Monday, 2014-10-13

*** baoli has joined #openstack-infra00:00
*** mriedem has quit IRC00:03
*** achanda has quit IRC00:03
*** baoli has quit IRC00:04
*** hdd has joined #openstack-infra00:07
*** baoli has joined #openstack-infra00:10
*** ChuckC___ has joined #openstack-infra00:13
*** baoli has quit IRC00:14
*** baoli has joined #openstack-infra00:30
*** baoli has quit IRC00:34
*** tsg has left #openstack-infra00:47
ianwclarkb / fungi / modred : he's one idea for the cenots7 xattr tar issue -- build a version and use it.  this is a problem that disappears as we move to later distros00:54
*** harlowja_at_home has joined #openstack-infra00:59
*** unicell has joined #openstack-infra01:01
clarkbianw I think I will just have to upgrade the nodepool seevee01:03
*** yaguang has joined #openstack-infra01:03
clarkbianw really I dont know why we qcow2 to tar then back again01:04
ianwclarkb: well, let me know if i can help in any way01:08
ianwclarkb: working around might give us confidence there's no more hidden issues.  although wrangling +2's and getting a release might be more difficult than upgrading01:10
*** baoli has joined #openstack-infra01:11
clarkbya my change is completely derped by ramdisk element01:12
*** harlowja_at_home has quit IRC01:17
*** r1chardj0n3s is now known as r1chardj0n3s_afk01:19
ianwclarkb: yeah, that's another one for the test matrix :(01:20
*** otherwiseguy has joined #openstack-infra01:22
*** hdd has quit IRC01:29
*** otherwiseguy has quit IRC01:33
*** bknudson has quit IRC01:33
*** achanda has joined #openstack-infra01:35
*** otherwiseguy has joined #openstack-infra01:36
*** r1chardj0n3s_afk is now known as r1chardj0n3s01:48
*** baoli has quit IRC01:51
*** baoli has joined #openstack-infra01:52
*** baoli has quit IRC01:57
*** otherwiseguy has quit IRC02:00
*** jpeeler has joined #openstack-infra02:03
openstackgerritDavanum Srinivas (dims) proposed a change to openstack/requirements: suds-jurko - Alternative to suds library for py 3.x support
*** bradjones has quit IRC02:12
*** bswartz has joined #openstack-infra02:13
*** otter768 has joined #openstack-infra02:14
*** mwagner_lap has joined #openstack-infra02:14
*** koolhead17 has joined #openstack-infra02:18
*** baoli has joined #openstack-infra02:21
*** dims has quit IRC02:22
*** baoli_ has joined #openstack-infra02:23
*** baoli has quit IRC02:26
*** baoli has joined #openstack-infra02:27
*** baoli_ has quit IRC02:28
*** zhiyan|afk is now known as zhiyan02:28
*** Longgeek has joined #openstack-infra02:30
*** baoli has quit IRC02:37
*** hdd has joined #openstack-infra02:47
*** baoli has joined #openstack-infra02:49
*** Longgeek has quit IRC02:50
*** david-lyle has joined #openstack-infra02:51
*** baoli has quit IRC02:54
*** shashankhegde has joined #openstack-infra02:55
*** krtaylor has quit IRC03:05
*** krtaylor_ has joined #openstack-infra03:05
*** baoli has joined #openstack-infra03:08
*** baoli_ has joined #openstack-infra03:09
*** baoli has quit IRC03:12
*** baoli_ has quit IRC03:19
*** ashaeron has joined #openstack-infra03:24
*** Longgeek has joined #openstack-infra03:27
*** shashankhegde has quit IRC03:34
*** r1chardj0n3s is now known as r1chardj0n3s_afk03:36
*** baoli has joined #openstack-infra03:45
*** baoli has quit IRC03:50
*** otter768 has quit IRC03:51
*** koolhead17 has quit IRC03:52
*** yfried|afk is now known as yfried04:08
*** shashankhegde has joined #openstack-infra04:11
*** r1chardj0n3s_afk is now known as r1chardj0n3s04:12
*** baoli has joined #openstack-infra04:13
*** baoli_ has joined #openstack-infra04:14
*** baoli has quit IRC04:18
*** baoli_ has quit IRC04:18
*** armax has joined #openstack-infra04:22
openstackgerritWayne Warren proposed a change to openstack-infra/jenkins-job-builder: Accept colon-delimited list of paths in args.
*** baoli has joined #openstack-infra04:30
*** baoli has quit IRC04:35
*** koolhead17 has joined #openstack-infra04:44
*** ashaeron has quit IRC04:46
*** armax has quit IRC04:53
*** _nadya_ has joined #openstack-infra04:56
*** zaphod42 has quit IRC05:00
*** baoli_ has joined #openstack-infra05:00
*** zaphod42 has joined #openstack-infra05:00
*** yfried is now known as yfried|afk05:03
*** baoli_ has quit IRC05:04
*** zaphod42 has quit IRC05:05
*** hdd has quit IRC05:08
*** yfried|afk is now known as yfried05:12
*** chandankumar has joined #openstack-infra05:12
*** Longgeek has quit IRC05:16
*** david-lyle has quit IRC05:21
*** remote_morgan_ has quit IRC05:24
*** baoli has joined #openstack-infra05:26
*** afazekas has quit IRC05:28
*** yfried is now known as yfried|afk05:29
*** yfried|afk is now known as yfried05:30
*** baoli has quit IRC05:30
*** ivar-lazzaro has joined #openstack-infra05:33
*** rushiagr_away is now known as rushiagr05:33
*** yfried is now known as yfried|afk05:42
*** Longgeek has joined #openstack-infra05:46
*** armax has joined #openstack-infra05:48
*** loki184 has joined #openstack-infra05:49
*** armax has quit IRC05:49
*** yfried|afk is now known as yfried05:52
*** baoli_ has joined #openstack-infra05:55
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Make JJB python 3 compatible
*** ashaeron has joined #openstack-infra05:56
*** ivar-lazzaro has quit IRC05:57
*** baoli_ has quit IRC06:00
*** armax has joined #openstack-infra06:01
*** yfried is now known as yfried|afk06:05
*** shashankhegde has quit IRC06:05
*** afazekas has joined #openstack-infra06:09
*** yolanda has joined #openstack-infra06:11
*** enikanorov__ has joined #openstack-infra06:14
*** k4n0 has joined #openstack-infra06:16
*** enikanorov has quit IRC06:16
*** katyafervent_awa is now known as katyafervent06:18
*** Longgeek has quit IRC06:20
*** jcoufal has joined #openstack-infra06:21
*** _nadya_ has quit IRC06:22
*** HeOS has quit IRC06:23
*** baoli_ has joined #openstack-infra06:23
*** mmedvede has quit IRC06:23
*** armax has quit IRC06:24
*** Longgeek has joined #openstack-infra06:24
*** yfried|afk is now known as yfried06:25
*** baoli_ has quit IRC06:27
*** achanda_ has joined #openstack-infra06:33
*** dizquierdo has joined #openstack-infra06:36
*** achanda has quit IRC06:37
*** camunoz has quit IRC06:50
*** _nadya_ has joined #openstack-infra06:53
*** camunoz has joined #openstack-infra06:53
*** achanda_ has quit IRC06:58
*** _nadya_ has quit IRC07:01
*** chandankumar has quit IRC07:01
*** armax has joined #openstack-infra07:10
*** chandankumar has joined #openstack-infra07:10
*** baoli_ has joined #openstack-infra07:10
openstackgerritJaroslav Henner proposed a change to openstack-infra/jenkins-job-builder: Allow multiple comment-added events in gerrit trig.
*** e0ne has joined #openstack-infra07:13
*** mpaolino has joined #openstack-infra07:14
*** michchap has quit IRC07:14
*** armax has quit IRC07:15
*** jlibosva has joined #openstack-infra07:15
*** baoli_ has quit IRC07:15
*** michchap has joined #openstack-infra07:17
*** HeOS has joined #openstack-infra07:23
*** ihrachyshka has joined #openstack-infra07:24
*** achanda has joined #openstack-infra07:27
*** camunoz is now known as camunoz_gone07:29
*** luqas has joined #openstack-infra07:31
*** ttx has quit IRC07:33
*** ttx has joined #openstack-infra07:34
*** ttx has quit IRC07:34
*** ttx has joined #openstack-infra07:34
*** Guest51022 has joined #openstack-infra07:35
*** liu has joined #openstack-infra07:35
*** coolsvap has joined #openstack-infra07:36
*** coolsvap has quit IRC07:36
*** baoli has joined #openstack-infra07:36
liuHello,I am a newer to openstack infra07:37
*** coolsvap has joined #openstack-infra07:37
liuAnd I have a simple question: Does one company can only have one third party CI account?07:38
*** arxcruz has joined #openstack-infra07:39
*** coolsvap has quit IRC07:41
*** baoli has quit IRC07:41
*** achanda has quit IRC07:41
*** coolsvap has joined #openstack-infra07:41
*** coolsvap has quit IRC07:43
*** coolsvap has joined #openstack-infra07:45
*** dizquierdo has quit IRC07:45
*** yfried is now known as yfried|afk07:49
*** salv-orlando has joined #openstack-infra07:49
*** viktors|afk is now known as viktors07:52
*** _nadya_ has joined #openstack-infra07:53
*** jistr has joined #openstack-infra07:54
*** mmaglana has quit IRC07:57
*** derekh has joined #openstack-infra07:57
*** jlibosva has quit IRC07:58
*** jlibosva has joined #openstack-infra07:59
*** andreykurilin_ has joined #openstack-infra08:01
*** yfried|afk is now known as yfried08:02
*** jpich has joined #openstack-infra08:05
*** baoli_ has joined #openstack-infra08:06
*** ihrachyshka_ has joined #openstack-infra08:08
*** ihrachyshka has quit IRC08:08
*** baoli_ has quit IRC08:10
*** imcsk8 has quit IRC08:12
*** imcsk8 has joined #openstack-infra08:12
*** jistr has quit IRC08:14
*** andreykurilin_ has quit IRC08:15
*** jistr has joined #openstack-infra08:18
*** jlibosva has quit IRC08:23
*** dizquierdo has joined #openstack-infra08:23
*** jlibosva has joined #openstack-infra08:23
*** jedimike has joined #openstack-infra08:24
*** alexpilotti has joined #openstack-infra08:32
*** baoli has joined #openstack-infra08:32
*** baoli has quit IRC08:36
*** jgallard has joined #openstack-infra08:37
*** ihrachyshka_ is now known as ihrachyshka08:40
*** baoli_ has joined #openstack-infra08:40
*** IvanBerezovskiy has joined #openstack-infra08:42
BobBallno liu08:44
BobBallCitrix has atleast two08:44
BobBallwe have one per useful test - one for Nova and a seperate one for Neutorn08:45
*** flaper87|afk is now known as flaper8708:45
*** baoli_ has quit IRC08:45
liuthanks BobBal08:46
liuwe just decided to use the only one account for CI test08:46
*** coolsvap has quit IRC08:47
*** pkoniszewski has joined #openstack-infra08:47
*** Longgeek has quit IRC08:47
*** Longgeek has joined #openstack-infra08:47
*** k4n0 has quit IRC08:50
*** ildikov has joined #openstack-infra08:53
*** mmaglana has joined #openstack-infra08:54
*** k4n0 has joined #openstack-infra08:55
*** mmaglana has quit IRC08:58
*** baoli has joined #openstack-infra08:59
*** liu has quit IRC09:00
*** baoli_ has joined #openstack-infra09:01
openstackgerritJeffrey Zhang proposed a change to openstack-infra/project-config: Add openstack-salt-formula project
*** coolsvap has joined #openstack-infra09:01
*** baoli__ has joined #openstack-infra09:04
*** baoli has quit IRC09:04
*** baoli_ has quit IRC09:05
*** yfried has quit IRC09:05
*** berendt has joined #openstack-infra09:06
*** baoli__ has quit IRC09:08
*** jp_at_hp has joined #openstack-infra09:08
*** ihrachyshka has quit IRC09:13
*** ociuhandu has joined #openstack-infra09:14
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard-webclient: Small fix for Preferences page
*** Guest51022 has quit IRC09:14
*** zz_johnthetubagu is now known as johnthetubaguy09:15
*** rlandy has joined #openstack-infra09:20
*** yfried has joined #openstack-infra09:24
*** Longgeek has quit IRC09:25
*** cdent has joined #openstack-infra09:26
*** baoli has joined #openstack-infra09:31
*** armax has joined #openstack-infra09:32
*** baoli_ has joined #openstack-infra09:33
*** yfried is now known as yfried|afk09:34
*** k4n0 has quit IRC09:35
*** baoli has quit IRC09:36
*** coolsvap has quit IRC09:36
*** chandankumar has quit IRC09:36
*** Longgeek has joined #openstack-infra09:37
*** baoli_ has quit IRC09:38
*** yfried|afk is now known as yfried09:39
*** achuprin_ has joined #openstack-infra09:39
*** pblaho has joined #openstack-infra09:41
*** amuller has joined #openstack-infra09:42
*** armax has quit IRC09:42
*** jerryz_ has joined #openstack-infra09:43
*** koolhead17 has quit IRC09:44
*** ociuhandu_ has joined #openstack-infra09:47
*** coolsvap has joined #openstack-infra09:48
*** ociuhandu has quit IRC09:49
*** ociuhandu_ is now known as ociuhandu09:49
*** dims has joined #openstack-infra09:50
*** jerryz_ has quit IRC09:50
viktorshi folks! Can somebody add +A to patch (Add PyMySQL to global requirements) which already got two +2 ?09:50
