobs: https://review.openstack.org/#/c/136909/11:58
SridharGsergsh: the bug is marked as tags:added: juno-backport-potential11:58
sergshSridharG, what does it mean? It MIGHT be backported?11:59
SridharGsergsh: so we can cherry pick and propose the same fix stable/juno branch.11:59
sergshSridharG, great, and who can do it? :)11:59
SridharGsergsh: normally the owner of the original patch does it, but AFAIK anyone can propose the cherry pick (if the owner is busy with other stuff).12:00
sdaguekashyap: the requirements sync end up being kind of important to actually ensure a consistent environment.12:01
*** adalbas has joined #openstack-qa12:01
* afazekas clicked on the cherry-pick button12:02
kashyapsdague, I see, noted.12:02
sergshSridharG, ok, thanks, I'll talk to Dane12:02
*** vigneshvar has joined #openstack-qa12:02
*** yfried is now known as yfried|afk12:02
sergshafazekas, thanks :)12:03
SridharGsergsh: wlc.12:03
*** markus_z has quit IRC12:03
sdaguekashyap: maybe if we start using a virtual env instead of installing services into system libs we can get rid of it12:04
kashyapsdague, Yeah. I bring it up because, sometimes, I have to manually do a 'git checkout *requirements.txt' then, a script that traverses all git repos to do a git pull12:06
SridharGafazekas: a tiny patch. Can you please review when you get some time. https://review.openstack.org/#/c/129993/112:07
*** yfried|afk is now known as yfried12:08
kashyapsdague, While I have your attention, RECLONE=yes in local.conf annoyingly removes branch tracking, any reason why?12:08
kashyapWhich results in me having to manually set upstream tracking $ git branch --set-upstream master origin12:09
sdaguekashyap: don't know about that one12:10
kashyapAnyhow, /me stopped using RECLONE=yes & do a git pull of project repos via a silly script12:10
sdagueso that should be fixed12:10
sdaguein theory there is also an UNDO_REQUIREMENTS which should be true by default, which does the first thing12:10
sdaguebut... a bunch of this is subtle12:10
yfriedsergsh: btw, if you have a test to recreate a bug and test for it, it's good to have a "related-bug" in commit message, or even add tempest to the bug and add "closes-bug" the the commit12:10
kashyapAh, didn't know of UNDO_REQ. . ., thank you, will test in one my runs to see how it goes.12:12
sergshyfried, closing bug - but it's not tempest bug..?12:12
sdaguekashyap: it should be set by default, but the whole dance around requirements is very subtle, if you look at the implementation12:12
yfriedsergsh: ^ as I said, you can add tempest to the bug page12:13
sergshyfried, ah, sorry, got it :)12:13
yfriedsergsh: https://review.openstack.org/#/c/124722/12:13
yfriedsergsh: see the attached bug for example12:13
kashyapsdague, Yep, will have to take a look. On a different note, do you use 'devpi' in your test environments?  I recently stumbled across it via danpb's blog post, seems like a very useful method.12:14
sdaguekashyap: I do not, but I think it would be really valuable to integrate into devstack12:14
openstackgerritSergey Shnaidman proposed openstack/tempest: Test delete of SLAAC subnet without port delete  https://review.openstack.org/13690912:14
openstackgerritUdi Kalifon proposed openstack/tempest: Support creating users with a default_project_id  https://review.openstack.org/13592212:15
kashyapsdague, Nice idea, probably someone experienced with it should bring this topic up on openstack-dev list.12:15
*** jhenner has joined #openstack-qa12:16
*** tkammer has quit IRC12:16
*** markus_z has joined #openstack-qa12:16
*** tochi has quit IRC12:20
sdaguekashyap: honestly, I think people just need to do it12:21
kashyapsdague, Probably. I'm just not experienced enough with this stuff (requirements management) to speak definitively.12:23
*** salv-orlando has quit IRC12:25
*** tkammer has joined #openstack-qa12:26
*** vigneshvar has quit IRC12:30
*** avico has quit IRC12:30
*** luqas has quit IRC12:31
openstackgerritAbhishek G M proposed openstack/tempest: Test to create bulk port  https://review.openstack.org/13703812:42
*** weshay has joined #openstack-qa12:42
*** Longgeek_ has joined #openstack-qa12:44
*** yamamoto has joined #openstack-qa12:44
*** Longgeek has quit IRC12:44
openstackgerritDenis M. proposed openstack-dev/devstack: Copy datastore templates when installing Trove  https://review.openstack.org/13703912:44
openstackgerritKyle Mestery proposed openstack-dev/devstack: Update the ODL support to Helium SR1.  https://review.openstack.org/13430912:46
*** yamamoto has quit IRC12:49
*** OmPrakashPandey has quit IRC12:49
*** mkollaro has joined #openstack-qa12:54
openstackgerritKirill Shileev proposed openstack/tempest: Add scenario test related to the way  IPv6 addressed assigned to vNIC  https://review.openstack.org/11233613:00
*** gszasz_ has joined #openstack-qa13:06
*** gszasz has quit IRC13:07
*** yamamoto has joined #openstack-qa13:08
*** gszasz_ has quit IRC13:08
*** jordanP has quit IRC13:10
*** mpaolino has quit IRC13:12
*** yamamoto has quit IRC13:13
*** mdbooth has quit IRC13:15
*** neeti has quit IRC13:16
openstackgerritSean Dague proposed openstack/tempest: remove xml_utils and all things that depend on it  https://review.openstack.org/13682213:17
openstackgerritYair Fried proposed openstack/tempest: Adds scenario for network-large-ops  https://review.openstack.org/13678913:17
*** JayJ has joined #openstack-qa13:17
yfriedsergsh: pushed a fix ^13:17
yfriedsergsh: it ran on my setup13:17
openstackgerritChmouel Boudjnah proposed openstack/qa-specs: Add devstack external plugin spec  https://review.openstack.org/13705413:19
openstackgerritSean Dague proposed openstack/tempest: rm pyc files before test runs  https://review.openstack.org/13705513:19
sergshyfried, can you also address my inline comments? theres is a typo in test name13:21
*** chandankumar has quit IRC13:21
*** JayJ has quit IRC13:22
*** JayJ has joined #openstack-qa13:22
*** Longgeek has joined #openstack-qa13:24
yfriedsergsh: sure13:24
yfriedsergsh: didn't see it13:24
*** gszasz has joined #openstack-qa13:25
*** markus_z has quit IRC13:25
*** JayJ has quit IRC13:27
*** Longgeek_ has quit IRC13:27
*** JayJ has joined #openstack-qa13:27
openstackgerritKirill Shileev proposed openstack/tempest: Adds scenario for IPv6 addresses  https://review.openstack.org/11233613:28
sergshyfried, http://pastebin.com/izyUA7kM13:29
openstackgerritDenis M. proposed openstack-dev/devstack: Copy datastore templates when installing Trove  https://review.openstack.org/13703913:29
*** JayJ has quit IRC13:32
*** JayJ has joined #openstack-qa13:32
openstackgerritEric Harney proposed openstack-dev/devstack: Ceph: undefine secret from nova.conf when deleting secret  https://review.openstack.org/12543213:33
openstackgerritDenis M. proposed openstack-dev/devstack: Increate time on start for instance datastore  https://review.openstack.org/13706213:37
*** topshare has joined #openstack-qa13:38
*** markus_z has joined #openstack-qa13:38
nfedotovHello all! Does somebody know... Is Devstack going to start using Cirros 0.3.3 image by default? Sorry for off-topic13:39
*** mpaolino has joined #openstack-qa13:41
*** nosnos has quit IRC13:48
*** enikanorov__ has quit IRC13:48
*** nosnos has joined #openstack-qa13:49
openstackrecheckopenstack/ceilometer change: https://review.openstack.org/136781 failed gate-ceilometer-pep8 in the gate queue with an unrecognized error13:52
