Friday, 2014-10-10

*** harlowja has quit IRC00:02
*** harlowja has joined #openstack-infra00:02
*** HenryG has quit IRC00:04
*** achanda has quit IRC00:08
*** beekneemech has joined #openstack-infra00:09
*** Sukhdev has quit IRC00:09
*** bnemec has quit IRC00:09
*** MaxV has joined #openstack-infra00:09
*** melwitt has quit IRC00:10
openstackgerritA change was merged to stackforge/gertty: Rename gerrit-gertty.yaml to googlesource-gertty.yaml
openstackgerritA change was merged to stackforge/gertty: Fix help string for --version
openstackgerritA change was merged to stackforge/gertty: Unify small vs. capital letters in help output for consistency
*** flaviof is now known as flaviof_zzz00:13
*** MaxV has quit IRC00:14
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Filter tasks by project, project_group, and various other fields
*** packet has joined #openstack-infra00:16
*** yamahata has joined #openstack-infra00:16
*** yamamoto has joined #openstack-infra00:16
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Add ansible node launching
*** achanda has joined #openstack-infra00:20
*** packet has quit IRC00:20
*** bdpayne has quit IRC00:20
*** david-lyle has joined #openstack-infra00:25
ianwfungi / clarkb : ok, one other issue fixed in, should be ok ... similar to the RELEASE var thing, if it turns out to be unstable maybe we should pin to a version and update asynchronously00:26
jeblairianw: yeah, i think it's worth trying it out and seeing how much of a problem it is; and if it is a big problem, pin.00:28
jeblairianw: but i think dib wants to be continuously deployable, so i think it should be safe00:28
jeblairianw: and i'm okay with helping find out if it isn't safe in the (hopefully rare) cases something slips through00:28
fungiworst case, we'll "fail safe" (e.g. keep using old images while we can't build a replacement)00:29
*** TheJulia has quit IRC00:30
*** salv-orlando has quit IRC00:31
*** zaphod42 has quit IRC00:31
*** zaphod42 has joined #openstack-infra00:31
*** HeOS has joined #openstack-infra00:32
*** zaphod42 has quit IRC00:32
*** achanda has quit IRC00:34
*** groknix has joined #openstack-infra00:34
*** HeOS has quit IRC00:36
*** groknix_ has quit IRC00:38
openstackgerritK Jonathan Harker proposed a change to openstack-infra/nodepool: Add missing dependencies to README.rst
*** hdd has joined #openstack-infra00:41
*** mmedvede has quit IRC00:42
*** packet has joined #openstack-infra00:43
*** rwsu has quit IRC00:47
*** packet has quit IRC00:48
*** Ryan_Lane has quit IRC00:50
*** hdd has quit IRC00:52
*** groknix_ has joined #openstack-infra00:54
*** otter768 has joined #openstack-infra00:55
ianwfungi: more worried about little things that don't fail the build, but build a broken image00:55
ianwfungi: anyway, i'm in contact with the centos image people; one thing on the todo list is to incorporate diskimage-builder into the pre-release checks before they publish new .qcows00:56
fungiyeah, we could eventually consider acceptance tests for images i guess00:56
ianw(have to get it working here first :)00:56
*** groknix has quit IRC00:56
*** JayJ has joined #openstack-infra00:56
*** david-lyle has quit IRC00:57
*** spzala has quit IRC00:57
*** yongli has joined #openstack-infra01:00
*** unicell has joined #openstack-infra01:00
*** groknix_ has quit IRC01:00
*** JayJ has quit IRC01:02
*** amcrn has quit IRC01:02
*** JayJ has joined #openstack-infra01:03
*** shashankhegde has quit IRC01:04
*** jerryz_ has joined #openstack-infra01:07
*** JayJ has quit IRC01:07
*** JayJ has joined #openstack-infra01:08
*** MaxV has joined #openstack-infra01:10
*** JayJ has quit IRC01:12
*** JayJ has joined #openstack-infra01:13
*** jerryz_ has quit IRC01:14
*** MaxV has quit IRC01:15
openstackgerritA change was merged to stackforge/gertty: Remove call of 'python testr' in tox.ini