*** flaper87 is now known as flaper87|afk09:51
*** k4n0 has joined #openstack-infra09:54
*** dims has quit IRC09:55
*** k4n0 has quit IRC09:55
*** pelix has joined #openstack-infra09:55
*** mmaglana has joined #openstack-infra09:55
*** k4n0 has joined #openstack-infra09:56
*** mmaglana has quit IRC10:00
*** pcm_ has joined #openstack-infra10:00
*** ihrachyshka has joined #openstack-infra10:02
*** mfink has quit IRC10:05
*** bradjones has joined #openstack-infra10:06
*** pcm_ has quit IRC10:06
*** koolhead17 has joined #openstack-infra10:07
*** koolhead17 has joined #openstack-infra10:07
*** pcm_ has joined #openstack-infra10:07
*** yfried is now known as yfried|afk10:08
*** _nadya_ has quit IRC10:10
*** pkoniszewski has quit IRC10:10
*** qba73 has joined #openstack-infra10:12
*** baoli has joined #openstack-infra10:13
*** mmaglana has joined #openstack-infra10:13
*** baoli_ has joined #openstack-infra10:15
*** baoli has quit IRC10:17
*** mmaglana has quit IRC10:18
*** ociuhandu_ has joined #openstack-infra10:19
*** baoli_ has quit IRC10:19
*** ociuhandu has quit IRC10:19
*** ociuhandu_ is now known as ociuhandu10:19
*** koolhead17 has quit IRC10:24
*** jgallard has quit IRC10:35
*** chandankumar has joined #openstack-infra10:36
*** ociuhandu has quit IRC10:39
*** koolhead17 has joined #openstack-infra10:42
*** koolhead17 has joined #openstack-infra10:42
*** _nadya_ has joined #openstack-infra10:43
*** koolhead17 has quit IRC10:44
*** unicell has quit IRC10:46
*** _nadya_ has quit IRC10:47
*** pcm_ is now known as pc_m10:47
*** jcoufal has quit IRC10:49
*** jcoufal has joined #openstack-infra10:50
*** Longgeek has quit IRC10:53
*** yfried|afk is now known as yfried10:58
*** Longgeek has joined #openstack-infra10:59
*** koolhead17 has joined #openstack-infra10:59
*** loki184 has quit IRC11:09
*** mmaglana has joined #openstack-infra11:14
*** baoli has joined #openstack-infra11:15
*** slagle has joined #openstack-infra11:16
*** mmaglana has quit IRC11:18
*** baoli has quit IRC11:20
*** hashar has joined #openstack-infra11:21
*** koolhead17 has quit IRC11:21
pelixis it possible to turn on python 3 jobs to be voting on projects individually?11:21
*** armax has joined #openstack-infra11:25
*** koolhead17 has joined #openstack-infra11:25
*** dmakogon_ is now known as denis_makogon11:25
*** habib has joined #openstack-infra11:26
*** AJaeger has joined #openstack-infra11:26
*** AJaeger has joined #openstack-infra11:26
*** koolhead17 has quit IRC11:29
openstackgerritDarragh Bailey proposed a change to openstack-infra/project-config: Make py33 job voting for jenkins-job-builder
*** habib has quit IRC11:32
*** hashar has quit IRC11:33
*** armax has quit IRC11:38
*** cnesa18 has joined #openstack-infra11:39
*** katyafervent is now known as katyafervent_awa11:41
*** katyafervent_awa is now known as katyafervent11:43
*** rfolco has joined #openstack-infra11:44
*** cdent has quit IRC11:44
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Add support for build parameter of type "Run"
*** davinchi has joined #openstack-infra11:48
*** IvanKliuk has joined #openstack-infra11:48
davinchihi there. I'm getting this screen when trying to update contacts information
*** jistr has quit IRC11:49
*** IvanKliuk has left #openstack-infra11:49
*** davinchi is now known as pboldin11:51
*** pblaho has quit IRC11:52
*** baoli has joined #openstack-infra11:52
*** doude has joined #openstack-infra11:53
*** baoli has quit IRC11:57
*** Longgeek has quit IRC11:58
*** katyafervent is now known as katyafervent_awa11:59
*** katyafervent_awa is now known as katyafervent11:59
*** jlibosva has quit IRC11:59
*** katyafervent is now known as katyafervent_awa12:00
*** dims has joined #openstack-infra12:01
*** _nadya_ has joined #openstack-infra12:01
*** jlibosva has joined #openstack-infra12:01
*** mrmartin has joined #openstack-infra12:02
pboldinor should i use launchpad to file a bug?12:03
*** Longgeek has joined #openstack-infra12:05
AJaegerpboldin, did you follow and accepted the CLA?12:07
AJaegerpboldin, and did you join the OpenStack foundation?12:08
AJaegeryou need to use the same email address for both gerrit and foundation12:08
*** Guest51022 has joined #openstack-infra12:11
*** katyafervent_awa is now known as katyafervent12:11
*** baoli has joined #openstack-infra12:11
*** baoli has quit IRC12:11
jpichI think queorgullo on #openstack might be a spam account, I got a PM with garbage from them the second I posted something in the channel12:11
*** jistr has joined #openstack-infra12:12
*** cnesa18 has quit IRC12:12
*** dprince has joined #openstack-infra12:13
pboldinAJaeger: thank you. it would be nice to state that joining is *required* in order to submit a contact information. otherwise the errorcode and message is misleading12:14
*** packet has joined #openstack-infra12:14
*** mmaglana has joined #openstack-infra12:15
AJaegerpboldin, the wiki page says "required". Which page stated it otherwise?12:15
pboldinAJaeger: it was not. but i would make this 'required' word bold and stated explicitly that "you won't be able to fill a contact information on gerrit until you joined community"12:16
AJaegerpboldin, I agree that the process is too complicated, there're a few people working on alternatives12:16
AJaegerpboldin, I made the "required" now bold...12:17
pboldinAJaeger: thank you12:18
*** dprince has quit IRC12:18
*** dprince has joined #openstack-infra12:19
*** doug-fish has joined #openstack-infra12:19
*** weshay has joined #openstack-infra12:19
*** mmaglana has quit IRC12:20
*** _nadya_ has quit IRC12:20
*** Guest51022 has quit IRC12:20
*** k4n0 has quit IRC12:21
*** katyafervent is now known as katyafervent_awa12:21
*** _nadya_ has joined #openstack-infra12:21
*** katyafervent_awa is now known as katyafervent12:22
*** armax has joined #openstack-infra12:22
*** mika has quit IRC12:22
*** toabctl has quit IRC12:22
*** aysyd has joined #openstack-infra12:23
*** IvanBerezovskiy has quit IRC12:23
*** mika has joined #openstack-infra12:23
*** mika has joined #openstack-infra12:23
*** toabctl has joined #openstack-infra12:24
*** rakhmerov has quit IRC12:24
*** yaguang has quit IRC12:25
*** IvanBerezovskiy has joined #openstack-infra12:26
*** radez_g0n3 is now known as radez12:27
*** rakhmerov has joined #openstack-infra12:28
*** _nadya_ has quit IRC12:28
*** katyafervent is now known as katyafervent_awa12:30
openstackgerritEvgeniy L proposed a change to openstack-infra/project-config: Fuel-plugins repository for Fuel
*** mfink has joined #openstack-infra12:34
*** kgiusti has joined #openstack-infra12:36
*** yfried is now known as yfried|afk12:36
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: fix for Bug #1366690, handling windows command prompt properly
uvirtbotLaunchpad bug 1366690 in git-review "Failure when pushing many commits with certain UTF-8 characters" [Undecided,Fix committed]
*** bswartz has quit IRC12:39
*** bknudson has joined #openstack-infra12:39
*** yfried|afk is now known as yfried12:43
*** slagle_ has joined #openstack-infra12:44
openstackgerritA change was merged to openstack-infra/storyboard: Disable login for users
*** mriedem has joined #openstack-infra12:48
openstackgerritA change was merged to openstack-infra/storyboard: Added functional tests for Project api.
*** baoli has joined #openstack-infra12:48
*** jgallard has joined #openstack-infra12:49
*** habib has joined #openstack-infra12:51
openstackgerritA change was merged to openstack-infra/storyboard: Search Project by Project Group
*** cpowell has joined #openstack-infra12:56
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: fix for Bug #1366690, handling windows command prompt properly
uvirtbotLaunchpad bug 1366690 in git-review "Failure when pushing many commits with certain UTF-8 characters" [Undecided,Fix committed]
*** bswartz has joined #openstack-infra13:01
*** mase_x200 has joined #openstack-infra13:04
*** slagle has quit IRC13:05
*** slagle_ is now known as slagle13:06
*** dustins has joined #openstack-infra13:07
*** nkinder has quit IRC13:10
*** _nadya_ has joined #openstack-infra13:11
*** zz_jgrimm is now known as jgrimm13:11
*** mmedvede has joined #openstack-infra13:11
*** dims has quit IRC13:14
*** jcoufal has quit IRC13:15
*** dims has joined #openstack-infra13:15
*** homeless has joined #openstack-infra13:15
*** katyafervent_awa is now known as katyafervent13:15
*** jcoufal has joined #openstack-infra13:15
*** mmaglana has joined #openstack-infra13:16
*** mmaglana has quit IRC13:20
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: fix for Bug #1366690, handling windows command prompt properly
uvirtbotLaunchpad bug 1366690 in git-review "Failure when pushing many commits with certain UTF-8 characters" [Undecided,Fix committed]
*** jlibosva has quit IRC13:22
*** _buzz has joined #openstack-infra13:22
*** jlibosva has joined #openstack-infra13:23
*** nelsnelson has joined #openstack-infra13:25
*** armax has quit IRC13:28
*** ildikov is now known as ildikov_afk13:28
*** nelsnelson has quit IRC13:31
*** erlon_ has joined #openstack-infra13:32
*** sdake has joined #openstack-infra13:32
*** sdake has quit IRC13:32
*** sdake has joined #openstack-infra13:32
*** nelsnelson has joined #openstack-infra13:33
*** berendt has quit IRC13:34
*** plars has quit IRC13:34
*** sdake has quit IRC13:34
*** sdake has joined #openstack-infra13:35
*** mase_x200 has quit IRC13:35
*** plars has joined #openstack-infra13:36
*** mpaolino has quit IRC13:38
*** otherwiseguy has joined #openstack-infra13:38
*** ashaeron has quit IRC13:41
*** habib has quit IRC13:41
*** habib has joined #openstack-infra13:42
*** yfried is now known as yfried|afk13:42
*** yfried|afk is now known as yfried13:43
*** rushiagr is now known as rushiagr_away13:48
*** julim has joined #openstack-infra13:48
*** JayJ has joined #openstack-infra13:49
*** JayJ has quit IRC13:51
*** vhoward has joined #openstack-infra13:51
*** mpaolino has joined #openstack-infra13:52
*** katyafervent is now known as katyafervent_awa13:52
*** bcourt has joined #openstack-infra13:52
*** otherwiseguy has quit IRC13:53
*** coolsvap has quit IRC13:53
*** rainya has joined #openstack-infra13:55
*** rainya has quit IRC13:55
*** mpaolino has quit IRC13:57
*** mfer has joined #openstack-infra13:58
openstackgerritA change was merged to openstack-infra/git-review: Align git-review and python -m git_review.cmd behaviors
*** nkinder has joined #openstack-infra13:59
*** hdd has joined #openstack-infra13:59
*** chuckC_ has quit IRC14:01
openstackgerritA change was merged to openstack-infra/git-review: -F/--force-rebase has no effect if rebase is disabled by config
openstackgerritJulien Danjou proposed a change to openstack-infra/project-config: Import monitoring-for-openstack
*** koolhead17 has joined #openstack-infra14:05
*** Guest20248 is now known as redrobot14:06
*** habib has quit IRC14:06
*** habib has joined #openstack-infra14:07
openstackgerritA change was merged to openstack-dev/hacking: Updated from global requirements
*** xyang0 has joined #openstack-infra14:08
*** ChuckC___ has quit IRC14:08
*** chuckC_ has joined #openstack-infra14:08
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Consolidated functional test data bootstrap
*** miqui has joined #openstack-infra14:13
*** mmaglana has joined #openstack-infra14:16
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Consolidated functional test data bootstrap
*** thedodd has joined #openstack-infra14:20
*** tonytan4ever has joined #openstack-infra14:20
*** marcusvrn_ has joined #openstack-infra14:20
*** markmcclain has joined #openstack-infra14:21
*** mmaglana has quit IRC14:21
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Filter tasks by project, project_group, and various other fields
*** rwsu has joined #openstack-infra14:22
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: Handling windows command prompt properly
*** atiwari has joined #openstack-infra14:23
*** marun has joined #openstack-infra14:24
*** cnesa18 has joined #openstack-infra14:25
openstackgerritCedric Brandily proposed a change to openstack-infra/git-review: Prefer git-config over git-review config files
*** beekneemech is now known as bnemec14:25
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Filter stories by project_id, status, project_group_id, more.