*** salv-orlando has joined #openstack-qa13:53
*** nosnos has quit IRC13:54
openstackgerritKirill Shileev proposed openstack/tempest: Adds scenario for IPv6 addresses  https://review.openstack.org/11233613:56
*** markus_z has quit IRC13:57
*** enikanorov_ has joined #openstack-qa13:57
*** dims has quit IRC13:57
*** yamahata has joined #openstack-qa13:58
*** dims has joined #openstack-qa13:58
openstackgerritSean Dague proposed openstack/tempest: Enable block_migration_for_live_migration by default  https://review.openstack.org/13514213:59
openstackgerritYair Fried proposed openstack/tempest: Adds scenario for network-large-ops  https://review.openstack.org/13678914:00
yfriedsergsh: take a look ^14:01
sdaguenfedotov: if there is a new cirros, patches welcomed to bump it up14:02
*** dustins has joined #openstack-qa14:06
*** beagles is now known as beagles_brb14:10
*** markus_z has joined #openstack-qa14:10
*** toure has joined #openstack-qa14:12
*** yamamoto has joined #openstack-qa14:14
*** dkranz has joined #openstack-qa14:14
*** dustins has quit IRC14:18
*** yamamoto has quit IRC14:18
*** dustins has joined #openstack-qa14:20
*** dustins has quit IRC14:20
nfedotovsdague: The reason I ask is because the patchset already exists. https://review.openstack.org/#/c/122936/ . Anyway, thanks. Maybe it would be better to create new patchset for it14:21
sdaguenfedotov: ok, but it's not passing tests14:21
sdaguedid you look into why?14:21
sdagueianw's -1 is because it's not passing tests14:21
*** rushiagr is now known as rushiagr_away14:24
*** mriedem has joined #openstack-qa14:26
yfriedjogo: around?14:27
*** jlibosva has joined #openstack-qa14:28
sdagueyfried: this is very early for jogo14:28
*** jenrh has joined #openstack-qa14:32
mriedemany ideas if we have a query for this issue already? http://goo.gl/DdBdqx14:34
mriedemcinder and nova logs don't have any errors,14:35
mriedembut must be something blocking a volume operation somewhere in there14:35
openstackgerritMichele Paolino proposed openstack-dev/devstack: Support for libvirt/QEMU tar releases  https://review.openstack.org/10871414:36
sdaguemriedem: so... honestly, I expect this is vaguely related to the ML post I put out there14:37
sdaguewe're increasing the # of volume tests14:37
sdaguehitting critical sections in cinder14:37
sdagueand ... sadness14:37
mriedemwe've had known lvm issues in the gate for awhile causing locking issues, re: the kernel thing14:38
mriedemso i can buy that14:38
mriedemwould be nice to see something in the logs14:38
sdagueyeh... agreed14:38
*** nmagnezi has quit IRC14:40
mriedemi see a 3 minute gap here in the volume delete flow14:40
mriedemstart: http://logs.openstack.org/90/131490/4/gate/gate-grenade-dsvm-partial-ncpu/0537e3b/logs/old/screen-c-vol.txt.gz#_2014-11-25_03_41_22_46614:40
mriedemnext mention: http://logs.openstack.org/90/131490/4/gate/gate-grenade-dsvm-partial-ncpu/0537e3b/logs/old/screen-c-vol.txt.gz#_2014-11-25_03_44_41_03214:40
mriedemtempest failed at 3:44: http://logs.openstack.org/90/131490/4/gate/gate-grenade-dsvm-partial-ncpu/0537e3b/logs/grenade.sh.txt.gz#_2014-11-25_03_44_48_21714:41
mriedemsdague: would you be opposed to having an e-r query with this?14:41
mriedemmessage:"(TestVolumeBootPattern:_run_cleanups) Failed to delete volume" AND message:"within the required time" AND tags:"console"14:42
sdaguemriedem: nope, sounds good to me14:42
mriedemit's pretty generic i know14:42
mriedemok14:42
*** om has joined #openstack-qa14:42
sdaguerealistically I'm pretty sure that this is because cinder deletes are taking *for ever*14:42
sdaguewhich... I'm not sure why14:42
mriedemwell, i've seen some of the commands take between a couple seconds and 2 minutes,14:42
mriedemin the periodic task that checks volume status14:42
sdagueright... that's that lvm query that stalls14:43
mriedemi think the delete flow is hitting an i/o block or process lock or something when a periodic task is running14:43
sdaguemaybe, that's the thing14:43
sdagueyeh, what's actually crazy is that it's the blockdev query14:43
sdagueit's not actually an action14:43
mriedemthis is the bug i think we were tracking https://bugs.launchpad.net/cinder/+bug/137351314:43
sdagueyeh, so basically we need to figure out which cinder cores can help dive in, or we just start turning off tests14:44
mriedemtest_volume_boot_pattern would be my first candidate :)14:47
mriedembetween the volume delete hangs and the ssh bullshit, i hate that test14:48
mriedemi'm pretty sure test_volume_boot_pattern sabotaged my mic14:48
*** yfried is now known as yfried|afk14:48
sdagueheh14:48
sdaguemriedem: push the skip14:48
sdagueI'll +2 it14:48
*** markus_z has quit IRC14:49
sdaguethat test we've turned off a ton in the past14:49
mriedemk, getting the e-r query up first14:49
openstackgerritMerged openstack/tempest: Use standard assertVolumeStatusWait for volume detachment  https://review.openstack.org/13026214:51
*** rushiagr_away is now known as rushiagr14:51
*** markus_z has joined #openstack-qa14:51
*** yfried|afk is now known as yfried14:51
mtreinishjogo: awesome14:53
*** tosky has quit IRC14:56
*** markus_z has quit IRC14:57
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for cinder volume delete hang bug 1396186  https://review.openstack.org/13709014:57
*** tosky has joined #openstack-qa14:57
mriedemsdague: mtreinish: ^14:58
mtreinishmriedem: +A14:59
*** beagles_brb is now known as beagles15:00
openstackgerritMerged openstack/tempest: Sync latest log module from oslo-incubator  https://review.openstack.org/13160115:00
*** markmcclain has joined #openstack-qa15:01
openstackgerritMerged openstack/tempest: enable cinder v2 api for test_volumes_backup  https://review.openstack.org/13517915:01
*** markus_z has joined #openstack-qa15:01
*** topshare has quit IRC15:02
*** zz_jgrimm is now known as jgrimm15:03
mriedemsdague: replied to ML, working on that skip patch now15:03
*** raja has joined #openstack-qa15:03
sdaguemriedem: danke15:03
*** raja is now known as Guest8372815:04
mriedemthis took 50 seconds15:04
mriedem2014-11-25 03:41:48.993 7804 DEBUG cinder.openstack.common.processutils [-] Running cmd (subprocess): sudo cinder-rootwrap /etc/cinder/rootwrap.conf env LC_ALL=C vgs --noheadings --unit=g -o name,size,free,lv_count,uuid --separator : --nosuffix stack-volumes-lvmdriver-1 execute /opt/stack/old/cinder/cinder/openstack/common/processutils.py:15815:04
sdaguemtreinish: so I -2ed a jogo patch to change a tempest default that hit the gate this morning, I'm pretty convinced that was not the right way to change the live migration configs15:04
sergshyfried, http://paste.ubuntu.com/9233767/15:04
sdaguemtreinish: https://review.openstack.org/#/c/135142/15:04
openstackgerritJiri Tomasek proposed openstack/tempest: Dashboard login page assert string fix  https://review.openstack.org/13709215:05
*** rbak has joined #openstack-qa15:05
mriedemthis took almost 2 minutes:15:05
mriedem2014-11-25 03:42:44.858 7804 DEBUG cinder.openstack.common.processutils [req-81b41283-0928-4572-891e-c45ef4667e64 4eb70d740cda49618ccb9ac9b64ca824 5a2e289cbd224a84a78f496717d98684 - - -] Running cmd (subprocess): sudo cinder-rootwrap /etc/cinder/rootwrap.conf env LC_ALL=C lvs --noheadings --unit=g -o vg_name,name,size --nosuffix stack-volumes-lvmdriver-1 execute /opt/stack/old/cinder/cinder/openstack/common/processutils.py:15815:05
sdaguemriedem: yeh, then it failed with a timeout, right?15:06
mriedemsdague: yup15:07
Guest83728I have a devstack running. i want to work on "hot swap volume" feature15:07
mriedemso that's 2 commands taking ~3 minutes total which is the gap and then the timeout15:07