*** JayJ has quit IRC01:17
*** JayJ has joined #openstack-infra01:18
*** JayJ has quit IRC01:22
*** JayJ has joined #openstack-infra01:23
*** JayJ has quit IRC01:28
*** JayJ has joined #openstack-infra01:28
*** mmedvede has joined #openstack-infra01:29
*** ivar-lazzaro has quit IRC01:30
*** JayJ has quit IRC01:33
*** JayJ has joined #openstack-infra01:33
*** weshay has quit IRC01:34
*** sunrenjie has joined #openstack-infra01:38
*** JayJ has quit IRC01:38
*** JayJ has joined #openstack-infra01:38
*** nosnos has joined #openstack-infra01:39
*** JayJ has quit IRC01:40
*** IAX has quit IRC01:42
*** baoli has joined #openstack-infra01:44
*** packet has joined #openstack-infra01:44
*** rainya has joined #openstack-infra01:47
*** patrickeast has quit IRC01:48
*** JayJ has joined #openstack-infra01:48
*** zns has joined #openstack-infra01:49
*** packet has quit IRC01:49
*** wenlock has joined #openstack-infra01:50
*** JayJ has quit IRC01:53
*** JayJ has joined #openstack-infra01:53
*** JayJ has quit IRC01:58
*** JayJ has joined #openstack-infra01:58
*** JayJ has quit IRC01:59
*** rainya has quit IRC02:01
*** zz_dimtruck is now known as dimtruck02:02
*** baoli has quit IRC02:02
*** SumitNaiksatam has quit IRC02:10
*** MaxV has joined #openstack-infra02:11
*** winston-d_zZZ is now known as winston-d02:12
*** MaxV has quit IRC02:16
*** tgohad has quit IRC02:16
*** HeOS has joined #openstack-infra02:20
*** ramashri_ has quit IRC02:21
*** SumitNaiksatam has joined #openstack-infra02:23
*** HeOS has quit IRC02:25
*** HenryG has joined #openstack-infra02:26
*** otherwiseguy has joined #openstack-infra02:27
*** ramashri has joined #openstack-infra02:27
*** Longgeek has joined #openstack-infra02:33
*** hdd has joined #openstack-infra02:35
*** harlowja is now known as harlowja_away02:37
openstackgerritSteve Martinelli proposed a change to openstack-infra/elastic-recheck: Add query for bug 1379201
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’]" [Undecided,Confirmed]
*** hdd has quit IRC02:39
*** zns has quit IRC02:41
*** achanda has joined #openstack-infra02:42
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for bug 1379201
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’]" [Undecided,Confirmed]
*** sunrenjie has quit IRC02:45
*** armax has quit IRC02:50
*** ayoung-afk has quit IRC02:52
*** homeless has quit IRC02:55
*** liusheng has quit IRC02:55
*** liusheng has joined #openstack-infra02:56
*** achanda has quit IRC02:58
*** openstackgerrit has quit IRC02:58
*** SumitNaiksatam has quit IRC02:58
*** achanda has joined #openstack-infra02:59
anteayaopenstackgerrit has left us03:00
*** SumitNaiksatam has joined #openstack-infra03:00
*** SumitNaiksatam has quit IRC03:02
*** mmedvede has quit IRC03:04
*** ayoung-afk has joined #openstack-infra03:04
*** yamahata has quit IRC03:05
*** yamahata has joined #openstack-infra03:06
arosen1Everyone and then I see this weird issue with my jenkins/zuul.   I have a pre_test_hook  that does:03:07
*** hdd has joined #openstack-infra03:08
arosen1   git clone $ZUUL_URL/$ZUUL_PROJECT;   cd congress   git fetch $ZUUL_URL/$ZUUL_PROJECT $ZUUL_REF03:08
arosen1which ends up doing a:  git fetch refs/zuul/master/Zd93ec7e072684cef86f12cecb3b2396f03:08
arosen1though from time to time that doesn't match up with the patch that triggered the build.03:09
arosen1have you guys seen this occur before?03:09
*** MaxV has joined #openstack-infra03:12
*** shashankhegde has joined #openstack-infra03:12
*** harlowja_away has quit IRC03:12
*** MaxV has quit IRC03:17
anteayaarosen1: not ignoring you, I don't know the answer03:25