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Consolidated functional test data bootstrap
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Filter tasks by project, project_group, and various other fields
*** unicell has joined #openstack-infra14:26
*** david-lyle has joined #openstack-infra14:26
*** _nadya_ has quit IRC14:31
*** chandankumar has quit IRC14:34
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: Handling windows command prompt properly
*** packet has quit IRC14:36
*** doug-fish1 has joined #openstack-infra14:36
*** doug-fish has quit IRC14:36
*** lttrl has joined #openstack-infra14:38
*** zz_dimtruck has joined #openstack-infra14:40
*** zz_dimtruck is now known as dimtruck14:40
*** signed8bit has joined #openstack-infra14:41
*** armax has joined #openstack-infra14:42
AJaegerfungi, clarkb, are you around?14:46
*** chandankumar has joined #openstack-infra14:46
* fungi is14:46
fungi(quietly trying to get work done)14:46
AJaegermorning, fungi!14:46
AJaegerOh, sorry...14:46
AJaegerannegentle tagged a few hours ago openstack-doc-tools 0.20 - and now all builds fail for the manuals ;(14:47
fungino need to apologize. if you need something, that's why i'm here14:47
AJaeger is one failure...14:47
AJaegerfungi, great, thanks!14:47
AJaegerdo you have an idea why it can't download it?14:47
*** rwsu has quit IRC14:47
clarkbI am almost awake14:48
fungiAJaeger: checking14:49
AJaegertake your time, clarkb ;)14:49
fungiAJaeger: at first glance, bandersnatch mirrored the tarball but not the wheel, but indexed both and the job is failing trying to retrieve the latter14:49
* mordred is not awake really at all14:49
mordredfungi: lovely14:49
*** thedodd has quit IRC14:49
AJaegerfungi, yes, that explains the failure.14:50
*** pcrews has joined #openstack-infra14:50
fungiseems i can download the wheel from pypi.python.org14:50
fungiprobably either or but i'll see if i can narrow down the symptoms14:51
*** packet has joined #openstack-infra14:52
AJaegerfungi:  I'll cycle home in 5 mins and send an email to the doc list telling them about the issue. If there's anything I need to do, please tell me, I'll scrollback later14:52
AJaegerfungi, thanks!14:54
fungiseems to be similar to issue 51 in that all of our mirrors added the tarball and updated the index for the wheel, but none retrieved the wheel14:54
anteayaI'm trying to observe Thanksgiving Monday14:54
anteayaso far I'm off to a poor start but will try harder14:55
*** rushiagr_away is now known as rushiagr14:55
fungianteaya: go enjoy your holiday14:55
*** achanda has joined #openstack-infra14:56
*** thedodd has joined #openstack-infra14:56
fungiit's "columbus day" here, which i've never been entirely certain how to observe... other than perhaps celebrating the barbaric bloodbaths of my hegemonic european invader ancestors who enslaved and killed the indigenous inhabitants of this continent14:56
*** habib has quit IRC14:56
*** armax has quit IRC14:56
AJaegerNow I'm feeling less guilty to have pinged fungi ;)14:57
AJaegeranteaya, enjoy!14:57
jrollfungi: easy, go sailing :P14:57
anteayafungi: yeah I'm at a loss on how to observe that one14:57
clarkbfungi it is a bank holiday so larissa has it off which means I celebrate by running errands14:57
fungijroll: ha!14:57
anteayaAJaeger: thanks14:57
*** kgiusti has quit IRC14:58
*** datsun180b has joined #openstack-infra14:58
*** armax has joined #openstack-infra14:58
*** IAX has joined #openstack-infra14:58
*** habib has joined #openstack-infra14:58
*** miarmak has joined #openstack-infra14:59
AJaegerok, I'm back from home in an hour or so...14:59
*** baoli has quit IRC14:59
fungiAJaeger: i'll let you know what i find, but i should have it worked around by then14:59
*** habib has quit IRC14:59
krotscheckStoryBoard Meeting in #openstack-meeting15:00
AJaegerthanks, fungi!15:00
anteayaclarkb: happy errands15:00
*** kgiusti has joined #openstack-infra15:00
*** habib has joined #openstack-infra15:00
*** baoli has joined #openstack-infra15:01
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: Handling windows command prompt properly
*** habib_ has joined #openstack-infra15:02
*** habib_ has quit IRC15:03
SergeyLukjanovfolks, I'm really sorry, but I have now very small time windows (only personal free time) to work on infra stuff and so, I'd like to ask to not expect from me some periodic stuff done well (like 3rd party accounts creation)15:03
SergeyLukjanovjeblair, fungi, clarkb, mordred, anteaya ^^15:04
*** habib has quit IRC15:04
*** krtaylor_ is now known as krtaylor15:04
SergeyLukjanovso, I still have enough time to review at least15:04
anteayaSergeyLukjanov: thanks for letting us know15:04
clarkbI can make a pass at account creation15:04
anteayaSergeyLukjanov: you need some free time to yourself, you shouldn't be working in your free time15:04
anteayaclarkb: thanks15:05
krtaylortime to implement self service process  :)15:05
*** thedodd has quit IRC15:05
clarkbanteaya its a holiday so maybe tomorrow we put a list together with names?15:05
anteayaclarkb: I'm here to help you with what you need15:05
anteayaclarkb: let's do it tomorrow15:05
clarkbkrtaylor I still think moving them to openid is the way to go :)15:05
clarkbanteaya sounds good15:05
*** thedodd has joined #openstack-infra15:05
anteayakrtaylor: well these folks haven't been able to manage to read teh requirements and impliment them15:05
anteayakrtaylor: self service account creation feels like it is quite a ways off15:06
SergeyLukjanovunfortunately, my current job resposibilities doesn't provide enough free time for infra, but I really want to continue working on it, so... anteaya, clarkb15:06
krtayloryeah, there is that, sounds like  agood summit discussion15:06
anteayaSergeyLukjanov: we understand, thanks for telling us15:06
*** achanda has quit IRC15:06
anteayaSergeyLukjanov: hopefully your responsibilites will change again and you will have time for us again15:06
krtayloranteaya, btw, re: the summit session, can I get the name changed, I didnt understand the gerrit reference15:07
anteayaSergeyLukjanov: also feel free to send DinaBelova over :D15:07
anteayakrtaylor: I picked the name15:07
mferfolks, i've got a stackafoge project with enough core review and workflow + but it's not trying to merge. anyone have an idea why? review is at
SergeyLukjanovanteaya, :)15:07
DinaBelovaanteaya, hehe15:07
DinaBelovawhy not actually :D15:07
DinaBelovawe'll see15:07
anteayakrtaylor: the name is in reference to the fact that this session is for and about operators who have gerrit third party accounts15:07
anteayakrtaylor: and the description is identifying that items to be discussed are on the agenda which is decided at third paraaty meettings15:08
krtayloranteaya, hm, I see - I was just thinking something like "Improving third-party CI testing"15:08
anteayakrtaylor: since about 300 new people will show up expecting us to fix their jenkins in this session and we won't15:09
anteayakrtaylor: no15:09
anteayasee my above point15:09
anteayathis is for people who have third party gerrt accounts15:09
krtayloranteaya, yes, I agree with the fixing, not the focus15:09
clarkbmfer hard to say when the reviews are more than a month old15:09
*** rodrigods has joined #openstack-infra15:09
rodrigodshello, is this issue here already known?15:10
mferclarkb what can I do with it now?15:10
*** thedodd has quit IRC15:10
krtaylorI was just thinking about following what clarkb did last, it seemed to work ok, but the description was clear that it was not for fixing15:10
clarkbmfer try approving it again and/or recheck that will push it through15:10
anteayakrtaylor: I don't understand what you are saying15:10
krtayloranteaya, I am just concerned that someone will show up expecting to hear about gerrit changes15:10
anteayakrtaylor: if that actually happens how about you let me deal with it15:11
anteayasince noone has ever shown up in channel asking that kind of question15:11
mferclarkb and it just worked. i tried that before and nothing. this time it worked. huh15:12
mferah well15:12
clarkbrodrigods ya pypi mirroring has an issue with a dependency there. fungi is working on it and Ajaeger sent details to the mail list15:12
krtayloranteaya, I would just see the title and not expect what the agenda reflects, thats all, and I might not even go15:12
krtayloranteaya,  not sure what question you are referring to15:13
*** julim has quit IRC15:13
anteayakrtaylor: if you saw third party on the title and didn't read the description and the etherpad then yes, I am glad that folks too lazy to read documentation would be detered15:13
fungiclarkb: rodrigods: yeah, it's looking like is still a problem in bandersnatch 1.6.1 so i'm updating the issue with a fresh log example15:13
krtayloranteaya, hehheh, good point15:13
anteayakrtaylor: I trust you don't need me today for the third party meeting15:14
*** pboldin has quit IRC15:14
anteayaand that I may try to scrape together as much day off out of today as I can manage?15:15
krtayloranteaya, I was planning to go over the summit topics, prepare an outline for the folks interested in the session15:15
krtayloranteaya, did not relize, absolutely, take off15:15
*** xyang0 has quit IRC15:15
krtayloror realize15:15
anteayawell the summit topics have been prioritized15:16
anteayaand the etherpad is there15:16
anteayaI didn't expect more to be done with that etherpad15:16
anteayawhich is why we worked on it last monday's meeting15:16
openstackgerritMarton Kiss proposed a change to openstack-infra/config: Add mod_security to openstackid
*** mmaglana has joined #openstack-infra15:17
*** julim has joined #openstack-infra15:17
*** _buzz has quit IRC15:18
*** davinchi has joined #openstack-infra15:18
*** shashankhegde has joined #openstack-infra15:19
*** koolhead_ has joined #openstack-infra15:20
*** mmaglana has quit IRC15:21
*** koolhead17 has quit IRC15:24
*** zaphod42 has joined #openstack-infra15:24
*** ayoung has joined #openstack-infra15:26
*** thedodd has joined #openstack-infra15:27
*** thedodd has quit IRC15:27
*** thedodd has joined #openstack-infra15:27
*** ihrachyshka has quit IRC15:29
*** rainya has joined #openstack-infra15:31
*** yolanda_ has joined #openstack-infra15:31
fungimrmartin: ^ i guess our discussion last week about how mod_security is basically useless and risks causing random breakage wasn't enough to dissuade you?15:33
*** _buzz has joined #openstack-infra15:35
*** koolhead_ has quit IRC15:36
mrmartinfungi hi15:37
*** unicell has quit IRC15:39
fungimrmartin: is a temporary measure until someone completes a code audit on the openstackid software, or something we want to maintain indefinitely?15:39
mrmartinI suggest to reach a consensus in this topic from infra team as a whole15:39
mrmartinthe patch is there, so if we want to merge, let's do it, if not just throw away15:40
*** yfried has quit IRC15:41
*** yfried_ has joined #openstack-infra15:41
fungimrmartin: good idea. we can either see if we can get a rough consensus in here, or we can add it to the team meeting agenda for tomorrow15:41
fungithanks for bringing it up!15:41
mrmartinBasically I operated a modsec module in front of a Drupal site before, and I want to underline the operate word here.15:41
*** med_ has joined #openstack-infra15:42
*** med_ has quit IRC15:43
*** med_ has joined #openstack-infra15:43
mrmartinfungi, when you want to do the groups prod instance resize?15:43
*** miarmak has quit IRC15:43
*** koolhead17 has joined #openstack-infra15:44
fungimrmartin: i can do it right now if you want, or pick a time later today, or later this week15:44
fungiyou suggested upgrading to 4gb ram right?15:44
mrmartinYes, 4g. I vote on doing later today, because almost 18.00o'clock here, and I have some family duties, but will be available from 21.00CET (12.00PDT)15:46
*** rainya has quit IRC15:46
mrmartinfungi, and plz. add your opinion to
*** markmcclain has quit IRC15:48
fungimrmartin: i'll be glad to do it at that time... i just have to figure out what that time is for me ;)15:49
fungi1900 utc?15:50
fungiit looks like 1900 utc anyway15:50
*** yfried_ has quit IRC15:50
mrmartinfungi, yep, but feel free to going eat, I'll be here15:50
*** yfried__ has joined #openstack-infra15:50
*** Longgeek has quit IRC15:51
fungii don't expect to need to eat around that time since it'll be the middle of the afternoon where i live15:52
*** viktors is now known as viktors|afk15:52
*** rainya has joined #openstack-infra15:54
*** dangers_away is now known as dangers15:56
*** jlibosva has quit IRC15:57
*** doug-fish has joined #openstack-infra15:59
zaropelix: python 3 jobs?15:59
*** kgiusti has quit IRC15:59
*** doug-fish1 has quit IRC15:59
*** arxcruz has quit IRC16:00
*** baoli has quit IRC16:00
*** ZZelle has quit IRC16:00
*** ZZelle has joined #openstack-infra16:01
*** xyang0 has joined #openstack-infra16:01
openstackgerritEmilien Macchi proposed a change to openstack-infra/project-config: Add meetbot to #puppet-openstack channel
EmilienMfungi: not sure I did it the right way, you may want to have a quick look ^16:01
*** andreykurilin_ has joined #openstack-infra16:02
*** superdan is now known as dansmith16:03
*** gyee has joined #openstack-infra16:03
*** jgallard has quit IRC16:04
*** baoli has joined #openstack-infra16:05
*** julim has quit IRC16:05
jgriffithfungi: any chance you might help me figure out what's up with my LaunchPad/Ubuntu One ID's?16:06
jgriffithfungi: I went in to a patch to remove my -1, and instead it entered a new entry16:06
jgriffithfungi: would appear I have multiple personalities :)16:06
*** kgiusti has joined #openstack-infra16:07
fungijgriffith: that's... ubunfortunate16:08
*** julim has joined #openstack-infra16:08
fungijgriffith: just a sec and i'll see what i can find in the gerrit db16:08
*** rainya has quit IRC16:08
*** asselin has joined #openstack-infra16:08
fungijgriffith: did you recently make any changes to your lp profile? that's what we've usually seen trigger this in the past16:09
*** erlon_ has quit IRC16:09
jgriffithfungi: I did :(16:09
jgriffithfungi: I changed my primary email address16:09
*** jpich has quit IRC16:10
fungiEmilienM: -2'd because it looks like you proposed a change to a file in the wrong project... see my comment on 128004 for details16:10
jgriffithfungi: I realize that's likely the cause, just trying to figure out how to fix it16:10
*** yfried__ has quit IRC16:10
clarkbok mostly properly awake now16:10
jgriffithfungi: if I need to revert back to my previous email address maybe?16:10
fungijgriffith: no, i'll patch it up in gerrit. but if just changing your e-mail address causes it to create a new id url, i consider lp's openid provider remarkably broken16:11
*** dmsimard_away is now known as dmsimard16:11
jgriffithfungi: I'm wondering if it might be related to Ubuntu One accounts as well16:11
jgriffithfungi: I had one of those prior to LP switching to it16:11
fungijgriffith: entirely possible16:12
*** roaet has quit IRC16:12
fungijgriffith: what does your new account say the numeric account id is at
*** roaet_ has joined #openstack-infra16:12
jgriffithfungi: 224316:13
fungioh, huh... that's your old account still16:13
jgriffithThus my confusion16:13
jgriffithRegistered 201116:13
jgriffithand has my updated email address16:13
fungijgriffith: where'16:14
fungiwhere's a change you added a second comment row on?16:14
jgriffithfungi: and interestingly the details on both are the same when I select them :)16:15
fungioh weird16:15
*** krtaylor has quit IRC16:15
fungiso that was made by an old account of yours, id 104316:15
jgriffithfungi: it's kind of a neat trick if I could do it inentionally :)16:15
fungicreated in gerrit 2011-07-22 19:13:4416:15
*** stevemar has joined #openstack-infra16:15
jeblairjgriffith: 2x+2s ftw, right? :)16:15
jgriffithjeblair: exactly!16:16
jgriffithfungi: oh... my very first LP account16:16
fungijgriffith: it does seem to still have an old openid associated with it, so i'm going to move that to your new account16:16
jgriffithfungi: awesome16:16
fungi"new" account (from later in 2011)16:16
jgriffithfungi: strange that got picked up16:16
jgriffithfungi: yeah, that's perfect16:16
fungii find that extremely strange16:16
pelixzaro: yeap, figured since we're not python3 compatible in JJB, we could enable voting on the job -
fungiokay, if for some reason it does decide to log you back in with that older openid url, it will now log you into your correct account16:17
*** mmaglana has joined #openstack-infra16:18
fungijgriffith: ^16:18
jgriffithfungi: awesome!  thanks for the help16:18
fungijgriffith: i can also delete that stray -1 vote from the change you linked, if you like16:18
fungiso as to avoid confusing reviewers16:18
openstackgerritJaime Flynn proposed a change to openstack-infra/jenkins-job-builder: Added flexible publish support
*** ildikov_afk has quit IRC16:19
jgriffithfungi: dang it...16:19
jgriffithfungi: still duplicate me's16:19
fungipelix: please test against python 3.4. we want to move off the 3.3 jobs so it would be good not to add voting for 3.3 if it's broken with 3.416:19
*** IvanBerezovskiy has left #openstack-infra16:20
fungijgriffith: yeah, as i said i can delete that -1 for you. seems to have been added with your old account. i'll also disable the old account to keep people from accidentally selecting it in reviewer selection drop-down/autocomplete16:21
*** cnesa18 has quit IRC16:21
jgriffithfungi: Ohh... sorry, missed that part16:21
jgriffithOk, thanks again16:21
fungijgriffith: i've disabled your old 1043 account id, and also manually cleared your old account's lingering -1 vote from that particular review16:22
*** groknix has joined #openstack-infra16:22
jgriffithfungi: you rock... thanks!16:23
fungimy pleasure16:23
pelixfungi: I test locally against python 3.4 as I don't have 3.3 installed ;)16:23
fungipelix: even better!16:24
*** yfried__ has joined #openstack-infra16:24
pelixshould I just update the jobs to 3.4 as well?16:24
fungipelix: thanks for checking16:24
fungipelix: you can if you like, though i hope to soon move everyone's 3.3 jobs to 3.4... we've just got a few lingering issues holding it back we need to get to cleared up first16:24
*** krtaylor has joined #openstack-infra16:24
*** jcoufal has quit IRC16:25
*** groknix has quit IRC16:25
clarkbspeaking of, I wonder how that bug is doign16:27
*** jistr has quit IRC16:29
*** davinchi has quit IRC16:31
fungiwe still need to open one requesting an sru for the commit which fixes that other bug preventing rootwrap's tests from working too16:31
fungiand there are a couple more projects down at the end of the broken list which i haven't had a chance to look into the errors on yet16:31
clarkbso one thing I have noticed about changing computers more often is I lose track of etherpads much more easily16:32
clarkbis it called python34-transition?16:32
*** _nadya_ has joined #openstack-infra16:33
zaropelix: LGTM16:34
uvirtbotLaunchpad bug 1367907 in python3.4 "Segfault in gc with cyclic trash" [High,Fix released]16:34
clarkbdoes that mean we need to use that ppa?16:34
fungiclarkb: i think the maintainers use the ppa as an intermediate step to getting the sru in stable16:34
clarkboh that makes sense16:35
fungiclarkb: but not sure what the timing usually is for that... zul ^ ? is that correct?16:35
clarkbfungi: maybe we should pull the package from the PPA on a node and run it through some testing on our side?16:35
fungiclarkb: definitely16:35
*** cody-somerville has joined #openstack-infra16:35
*** cody-somerville has quit IRC16:35
*** cody-somerville has joined #openstack-infra16:35
*** e0ne has quit IRC16:35
*** dizquierdo has quit IRC16:36
zaroclarkb: you satisfied with gearman-plugin changes?