Guest83728can anyone help me on this pls15:07
mriedemvgs and lvs, which is what i've seen before15:07
mtreinishsdague: hmm, that might be, let me check what the option does15:07
*** yamahata has quit IRC15:08
mtreinishI was fine with changing the other live_migration flag to default true because that's really the hypervisor opt (we still check host count)15:08
sdaguemtreinish: how is hostcount checked?15:08
*** avico has joined #openstack-qa15:09
*** Guest83728 has left #openstack-qa15:10
mtreinishsdague: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/api/compute/test_live_block_migration.py#n9115:10
mtreinishit does a list-hosts and checks how many compute nodes are there15:10
sdagueok15:10
sdagueso ... test_live_migration should live in an admin tree then15:10
mtreinishit should15:11
sdaguewhich it probably should have anyway15:11
*** rajashekar_vedam has joined #openstack-qa15:11
mtreinishI really don't know why it was where it was in the tree15:11
*** afazekas has quit IRC15:11
rajashekar_vedamIs there any "hot swap volume" feature in cinder?15:11
sdagueok, well it can't pass until the precise jobs get fixed15:11
openstackgerritSean Dague proposed openstack/tempest: Enable block_migration_for_live_migration by default  https://review.openstack.org/13514215:13
*** mwagner_lap has quit IRC15:13
mtreinishsdague: but for the block migration flag I'm not exactly sure what the default should be, it's a bit different15:13
openstackgerritMerged openstack-infra/elastic-recheck: Add query for cinder volume delete hang bug 1396186  https://review.openstack.org/13709015:13
sdagueok, well put your comments on it15:13
yfried sergsh: how many VMs are you building?15:13
sergshyfried,1015:14
yfriedsergsh: does your setup support this large amount?15:14
yfriedsergsh: what version are you using?15:14
sergshyfried, should support, I'll try with 2, let's see15:15
yfriedsergsh: is this devstack? all in one?15:16
rajashekar_vedamIs there any "hot swap volume" feature in cinder?15:16
*** yamahata has joined #openstack-qa15:16
*** bkopilov has quit IRC15:16
*** zz_dimtruck is now known as dimtruck15:16
sergshyfried, yes15:16
sergshyfried, version of what?15:17
yfriedsergsh: openstack15:17
openstackgerritMatt Riedemann proposed openstack/tempest: Skip test_volume_boot_pattern until bug 1396186 is fixed  https://review.openstack.org/13709615:18
mriedemsdague: there you go ^15:18
sergshyfried, a few days ago version, wait a moment, I'll install the latest15:18
*** jordanP has joined #openstack-qa15:19
*** openstackrecheck has quit IRC15:21
*** openstackrecheck has joined #openstack-qa15:21
mtreinishmriedem: are you going to push up a revert for the e-r query we just landed if it's covered by another one?15:23
mriedemmtreinish: yeah15:23
mriedemoops15:23
openstackgerritMatt Riedemann proposed openstack/tempest: Skip test_volume_boot_pattern until bug 1373513 is fixed  https://review.openstack.org/13709615:23
mtreinishheh, we've got enough queries it's hard to know what's in all of them15:23
mriedemi had looked at the other but only noticed the first part before the OR15:24
mtreinishI'm wondering if we should have a test that looks for the same set of matches between queries15:24
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Revert "Add query for cinder volume delete hang bug 1396186"  https://review.openstack.org/13709815:25
mriedemmtreinish: there is the revert15:25
*** jordanP has quit IRC15:25
*** jordanP has joined #openstack-qa15:26
mtreinishmriedem: ok, +A15:26
*** jordanP has quit IRC15:26
mriedemmtreinish: this didn't have enough related bugs for you? :) https://review.openstack.org/#/c/137096/15:26
*** jordanP has joined #openstack-qa15:27
yfriedsergsh: don't bother. I'm testing on juno15:28
*** achanda has joined #openstack-qa15:29
sdaguemtreinish: see ML for a series of test definition related emails this morning. Mostly trying to trim back things that don't provide value.15:29
sergshyfried, but should it work on kilo also?15:29
openstackgerritMatthew Treinish proposed openstack-infra/subunit2sql: Add cli option to subunit2sql to specify run_id  https://review.openstack.org/13129815:29
yfriedsergsh: no reason why not.15:30
*** gszasz has quit IRC15:30
yfriedsergsh: are you using a VM?15:30
sergshno, hadware15:30
yfriedsergsh: have you disabled the limits?15:30
yfriedsergsh: quotas?15:30
sergshI think it has default 10 instances, no?15:30
*** bkopilov has joined #openstack-qa15:31
*** achanda has quit IRC15:32
openstackgerritKirill Shileev proposed openstack/tempest: Adds scenario for IPv6 addresses  https://review.openstack.org/11233615:33
yfriedsergsh: there are quotas in neutron that would be exceeded if you are creating 10 machines15:36
*** rajashekar_vedam has quit IRC15:36
yfriedsergsh: see my commit message15:36
openstackgerritMerged openstack/tempest: Security Groups  API Tests Enhancements  https://review.openstack.org/10765415:38
openstackgerritMerged openstack/tempest: Updated from global requirements  https://review.openstack.org/13599415:38
*** om has quit IRC15:38
openstackgerritMerged openstack-infra/elastic-recheck: Revert "Add query for cinder volume delete hang bug 1396186"  https://review.openstack.org/13709815:40
openstackgerritChmouel Boudjnah proposed openstack/qa-specs: Add devstack external plugin spec  https://review.openstack.org/13705415:41
yfrieddkranz: ping15:43
dkranzyfried: busy now, what's up?15:43
yfrieddkranz: just wanted to ping you about the client-return-value bp. I said I'd help you with it and I saw it just got merged15:44
dkranzyfried: This needs to merge: https://review.openstack.org/#/c/130645/15:46
dkranzyfried: Then the spec has an etherpad to track work on each service.15:46
dkranzyfried: You just pick one, and change all client and test uses15:47
yfrieddkranz: ok. will start tomorrow15:48
*** achanda has joined #openstack-qa15:48
yfriedmtreinish: dkranz: btw, who's in charge of the tempset cleanup util?15:49
dkranzyfried: dpatterson15:49
dkranzyfried: He said there is a patch up for review that fixes the issue you saw15:50
*** openstackrecheck has quit IRC15:50
*** openstackrecheck has joined #openstack-qa15:50
yfrieddkranz: what TZ is he?15:51
dkranzyfried: us east15:51
*** avico has quit IRC15:51
dkranzyfried: same as me15:51
yfrieddkranz: I just discovered stackforge/ospurge which cleans up all resources in a tenant. seems like a duplication of effort. what's the policy on using it in tempest?15:52
*** avico has joined #openstack-qa15:53
mesteryHi QA folks! If I could get some time for this super simple review https://review.openstack.org/#/c/134309/. It enables an updated version of ODL in devstack, which fixes IPV6 there and lets the Tempest API tests run again. We need this to land to re-enable our CI.15:54
mesteryThanks for the help!15:54
openstackgerritSean Dague proposed openstack/tempest: delete all compute xml tests  https://review.openstack.org/13679215:56
openstackgerritSean Dague proposed openstack/tempest: part 2 of delete xml  https://review.openstack.org/13681515:56
openstackgerritSean Dague proposed openstack/tempest: purge xml clients from tempest.clients  https://review.openstack.org/13682115:56
openstackgerritSean Dague proposed openstack/tempest: remove xml_utils and all things that depend on it  https://review.openstack.org/13682215:56
openstackgerritSean Dague proposed openstack/tempest: use built in xml.etree instead of lxml  https://review.openstack.org/13711315:56