*** nosnos has quit IRC03:26
*** nosnos has joined #openstack-infra03:26
*** armax has joined #openstack-infra03:27
*** nosnos has quit IRC03:30
*** antonym has quit IRC03:31
*** SumitNaiksatam has joined #openstack-infra03:31
*** antonym has joined #openstack-infra03:33
tchaypoanteaya: it's unforunate that the URL trims that quote in a way that makes it seem to mean something quite different03:37
tchaypostevenk points out that it's also unfortunate that my unfortunate lacks a t03:38
*** unicell has quit IRC03:39
*** wenlock has quit IRC03:40
arosen1anteaya:  no worries i know it's late :P03:50
*** andreykurilin_ has joined #openstack-infra03:53
*** armax has quit IRC03:53
*** pcrews has quit IRC03:55
*** otherwiseguy has quit IRC04:05
*** Ryan_Lane has joined #openstack-infra04:06
*** garyh has quit IRC04:09
*** andreykurilin_ has quit IRC04:10
*** dmsimard is now known as dmsimard_away04:10
*** r1chardj0n3s is now known as r1chardj0n3s_afk04:12
*** MaxV has joined #openstack-infra04:12
*** otter768 has quit IRC04:15
*** MaxV has quit IRC04:17
*** harlowja_at_home has joined #openstack-infra04:18
*** koolhead17 has joined #openstack-infra04:19
*** unicell has joined #openstack-infra04:22
*** harlowja_at_home has quit IRC04:22
*** harlowja_at_home has joined #openstack-infra04:24
*** mmaglana has quit IRC04:25
*** hdd has quit IRC04:27
*** nosnos has joined #openstack-infra04:32
*** Ryan_Lane has quit IRC04:35
*** yongli has quit IRC04:36
*** vigneshvar has joined #openstack-infra04:36
*** yongli has joined #openstack-infra04:37
*** jgrimm is now known as zz_jgrimm04:39
*** harlowja_at_home has quit IRC04:40
*** garyh has joined #openstack-infra04:41
*** vigneshvar has quit IRC04:41
*** stevemar has quit IRC04:43
*** garyh has quit IRC04:51
*** otherwiseguy has joined #openstack-infra04:57
*** vigneshvar has joined #openstack-infra04:57
*** achanda has quit IRC04:58
*** achanda_ has joined #openstack-infra05:01
*** r1chardj0n3s_afk is now known as r1chardj0n3s05:02
*** otherwiseguy has quit IRC05:11
*** MaxV has joined #openstack-infra05:13
*** MaxV has quit IRC05:18
*** loki184 has joined #openstack-infra05:18
*** rushiagr_away is now known as rushiagr05:26
*** r1chardj0n3s is now known as r1chardj0n3s_afk05:29
*** dimtruck is now known as zz_dimtruck05:39
*** achuprin_ has quit IRC05:39
*** mrmartin has joined #openstack-infra05:42
*** MaxV has joined #openstack-infra05:45
anteayatchaypo: yes very unfortunate05:46
* anteaya confirms jgriffith05:47
*** loki184 has quit IRC05:48 has been rearranged05:48
anteayalovely timing05:48
anteayaanyone with a link to the foundation gui of profiles05:48
* anteaya shakes her head05:48
*** packet has joined #openstack-infra05:49
*** garyh has joined #openstack-infra05:49
anteayayeah so maybe a link to the foundation on the home page?05:50
*** mmaglana has joined #openstack-infra05:53
*** packet has quit IRC05:54
*** andreaf has joined #openstack-infra05:55
*** ayoung-afk has quit IRC05:56
anteayajgriffith: one of your responses has a > at the front, intentional?05:57
* anteaya leaves it in for now05:57
*** arosen1 has left #openstack-infra05:58
*** loki184 has joined #openstack-infra06:01
*** garyh has quit IRC06:01
*** yamamoto has quit IRC06:03
* anteaya notices David Lyle's candidacy announcement that made it in before the deadline and that nominations are now closed06:06
anteayawill confirm david-lyle in a minute06:06
StevenKanteaya: How much before the deadline, out of interest?06:13
anteayadeadline for nominations has past06:14
anteayaso -15 minutes06:14
StevenKanteaya: Sorry, I was unclear. I meant how long before the deadline did David Lyle post his in06:14