clarkbzaro: I am properly reviewing them now, but I was pretty happy with it when I looked last16:37
*** luqas has quit IRC16:38
*** _buzz has quit IRC16:40
clarkbzaro: I am still a little worried about the onTemporarily offlined state16:40
*** cpowell has quit IRC16:41
*** ramashri has joined #openstack-infra16:41
clarkbzaro: because in the gui you can set that state while a job is running. But your testing indicates the event doesn't fire until after the job is complete?16:41
*** rwsu has joined #openstack-infra16:42
clarkbzaro: in any case we can go with the results of your testing, then upgrade one node at a time and make sure nothing funny happens when run by zuul16:42
*** cpowell has joined #openstack-infra16:42
*** eglynn-officeafk is now known as eglynn-office16:43
clarkbstarting with jenkins-dev16:43
*** AJaeger_ has joined #openstack-infra16:43
zaroclarkb: i can verify the exact time when the event fires if it will make you feel better.16:44
clarkbzaro: as long as you are sure it is after the running job completes16:44
clarkbI think that is the important requirement16:45
*** markmcclain has joined #openstack-infra16:45
zaroclarkb: will verif in a min.16:45
*** markmcclain has joined #openstack-infra16:45
*** davinchi has joined #openstack-infra16:46
*** funzo_ has joined #openstack-infra16:46
*** funzo has quit IRC16:47
mriedemstevemar: why was this abandoned?
openstackgerritvinod kumar proposed a change to openstack-infra/git-review: Handling windows command prompt properly
*** BobBall is now known as BobBall_AWOL16:48
fungimriedem: per my comment on the bug, it only ever impacted two check runs on one single patch to the one project which was misconfigured. may not be worth tracking16:48
mriedemfungi: oh, there was a hit in the gate16:49
mriedemaccording to logstash16:49
mriedembut only 4 hits total, yeah16:49
fungimriedem: aha, those were newer than when i checked16:49
mriedemapparently something is killing the gate and we aren't tracking it
mriedem60% categorization is pretty low16:49
fungimriedem: but still the same patch16:49
*** doude has quit IRC16:49
fungimriedem: i guess the gate failure and subsequent recheck were the only ones indexed when i first searched16:50
*** HeOS has quit IRC16:51
jogomordred: yeah, hoping to have that fixed in the next hacking release. Python doesn't make that logic easy16:51
fungibut yeah, small contribution to the failure aggregate16:51
*** patrickeast has joined #openstack-infra16:51
stevemarmriedem, yay for not having to answer16:51
uvirtbotLaunchpad bug 1379201 in openvswitch "openvswitch-datapath-dkms 1.4.6-0ubuntu1.12.04.3: openvswitch kernel module failed to build [error: too many arguments to function ‘ip_select_ident’]" [Medium,Fix committed]16:51
jogois it possible for us to test out that fix?16:51
stevemarfungi, sorry about the mixup on that bug :\16:52
jogoas precise neutron is hurting16:52
*** gyee has quit IRC16:52
clarkbjogo: yes, it should already be fixed16:52
*** erlon_ has joined #openstack-infra16:52
* clarkb checks image build times16:52
jogoclarkb: cool, still seeing hits in logstash though16:52
jogonot sure what that means16:52
AJaeger_fungi, thanks for fixing pypi for us!16:53
*** yfried__ has quit IRC16:53
*** yfried__ has joined #openstack-infra16:53
clarkbya looks like the oldest devstack-precise image is about 25 hours old16:53
*** funzo_ has quit IRC16:53
*** funzo_ has joined #openstack-infra16:54
*** andreykurilin_ has quit IRC16:55
*** chandankumar has quit IRC16:55
*** andreykurilin_ has joined #openstack-infra16:55
jogoclarkb: so that means the next image should get the potential fix?16:55
*** chandankumar has joined #openstack-infra16:55
*** cdent has joined #openstack-infra16:56
mriedemfungi: jogo: i suspect this is what's killing us in the gate right now
uvirtbotLaunchpad bug 1373513 in cinder "Lvm hang during tempest tests" [High,Confirmed]16:56
fungijogo: no, i think all the nodes should currently be getting booted from virtual machines pinned to the previous kernel version... i'm going to hold one in the same region as a recent failure to confirm16:56
clarkbfungi: jogo: we do run uname -a on tests now. So I think we want to find the failed runs and check that value16:56
mriedemwell, untracked thing killing us16:56
fungiclarkb: i thought we did... but couldn't find it getting called in the console log16:57
clarkbits possible that maybe devstack is updating the kernel headers for us16:57
clarkband dkms is failing anyways?16:57
clarkbfungi: hrm maybe I reviewed the change but it didn't merge /me looks16:57
*** stevemar has quit IRC16:57
clarkbno its there in functions.sh16:58
fungifor reference, example failure from half an hour ago:
clarkbfungi: in setup_host16:58
*** _nadya_ has quit IRC16:58
clarkb wrong kernel16:58
fungiaha, yep16:58
funginewer kernel16:59
*** sarob has joined #openstack-infra16:59
davinchiguys, there is a error in 'API  version' concept description
davinchi> Backwards compatible changes are indicated by an increased minor number  and incompatible changes are indicated by an increased minor number16:59
fungidavinchi: you probably want to bring that up in #openstack-oslo16:59
clarkbfungi: so the image build isn't doing what we expect for some reason. Probably need to go review those logs17:00
jedimikeovs_packages='make fakeroot dkms openvswitch-switch openvswitch-datapath-dkms linux-headers-3.2.0-70-generic'17:00
*** derekh has quit IRC17:01
fungiclarkb: i'm going to grab a sample node first... hopefully this isn't yet another occurrence of the "rackspace boots from old deleted snapshots" bug17:01
clarkbfungi: ok, I am digging through logs17:02
dtroyerfungi: don't hold your breath…all of the failures like that I looked at over the weekend were rax nodes17:02
fungiclarkb: random sample says they're all 3.2.0-7017:02
fungiso answer is likely in the logs17:02
openstackgerritAndreas Jaeger proposed a change to openstack/requirements: Update openstack-doc-tools to 0.20
fungimaybe something about jedimike's fix is working for hpcloud but not rax17:03
jedimikefungi, clarkb, not sure if it helps, but ovs_packages includes the 70-generic kernel, and the fix we put in place removes and replaces the -virtual kernel17:03
*** mdbooth has quit IRC17:03
*** rainya has joined #openstack-infra17:03
*** harlowja_away is now known as harlowja17:03
dhellmannfungi: we've unfrozen requirements, right? I was going to approve if so17:05
fungidhellmann: yep17:05
fungidhellmann: let the mayhem commence17:05
koolhead17ttx: around17:05
koolhead17hello everyone17:05
dhellmannfungi: thanks (hat tip clarkb for responding in openstack-dev first :-)17:05
*** ayoung has quit IRC17:05
dhellmannfungi: mwuahaha?17:06
AJaeger_dhellmann: could you review, please? This is important for docs17:06
*** thedodd has quit IRC17:07
dhellmannAJaeger_: the newer version is *required* or *desired*?17:07
*** yfried__ has quit IRC17:07
AJaeger_dhellmann: required17:07
*** andreykurilin_ has quit IRC17:07
dhellmannAJaeger: I only ask because we're trying to be more careful that we only bump minimum versions when we actually need to, not just when we have new releases17:07
dhellmannAJaeger: ok, good17:07
*** tonytan4ever has quit IRC17:07
AJaeger_dhellmann: that's why we skipped 0.19 - that one was only desired ;)17:08
*** lttrl has quit IRC17:08
dhellmannAJaeger: cool :-)17:08
fungiclarkb: !!! could it be bootloader differences in rax?17:08
AJaeger_dhellmann: should I make it clearer next time? I thought it was...17:08
*** markmcclain has quit IRC17:08
fungiclarkb: we install -69 but pv-grub is still selecting -70?17:08
dhellmannAJaeger: re-reading you were clear, I'm just being paranoid17:08
dhellmannAJaeger: and since you were here to ask...17:09
clarkb'Ignoring Xen kernel on non-Xen host: vmlinuz-3.2.0-69-virtual' is the only odd thing I see17:09
*** markmcclain has joined #openstack-infra17:09
clarkbfungi: but we remove -70 first17:09
AJaeger_dhellmann: sure ;)17:09
*** denis_makogon_ has joined #openstack-infra17:09
fungiPackage linux-image-3.2.0-70-virtual is not installed, so not removed17:09
clarkbfungi: rax would have to be providing the kernel out of band for that to work right?17:09
jedimikefungi, shows that the -70-generic kernel is explicitly installed, search for ovs_packages17:10
clarkbfungi: wait not installed == not removed?17:10
fungioh, that's not the image package though17:10
*** mdbooth has joined #openstack-infra17:10
zaroclarkb: the event fires when the command is issued to put the slave temp offline, in other words the event fires before the job completes.17:10
clarkbzaro: right so we will break everything with that change17:10
fungiclarkb: linux-image-3.2.0-70-generic is left installed on the image, looks like17:11
clarkbzaro: we should probably just not run anything on that event?17:11
clarkbfungi: jedimike: do we need to purge?17:11
jedimikeclarkb, no, the fix from last week presumed the kernel was -virtual, so that's what it removes, but rax is using -generic17:12
clarkboh I see17:12
clarkbdifferent names17:12
clarkbcan we safely just remove both of them17:12
clarkband let apt sort it out for us?17:12
fungiclarkb: entry index 0 in grub menu.lst is for /boot/vmlinuz-3.2.0-70-generic17:12
jedimikeyes, but we don't really know which one we need to install, -generic or -virtual17:12
jedimikewe can check before we remove it...17:13
jedimikeso if `uname -r` ends in vitural or generic, we can do the right thing17:13
jeblairclarkb: why will it break everything?  (i'm still catching up)17:13
fungijedimike: clarkb: and we've booted -70-generic according to uname -a (not -70-virtual which we removed)17:13
clarkbjeblair: because that handler offlines the gearman worker17:13
*** hdd has quit IRC17:13
clarkbwhich should stop running the job right?17:13
zaroclarkb: i'm not convinced of that.  let me investigate a littl further.17:13
jedimikefungi, so we remove 70-virtual, install -69-virtual, and something else is taking us to 70-generic?17:14
fungijedimike: clarkb: so we should probably just remove both package names17:14
clarkbzaro: at the very least it will prevent reporting the job back I would expect17:14
jeblairclarkb: yeah, i was thinking (without looking at code) that it would just deregister, but if it drops the gearman connection, yes very bad.17:14
jeblairzaro: ^17:14
fungijedimike: it appears that -70-generic and -70-virtual were both installed, we only trmoved -70-virtual and so booted -70-generioc17:14
jedimikefungi, oh! I didn't expect that17:14
jeblairthis is like, totally the reason we want consistent images ;)17:15
fungijedimike: yeah, one more nail in the coffin for cloud-provider-specific base images17:15
fungijeblair: YES17:15
jedimikefungi, i'll submit a new change that removes both -generic and -virtual17:15
*** jheroux has joined #openstack-infra17:15
clarkbjeblair: fungi: on that note I think I have my dib change sorted17:15
clarkbhopefully that can merge this week, then we have that open as an option for near term image stuff + nodepool17:16
zaroclarkb: yeah. i think you are right, but just want to verify.17:16
*** mdbooth has quit IRC17:17
clarkbjeblair: fungi: also I was planning to take a stab at making a new nodepool server this week. I should be able to spin up a trusty node without interferring with anything due to firewall rules right? then I can poke at the puppeting on trusty (which should work just fine) before swapping the daemons out17:17
*** yamahata has joined #openstack-infra17:17
openstackgerritMike Heald proposed a change to openstack-infra/project-config: Remove broken 70-generic kernel as well as 70-virtual
clarkbI will have to build it as a non performance node until we split out image workers due to disk needs17:18
fungiclarkb: yeah, i think nodepool could even run entirely parallel masters as long as we adjusted the provider quotas and minimums on them accordingly, right?17:19
jedimikefungi, ^^ removes -virutal and -generic before installing 69-virtual17:19
clarkbfungi: I think we may need to tweak how nodepool reads the gearman queues in that case17:19
fungii mean, it might get weirded out when it sees events zmq events for nodes it doesn't know about, but aside from that...17:19
*** rainya has quit IRC17:19
jeblairclarkb: why not use a cinder volume?17:20
fungiclarkb: ahh, right, demand will get double-counted17:20
clarkbjeblair: because dib is silly is basically the answer17:20
clarkbjeblair: you would have to mount on ~root and /tmp and /opt/nodepool_dib17:20
jeblairclarkb: uhoh.  can you elaborate?17:20
jeblairclarkb: can we bind mount a volume to all 3 places?  (or use 3 cinder vols?)17:21
jeblairclarkb: i think we decided we could symlink ~root/.cache17:21
fungiclarkb: any reason not to go with lifeless's symlink suggestion instead for ~root/.cache ?17:21
jeblair(in fact, i thought we did that)17:21
*** ayoung has joined #openstack-infra17:21
fungier, what jeblair said17:21
clarkbthere are three locations we need disk. ~root (or ~nodepool) for the cache of all the things. /tmp because it shoves data in there temporarily and local testing found that falling back on my 1GB tmpfs completely failed on /tmp17:21
clarkband finally the destination where the images end up needs disk too17:21
clarkbcurrently that is /opt/nodepool_dib17:22
jeblairclarkb: i think he said it obeyed TMPDIR too17:22
*** mdbooth has joined #openstack-infra17:22
fungiyeah, so we could set that to somewhere in /opt17:22
jeblairclarkb: so we should be able to symlink and set TMPDIR= to /opt/nodepool_dib/something17:22
clarkbjeblair: fungi: so there are things we can do, but I don't know that we gain much by complicating out fstab17:22
jeblairclarkb: that ^ way does no fstab changes17:22
fungiclarkb: a symlink and setting tmpdir don't really complicate the fstab do they? (then we only need to mount /opt on cinder)17:23
clarkbjeblair: fungi: well the original consideration was 3 volumes :)17:23
clarkbjeblair: fungi: would we just use the ephemeral disk on /opt as is our default and set the symlink and env var?17:23
clarkbor do you want a cinder volume to back that instead?17:23
jeblairclarkb: i don't want a cinder volume unless the ephemeral disk is too small17:23
clarkbI don't think ephemeral disk is too small.17:24
fungii think ephemeral would be fine if we don't want to persist anything there across reboots17:24
clarkblet me check rax flavors17:24
*** rushiagr is now known as rushiagr_away17:24
jeblairfungi: i think ephemeral persists across reboots, it's just not backed up, etc, right?17:24
clarkbjeblair: that is my understanding17:25
jeblair(and i would not expect it to persist across rebuilds)17:25
clarkbgoes away if you delete the node, persists across reboots17:25
fungioh, got it17:25
*** _buzz has joined #openstack-infra17:25
fungiregardless, i wouldn't want to put anything important in there, but what we're discussing seems fair game17:25
*** qba73 has quit IRC17:26
*** e0ne has joined #openstack-infra17:26
clarkbif we keep the same ram size and move to performance that is the 8GB flavor with 80GB of ephemeral disk17:26
clarkb80GB should be enough. we are using 45GB on / total on the current node17:27
*** amuller has quit IRC17:27
* clarkb does rough math17:28
clarkbeach image is worth about 8GB of disk if we have a qcow2 and vpc17:28
clarkbwe have devstack-precise, devstack-trusty, devstack-f20, devstack-centos7, bare-trusty, bare-precise, bare-centos6 images17:29
clarkbthat is ~7 total images for a super conservative number of ~60GB of disk17:29
clarkbbut we have to keep old image around while we build new image so we may actually need double that number?17:29
*** yfried__ has joined #openstack-infra17:30
fungiif it builds those in parallel, how much scratch space is it going to suck up? and what's the cache utilization look like?17:30
AJaeger_dhellmann: regarding your review of : I think that's something to discuss for Paris.17:30
jeblairwe should keep moving toward reducing the number of image variants we have :)17:30
clarkbjeblair: agreed17:30
AJaeger_dhellmann: we like for docs the convience of syncing version numbers for necessary changes - but that means also adding new packages...17:31
jeblairclarkb, fungi: so we can keep moving to dib one image at a time and see how the numbers add up17:31
dhellmannAJaeger_: ah, ok, that might be a new thing we need to build. "sync requirements that aren't needed for the integrated release into some other projects"17:31