openstackgerritAysy Anne A. Duarte proposed openstack-infra/elastic-recheck: Add query for insufficient free space  bug 1386249  https://review.openstack.org/13211815:56
sdaguemtreinish: https://review.openstack.org/#/c/137055/15:57
sdaguemestery: so.... I'll +2 that *if* someone from ODL is going to convert this to a devstack plugin once we get that speced out15:58
sdaguebecause I actually don't think opendaylight should be a top level construct in devstack15:58
mesterysdague: Absolutely! At one point we wanted it a top-level construct, but I agree, making it a plugin gets a huge +1 from me.15:58
mesterysdague: Is there a spec for devstack plugins proposed yet?15:59
mtreinishmestery: https://review.openstack.org/13705415:59
mesterymtreinish: Thanks!15:59
mesterymtreinish sdague: That's a fresh spec :)16:00
*** SridharG has left #openstack-qa16:01
sdaguemestery: it is16:01
sdaguewhich is why I'll still give you a +216:01
mtreinishmestery: heh, that's how I know about it, I just saw it pop up somewhere16:01
mesterylol16:01
sdagueyeh it started with a chmouel email from earlier this week16:01
*** achanda has quit IRC16:02
mesterysdague: Thanks, we'll work to make sure ODL fits into this once it's ready, I've added it to my review list.16:02
sdaguemtreinish: https://review.openstack.org/#/c/137055/16:02
sdaguemostly because one of the rmxml patches died because we didn't have that16:03
*** nfedotov has quit IRC16:07
*** jlibosva has quit IRC16:08
dkranzyfried: Yes there is duplication and that was discussed. But sometimes it costs more to reduce duplication than to have it.16:12
*** chandankumar has joined #openstack-qa16:12
dkranzyfried: I believe this code belongs in tempest. If some one wants to take the other code, make it work in tempest, and remove the current cleanup  I have no issus with that.16:13
dkranzyfried: Assuming it was better than what we have now.16:13
*** yfried has quit IRC16:14
*** yamamoto has joined #openstack-qa16:16
*** bkopilov has quit IRC16:17
*** matrohon has quit IRC16:20
*** yamamoto has quit IRC16:21
*** avico has quit IRC16:23
*** X019 has quit IRC16:23
*** X019 has joined #openstack-qa16:24
*** JayJ has quit IRC16:29
*** om has joined #openstack-qa16:29
*** jenrh has quit IRC16:29
*** obondarev has quit IRC16:32
*** tkammer has quit IRC16:35
openstackgerritchandan kumar proposed openstack/tempest: Enable cinder v2 api tests in volume quotas  https://review.openstack.org/13488916:36
*** X019 has quit IRC16:37
*** sergsh has quit IRC16:40
openstackgerritRob Crittenden proposed openstack/tempest: Add ca_certificates_file option to test against SSL servers  https://review.openstack.org/13679316:41
*** shashankhegde has joined #openstack-qa16:42
*** X019 has joined #openstack-qa16:43
jgriffithsdague: looks like about 6 common cinder failures16:46
sdaguejgriffith: yeh16:46
jgriffithsdague: 3 inparticular with a super high freq16:46
*** mpaolino has quit IRC16:47
sdaguewhich is why I'm concerned on piling on new tests when those aren't understood16:47
jgriffithsdague: see what thingee has to say but I'll put it on my plate to see if I can't start whitling some of them16:47
sdaguethat would be awesome16:47
jgriffithsdague: encryption is def broken based on the initial data16:47
jgriffithsdague: but it looks like it's just because it's horribly slow16:47
jgriffithanyway16:47
jgriffithsdague: I'll sign up to see if I can help16:48
sdagueso I think there is also a root cause that lvs scans can go into an extremely long (2 minute) timeout if something happens underneath16:48
jgriffithsdague: yeah, that whole thing is still killing me16:48
sdagueI kind of wonder if we can do an equivalent thing on a shorter timeout16:48
sdaguebecause when it succeeds, it's a couple of seconds16:48
jgriffithsdague: it seems like there's something going on in parallel that's causing that16:48
sdagueat most16:48
jgriffithif that makes sense?16:48
sdagueyep, very well could be16:49
jgriffithsdague: in other words, I have a rig running just volume tests and not triggering it16:49
jgriffithsdague: but once I get my lab back I want to do some better data colection on that16:49
jgriffithcurrently I have no gear, they "lost" my servers in the move last week :(16:49
sdaguejgriffith: yeh, I wonder what else needs to be in the mix16:49
sdaguebummer16:49
jgriffithsdague: I'm going to try and light bulb that, but it's going to surely take a while16:50
jgriffithsdague: like mid next week best case16:50
sdagueyeh, no worries16:50
jgriffithk16:50
sdaguemostly I wanted to get the conversations rolling here to make sure they hadn't gotten lost16:50
jgriffithI'll keep you posted, or if you come across something new ping me please16:50
jgriffithcommunication will go a long way on this I thing16:50
*** sergsh has joined #openstack-qa16:50
jgriffithno, certainly not lost16:50
jgriffithjust horrible timing :)16:50
sdague:)16:51
*** sergsh is now known as sergsh_away16:51
*** chandankumar has quit IRC16:51
jgriffithof course it's never a good time :)16:53
*** e0ne has quit IRC16:54
*** nmagnezi has joined #openstack-qa16:58
*** Longgeek has quit IRC16:58
*** amotoki has quit IRC17:00
*** markus_z has quit IRC17:01
*** yfried has joined #openstack-qa17:01
*** salv-orlando has quit IRC17:01
yfriedmtreinish: dkranz: sdague: could you please take a look? https://review.openstack.org/#/c/136611/17:02
yfriedmkoderer: ^17:02
*** om has quit IRC17:02
*** kmartin has quit IRC17:05
*** dims has quit IRC17:09
*** avico has joined #openstack-qa17:10
*** markmcclain has quit IRC17:11
*** yamahata has quit IRC17:14
*** StevenK has quit IRC17:18
*** StevenK has joined #openstack-qa17:20
mtreinishjogo: when you're around: http://paste.openstack.org/show/138278/ top 25 failure counts for a ~week17:21
*** shashankhegde has quit IRC17:21
mtreinishalthough the queue filtering isn't perfect yet so a few check jobs still slip in17:21
sdaguemtreinish: you should do it by percentage17:24
sdagueabsolutes aren't quite as useful as percentages17:24
mtreinishsdague: well it's just a db query, I was going to do a percentage on the db api side17:25
clarkbone interesting thing about those numbers i sthe tests we run more fail less17:25
sdagueclarkb: well... it's a slice17:26
sdaguemtreinish: can I get access to the sql connection from here (i.e. is the proxy up?)17:26
openstackgerritSean Dague proposed openstack/tempest: use built in HTMLParser instead of lxml  https://review.openstack.org/13711317:26
mtreinishsdague: nope not yet, the review is still pending: https://review.openstack.org/122311 and there are some new hiera values for it too17:27
sdagueclarkb: honestly, at limit 25 out of 2200 tests, I wouldn't read too much into it17:27
openstackgerritMerged openstack/tempest: Unified interface for ScenarioTest and NetworkScenarioTest  https://review.openstack.org/13679117:27
clarkbsdague: but its sorted by fail right?17:28
sdagueclarkb: by absolute fail #17:28
sdaguewhich doesn't take into account percentages17:28
mtreinishsdague, clarkb: at least for the top 2 those are for a stable bug that popped up yesterday (or over the weekend) for a test added recently (I think)17:28
mtreinishwhich would explain the smaller numbers17:28
*** jordanP has quit IRC17:29
adam_gmtreinish, which bug is that?17:30
mtreinishthe one which stands out most to me is the heat cfn init failure number which was like 25% before we skipped it17:30
jogomtreinish: nice17:30
mtreinishadam_g: https://bugs.launchpad.net/tempest/+bug/1395368 (or least that's the bug I saw for it)17:31
sdagueyeh, which, realistically is just that fedora guests 2nd level are untenable17:32