StevenKHeh, eight minutes. Nice.06:16
anteayalook at the archives:
anteayayeah about that06:17
*** achanda_ has quit IRC06:17
StevenKanteaya: I figured I could stop being lazy and look in Thunderbird :-)06:17
*** mmaglana has quit IRC06:17
*** mmaglana has joined #openstack-infra06:18
anteayahey that works too06:18
*** shashankhegde has quit IRC06:19
*** mmaglana has quit IRC06:22
anteayaconfirmed that all tc candidates have been confirmed06:24
*** achanda has joined #openstack-infra06:29
*** luqas has joined #openstack-infra06:30
*** yamamoto has joined #openstack-infra06:34
*** k4n0 has joined #openstack-infra06:34
andreaf"Toggle CI" in gerrit stopped working when using the new screen - it seems I'll have to go back to the old one :(06:34
anteayaandreaf: can you file a bug please?06:35
andreafanteaya: ok06:35
*** achanda has quit IRC06:38
*** MaxV has quit IRC06:38
*** jcoufal has joined #openstack-infra06:39
andreafanteaya: - it's not very detailed but I don't know what more to say, it should be fairly easy to reproduce06:39
uvirtbotLaunchpad bug 1379659 in openstack-ci ""Toggle CI" in gerrit UI does not work when using the new screen" [Undecided,New]06:39
anteayathe title is very descriptive, thank you06:41
anteayathanks for adding the browser information, what version of chrome?06:41
*** yamamoto has quit IRC06:44
*** flaper87|afk is now known as flaper8706:46
*** ramashri has quit IRC06:47
*** garyh has joined #openstack-infra06:49
*** reed has joined #openstack-infra06:50
*** packet has joined #openstack-infra06:50
*** kevinbenton has quit IRC06:50
*** packet has quit IRC06:54
*** doude has joined #openstack-infra06:56
*** packet has joined #openstack-infra06:58
*** garyh has quit IRC06:59
*** coolsvap|afk is now known as coolsvap07:02
*** packet has quit IRC07:03
*** andreaf has quit IRC07:06
*** oomichi has joined #openstack-infra07:07
*** oomichi has quit IRC07:07
*** arxcruz has joined #openstack-infra07:10
*** esker has joined #openstack-infra07:13
anteayaand we're off:
*** SumitNaiksatam has quit IRC07:16
anteayavote vote vote07:16
*** SumitNaiksatam has joined #openstack-infra07:17
*** skolekonov has joined #openstack-infra07:18
*** ashaeron has joined #openstack-infra07:21
*** Guest51022 has joined #openstack-infra07:21
*** pkoniszewski has joined #openstack-infra07:21
*** fandi has joined #openstack-infra07:21
*** Ryan_Lane has joined #openstack-infra07:22
*** k4n0 has quit IRC07:24
*** Ryan_Lane has quit IRC07:24
anteayatired now, I think I will go to sleep07:24
*** yamamoto has joined #openstack-infra07:24
*** liusheng has quit IRC07:28
*** k4n0 has joined #openstack-infra07:28
*** yamamoto has quit IRC07:29
*** liusheng has joined #openstack-infra07:30
*** mrmartin has quit IRC07:31
*** mrmartin has joined #openstack-infra07:31
*** afazekas has joined #openstack-infra07:33
reedanteaya, wow, still up? I'm in Europe, what's your excuse? :) thank you for your work, amazing07:34
hogepodgereed o/07:35
*** MaxV has joined #openstack-infra07:35
reedhogepodge, you too? an epidemic on sleep deprivation on the west coast?07:36
*** tty57 has joined #openstack-infra07:37
*** frank_zhou has joined #openstack-infra07:38
*** esker has quit IRC07:40
*** yfried has quit IRC07:44
*** jpich has joined #openstack-infra07:47
*** tty57 has quit IRC07:47
*** garyh has joined #openstack-infra07:49
*** frank_zhou has quit IRC07:50
*** salv-orlando has joined #openstack-infra07:52
*** jistr has joined #openstack-infra07:56
*** arxcruz has quit IRC07:57
*** garyh has quit IRC07:59
*** packet has joined #openstack-infra07:59
*** arxcruz has joined #openstack-infra08:00
*** ihrachyshka has joined #openstack-infra08:01
*** oomichi has joined #openstack-infra08:01
*** skraynev has quit IRC08:02
*** yamamoto has joined #openstack-infra08:02
*** packet has quit IRC08:03
*** jedimike has joined #openstack-infra08:04
*** yamamoto has quit IRC08:06
*** Longgeek has quit IRC08:07
jedimikeso I saw that is causing neutron gate job fails on too08:08