clarkbfungi: I started a du earlier that wasn't coming back so I killed it... I am started another in /home/nodepool/.cache17:31
jeblairclarkb, fungi: and switch to cinder if we need to after a few more; how's that sound?17:31
fungisounds good17:32
clarkbjeblair: yup17:32
clarkbthat works for me17:32
*** maxbit has joined #openstack-infra17:32
AJaeger_dhellmann: yeah, that's one way of doing it...17:33
clarkbfungi: jeblair: .cache is currentl 6.1GB17:33
*** mriedem has quit IRC17:33
ameadeHave you folks seen this issue on rackspace nodes? "libvirtError: XML error: Missing CPU model name"17:33
ameadeusing the devstack-trusty image?17:33
zaroclarkb: ok, confirmed.  you are correct.  disconnection will happen before job ends.  however i don't think it can be a noop otherwise the function will still be registered.17:33
dhellmannAJaeger_: I think the idea is to have a list of requirements as a signal to the distros. We used to also use that list to build a mirror, but we don't need to do that any more so we want to get away from listing things we aren't actually *requiring* for runtime. Discussing at the summit may be the quickest way to resolve. Want to file a cross-project session?17:34
*** SumitNaiksatam has joined #openstack-infra17:34
fungiameade: i don't find that in any of the job logs we index going back for at least the past week17:34
ameadefungi: hmm ok, must be something I'm doing17:35
ameadefungi: thanks, dont wanna chase my tail if it is something we are all hitting17:35
fungiameade: perhaps helps?17:35
ameadefungi: I think i need to use the pvhvm image in rax, looks like u guys are17:36
fungiameade: yep, we do that17:37
clarkbdhellmann: AJaeger_ with current mirrors non integrated projects don't need to worry about openstack/requirements unless they choose to be like an integrated project. And in that case they are stuck with what integrated projects want17:37
clarkbI feel like I am missing some context there, but not sure we need a summit session on that17:38
zaroclarkb: ahh, i think we are ok.  the gearman-plugin doesn't set issue the command to offline the node until the job completes.17:38
clarkbzaro: does it properly unregister the job?17:39
clarkbzaro: or would it do that after the job completes and before grabbing a new job?17:39
clarkbI think if ^ is the behavior we are fine17:39
*** jp_at_hp has quit IRC17:40
fungiclarkb: dhellmann: AJaeger_: i think the sanest explanation for global-requirements.txt is "enforce this on all projects you want to be sure can all be installed together on the same machine (or which might need to be packaged together in a given linux distribution) without conflicting over their dependencies"17:40
zaroclarkb: yes, I believe that is correct.17:41
AJaeger_dhellmann: I've proposed a session.17:41
dhellmannfungi, clarkb: we're talking about the documentation projects17:41
*** ivar-lazzaro has joined #openstack-infra17:41
*** ivar-lazzaro has quit IRC17:41
fungidhellmann: yep, i get that. i was taking those as an example of something which we probably don't need it enforced on17:41
AJaeger_fungi: so, how should openstack-manuals and openstack-doc-tools handled with requirements?17:42
zaroclarkb: so i believe this is how it works.  job starts -> gearman-plugin waits until it completes -> issues a tempOffline to the node -> function gets unregistered17:42
*** mriedem has joined #openstack-infra17:42
*** ivar-lazzaro has joined #openstack-infra17:42
AJaeger_fungi: patches are and
clarkbzaro: ok, if that is the case we should be fine. We should test that though :)17:42
zaroclarkb: yes, i will verify17:42
clarkbAJaeger_: if we go with fungi's criteria they would not be under the umbrella of global-requirements17:43
AJaeger_fungi: agreed, we don't need it enforced on - but I would love to have it since it allows us to use the convenience of syncing requiremnents everywhere.17:43
dhellmannfungi: right, that's why I was voting -1 too, but AJaeger_ pointed out the docs team is using the sync feature of the global requirements list between *their* projects, so I was suggesting we have a conversation about whether or not to generalize that17:43
dhellmannfungi, clarkb, AJaeger_ : it might not need a summit session, maybe just an infra spec?17:43
* dhellmann has summit on the brain17:43
* AJaeger_ is fine with an infra-spec as well....17:43
fungidhellmann: AJaeger_: right, this came up in a slightly different context already with solum. they wanted to depend on things not listed in global requirements but also get sync proposals for things listed in global requirements17:44
AJaeger_fungi: yep, some concept.17:44
dhellmannfungi: ah, right, I forgot about that case17:45
fungidhellmann: AJaeger_: it seems like it's unlikely to be worth the added complexity, but regardless the biggest hurdle will be coming up with a way to signal which requirements you do or don't want synchronized from the global list17:45
AJaeger_for documentation I'm fine with getting everything from global-requirements.17:45
dhellmannAJaeger_: how many different repositories are you syncing requirements into?17:45
fungidhellmann: AJaeger_: and then making sure we don't accidentally introduce a loophole into the enforcement allowing projects which *should* be gated against whether their requirements are in the global list slip in requirements which aren't17:45
dhellmannAJaeger_: the problem is, we no longer want to put documentation requirements into the global list at all17:46
*** tonytan4ever has joined #openstack-infra17:46
dhellmannfungi: maybe a separate repo for those projects?17:46
davinchipaste is 500:
AJaeger_dhellmann: I quickly counted 10. Might be one or two more17:47
*** ociuhandu has joined #openstack-infra17:47
fungidhellmann: AJaeger_: however, i'm also fine with an interpretation which says that we should expect anyone to be able to rebuild our manuals from source on the same servers/distributions where they run our software17:47
*** gyee has joined #openstack-infra17:47
dhellmannAJaeger_: yeah, I think it's the idea of not having to think about it. I'm not opposed to keeping them, I'm just trying to explain the new thinking about removing them.17:47
clarkbdavinchi: if you retry it should get through. There are mysql connection issues since our move to a trove hosted DB17:47
fungidavinchi: try going to and then force-reloading
dhellmannfungi: yeah, that feels right to me, too17:48
AJaeger_fungi: the problem we have now is that we need extra packages added to global-requirements, just for documentation.17:48
davinchifungi, clarkb: it works, thanks17:48
AJaeger_dhellmann: Yeah, I get the new thinking but wonder what I can do for the 10+ doc repos.17:49
* AJaeger_ could sync manually - but that's not nice...17:49
fungidhellmann: AJaeger_: so that may just be the argument we use for including docs build dependencies in global-requirements.txt (for example, debian won't want to package the published output of our docs builds, if they package our manuals they'll want to render them again from source)17:49
AJaeger_fungi: indeed, debian is building the docs themselves...17:50
*** achanda has joined #openstack-infra17:50
dhellmannAJaeger_: oh, that was 10 repos, I thought you meant requirements!17:50
*** hdd has joined #openstack-infra17:50
clarkbwhere this feels weird to me is the docs are not tied to any specific release17:51
AJaeger_dhellmann: yes, syncing to 10 repos.17:51
clarkbwe have a single doc that is? (the install guide)17:51
dhellmannAJaeger_: ok, if debian is actually building the docs then I agree we should have the requirements17:51
fungiso ultimately, i'm unconvinced that documentation projects are any different from other official projects in this regard. they all have some combination of test-time, build-time, install-time and run-time dependencies which should be able to be satisfied from some consistent set of packages somewhere17:51
* dhellmann returns to the review list17:51
clarkbwe can probably get away with it, but it seems like we are mixing continuously released docs with server releases17:51
*** baoli has quit IRC17:51
*** baoli has joined #openstack-infra17:51
*** baoli has quit IRC17:52
clarkbgranted we tie python clients to global reqs because of the hard dependencies there but they are not released with the servers either so maybe that is a non issue17:52
fungii think as long as the stable docs can be built from the same requirements as the stable servers, then that's probably fine?17:52
*** baoli has joined #openstack-infra17:52
AJaeger_dhellmann: we have three requests for docs right now: and
clarkbfungi: ya I think its fine as long as the docs don't try to conflict reqs with stable releases17:53
clarkbwhich is the intent? so should be just fine17:53
AJaeger_fungi, clarkb : Let's write this down as policy ;)17:53
fungiwell, we should mostly consider the stable branches of openstack/requirements subject to similar rules as the stable branches of servers17:53
fungiwhich means no introducing new requirements into stable/icehouse of openstack/requirements just because we want to start using newer tooling to build their documentation17:54
AJaeger_so far, we did not update stable branches of requirements for documentation17:54
fungibut as long as that's understood, should be fine17:54
AJaeger_fungi: that's what we try to do - and so far it worked out ;)17:55
* AJaeger_ removes the proposed session again from the summit cross-project etherpad17:55
*** baoli has quit IRC17:57
*** david-lyle has quit IRC17:58
*** thedodd has joined #openstack-infra17:58
*** david-lyle has joined #openstack-infra17:58
zaroclarkb: ok, i've verified. it works exactly as i described.18:00
*** dprince has quit IRC18:02
*** Ryan_Lane has joined #openstack-infra18:03
clarkbzaro: great thank you18:03
clarkbso my +2 is safe to leave in place18:03
*** denis_makogon has quit IRC18:05
*** denis_makogon has joined #openstack-infra18:05
*** denis_makogon has quit IRC18:05
*** denis_makogon_ is now known as denis_makogon18:05
clarkbthe precise neutron fix is waiting for a precise node to be tested on18:05
*** dmakogon_ has joined #openstack-infra18:06
*** chuckC_ has quit IRC18:06
clarkbthen we will need to rebuild rax images18:06
*** yamahata has quit IRC18:06
openstackgerritA change was merged to openstack-infra/project-config: Remove broken 70-generic kernel as well as 70-virtual
openstackgerritA change was merged to openstack/requirements: Add PyMySQL to global requirements
clarkbfungi: ^ that just merged I can kick off an image rebuild in rax-dfw really quickly18:12
clarkbmeh I will do all three18:12
*** ChuckC has quit IRC18:13
* clarkb has forgotten to wait for puppet to update nodepool.o.o18:13
jogoclarkb: thanks for digging in18:13
clarkbthere will be a redundant image build. But only one :) I remembered after the first one started18:14
*** eglynn-office is now known as eglynn-officeafk18:16
*** ChuckC_ has joined #openstack-infra18:16
mordredclarkb: :)18:17
nibalizerjesusaurus: you said you found some steps that needed to be added to ?18:19
*** e0ne has quit IRC18:21
jesusaurusnibalizer: well, just for running locally I needed to `apt-get install libxml2-dev libxslt-dev`18:21
pleia2gosh, I'm just -1 all over the place this morning, sorry!18:21
jesusaurusthough those might be getting installed somewhere along the way in the puppet18:21
clarkbjesusaurus: those will be required if installing python-*clients from source (any of them that depend on keystoneclient need to link against those libs)18:22
fungialso, as we discovered recently, installing the python-lxml package from the distro instead to avoid having to install those headers conflicts with use of pip install -U (since pip will just unilaterally retrieve newer lxml from pypi and crap on your distro packaged version)18:23
jesusaurusoh, fun18:24
nibalizerfungi: so the current state of nodepool slaves is that they have those packages or no?18:25
fungiso your options are to either not use pip install -U or only install python libs you need from pypi, ever18:25
dtroyerI'm going to hunt down whoever added those to keystoneclient and glare at them mercilessly in Paris…18:25
nibalizerfungi: but thats not the problem I have, the problem i have is that those system libs are needed to build the nokogiri gem, which is in turn needed by NIBZ_PUPPET_STUFF18:25
clarkbdtroyer: ++18:26
funginibalizer: we install those xml dev packages in enough places in our manifests it's probably best to just install them conditionally so as to avoid collisions over duplicate resource declarations18:27
*** ildikov has joined #openstack-infra18:27
nibalizerfungi: well i was just gonna add that install line to the job
*** weshay has quit IRC18:28
funginibalizer: between uninstall-puppet and revoke-sudo i guess?18:29
nibalizeroh i would have to do it before sudo, i guess, well it doesn't hurt to add a line18:29
*** tonytan4ever has quit IRC18:29
*** ChuckC_ has quit IRC18:30
*** jedimike has quit IRC18:30
*** flaper87|afk is now known as flaper8718:31
*** nkinder has quit IRC18:31
openstackgerritSpencer Krum proposed a change to openstack-infra/project-config: Use beaker to do acceptance testing of Puppet
jeblairadding it to the job makes it easier for us to simplify node types later18:32
openstackgerritSpencer Krum proposed a change to openstack-infra/project-config: Use beaker to do acceptance testing of Puppet
nibalizerand now its in the job18:32
fungiincreases job run-time a bit, but unless you're talking about a giant bomb of packages, not really a big deal18:33
jeblairkrotscheck: check out!/story/30218:33
jeblairkrotscheck: when it loads, the project names show up in slow motion, i'm guessing it does a rest call for each18:34
krotscheckjeblair: Yeah, we’ve got a task to add a data caching layer and/or a batch mechanism.18:35
krotscheckjeblair: We’ve got the same issue with multiple copies of the same user being resolved in a conversation. Sometimes, the same user is being loaded lots and lots of times.18:36
jeblairkrotscheck: by cache, you mean client-side?  so that helps this case because it's on the same project?18:36
jeblairkrotscheck: i assume batch would help cases where there were lots of different projects?18:36
krotscheckjeblair: Both. There’s two approaches.18:37
*** _nadya_ has joined #openstack-infra18:37
*** HeOS has joined #openstack-infra18:37
krotscheckjeblair: The first one is to make angular add a little more of a delay when it collects GET requests, so that multiple scope redraws don’t trigger individual loads.18:37
jeblair(and yeah, this suffers from the user problem you describe as well -- this page is 70 requests, half of them for project names, half of them for spencer's name)18:37
*** rainya has joined #openstack-infra18:38
krotscheckjeblair: An alternative to that could be something angular-data, which creates a local cache of all the data and only performs a GET if you explicitly invalidate a record.18:38
*** baoli has joined #openstack-infra18:39
*** _buzz has quit IRC18:39
krotscheckjeblair: The second one is to teach the client and the API to batch up requests in a single multibody request with content-type: application/http.18:39
asselinclarkb, so the pre-clean cannot be done w/o the change. however, I do see how I can change my pre-clean to post-clean and do it outside of d-g.18:39
*** rainya has quit IRC18:39
clarkbasselin: I guess my worry here is that we shouldn't be encoding everything in d-g just because we can18:40
*** baoli has quit IRC18:40
*** amcrn has joined #openstack-infra18:40
clarkbasselin: if third parties have additional job requirements like cleanup that can happen at the job level18:40
jeblairkrotscheck: i think a solution that prevents this particular case from doing repetitive lookups of the same data would be a good first order solution...18:40
*** baoli has joined #openstack-infra18:40
asselinclarkb, yes I agree. I was just following the hook approach which is very convenient way to manage jobs for 3rd party.18:41
jeblairkrotscheck: that is to say, that batching is probably a good idea, but only after we make sure that we don't batch 30 requests for the same data :)18:41
krotscheckjeblair: I agree18:41
clarkbasselin: I don't think its any more or less convenient. The jobs have to be updated either way18:41
*** rainya has joined #openstack-infra18:42
*** ociuhandu has quit IRC18:42
openstackgerritA change was merged to openstack/requirements: Bump eventlet dependency to 0.15.2.