openstackrecheckopenstack/cinder change: https://review.openstack.org/136438 failed because of: gate-cinder-docs: https://bugs.launchpad.net/bugs/129800617:33
*** jlanoux has quit IRC17:33
*** dustins has joined #openstack-qa17:33
*** dustins has quit IRC17:34
*** beagles is now known as beagles_brb17:34
*** qba73 has quit IRC17:38
*** dustins has joined #openstack-qa17:38
*** e0ne has joined #openstack-qa17:40
jogoadam_g: ping17:41
jogomtreinish: will that data be published in an automatic report somewhere?17:41
*** shashankhegde has joined #openstack-qa17:41
*** andreykurilin_ has joined #openstack-qa17:42
jogoadam_g: latest live migration test run17:42
*** dustins has quit IRC17:42
mtreinishjogo: maybe, but it's not ready for that though17:42
mtreinishwe still need to get public read-only access to the db17:42
*** melwitt has joined #openstack-qa17:42
*** dustins has joined #openstack-qa17:42
mtreinishright now it's just on the trove node on rax which is on the "private" network17:42
jogomtreinish: ahh17:43
mtreinishthe queries work by relay through fungi :)17:43
jogo^_^17:43
*** wuhg has quit IRC17:43
*** luqas has joined #openstack-qa17:43
adam_gjogo, hey17:44
*** X019 has quit IRC17:45
*** avico has quit IRC17:45
*** luqas has quit IRC17:45
jogoadam_g: still seeing the same failures17:46
jogobut now have auth log17:46
*** shashankhegde has quit IRC17:46
adam_gjogo, oh, anything useful?17:46
fungimtreinish: i am not an api17:46
fungi;)17:46
fungi(i am a meat popsicle)17:47
jogoadam_g: sshd[6170]: Connection closed by 10.208.34.141 [preauth]17:48
jogoadam_g: is all I could find17:48
jogoadam_g: and the second node is still having name resolution issues17:49
*** harlowja_away is now known as harlowja17:49
adam_gjogo, that looks more like a hostkey issue but those errors are logged with key error messages17:49
*** marrusl has quit IRC17:49
adam_gjogo, i have a pair of rax servers i prepped with the same networking that i can bring up and restack soon if you want to poke at more directly (in case you dont have something similar going)17:50
jogoadam_g: I don't17:50
jogothat would be usefu17:50
jogouseful17:50
jogoI wonder if this an ssh-agent issue17:50
adam_gthere shouldn't be any ssh agent running there?17:51
*** dimsum__ has joined #openstack-qa17:51
jogoadam_g: I don't know17:52
*** marrusl has joined #openstack-qa17:52
jogohow else does it find the ssh keys to use17:52
*** mkollaro has quit IRC17:53
*** mwagner_lap has joined #openstack-qa17:53
adam_gjogo, i would have assumed python-libvirt/libvirt uses the users default path to key ($HOME/.ssh/id_$foo) but actually--ssh agent *might* explain  why i was seeing it succeed in one direction and not the other last week17:55
adam_gbut i dont usually forward my agent, hmm17:55
jogoadam_g: is your two node env set up17:56
adam_gjogo, rebuilding it now17:56
*** directxman12 has joined #openstack-qa17:57
jogodirectxman12: o/ thanks17:57
*** X019 has joined #openstack-qa17:57
jogohttp://logs.openstack.org/04/136504/3/experimental/check-tempest-dsvm-aiopcpu/f7eea47/logs/screen-n-cpu.txt.gz?level=TRACE#_2014-11-25_06_42_44_06017:57
jogoDinaBelova: testing libvirt live block migration with qemu+ssh and no shared storage17:58
jogodirectxman12: woops ^17:58
jogoand ssh should be working between the two nodes17:58
jogoas its used earlier in the testing17:59
directxman12jogo: if you su into the libvirt user, are you able to connect using `virsh -c 'qemu+ssh://xyz'`17:59
jogodirectxman12: these are throw away gate nodes don't have one lying around17:59
*** rushiagr is now known as rushiagr_away17:59
jogowe are trying 'qemu+ssh://stack@xyz' since the stack user has ssh keys set up18:00
jogoadam_g: is bringing up a test env to reproduce this18:00
directxman12jogo: ok18:01
jogodirectxman12: how does libvirt find the ssh keys?18:02
yfriedjogo: please take a look https://review.openstack.org/#/c/136981/18:02
yfriedjogo: it's the bug we've discussed last week18:02
*** safchain has quit IRC18:03
directxman12jogo: but IIRC, live migration involves libvirt daemons connecting to each other, so I think it is going to be connecting as whatever user the libvirt daemon is running as18:03
jogoyfried: thanks! its on the list18:03
yfriedjogo: what list?18:03
jogoyfried: list of reviews that I need to do18:03
yfriedjogo: oh. YOUR list :)18:04
adam_gdirectxman12, you can specify user in qemu+ssh://$user@host and, as long as user is member of libvirtd group all should be okay18:04
yfriedjogo: I have some more for you re that test :)18:04
adam_gat least, IIRC :)18:04
directxman12adam_g: yeah, I know that18:04
*** e0ne has quit IRC18:05
*** suneelb has joined #openstack-qa18:06
*** mriedem has quit IRC18:06
*** e0ne has joined #openstack-qa18:06
directxman12adam_g, jogo: hold on, I'm checking something18:07
*** kmartin has joined #openstack-qa18:08
*** bkopilov has joined #openstack-qa18:09
*** btran has quit IRC18:09
*** jamielennox|away is now known as jamielennox18:11
openstackrecheckopenstack/cinder change: https://review.openstack.org/131987 failed gate-grenade-dsvm in the gate queue with an unrecognized error18:12
*** shashankhegde has joined #openstack-qa18:13
*** e0ne has quit IRC18:14
*** dbel has quit IRC18:16
*** e0ne has joined #openstack-qa18:16
directxman12jogo, adam_g: what migration flags are you using18:16
adam_gjogo, ^18:17
*** jpich has quit IRC18:18
directxman12jogo: also, just to confirm: the libvirt user on the source node can SSH into the stack user on the target node?18:20
directxman12jogo: (or whatever user libvirtd is running as)?18:20
jogodirectxman12: default18:21
jogodirectxman12: default nova settings18:21
*** beagles_brb is now known as beagles18:22
jogodirectxman12: maybe adam_g can confirm the ssh parat18:22
clarkbno libvirt user won't ssh as stack18:22
clarkbits stack to stack18:22
*** mkollaro has joined #openstack-qa18:22
clarkbso that is probably your issue18:22
directxman12jogo: ok, so that means tunnelled p2p migration -- https://libvirt.org/migration.html#scenariotunnelpeer2peer218:22
clarkb(at least if my reading of the code last week is still valid)18:22
*** e0ne has quit IRC18:22
directxman12jogo: that's probably your issue -- if you look at the above link, the way it works is that the libvirt daemon uses the uri provided to connect to the other libvirt daemon18:23
jogodirectxman12: that make sense as ssh is used elsewhere and working18:25
adam_gjogo, ubuntu@(166.78.143.58, 166.78.62.51)  theres a tmux session running on the first18:25
*** aix has quit IRC18:25
directxman12jogo: so I'd just make sure whomever libvirtd is running as can SSH to the target user (you probably just want to SSH into the same user to keep it simple)18:26
jogoroot     19582     1 19582  0.5  0.1 820416 14760 /usr/sbin/libvirtd -d18:27
adam_gjogo, devstack hasn't been run there yet18:28
sdaguemtreinish: you should put xml death on the project meeting for today18:29
jogoadam_g: ahh18:29
sdaguebecause the patches seem to be passing18:29
jogodirectxman12: so virsh -c 'qemu+ssh:// seems to be working18:30
jogoadam_g: is ssh as root set up?18:31
adam_gjogo, no18:31
jogoadam_g: good18:31
*** patrickeast has joined #openstack-qa18:31
jogoadam_g: as that is the user running libvirt18:31
jogoaccording to http://logs.openstack.org/04/136504/3/experimental/check-tempest-dsvm-aiopcpu/f7eea47/logs/10.208.35.18/ps.txt.gz18:31
adam_gjogo, do you have any patches to devstack that should be pulled in before kicking it off?18:31