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’]" [Undecided,Confirmed]08:08
jedimike shows all the dsvm-neutron jobs failing with that error08:09
*** mpaolino has joined #openstack-infra08:11
*** skraynev has joined #openstack-infra08:12
* dstufft finds it humerous that after he was asked to vote for the TC, the website then suggested he vote on his favorite breaking bad character08:13
*** hashar has joined #openstack-infra08:15
*** derekh has joined #openstack-infra08:15
*** kashyap has joined #openstack-infra08:16
*** berendt has joined #openstack-infra08:22
*** garyh has joined #openstack-infra08:30
*** yamamoto has joined #openstack-infra08:32
*** e0ne has joined #openstack-infra08:34
*** yamamoto has quit IRC08:37
jpichAny idea what could be preventing a job from creating a directory in the devstack tree? . This from a new job definition so I may well have missed something in the set up08:37
*** AJaeger has joined #openstack-infra08:39
*** pblaho has joined #openstack-infra08:41
*** garyh has quit IRC08:41
*** zz_johnthetubagu is now known as johnthetubaguy08:46
*** HeOS has joined #openstack-infra08:52
*** fandi has quit IRC09:02
*** pkoniszewski has quit IRC09:03
*** Longgeek has joined #openstack-infra09:05
*** dizquierdo has joined #openstack-infra09:06
HeOSHi everyone! Who can tell me about your workflow in gerrit? In particular I want to know all about procedure submit/merge after review received the necessary scores (CR,V,W).09:06
*** Longgeek has quit IRC09:07
jedimikeHeOS, hi! If the review received 2+2's, and a core reviewer Approved it, then zuul should automatically merge the change, and you're done!09:07
*** Longgeek has joined #openstack-infra09:08
HeOSjedimike, thanks.09:09
*** liam_ has joined #openstack-infra09:10
*** jp_at_hp has joined #openstack-infra09:10
*** garyh has joined #openstack-infra09:12
*** dalgaaf__ has joined #openstack-infra09:12
AJaegerHeOS: for all the glory details, read
*** dizquierdo has quit IRC09:14
*** rlandy has joined #openstack-infra09:15
*** mpaolino has quit IRC09:17
*** _nadya_ has joined #openstack-infra09:17
*** fandi has joined #openstack-infra09:19
*** garyh has quit IRC09:22
*** chandankumar has joined #openstack-infra09:23
*** dizquierdo has joined #openstack-infra09:23
*** flaviof_zzz is now known as flaviof09:27
*** parimala_ has joined #openstack-infra09:28
*** yamamoto has joined #openstack-infra09:32
pblahohi all... can someone clarify for me what NOT_REGISTERED as results for jenkins job means? see
tchaypo(and if pblaho is asking in the wrong place, it's my fault for sending him here, so please direct flames at me)09:33
*** lttrl has quit IRC09:34
*** yamamoto has quit IRC09:36
pblahotchaypo: there will be no flames here I hope... people are nice and will point me in another channel if necessary :-)09:38
*** hashar has quit IRC09:38
*** nosnos has quit IRC09:39
*** nosnos has joined #openstack-infra09:40
*** _nadya_ has quit IRC09:41
parimala_Hi All, I am a newbie,09:44
*** nosnos has quit IRC09:44
tchaypoparimala_: Welcome :)09:45
*** dizquierdo has quit IRC09:45
parimala_Need help in assessing bugs which a newcomer can fix? any help?09:45
*** oomichi has quit IRC09:45
*** hashar has joined #openstack-infra09:47
parimala_I looked at the "low-hanging-fruit" bugs , but they are quite confusing (for a newbie)09:47