asselinclarkb, what is your opinion of the refactoring? ?18:43
*** cdent has quit IRC18:45
clarkbasselin: I think the refactor is fine.18:45
pleia2hooray, Zanata will be using Wildfly as their default community platform moving forward (rather than the old community-supported version of jboss)18:47
pleia2and at least camunoz_gone will be at the summit too :)18:47
fungipleia2: well fought!18:47
clarkbasselin: though I haven't had time to properly review it yet18:47
asselinclarkb, ok, i will refactor my jobs to not use the pre-post clean. I used pre-clean to sanitize files before they're gziped, but I can unzip them and rezip them after d-g, and should be able to do other cleanup there with your sugggestion.18:47
harlowjaqq, just to verify a question on a review, the 'thick_slaves' are rebuilt for every test run, so if say redis is interacted with for a test run, it won't have to worry about the same data existing there for another test run?18:47
clarkbasselin: I would get a second opinion from other cores before you go off of what I have said18:47
fungiharlowja: yes18:48
clarkbasselin: just because I may be in the minority and I can live with that :018:48
clarkber :)18:48
fungiharlowja: that said, cleaning up after your tests is helpful to developers who may want to run them18:48
asselinclarkb, ok, I'll hold off. but thanks for the suggestion. it's nice knowing another way to do it.18:48
mmedvedeclarkb, thank you for looking into the patches. I would see if adding  functionality we need at job level would work.18:49
*** mpaolino has joined #openstack-infra18:49
*** mpaolino has quit IRC18:49
*** AzherKhan has joined #openstack-infra18:50
*** tonytan4ever has joined #openstack-infra18:51
*** e0ne has joined #openstack-infra18:51
*** ildikov has quit IRC18:52
openstackgerritA change was merged to openstack/requirements: Add ha-guide to projects
jeblair    There appears to be a strong desire to stop gating every project against every other project. To make this more doable we should automate the construction of dependency graphs. Discussion of how we can do that would probably be a good summit topic18:55
jeblairclarkb: ^ did you write that?18:55
asselinmmedvede, clarkb actually in both cases the functionality is defined at the job level (at least for my stuff). i.e. I'll either define the hook functionality, or just call it directly after saving the d-g return code.18:55
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests
openstackgerritA change was merged to openstack/requirements: updated requirements repo readme
*** Sukhdev has joined #openstack-infra18:57
clarkbjeblair: ya18:57
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for libvirt domain destroy bug 1380741
uvirtbotLaunchpad bug 1380741 in nova "Error from libvirt during destroy. Code=38 Error=Failed to terminate process with SIGKILL: Device or resource busy" [High,Confirmed]
jeblairclarkb: i may not understand what you mean, but i don't think anything substantial in that realm is necessary... i think the 'big-tent' discussion is moving more towards having a smaller set of projects share integration tests18:58
clarkbjeblair: basically everyone is going around talking about redoing gating. And I think for that to be succesful we need automation around the graphs that determine what is necessary18:58
openstackgerritAdam Gandelman proposed a change to openstack-infra/project-config: Adds ironic grenade forward testing
jeblairclarkb: and then potentially other projects having pairwise dependencies18:58
clarkbjeblair: its moved on from there. so rather than having a single tent projects like ironic want to gate against only keystone and nova18:58
clarkbexcept that maybe ironic really needs to gate against glance too and we forgot about it18:59
jeblairclarkb: i'm guessing you responded before i wrote that second sentence? :)18:59
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests
clarkbmy concern here is our dependency graphs are really confusing and huge. We cannot simply just say gate X with Y18:59
jeblairi understand jogo wrote a tool to find code-level dependencies18:59
mmedvedeasselin, now thinking about it, I need to revisit it first. I remember I had something I could not define at the job level if there is no ability to change what happens before the cleanup.18:59
clarkbjeblair: not really. its majorly hand currated18:59
clarkbso I think what I have in mind is some way of describing the dependencies in a way that makes jogo's tool more automagic19:00
asselinmmedvede, ok, I'm interested to  know what you find. Thanks.19:00
clarkbright now its look for deps on python-*client and manually curate what that means19:00
jogoclarkb: yeah, right now I am manually making the list of dependencies19:00
*** rainya has quit IRC19:00
jogoin part because its not always clear what the nature of the dependency is19:01
jogoclarkb: patches to are welcome19:02
jeblairi think that's the key -- the (perhaps hypothetical) interaction between two programs that deal with the nova db directly is the kind of interaction that should share a gate job, whereas two programs that merely use their published/released apis may need it less.19:02
jeblairand the first one is harder to detect than the second19:02
jogojeblair: we have a different thing right now as well19:02
mmedvedeasselin, correct me if I am wrong, but a lot of log preparation (i.e. compression) happens during cleanup. So now if we need to add additional log file, it is better to be done before the cleanup, but after d-g. There is no way to do it without the hook there.19:02
openstackgerritA change was merged to openstack/requirements: Fix nits in README.rst
jogojeblair: take cinder + nova19:02
jogoas a consumer of APIs to use cinder, you need to talk to a nova endpoint as well19:03
asselinmmedvede, yes, you can uncompress and recompress.19:03
jeblairjogo: don't i know it :(19:03
jogoas apposed to using glance, which can use swift as a backend. but the end user doesn't need to see it19:03
jeblair(50% of using cinder is using nova)19:03
jogojeblair: for now I think those are the connections which really need co-gating19:04
jogobecause the services are coupled tighter19:04
jeblairjogo: which?  both?19:04
jeblairglance+swift and also cinder+nova?19:04
*** thedodd has quit IRC19:04
*** ChuckC_ has joined #openstack-infra19:05
jogonot glance+swift19:05
jogobecause glance is just consuming the swift API19:05
jogobut cinder+nova is not that relationship19:05
jogonot just19:05
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: Add a --version command line flag to show JJB version
jeblairi think the evolution of this may cause us to formally define these relationships and establish criteria we use for different kinds of gating19:06
jeblairclarkb: is that what you are getting at ^?19:06
*** ChuckC_ has quit IRC19:06
mmedvedeasselin, correct, but doing recompression  seems hacky. It would be much cleaner with the additional hook, or if the post d-g hook did always run.19:06
asselinmmedvede, I ended up doing this regardless. not really a huge performance penalty. About 1s each:
*** ChuckC_ has joined #openstack-infra19:06
clarkbjeblair: I do not want it to be ad hoc. ad hoc is semi what we have now and it is a mess19:06
jeblairclarkb: now i think i understand why you want a summit session about it :)19:07
jogoclarkb jeblair: graph
jogogenerated from the yaml file19:07
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for libvirt domain destroy bug 1380741
uvirtbotLaunchpad bug 1380741 in nova "Error from libvirt during destroy. Code=38 Error=Failed to terminate process with SIGKILL: Device or resource busy" [High,Confirmed]
jogonote the data I am using most likely has bugs in it19:08
*** afazekas has quit IRC19:09
*** chuckC__ has joined #openstack-infra19:09
*** julim has quit IRC19:09
*** johnthetubaguy is now known as zz_johnthetubagu19:09
mmedvedeasselin, I went through history and did not find a good reason why an existing post test hook does not run on failure. Do you know a reason?19:10
*** rainya has joined #openstack-infra19:10
jeblairclarkb, jogo: i may be more comfortable with ad-hoc than you are -- eg, "we discovered an unknown dependency/breakage.  do we: a) add a functest to avoid future api breakage, or b) acknowledge a tight integration between the two components and join them in a shared gate job?"19:10
jeblairclarkb, jogo: but i fully acknowledge we're exploring new territory and am pretty convinced we'll be talking about this at the summit19:11
jogojeblair: well there are few things we can to help detect things19:11
jogoMake sure we account for use of python-*client in projects (as in if a project uses python-*client) there is some sort of dep19:11
jeblairjogo: also, i think your tool/graph is really helpful to help us figure out where we are now, because probably very few of us actually know that :)19:12
*** ChuckC_ has quit IRC19:12
*** chuckC__ is now known as chuckC_19:12
*** ChuckC___ has joined #openstack-infra19:12
jogowe can turn services off that aren't being used in the testing env. So to add a new dependency its an explicit process19:12
fungiand also, because it will make great slides for future conference talks! ;)19:12
*** armax has quit IRC19:12
jogojeblair: yeah, I did it so I could better understand how thins were related and to help further the discussion19:13
jogofungi: yup!19:13
fungimaybe we can finally replace that tired old "openstack is complicated" diagram with one we can mechanically keep updated ;)19:13
jogofungi: actually I was thinking about that19:13
jogowe can do something like this per service as well19:13
jogoturns out python has really easy to use graphing tools19:13
jeblairwe should run this job periodically and publish it so we can just direct link to the graphic :)19:13
fungijogo: yes, did you use dot?19:14
jogofungi: yeah19:14
* jeblair loves graphviz19:14
fungilooked like it19:14
* fungi too19:14
*** davinchi has quit IRC19:14
*** achanda has quit IRC19:15
jogoI would be happy to move this repo into something on btw19:15
jogoif we think it is useful19:15
* fungi totally needs a printed curcuit autorouter extension for graphviz/dot19:15
*** rainya has quit IRC19:15
jeblairfungi: it looks like you decided to perform the rename config -> system config last friday; did something happen?19:15
fungijeblair: we decided to check with the release manager19:15
*** rainya has joined #openstack-infra19:16
dtroyerclarkb: just curious, do you have a rough eta on the rebuilt rax precise images?  'tomorrow' is acceptable if that is the case…19:16
fungijeblair: who in turn was concerned with the possibility that prolonged impact to infra could interact badly with final release candidates happening over the weekend19:16
jeblairfungi: i think if you said "this will keep us from merging config changes that could break the project" he would have been excited for it to happen! ;)19:17
fungijeblair: so anyway, we conceded that this coming friday would be safer as the release would be over and done with19:17
clarkbjeblair I think the problem with ad hoc is it leaves us open to not testing important interactions19:17
clarkbdtroyer its mostly done already. only waiting on iad now. so within an hour19:18
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for lvm hang bug 1373513
uvirtbotLaunchpad bug 1373513 in cinder "Lvm hang during tempest tests" [High,Confirmed]
*** ryanpetrello has quit IRC19:18
*** ryanpetrello has joined #openstack-infra19:19
dtroyerclarkb: ok, thanks.19:19
*** ZZelle_ has joined #openstack-infra19:20
*** ChuckC___ has quit IRC19:20
*** tonytan4ever has quit IRC19:20
*** pelix has quit IRC19:20
*** berendt has joined #openstack-infra19:21
*** ociuhandu has joined #openstack-infra19:22
*** ramashri has quit IRC19:22
*** tonytan4ever has joined #openstack-infra19:22
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for lvm hang bug 1373513
uvirtbotLaunchpad bug 1373513 in cinder "Lvm hang during tempest tests" [High,Confirmed]
*** sdake_ has joined #openstack-infra19:23
*** hdd has quit IRC19:27
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Combine query for bug 1380741 into bug 1353939
uvirtbotLaunchpad bug 1380741 in nova "Error from libvirt during destroy. Code=38 Error=Failed to terminate process with SIGKILL: Device or resource busy (dup-of: 1353939)" [High,Confirmed]
uvirtbotLaunchpad bug 1353939 in nova "Rescue fails with 'Failed to terminate process: Device or resource busy' in the n-cpu log" [High,Confirmed]
*** ryanpetrello has quit IRC19:31
fungimrmartin: i'm ready to rebuild groups.o.o any time now... is this service still a beta test, or do i need to send a heads-up announcement to the mailing list?19:32
mrmartindon't need19:32
mrmartinlet's do it19:32
*** davinchi has joined #openstack-infra19:33
fungimrmartin: okay, powering off the current server19:33
*** TheJulia has joined #openstack-infra19:33
fungimrmartin: it's offline. building the 4gb replacement now19:33
mrmartincool, it is always exciting too see what was changed since we wrote the puppet scripts19:34
*** ChuckC has joined #openstack-infra19:34
*** rainya has quit IRC19:34
fungimrmartin: yes, hopefully it goes okay :/19:35
*** ryanpetrello has joined #openstack-infra19:35
fungii'll know in a couple more minutes since it's in the middle of puppeting right now19:35
openstackgerritA change was merged to openstack-infra/git-review: Define -T/--no-topic to disable review submit with topic
*** julim has joined #openstack-infra19:36
*** thedodd has joined #openstack-infra19:36
*** ChuckC___ has joined #openstack-infra19:36
jeblairkrotscheck: sorry i missed storyboard meeting this morning; i seem to be a little ill right now and i apparently needed some sleep19:38
krotscheckjeblair: No worries.19:38
*** sdake_ has quit IRC19:38
krotscheckjeblair: Do you have any specific questions on agenda items?19:38
*** nkinder has joined #openstack-infra19:39
*** _buzz has joined #openstack-infra19:39
jeblairkrotscheck: no; though i gathered you wanted some input from me on the infra migration19:39
*** yolanda_ has quit IRC19:39
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests
krotscheckjeblair: Not as such. The migration script is up and working and has been tested on the current batch of projects, however we’re still a few features short of being ready.19:40
*** yolanda has quit IRC19:40
*** AzherKhan has quit IRC19:41
krotscheckjeblair: So I guess if you want to pull the trigger early you can, but either way you can have the gun.19:41
*** rainya has joined #openstack-infra19:43
openstackgerritCedric Brandily proposed a change to openstack-infra/git-review: Add --submit-immediately/-S command to submit immediately after push
*** achanda has joined #openstack-infra19:45
fungimrmartin: it's booted up now. i've switched the address record in dns from to but you may still have the old one cached for a bit19:45
openstackgerritA change was merged to openstack-dev/cookiecutter: Add .venv to .gitignore
*** weshay has joined #openstack-infra19:48
*** homeless has quit IRC19:49
*** _nadya_ has quit IRC19:50
mrmartinfungi, it seems to be ok, I just login to run a scheduled tasks of event imports19:50
mrmartindo we need to change something with cacti, or this is related to ip change?19:52
fungimrmartin: the ip address change. it will take the cacti server a little while to notice the dns record changed19:52
*** cdent has joined #openstack-infra19:53
asselinmmedvede, I don't follow your comment from 12:10...19:53
*** cdent has quit IRC19:53
fungimrmartin: i'll keep an eye on that and give it a kick if it doesn't for some reason19:53
mrmartinok thanks19:53
*** groknix has joined #openstack-infra19:54
fungimrmartin: i've also just now set the reverse dns entries for it, and added an ipv6 aaaa record since it didn't have one before19:54
*** ryanpetrello_ has joined #openstack-infra19:55
*** prad has joined #openstack-infra19:55
*** ryanpetrello has quit IRC19:55
*** ryanpetrello_ is now known as ryanpetrello19:55
*** krtaylor has quit IRC19:56
openstackgerritA change was merged to openstack-infra/project-config: Use beaker to do acceptance testing of Puppet
mmedvedeasselin, the post_test_hook is not being run if the gate hook returns non-zero. And I can not see why. I though maybe you found out why.19:58
*** ramashri has joined #openstack-infra19:58
*** ryanpetrello has quit IRC19:58
*** krtaylor has joined #openstack-infra19:59
*** ryanpetrello has joined #openstack-infra20:00
rainyawhere are the Cats?!20:03
rainya(I have nothing else to offer today)20:03
AJaeger_mmedvede: see devstack-gate/devstack-vm-gate-wrap.sh20:03
*** whoops has joined #openstack-infra20:03
AJaeger_"if [ $GATE_RETVAL -eq 0 ] && function_exists "post_test_hook"; then"20:03
mrmartinfungi, I've checked the portal, seems to be ok now. The only thing I noticed was the cacti graphing.20:04
fungimrmartin: excellent. if the graph doesn't clear up in the next few minutes i'll look into it20:05
*** hdd has joined #openstack-infra20:05
*** ChuckC___ has quit IRC20:05
mrmartinok, I'm leaving now, cause tomorrow I'll travel. Drop me an email if you need any help regarding the portal.20:06
*** kgiusti has quit IRC20:06
fungimrmartin: will do--thanks again and safe travels!20:06
*** mpaolino has joined #openstack-infra20:06
fungimrmartin: oh, and the graph just updated showing the new maximum20:07
*** mpaolino has quit IRC20:07
*** ChuckC___ has joined #openstack-infra20:07
fungimrmartin: so looks like we're all set. unless i hear anything from you otherwise, i'll delete the old server in a few days20:07
mrmartincool, let's what happens. if it will allocate all of the 4GB, I need to investigate.20:07
mrmartinok, just going to linuxcon, so I guess they have wifi there.20:08
fungioh, yep. they totally will ;)20:08
*** david-lyle has quit IRC20:09
*** berendt has quit IRC20:10
dhellmannanteaya: do you have a few minutes to look over this change to the doc jobs for some oslo libs?