jogoadam_g: just yours 13619718:32
jogoadam_g: btw how was I able to log in to that machine? did you use my public key?18:32
mtreinishsdague: I did that yesterday18:33
adam_gjogo, yeah, 'ssh-import-id jogo'18:33
*** salv-orlando has joined #openstack-qa18:33
*** nmagnezi has quit IRC18:33
*** ranger81 has joined #openstack-qa18:33
jogoadam_g: nice trick I didn't know about that18:34
adam_gjogo, yeah, its super handy18:34
adam_gwell, among people who use launchpad18:34
directxman12jogo: so the options are to either make root be able to SSH into stack, or test using a different SSH setup.18:34
jogodirectxman12: if the test env we are using doesn't have ssh as root set up18:35
clarkbroot ability to ssh as stack is easy to set up18:35
clarkbjogo: it does18:35
jogodirectxman12: but "virsh -c 'qemu+ssh://" seems to be working18:35
clarkbjogo: thats how it works18:35
*** rushiagr_away is now known as rushiagr18:36
clarkbhrm unless I am mixing up jobs18:36
clarkbin any case root to stack should be simple18:36
jogoclarkb: yeah and easy to test if it set up or not18:36
directxman12jogo: FWIW, I generally reccomend using TLS auth instead.  It sidesteps some of the issues involved with SSH18:37
clarkbyou still have to set up kes18:37
clarkbcerts18:37
clarkbit just comes with different problems18:37
clarkbor am I mixing up auth methods?18:37
directxman12clarkb: you need TLS certs set up, but it only applies for libvirt, and you don't need to worry about root being able to SSH root, etc18:38
clarkbwe don't need root to ssh as root aiui18:38
directxman12clarkb: *root being able to SSH into anybody18:38
directxman12clarkb: sorry, mistype18:38
clarkbwe need root to ssh as a user that is in the libvirt group18:38
clarkbthat is super simple to set up18:38
directxman12clarkb: yep18:38
clarkbsince we already have that in the test setup18:38
directxman12clarkb: I mistyped18:38
clarkbsupporting TLS would require completely new stuff18:39
directxman12clarkb: fair enough18:39
clarkbwhich is probably worth doing at some point but imo the first and easiest thing to do is ssh18:39
*** yfried has quit IRC18:40
*** yfried has joined #openstack-qa18:40
*** dimtruck is now known as zz_dimtruck18:41
jogodirectxman12: this is the current error we are seeing18:41
jogohttp://logs.openstack.org/04/136504/3/experimental/check-tempest-dsvm-aiopcpu/f7eea47/logs/screen-n-cpu.txt.gz?level=TRACE#_2014-11-25_06_42_58_99818:41
jogo'Live Migration failure: operation failed: Failed to connect to remote libvirt URI qemu+ssh://stack@devstack-trusty-2-node-rax-dfw-3210485-414.slave.openstack.org/system: Cannot recv data: Permission denied (publickey).: Connection reset by peer18:41
jogoso that means libvirt is trying to ssh from root into stack?18:42
jogoand failing18:42
clarkbya so all you need to do is put private key in ~root and pubkey in stack authorized_keys18:43
openstackgerritAlok Kumar Maurya proposed openstack/tempest: Basic networks Scenario Test Enhancements  https://review.openstack.org/10662018:43
directxman12jogo: yep18:43
*** bkopilov has quit IRC18:45
directxman12jogo: and then you could `sudo virsh -c 'qemu+ssh://stack@xyz/system'` to double check18:45
directxman12jogo: after keys are deployed, etc18:45
jogoclarkb: like this? http://paste.openstack.org/show/13831618:46
jogoadam_g: trying directxman12's command18:46
*** achanda has joined #openstack-qa18:47
*** nfedotov has joined #openstack-qa18:47
*** mkollaro has quit IRC18:51
openstackgerritAlok Kumar Maurya proposed openstack/tempest: Basic networks Scenario Test Enhancements  https://review.openstack.org/10662018:52
adam_gjogo, stacking is done, theres an instance running on multi2. have at it18:52
*** achanda_ has joined #openstack-qa18:52
*** achanda has quit IRC18:52
*** johnthetubaguy is now known as zz_johnthetubagu18:53
mesteryHi QA folks! I just need one more core for this simple patch which enables us to move back to restoring the OpenDaylight CI: https://review.openstack.org/#/c/134309/ Any takers? Thanks!18:54
jogoadam_g: just produced the expected error18:57
adam_gjogo, cool18:57
*** harlowja is now known as harlowja_away18:58
adam_gjogo, are you gonna auth root<->root ?18:58
*** mwagner_lap has quit IRC18:59
directxman12jogo: did it work?18:59
jogodirectxman12: yes!19:00
jogoadded the same key to root and it worked19:00
directxman12jogo: good.  Glad I could help.19:01
sdaguemtreinish: also at the top of the stack of patches, we get to drop lxml from tempest - https://review.openstack.org/#/c/137113/19:01
jogodirectxman12: thanks19:01
jogohmm migration back failed19:01
jogooh I think I know why though19:01
directxman12jogo: did you set up the keys to got the other way?19:01
jogodirectxman12: host key failed19:02
jogomanually fixed that19:03
jogolooks like migration  back is working19:03
adam_gjogo, oh yeah-i didnt prep /etc/ssh with that stuff19:03
adam_gbut cool19:03
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13650419:03
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Add ssh key to root user  https://review.openstack.org/13717619:03
directxman12jogo: ok, good19:03
*** achanda_ has quit IRC19:04
jogoalthough I may have found an error in the tempest test19:04
jogolive migrate doesn't happen right away19:04
*** achanda has joined #openstack-qa19:04
jogoand the host status isn't changed instantly19:04
*** achanda has quit IRC19:05
jogoalthough migrating status does19:06
sdaguejogo: it is completely unsurprising that there are bugs in a test which isn't run :)19:07
*** achanda has joined #openstack-qa19:07
jogosdague: actually the test may be OK19:07
*** harlowja_away is now known as harlowja19:07
jogosdague: it waits for ACTIVE state19:07
directxman12jogo: hmm... it should happen pretty quickly, although depending on several factors it might take a little bit to finish up19:07
jogobut we will find out soon19:07
directxman12jogo: (the live migration itself, that is)19:08
jogodirectxman12: yeah live migration takes <1 minute in my test19:08
jogobut > 1 second19:08
yfriedsalv-orlando: ping19:09
salv-orlandoaloha yfried19:09
yfriedsalv-orlando: you are familiar with the floating-ip op-status (I believe you are the one who implemented it), right19:10
jogodirectxman12: thanks again for your help19:11
jogothat saved a lot of debugging19:11
*** markmcclain has joined #openstack-qa19:11
yfriedsalv-orlando: if the router is going down (set admin_state_up=False) by admin, shouldn't fip-op-status be DOWN as well?19:11
yfriedmarkmcclain: ^19:11
jogotesting it all together now: https://jenkins02.openstack.org/job/check-tempest-dsvm-aiopcpu/33/console19:11
directxman12jogo: no problem.  I'm always glad to help19:12
jogoclarkb: btw I am still seeing the second node not be sshable fairly often19:12
clarkbjogo: ya I think thats a nodepool bug19:12
clarkbpretty low on the priority list right now19:12
jogokk long as its known19:13
markmcclainyfried: eventually yes, but right now I don't believe that properly happens19:13
clarkbI think nodepool is deleting the subnode before deleting the primary node19:13
jogoclarkb: I think we are realistically at least a week or two off from even considering moving this test out of experimental19:13
clarkbjust need to change order of ops so that primary node is removed from jenkins first19:14
markmcclainyfried: for Kilo we're working on correcting issues like that19:14
yfriedmarkmcclain: should I file a bug on it?19:14
clarkbjogo: I disagree. I agree on your migration stuff though19:14
clarkbjogo: but thats one test out of a bunch19:14