*** unicell has quit IRC09:53
*** che-arne has joined #openstack-infra09:54
*** andreykurilin_ has joined #openstack-infra09:54
*** pelix has joined #openstack-infra09:57
*** dizquierdo has joined #openstack-infra09:59
*** packet has joined #openstack-infra10:01
*** kevinbenton has joined #openstack-infra10:03
*** packet has quit IRC10:05
AJaegerpblaho: NOT_REGISTERED means that there was basically no server to run the job - so something wrong in the configuration files.10:07
AJaegerThanks for reporting, pblaho. Let me check quickly..10:07
*** jgallard has joined #openstack-infra10:08
AJaegerpblaho: there's a small window when a new job is added for the first time where this might happen. I rechecked the patch and will look closer.10:09
AJaegerpblaho: this happened on the 7th of October just 15 mins after we merged the patch for the jobs in, so it should be fine now.10:10
AJaegerHi parimala_ !10:10
AJaegerparimala_: Do you want to help with infrastructure or with a specific project?10:10
parimala_Hi AJaeger :10:11
*** garyh has joined #openstack-infra10:12
parimala_Ajaeger, anything is OK10:12
*** jgallard has quit IRC10:14
AJaegerSorry, can't point you quickly to something but if you have a specific question, just ask.10:14
AJaegerAnd it's lunch time now - I'll be back soon.10:15
*** unicell has joined #openstack-infra10:16
pblahoAJaeger: thanks.... so simple recheck should be fine I suppose10:17
*** yolanda has joined #openstack-infra10:17
*** Guest51022 has quit IRC10:20
*** andreykurilin_ has quit IRC10:21
*** garyh has quit IRC10:22
*** alexpilotti has joined #openstack-infra10:28
*** dizquierdo has quit IRC10:30
*** yamamoto has joined #openstack-infra10:32
*** mmedvede has joined #openstack-infra10:36
*** yamamoto has quit IRC10:36
*** loki184 has quit IRC10:40
AJaegerparimala_: I'm back...10:41
parimala_Ajaeger, Hi10:42
*** dizquierdo has joined #openstack-infra10:42
AJaegerparimala_: looking at the infra bugs: yeah, those are not as trivial10:43
*** fandi has quit IRC10:43
parimala_Ajaeger : how do i choose bugs for fixing as a newbie10:44
AJaegerparimala_: just grab one that is not assigned.10:45
AJaegerThe low-hanging-fruit is normally an indicator that a newbie can do it.10:45
parimala_Ajaeger : I have no knowledge of openstack, openstack-infra, so how will I choose any bug10:46
AJaegerwhich Linux distro do you know best?10:46
*** pcm_ has joined #openstack-infra10:46
parimala_Ajaeger : Centos10:47
*** pcm_ has quit IRC10:48
AJaegerThen I suggest something completely different: We release next week the OpenStack Juno release and the documentation team is busy double checking that the Installatoin Guide works.10:48
AJaegerGo to and install from scratch your first OpenStack cloud.10:48
*** pcm_ has joined #openstack-infra10:48
*** pblaho has quit IRC10:49
AJaegerReport anything that does not work as bug (there's a bug icon on each page, just click it) and if you can fix it, even better...10:49
AJaegerAfter that you should have a much better overview on what OpenStack is, how it works and helped testing...10:49
parimala_Reporting bug is different, bug I need to fix a bug for OPW10:49
parimala_but I will look at the doc URL which u have given,10:50
*** habib has joined #openstack-infra10:50
AJaegerparimala_: so you're participating in OPW and have the task to fix a bug?10:51
AJaegerparimala_: take this bug: and see
uvirtbotLaunchpad bug 1344295 in openstack-manuals "Security Guide - Chapter 43. Trusted Images Intro" [Low,Confirmed]10:52
*** Longgeek has quit IRC10:52
AJaegerIf you need help, ask on #openstack-doc10:52
AJaegerThe repository is security-doc10:52
parimala_AJaeger: but this bug is already assigned10:54