mmedvedeAJaeger, thank you. My 'why' was about why there is the check in that prevents post test hook from running on test failure. I skimmed through git history for the wrap script and did not find a compelling reason to do so.20:11
*** bdpayne has joined #openstack-infra20:13
AJaeger_mmedvede: can't help further, sorry20:14
mmedvedeAJaeger, no worries, thx20:15
AJaeger_dhellmann: I think anteaya is out for today. Better ask fungi, clarkb or jeblair...20:15
fungidhellmann: yeah, they get a much better holiday up north today than we do20:15
fungidhellmann: taking a look20:16
*** flaper87 is now known as flaper87|afk20:18
*** emagana has joined #openstack-infra20:22
*** hdd has quit IRC20:23
openstackgerritA change was merged to openstack-infra/elastic-recheck: Combine query for bug 1380741 into bug 1353939
uvirtbotLaunchpad bug 1380741 in nova "Error from libvirt during destroy. Code=38 Error=Failed to terminate process with SIGKILL: Device or resource busy (dup-of: 1353939)" [High,Confirmed]
uvirtbotLaunchpad bug 1353939 in nova "Rescue fails with 'Failed to terminate process: Device or resource busy' in the n-cpu log" [High,Confirmed]
*** miqui has quit IRC20:23
dhellmannAJaeger_, fungi: oh, right, forgot today was a holiday for some, thanks!20:24
*** ramashri has quit IRC20:30
*** dduvall has joined #openstack-infra20:30
openstackgerritA change was merged to openstack-infra/project-config: Publish cliff and stevedore docs on
dduvallIs there a known issue with saving contact information via the gerrit site?  I'm repeatedly getting "Server Error: Cannot store contact information".20:33
fungidduvall: did you follow the instructions here in sequence?
davinchidduvall: you must register first20:33
davinchidduvall: as a contributor on the other site20:33
*** whoops has quit IRC20:33
fungidduvall: that error is usually an indicator that you didn't sign up as a foundation member first, or that you used a different foundation primary e-mail address from what you set in gerrit as your preferred e-mail address since it wants to link them20:34
*** baoli has quit IRC20:34
openstackgerritAndreas Jaeger proposed a change to openstack-infra/config: Fix indention in gerrit.config.erb
dduvallAh. I skipped over that; sorry.20:35
mwhudsonif i keep seeing failures due to bug 1379201 is there anything i can do other than keep retrying?20:35
uvirtbotLaunchpad bug 1379201 in openvswitch "openvswitch-datapath-dkms 1.4.6-0ubuntu1.12.04.3: openvswitch kernel module failed to build [error: too many arguments to function ‘ip_select_ident’]" [Medium,Fix committed]
mwhudsonor is it better to wait until there is a resolution?20:36
fungidduvall: no worries. if you have any other problems after you clear that up, though, please do let me or someone in here know20:36
*** shardy is now known as shardy_z20:36
*** _buzz has quit IRC20:36
dduvallWouldn't mind a more directed error message, but hopefully there aren't too many people like me making that mistake.  :)20:36
fungimwhudson: the fix we applied over the weekend apparently only worked for our hpcloud workers, so rackspace workers continued to exhibit the problem. we got an expanded fix in place a few hours ago and clarkb rebuilt the affected nodepool snapshots so it's (hopefully) taken care of now20:37
*** mmedvede has quit IRC20:37
mwhudsonfungi: ah ok20:38
mwhudsonfungi: i was kind of perplexed as to how this could only happen some of the time20:38
fungidduvall: yeah, the error message is actually hard-coded into gerrit, software we run but we didn't write, and the upstream patch to make that properly configurable is unfortunately nontrivial. instead we're working on finding a good way to decouple that requirement20:38
dduvallNo sweat; good to know.20:38
*** sdake_ has joined #openstack-infra20:39
fungimwhudson: yeah, rackspace pre-installs multiple kernel package flavors on their base images, so uninstalling one just caused the bootloader to fall back to the other we didn't notice (as opposed to hpcloud where we get to use ubuntu's official published cloud images)20:39
*** ramashri has joined #openstack-infra20:40
*** dmsimard is now known as dmsimard_away20:40
fungimwhudson: but now that we have access to glance in both providers, we're working on making and uploading our own consistent base images20:41
mwhudsonoh right, yes, i remember talking about this20:41
mwhudsonin the context of arm6420:41
fungimwhudson: an upcoming evolution of that is that we're probably going to decouple image building from the nodepool scheduler, so you may in theory be able to have an x86-based nodepool server with arm64-based image build workers20:42
lifelessit almost starts to sound like every other buildd at this point20:43
dduvallfungi, davinchi: all set; thanks.20:43
fungimwhudson: spec proposal at
mwhudsonfungi: what are you using to build the images?  live-build or something?20:43
davinchidduvall: had the same problem like 3 hours ago20:43
fungidduvall: great! happy hacking20:44
fungilifeless: indeed it does20:45
mwhudsonfungi: so that's "a big pile of shell"20:45
*** _buzz has joined #openstack-infra20:46
fungimwhudson: it is a moderate-sized pile of shell onto which you can, er, pile more shell20:46
*** julim has quit IRC20:46
fungi(and some descriptive data too)20:46
mwhudsonoh well, i guess it could be perl20:46
fungimwhudson: indeed. there's always a silver lining20:47
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: Added flexible publish support
mwhudson*cough* sbuild *cough*20:47
*** flaper87|afk is now known as flaper8720:49
fungimwhudson: yep, except for the bits where it gets used to set up "notdebian" (rhel/centos, fedora, suse, at some point soon apparently windows?)20:49
*** mrmartin has quit IRC20:49
*** radez is now known as radez_g0n320:49
mwhudsonyes, i have a bit of a blind spot there20:49
fungiyou and me both20:50
*** ramashri has quit IRC20:52
*** ramashri has joined #openstack-infra20:52
*** AJaeger_ has quit IRC20:53
*** david-lyle has joined #openstack-infra20:53
*** dane_leblanc has joined #openstack-infra20:55
fungiclarkb: i sampled a recent devstack-precise node in iad and it's running on the older kernel we downgraded into now20:55
fungiseems to be working as we intended20:56
*** Sukhdev has quit IRC20:56
*** rkukura has joined #openstack-infra20:57
lifelessmwhudson: it also has some python20:58
*** rkukura has quit IRC20:59
*** garyh has joined #openstack-infra20:59
*** rkukura has joined #openstack-infra20:59
*** david-lyle has quit IRC20:59
*** Sukhdev has joined #openstack-infra21:00
*** dustins has quit IRC21:01
*** cpowell has quit IRC21:01
*** marcusvrn_ has quit IRC21:02
*** mmedvede has joined #openstack-infra21:04
*** mfer has quit IRC21:04
*** sdake_ has quit IRC21:05
clarkbfungi great thanks21:05
*** amotoki has joined #openstack-infra21:06
*** dkliban is now known as dkliban_brb21:06
fungiclarkb: i didn't check all three regions, but assume that if one of them failed to build you'd have noticed21:07
clarkbya dfw and ord looked good before I ran off to lunch21:07
*** bswartz has quit IRC21:07
clarkbiad was the one I was waiting on21:07
fungiin that case, we're all set21:08
*** chandankumar has quit IRC21:09
*** groknix_ has joined #openstack-infra21:10
*** denis_makogon has quit IRC21:10
*** groknix has quit IRC21:13
clarkbgood morning21:15
*** maxbit has quit IRC21:15
*** e0ne has quit IRC21:18
*** rlandy has quit IRC21:21
*** sdake_ has joined #openstack-infra21:21
*** groknix_ has quit IRC21:23
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: add documentation for the delete command
clarkbhrm dfw nodes are still showing -70. I must not have started the second build there after the install devstack dependencies script updated. I am rebuilding to be double sure21:29
*** dizquierdo has joined #openstack-infra21:34
*** tonytan4ever has quit IRC21:34
*** sdake_ has quit IRC21:38
*** dizquierdo has quit IRC21:39
*** dkliban_brb is now known as dkliban21:39
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: add a JJB command reference
*** flaper87 is now known as flaper87|afk21:41
tchaypomorning el presidente21:42
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Remove query for bug 1317298
uvirtbotLaunchpad bug 1317298 in tempest "timeout in wait_for_resource_deletion in tempest teardown" [Undecided,New]
openstackgerritSteve Baker proposed a change to openstack-infra/project-config: Add python-jobs to heat-templates
*** _buzz has quit IRC21:46
*** dims has quit IRC21:50
*** jheroux has quit IRC21:51
clarkbTC members for 125796 did you all want to +1 that first?21:51
zarofungi: what do you think of David's comment here?
clarkbseems like low impact but I don't want to surprise everyone21:51
*** jraim has quit IRC21:52
waynrzaro: ah, I didn't realize that was you in the gerrit code review21:52
zarofungi: i've noticed other projects have already approved that chnage.21:52
zarowaynr: :)21:52
fungizaro: while i am indifferent about it being a required section, i do think that having a consistent name for it is beneficial21:53
*** jraim has joined #openstack-infra21:53
waynrto be honest I don't really understand the new jenkins failures21:53
openstackgerritSteve Baker proposed a change to openstack-infra/project-config: Add python-jobs to heat-templates
zarowaynr: your change?21:54
*** david-lyle has joined #openstack-infra21:55
zarowaynr: ahh, you need to rebase.21:55
waynroh i see21:55
zarowaynr: when the CI goes to merge it will rebase on top of master then run tests.  that's what's broken.21:56
zarofungi: wouldn't that change break the doc generation?21:57
*** Mithrandir has quit IRC21:57
fungizaro: i went ahead and approved 119910. the objections on the governance change are about whether to require that env, not against having a consistent name for it where it already exists21:57
*** datsun180b has quit IRC21:58
fungizaro: we don't run that tox env in jobs21:58
openstackgerritWayne Warren proposed a change to openstack-infra/jenkins-job-builder: Accept colon-delimited list of paths in args.