markmcclainyfried: I think we've already got one for it19:14
jogoclarkb: true, the smoke test version is pretty solid19:15
yfriedmarkmcclain: are you sure?19:15
*** btran has joined #openstack-qa19:15
markmcclainyfried: not a 100%, but that class of problem is one we're working to fix in the API refactoring work19:16
yfriedmarkmcclain: I'm filing a bug. feel free to mark as duplicate. it should serve for a test that we are working on19:16
*** achanda_ has joined #openstack-qa19:19
*** achanda has quit IRC19:20
*** X019 has quit IRC19:20
markmcclainyfried: that sounds good19:21
yfriedmarkmcclain: https://bugs.launchpad.net/neutron/+bug/139631019:22
openstackgerritJoshua Harlow proposed openstack-infra/elastic-recheck: Add query for bug 1395368  https://review.openstack.org/13665719:23
jogoclarkb: example failure https://jenkins02.openstack.org/job/check-tempest-dsvm-aiopcpu/33/console19:26
jogoclarkb: and neutron version failed for same reason19:27
yfriedmarkmcclain: salv-orlando your review of this test would help raise its profile. it's a pretty simple one https://review.openstack.org/#/c/106620/719:27
clarkbjogo: just use the ssh keys provided by nodepool19:27
clarkbjogo: they are already one every host19:28
jogoclarkb: derp19:28
jogoclarkb: that was my bug19:28
jogoclarkb: and its the same on all machines?19:31
clarkbyes19:31
*** suneelb has quit IRC19:31
*** suneelb has joined #openstack-qa19:31
*** suneelb has quit IRC19:32
*** matrohon has joined #openstack-qa19:32
*** suneelb has joined #openstack-qa19:34
*** mwagner_lap has joined #openstack-qa19:35
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13650419:36
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Add ssh key to root user  https://review.openstack.org/13717619:36
jogoclarkb: ^19:36
*** adalbas has quit IRC19:37
*** matrohon has quit IRC19:37
*** matrohon has joined #openstack-qa19:38
*** tosky has quit IRC19:40
*** mkollaro has joined #openstack-qa19:44
*** directxman12 has quit IRC19:44
openstackgerritAndrew Laski proposed openstack/tempest: Expand service validation to work for Nova cells  https://review.openstack.org/13718119:47
*** ranger81 has quit IRC19:48
*** toure has quit IRC19:48
*** achanda_ has quit IRC19:49
yfriedmtreinish: ping19:53
openstackgerritYair Fried proposed openstack/tempest: Adds scenario for network-large-ops  https://review.openstack.org/13678919:54
*** matrohon has quit IRC19:54
mtreinishyfried: pong19:58
yfriedmtreinish: https://review.openstack.org/13678919:58
yfriedmtreinish: this requires a new gate19:59
openstackgerritAlok Kumar Maurya proposed openstack/tempest: Basic networks Scenario Test Enhancements  https://review.openstack.org/10662019:59
yfriedmtreinish: is it possible to add one as a non-voting?20:00
yfriedmtreinish: I'm unfamiliar with the process20:00
*** ranger81 has joined #openstack-qa20:00
yfriedmtreinish: I would also like to see it replace the large-ops-neutron gates...20:00
*** suneelb has quit IRC20:01
mtreinishyfried: yes, you can add it non-voting, But I'm not sure we want that running everywhere. What are you trying to catch with that test?20:01
mtreinishjogo: ^^^20:01
*** suneelb has joined #openstack-qa20:02
yfriedmtreinish: see that neutron<->nova ops are able to handle multiple boot operations20:03
yfriedmtreinish: I'm not sure that 100 is the "sweet spot" for this. I'd settle for 50 and count my blessing if it passes20:04
yfriedmtreinish: this would be a good job to run in periodic post merge jobs we've discussed in Paris as it stresses both components greatly20:05
mtreinishyfried: so I understand that goal, but we can barely handle the like 10 we do during a normal gate run, and ssh failures are almost impossible to debug20:05
mtreinishI don't think this is something we want to run as part of check or gate, you can probably add it as an experimental job on neutron (and maybe tempest and nova)20:05
yfriedmtreinish: agreed20:06
*** achanda has joined #openstack-qa20:07
yfriedmtreinish: but If we get stuck on ssh, that would be great. this would mean that all VMs have reached ACTIVE state with network20:07
*** andreykurilin_ has quit IRC20:07
*** andreykurilin_ has joined #openstack-qa20:08
*** aysyd has quit IRC20:08
yfriedmtreinish: also, there's a tricky issue I've never logged because I as unable to properly document it - booting multiple VMs occasionally results in key injection issues. having this as non-voting job would help us/me track this better20:09
yfriedmtreinish: ^ this could be the source of the gate ssh issues that comes and goes20:10
*** shashankhegde has quit IRC20:10
*** om has joined #openstack-qa20:10
yfriedmtreinish: so, since the current gate won't run this job, what's the process? do we merge it and then create a gate for it in devsatck-gate or is there another flow?20:11
mtreinishyfried: it can't land until it's shown running somewhere by a ci system20:11
*** achanda has quit IRC20:11
mtreinishso you add the job as experimental20:12
mtreinishthen show it works there and it'll be able to land20:12
*** achanda has joined #openstack-qa20:12
yfriedmtreinish: but in order for it to be added as experimental it needs to land first, doesn't it???20:12
*** mwagner_lap has quit IRC20:13
yfriedmtreinish: how did you add the "large-ops" jobs to begin with?20:13
*** mwagner_lap has joined #openstack-qa20:13
*** jgrimm is now known as zz_jgrimm20:13
mtreinishyfried: you just add a job definition to the openstack-infra/project-config repo20:14
yfriedmtreinish: but will it land there, if the test doesn't exist?20:15
yfriedmtreinish: do you have an example?20:15
sdaguemtreinish: is there an across the board fix for the SG race?20:15
mtreinishyfried: its just a configuration repo, it doesn't run tempest20:16
mtreinishsdague: which sg race? But, my guess is no :)20:16
sdaguethe not waiting for sg to apply before we try to test connectivity20:16
mtreinishsdague: oh, yeah I just added a wait to the test for that, it seemed to fix the issue clarkb was seeing in multinode testing20:17
sdagueright, so I believe it's part of the issue we see in other tests20:17
sdagueso I'd like to see it addressed everywhere if possible20:17
mtreinishsdague: http://git.openstack.org/cgit/openstack/tempest/commit/?id=ff31bccce8f166d43e351e85a2e13d8d5277977220:17
sdagueok, but what about all the other tests that do that?20:18
yfriedmtreinish: dou you have an example for the experimental patch?20:18
mtreinishsdague: I think it was specific to that test adding a sg to an already booted instance20:18
*** zz_jgrimm is now known as jgrimm20:18
mtreinishIIRC the other tests set it before boot20:18
mtreinishyfried: you'll have to give me a min to dig one up from the git log20:19
mtreinishyfried: you can just look at the git log for examples too20:19
*** jgrimm is now known as zz_jgrimm20:19
yfriedmtreinish: tnx20:19
sdaguemtreinish: ah, ok20:19
*** achanda has quit IRC20:19
mtreinishsdague: I would double check though, I might be wrong20:20
*** achanda has joined #openstack-qa20:21
*** rushiagr is now known as rushiagr_away20:22
*** andreykurilin_ has quit IRC20:24
*** andreykurilin_ has joined #openstack-qa20:24
*** achanda has quit IRC20:24
*** achanda has joined #openstack-qa20:25
yfriedmtreinish: I see how to add an existing job the the experimental part. but how do I create a job? I'm no too familiar with JJB20:27
*** ranger81 has quit IRC20:29
yfriedjogo: maybe you are familiar with the process of creating a new job (experimental) in openstack-infra/project-config?20:31
openstackrecheckopenstack-dev/devstack change: https://review.openstack.org/136215 failed because of: gate-grenade-dsvm-partial-ncpu: unrecognized error, gate-tempest-dsvm-neutron-full: https://bugs.launchpad.net/bugs/137351320:33
*** e0ne has joined #openstack-qa20:45
*** nfedotov has left #openstack-qa20:45
*** mlavalle has joined #openstack-qa20:49
*** dpaterson has joined #openstack-qa20:52
*** zz_jgrimm is now known as jgrimm20:53
openstackgerritMerged openstack-dev/devstack: switch to use ceilometer specific connections  https://review.openstack.org/13571520:53
*** yfried is now known as yfried|afk20:53
*** jgrimm is now known as zz_jgrimm20:53
*** om has quit IRC20:54
*** shashankhegde has joined #openstack-qa20:55
openstackgerritMerged openstack-dev/devstack: Install bash completion for glanceclient  https://review.openstack.org/13594121:01
*** yfried|afk is now known as yfried21:02
openstackrecheckopenstack/cinder change: https://review.openstack.org/132698 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/135396221:03
jgriffithsdague: you or anybody have a lead on the clients failing for neutron on stable/icehouse?21:06
*** yfried has quit IRC21:07
mtreinishjgriffith: https://review.openstack.org/#/c/135903/ should fix it21:07
jgriffithmtreinish: ahh.. very nice21:08
jgriffithmtreinish: thanks!21:08
*** ranger81 has joined #openstack-qa21:08
jogogot live migrate working (partially)21:10
mtreinishjogo: what's partially?21:10
jogomtreinish: working on one direction21:10
jogomessed up the permissions of the ssh keys in the other direction21:10
mtreinishah, ok21:10
jogomtreinish: http://logs.openstack.org/04/136504/5/experimental/check-tempest-dsvm-aiopcpu/344a169/logs/testr_results.html.gz21:11
salv-orlandoyfried: a miracle happened and I reviewed a patch you requested (https://review.openstack.org/#/c/106620/) in less than 2 hours from your request21:15
*** X019 has joined #openstack-qa21:17
*** jamielennox is now known as jamielennox|away21:23
*** dkranz has quit IRC21:25
*** achanda has quit IRC21:28
jogoclarkb: cp: cannot stat ‘/etc/nodepool/id_rsa’: No such file or directory21:28
jogoclarkb: is id_rsa on both machines? or only one?21:28
*** jamielennox|away is now known as jamielennox21:28
openstackgerritSean M. Collins proposed openstack-dev/devstack: lib/ironic: Ensure sgabios package is installed before use  https://review.openstack.org/13721621:30
clarkbshould be all I thought. check docs21:32
*** X019 has quit IRC21:32
*** zz_jgrimm is now known as jgrimm21:32
jogoclarkb: docs and code point to it being both21:34
jogobut i don't know the codebase very well21:34
jogomaybe something is wrong with my code: https://review.openstack.org/#/c/137176/2/sub_node_prepare.sh,cm21:36
*** toure_ has joined #openstack-qa21:40
*** toure_ has quit IRC21:40
*** ajo has quit IRC21:41
openstackgerritMerged openstack/tempest: rm pyc files before test runs  https://review.openstack.org/13705521:44
*** e0ne has quit IRC21:47
*** ajo has joined #openstack-qa21:50
openstackgerritSergey Lukjanov proposed openstack-dev/devstack: Explicitly set Sahara service type to change it  https://review.openstack.org/13722021:51
*** matrohon has joined #openstack-qa21:53
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13650421:54
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Add ssh key to root user  https://review.openstack.org/13717621:54
openstackgerritAdam Gandelman proposed openstack-infra/devstack-gate: Enable nova-api-metadata for multi host nova-network  https://review.openstack.org/13617922:14
*** matrohon has quit IRC22:16
openstackgerritAdam Gandelman proposed openstack-infra/devstack-gate: Enable nova-api-metadata for multi host nova-network  https://review.openstack.org/13617922:16
*** dimsum__ has quit IRC22:16
*** dimsum__ has joined #openstack-qa22:17
*** melwitt has quit IRC22:17
*** melwitt has joined #openstack-qa22:18
*** JayJ has joined #openstack-qa22:18
*** yamamoto has joined #openstack-qa22:21
*** dimsum__ has quit IRC22:21
openstackgerritIan Wienand proposed openstack-dev/devstack: Move sgabios setup to hypervisor-ironic  https://review.openstack.org/13643622:22
*** yamamoto has quit IRC22:25
*** achanda has joined #openstack-qa22:28
*** mkollaro has quit IRC22:37
*** achanda has quit IRC22:39
*** e0ne has joined #openstack-qa22:42
*** e0ne has quit IRC22:42
openstackgerritDolph Mathews proposed openstack-infra/elastic-recheck: add CLI command for executing queries  https://review.openstack.org/13723422:47
*** dims_ has joined #openstack-qa22:49
adam_gjogo, you done with those cloud instances, gonna repurpose them22:50
*** dustins has quit IRC22:52
*** enikanorov_ has quit IRC23:01
*** enikanorov_ has joined #openstack-qa23:01
*** andreykurilin_ has quit IRC23:01
openstackgerritDolph Mathews proposed openstack-infra/elastic-recheck: add CLI command for executing elastic-recheck queries  https://review.openstack.org/13723423:01
openstackgerritChmouel Boudjnah proposed openstack/qa-specs: Add devstack external plugin spec  https://review.openstack.org/13705423:02
*** shashankhegde_ has joined #openstack-qa23:03
*** achanda has joined #openstack-qa23:03
openstackgerritChmouel Boudjnah proposed openstack/qa-specs: Add devstack external plugin spec  https://review.openstack.org/13705423:03
openstackgerritDolph Mathews proposed openstack-infra/elastic-recheck: add CLI command for executing elastic-recheck queries  https://review.openstack.org/13723423:05
*** shashankhegde has quit IRC23:06
*** shashankhegde_ is now known as shashankhegde23:06
*** mlavalle has quit IRC23:09
*** gfidente has quit IRC23:12
openstackgerritAdam Gandelman proposed openstack/qa-specs: Improve diskimage-builder in devstack  https://review.openstack.org/13610123:12
*** markmcclain1 has joined #openstack-qa23:15
*** markmcclain has quit IRC23:15
*** achanda has quit IRC23:17
jogoadam_g: go for it, thanks23:22
*** achanda has joined #openstack-qa23:23
*** X019 has joined #openstack-qa23:28
*** markmcclain1 has quit IRC23:29
*** markmcclain has joined #openstack-qa23:29
*** markmcclain has quit IRC23:30
openstackgerritSergey Shnaidman proposed openstack/tempest: Support for IPv6 for API continue  https://review.openstack.org/13723923:36
openstackgerritSergey Shnaidman proposed openstack/tempest: Support for IPv6 for API continue  https://review.openstack.org/13723923:39
*** tochi has joined #openstack-qa23:39
*** shashankhegde has quit IRC23:42
*** X019 has quit IRC23:46
*** oomichi has joined #openstack-qa23:46
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: testing aiopcpu with tempest full  https://review.openstack.org/13650423:47
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Set up ssh_known_host based on hostname  https://review.openstack.org/13659623:47
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Setup /etc/hosts for aiopcpu  https://review.openstack.org/13615823:47
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Add ssh key to root user  https://review.openstack.org/13717623:47
jogothat should do it23:47
jogoadam_g: I think I am very close if not there23:48
adam_gjogo, nice23:48
*** melwitt has quit IRC23:50
openstackgerritSergey Shnaidman proposed openstack/tempest: Support for IPv6 for API continue  https://review.openstack.org/13723923:51
openstackgerritSergey Shnaidman proposed openstack/tempest: Support for IPv6 for API continue  https://review.openstack.org/13723923:52
*** ajo has quit IRC23:54
*** rbak has quit IRC23:55

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