fungizaro: we run tox -e venv python build_sphinx21:58
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Rename doc environment to docs
zaroohh, ok. thanks.21:58
jeblairclarkb: i doubt the tc will have a problem with it, but i -1d on a technicality21:59
clarkbjeblair: thanks21:59
clarkbhrm why does group based policy need a repo just for its tests22:00
*** emagana has quit IRC22:00
fungithey learned how to create new repos recently, so when you have a new repo hammer every problem looks like a nail?22:01
*** emagana has joined #openstack-infra22:01
fungii agree the choice is non-obvious22:01
clarkbI am going to be meaner than SergeyLukjanov and -122:02
*** dimtruck is now known as zz_dimtruck22:03
*** markmcclain has quit IRC22:04
*** pc_m has quit IRC22:04
*** bknudson has quit IRC22:04
*** emagana has quit IRC22:05
*** emagana has joined #openstack-infra22:06
clarkbfungi: jeblair 127319 makes me think we should update all of our stable regexes to match proposed as well22:10
clarkbfungi: jeblair: or am I over thinking that?22:10
clarkbadam_g: you around? I think 127319 should remove the trailing .*22:11
waynrzaro: something i've been thinking about for JJB is allowing a list of default names to be specified, where parameters in higher-indexed defaults (further down on the list) override earlier values22:11
clarkbadam_g: left comment on review22:11
adam_gclarkb, oh, sure. thanks22:12
*** david-lyle_ has joined #openstack-infra22:12
fungiclarkb: hrm... maybe to avoid future omissions when jobs are added for stable branches before the release?22:12
fungiclarkb: not sure the added complexity is worth it22:12
clarkbfungi: well I think we end up testing the wrong things with proposed/ otherwise22:13
clarkbfungi: particularly in the case like adam_g's22:13
clarkbfungi: if we didn't put proposed in there we would stop testing anything until the proposed branch was renamed to stable22:13
fungi(in 4 days)22:13
openstackgerritAdam Gandelman proposed a change to openstack-infra/project-config: Run sideways ironic greande only on juno branches
*** sarob has quit IRC22:14
clarkbSergeyLukjanov: 127368 has a small thing from me. let me know if I am mistaken22:14
openstackgerritA change was merged to openstack-infra/project-config: Document jenkins/scripts/
clarkbadam_g: perfect +222:15
*** david-lyle has quit IRC22:15
openstackgerritA change was merged to openstack-infra/project-config: Add DNS Services to home page
fungiclarkb: yeah, if we're going to start limiting jobs to stable or omitting them from stable during rc, then maybe preemptively replacing stable with (proposed|stable) in every regex to future-proof against cut+paste omissions is worthwhile, but we didn't bother in previous releases either (when it was milestone-proposed) right?22:16
clarkbfungi: correct, but that omission identified issues with the milestone-proposed process that switching to proposed/* should correct22:16
clarkbfungi: we should probably do it as we make changes and not worry about it so much as a one shot thing22:17
openstackgerritA change was merged to openstack-infra/project-config: Add tripleo-ha job to diskimage-builder
openstackgerritA change was merged to openstack-infra/project-config: Add missing docimpact-group for Ironic config
*** sarob has joined #openstack-infra22:18
fungiclarkb: it still seems like this only applies for a couple weeks while jobs are being added in advance of the release, and are only omitting the new job on the soon-to-be-stable proposed branch. i guess it runs the risk that jobs added in that timeframe targeting the next release are broken come release day22:18
*** dmsimard_away is now known as dmsimard22:18
clarkbfungi: I alsothink we are not grenading properly22:18
*** dmsimard is now known as dmsimard_away22:19
openstackgerritA change was merged to openstack-infra/project-config: Added horizon channel
fungiclarkb: really? i could have sworn i checked the workspace setup for a grenade job once the first rc was cut22:19
fungiclarkb: but it's possible i am misremembering or checking wrong22:19
clarkbfungi: its doing icehouse to master right now22:19
clarkbfungi: not icehouse to proposed/juno22:20
fungioh, that22:20
clarkbat least that is my initial impression22:20
fungifor proposed/juno changes it was doing stable/icehouse to proposed/juno22:20
*** signed8bit has quit IRC22:21
clarkbok thats good22:21
fungitesting stable/icehouse to master instead of proposed/juno to master for a couple weeks doesn't seem particularly worrisome22:21
clarkbso only icehouse changes may be broken22:21
clarkband ya definitely less worrisome22:21
clarkbsince that side of the euqation moves less22:21
fungiafter all, if we somehow break the ability to upgrade from juno to master for a couple weeks, we do have ~6 months to solve that ;)22:22
clarkbwell no we can break icehouse to juno22:22
clarkbits the forward test that won't do what we want22:22
fungias for breaking icehouse changes, we never did get "forward grenade" working right?22:22
jeblairit might be good to change the forward check for icehouse to land on juno22:22
jeblairclarkb: yeah22:22
fungior has that been voting and i missed it?22:23
clarkbfungi: I think forward neutron was removed22:23
clarkbbut forward nova net is running22:23
clarkblet me check22:23
fungiwell, yes, i agree that needs to be correct22:23
fungidon't want to have people unable to upgrade from icehouse to juno come release day22:23
fungibecause, you know, they will probably want to22:23
clarkbnope I lied its just experimental22:23
clarkbso we do have that test, but only runs on experimental22:24
*** Mithrandir has joined #openstack-infra22:24
fungiso i guess we weren't properly guaranteeing that anyway, except insofar as it would then get noticed on grenade jobs for master once broken22:25
zarowaynr: hmm, that feature sounds like it exists, no?22:25
*** mriedem has quit IRC22:27
*** packet has quit IRC22:28
clarkbfungi: zaro: I am going to test jjb py3.4 compat really quickly and if that works I will approve thechange to make py33 voting22:29
clarkbfungi: does that mostly line up with what you talked to pelix about earlier?22:29
*** cody-somerville has quit IRC22:30
fungiclarkb: yeah, pelix said that in fact he only had 3.4 so that's how he'd been testing it all along22:30
clarkbeven better22:30
clarkbbut I have them running locally so may as well double check22:31
*** thedodd has quit IRC22:32
*** camunoz_gone is now known as camunoz22:35
*** baoli has joined #openstack-infra22:35
*** zz_dimtruck is now known as dimtruck22:35
*** mfink has quit IRC22:36
clarkbok py34 does indeed look good. I have approved the py33 voting test change22:37
*** david-lyle_ is now known as david-lyle22:40
*** dims_ has joined #openstack-infra22:41
clarkbok new dfw image is done22:43
clarkbI will check nodes once old ones can cycle out22:43
*** emagana has quit IRC22:48
*** baoli has quit IRC22:48
*** emagana has joined #openstack-infra22:49
openstackgerritA change was merged to openstack-infra/project-config: Fix Horizon integration job: permissions
clarkbjhesketh: what is the mechanism that makes 126482 work22:50
waynrzaro: oh? i guess i misunderstood how it already works...I thought you could only specify one defaults, for example "global"22:50
*** emagana has quit IRC22:50
*** emagana has joined #openstack-infra22:51
clarkbjhesketh: is touching the file sufficient to make the server process recalculate the sums?22:51
clarkbfungi: mordred jeblair 127652 would be a good one to review22:52
clarkbit will stop spinning us on he centos7-dib slaves22:52
jheskethclarkb: so I took a guess that it was, asked fungi to do so and it seemed to work22:54
jheskethso yes22:54
fungistrange but true!22:54
jheskethgerrit has the ability to reload the header files automatically, so it just needs to know to do so22:54
jheskethand that's when it does the weird 'let me munge in some sums' stuff22:54
clarkbfungi: any reason I shouldn't go ahead and approve that change then?22:55
clarkbSergeyLukjanov: gave it a second +2 but not +W. I am happy to +W22:55
*** groknix has joined #openstack-infra22:55
fungiclarkb: no, should be perfectly safe to approve22:56
jeblairjhesketh: oh neato, nice catch! :)22:56
*** ChuckC___ has quit IRC22:57
clarkbgreat /me does so22:57
openstackgerritDavanum Srinivas (dims) proposed a change to openstack-infra/config: Save last setting of Toggle CI button
*** david-lyle has quit IRC22:58
*** Sukhdev has quit IRC23:00
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Added project group list and detail view.
clarkbnibalizer: so I ran `git diff 0.2.0..3.2.0` in puppetlabs-ntp and the diff I get is much bigger than the one you link in 11766523:01
clarkbnibalizer: am I overthinking this?23:01
*** openstackgerrit has quit IRC23:03
*** groknix has quit IRC23:03
*** dkliban is now known as dkliban_afk23:04
*** openstackgerrit has joined #openstack-infra23:04
*** dane_leblanc has quit IRC23:05
openstackgerritDarragh Bailey proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests
*** davinchi has quit IRC23:06
*** Sukhdev has joined #openstack-infra23:07
*** david-lyle has joined #openstack-infra23:08
*** bswartz has joined #openstack-infra23:08
*** marun has quit IRC23:09
*** koolhead17 has quit IRC23:10
nibalizerclarkb: i showed  the diff of the ntp.conf23:12
nibalizerbetween current ntp and tge new ntp23:12
clarkbnibalizer: right, but a lot of things have changed in the module itself23:12
nibalizerso i showed the changes23:12
clarkbwhich we should care about right?23:12
nibalizeryup i believe it23:12
nibalizerwell our invocation of ntp module is "include ntp"23:13
nibalizerso as long as its still sane defaults, meh23:13
*** dmsimard_away is now known as dmsimard23:13
nibalizerntp is unfortunatley the 'science project' module23:13
clarkbmeaning it is experimental for puppet module patterns?23:14
nibalizerso if there is new patterns to try out in puppet.... yea23:14
* fungi is for science23:14
*** david-lyle has quit IRC23:14
nibalizerpoint is, should be fine to update23:14
clarkbok maybe I approve that one tomorrow early in the day rather than late on monday23:14
*** amotoki has quit IRC23:15
mordredclarkb, fungi: woot. I have all of the code working now for end-to-end node provisioning including cinder volumes, attaching, formatting and mounting them23:17
nibalizermordred: ooo23:18
clarkbmordred: fungi: 118924 is another one I will throw out there. I can babysit the approval if someone else +2s it23:18
clarkbmordred: nice23:18
mordrednibalizer, clarkb:
mordredthere's the example file for doing that ... I'll have a patch up soon update the infra patch23:19
clarkbmordred: what does "register:" do?23:20
mordredclarkb: all of the things appropriate don't-run-twice if they shouldn't, and most everything will appropriately just re-try the things that need to be retried23:20
mordredclarkb: it stores a thing in a variable for use in subsequent plays23:20
clarkbso 'node' is a varialbe iwth the uuid of that node?23:20
clarkbthat sort of thing?23:20
mordredso, the os_compute module returns a certain set of data - that is put into the node variable23:21
*** sarob has quit IRC23:21
mordredclarkb: it's actually a dict of data23:21
clarkbeven better23:21
mordredthere are some really sexy things going on with that register- the openstack key in the node dict contains all of the metadata that the openstack control plane knows about the node23:21
clarkboh I see then you use it under the add node public IP thing23:21
*** sarob has joined #openstack-infra23:22
mordredif you look down to register: volume - that one registers the output of os_compute_volume, which has the same info in openstack, except now with all of the node's cinder volumes too23:22
mordredclarkb: so after the add_host, the in-memory ansible inventory looks exactly like it would if you started ansible from scratch with teh node pre-existing23:22
clarkbso that is updating the openstack var not setting it?23:23
mordredit's the openstack key on the volume dict - different than the openstack key on the node dict23:23
mordredboth have the same basic info, but at the point in time that the os_compute module fired, the cinder volumes didn't exist23:24
mordredso I used the os_compute_volume output to seed the info needed by the subsequent plays23:24
clarkbI am slightly worried that the language here is ill suited to this sort of thing23:24
clarkbbut getting retries and all in one node boots will be really nice23:24
clarkbwhere does openstack.volumes come from?23:25
clarkbis that the key that volume.openstack kicks out?23:26
mordredit's just part of the metadata produced by the ansible openstack modules23:26
openstackgerritA change was merged to openstack-infra/project-config: Fuel-plugins repository for Fuel
jgriffithmordred: interesting....23:26
clarkba giant blue bird just flew into my window23:26
clarkbthen flew away23:27
jgriffithmordred: did you do a pull request?23:27
openstackgerritA change was merged to openstack-infra/project-config: Make py33 job voting for jenkins-job-builder
mordredjgriffith: on ansible? yes - but I've abandoned it a bit23:28
openstackgerritA change was merged to openstack-infra/config: Stop the elastic-recheck bot more gracefully
mordredjgriffith: I'm working ona  complete rewrite of the modules because what's in the ansible tree right now is completely unusable in a real workload23:28
jgriffithmordred: ahh...23:28
mordredand fixing it is too backwards incompatible23:28
jgriffithmordred: I started a version and was going to try and submit it this week23:28
openstackgerritA change was merged to openstack-infra/config: Reload the gerrit header with its dependencies
mordredoh - eeek. so - I've got a whole repo of them23:28
jgriffithmordred: but I'd rather work with you or leverage what you're doing23:28
mordredyes please!23:28
jgriffithmordred: haha... no worries, so do I :)23:28
jgriffithmordred: it's mostly been experimental for me23:28
jgriffithmordred: yeah... found it :)23:29
mordredjgriffith: I've got all of the logic that infra already uses in node launching23:29
openstackgerritA change was merged to openstack-infra/config: Set devstack-centos7-dib min ready to 0
mordredjgriffith: it's also building off of the
*** imcsk8 has quit IRC23:29
mordredso that it's not a ton of logic actually in the modules themselves that's not ansible specific23:29
jgriffithmordred: yeah... cool though23:29
openstackgerritA change was merged to openstack-infra/project-config: Allow rally-core to rebase changes in rally is the one that's the closest to looking like I want it to23:29
jgriffithmordred: similar to what i have, but you have better error checking :)23:29
mordredI've got a bunch of refactoring I want to do elsewhere23:30
mordredthat's almost never been said by anyone23:30
jgriffithmordred: LOL23:30
*** atiwari has quit IRC23:30
jgriffithmordred: ok... well, I'm happy to mess with it and give feedback23:30
mordredjgriffith: my goal is to eventually submit these to the ansible extras repo - but I want to make sure we've got a stable interface first23:30
mordredjgriffith: because ansible is VERY strict about breaking people23:30
jgriffithmordred: also if there's something you want someone to dive into let me know23:30
jgriffithmordred: roger that23:31
mordredjgriffith: well, first of all, if you wanted to make sure I'm doing all the right things in the cinder related modules, that would be neat :)23:31
jgriffithmordred: sure, I'm happy to dig into that23:31
mordredI think I need to filter the attachments repo in the meta returned here:
*** dangers is now known as dangers_away23:32
jgriffithmordred: I'll jump into it tomorrow if I can, if not maybe wed night23:32
mordredjgriffith: I have not double checked to make sure those are all only for the server in question23:32
mordredjgriffith: awesome23:32
*** openstackgerrit has quit IRC23:33
*** ociuhandu has quit IRC23:33
clarkbmordred: attachments repo?23:33
*** openstackgerrit has joined #openstack-infra23:34
fungiclarkb: 118924 lgtm. approve at will23:37
clarkbwoot thank you23:37
clarkbI am going to do that first thing ish tomorrow23:38
*** mancdaz has quit IRC23:39
*** zz_johnthetubagu has quit IRC23:39
*** salv-orlando has quit IRC23:40
clarkbanteaya: good catch on I would agree that the displayed path should match the linked path. Considering sdague is busy/out any chance you want to propose the fix?23:40
clarkber update that patch when you get back from holidays?23:40
*** rainya has quit IRC23:40
*** dmsimard is now known as dmsimard_away23:42
clarkbok finally dfw is looking better iwth the kernel version23:43
mordredclarkb: gah. not repo - attachments dict23:49
*** emagana has quit IRC23:50
*** ramashri has quit IRC23:50
*** emagana has joined #openstack-infra23:51
clarkbmordred: 11946623:51
clarkbcan haz update?23:51
*** bknudson has joined #openstack-infra23:55
*** emagana has quit IRC23:55
mordredclarkb: uhm. the comments on that change are wrong23:56
mordredclarkb: the change is removing
mordredthe other files point to
jrollmordred: you going to ansiblefest?23:57
clarkbdoes the pypi_mirror vhost not cover both?23:57
clarkbmordred: my understadning is that you are removing both /openstack and /simple23:57
clarkbmordred: or at least removing the management of both23:57
mordredclarkb: AH. I see it now23:57
mordredwill fix23:58
mordredjroll: I am not - I heard about it too late, and there's honestly not much content I'd find interesting23:58
mordredI'm tempted to just go up and say hi to michael23:58
jrollmordred: boo, our office is around the corner23:58
mordredit turns out - ansible is very simple to use and doesn't require much in the way of learning advanced tricks :)23:58
jrollwhich michael?23:58
*** patrickeast has quit IRC23:59
mordredthis may not serve them well on the conference circuit - there just isn't that much to talk about23:59
mordredjroll: dehann23:59
*** baoli has joined #openstack-infra23:59
jrollI see23:59
mordredjroll: when is it again?23:59

Generated by 2.14.0 by Marius Gedminas - find it at!