*** ihrachyshka has quit IRC10:55
*** ihrachyshka has joined #openstack-infra10:55
AJaegerparimala_: but no work done for 2 months, I'll delete the assignement...10:56
*** fandi has joined #openstack-infra10:56
*** yamamoto has joined #openstack-infra10:57
parimala_Ajaeger: I will look at both the Centos part and 134429510:57
*** lttrl has joined #openstack-infra10:58
AJaegerthanks, parimala_ ! Both are doc tasks - so ask on #openstack-doc if you need any help10:58
parimala_Ajaeger, Thanks a ton, it really helped, I will ping in #openstack-doc and to you if any help is needed,10:59
*** packet has joined #openstack-infra11:01
*** habib has quit IRC11:02
*** _nadya_ has joined #openstack-infra11:03
*** baoli has joined #openstack-infra11:04
AJaegerparimala_: If I'm around, I'm in #openstack-doc as well ;)11:05
*** packet has quit IRC11:05
parimala_Ok, thanks11:06
*** hashar has quit IRC11:09
*** parimala_ has left #openstack-infra11:09
*** yamahata has quit IRC11:09
*** ZZelle has quit IRC11:11
*** ZZelle has joined #openstack-infra11:11
*** garyh has joined #openstack-infra11:12
*** ociuhandu has joined #openstack-infra11:16
*** fandi has quit IRC11:16
*** lttrl has quit IRC11:23
*** garyh has quit IRC11:23
*** yfried has joined #openstack-infra11:25
*** rcarrillocruz has quit IRC11:26
*** tty57 has joined #openstack-infra11:27
*** packet has joined #openstack-infra11:30
*** Longgeek has joined #openstack-infra11:34
*** miqui has quit IRC11:34
*** bradjones has joined #openstack-infra11:36
*** ihrachyshka has quit IRC11:40
*** yamamoto has quit IRC11:40
*** isviridov_away is now known as isviridov11:40
*** k4n0 has quit IRC11:42
*** mfink has quit IRC11:43
*** jgallard has joined #openstack-infra11:51
*** ihrachyshka has joined #openstack-infra11:58
*** adalbas has joined #openstack-infra11:58
*** mbacchi has joined #openstack-infra11:58
*** jistr has quit IRC12:00
*** rcarrillocruz has joined #openstack-infra12:06
*** dprince has joined #openstack-infra12:08
*** mwagner_lap has joined #openstack-infra12:08
*** weshay has joined #openstack-infra12:09
*** garyh has joined #openstack-infra12:12
*** eglynn-office is now known as eglynn-officeafk12:12
*** JayJ has joined #openstack-infra12:14
*** Guest51022 has joined #openstack-infra12:15
*** JayJ has quit IRC12:19
*** jistr has joined #openstack-infra12:19
*** JayJ has joined #openstack-infra12:19
*** aysyd has joined #openstack-infra12:21
*** garyh has quit IRC12:22
AJaegerfungi, we've lost again the openstackgerrit IRC bot12:24
*** JayJ has quit IRC12:24
*** JayJ has joined #openstack-infra12:24
*** coolsvap is now known as coolsvap|afk12:26
*** JayJ has quit IRC12:29
*** JayJ has joined #openstack-infra12:30
*** hashar has joined #openstack-infra12:30
*** dkliban_afk is now known as dkliban12:30
*** dkranz has joined #openstack-infra12:32
*** kgiusti has joined #openstack-infra12:33
*** JayJ has quit IRC12:34
*** JayJ has joined #openstack-infra12:35
*** chandankumar has quit IRC12:36
*** dims has joined #openstack-infra12:37
*** JayJ has quit IRC12:39
*** JayJ has joined #openstack-infra12:40
*** eharney has quit IRC12:40
*** JayJ has quit IRC12:44
*** ayoung has joined #openstack-infra12:45
*** JayJ has joined #openstack-infra12:45
jedimikeI wish zuul had a "fail fast" feature. Like, if my change fails pep8 or the commit message is bad, I don't want zuul to waste an hour running tempest tests. Notify me immediately, let me fix the trivial stuff, then spend the hour running tempest tests12:46
*** mpaolino has joined #openstack-infra12:47
*** hdd has joined #openstack-infra12:47
mpaolinohello folks, I'm experiencing some problems while "git cloning" some of the OpenStack components (I'm using devstack)12:48
mpaolinois that happening also to someone else?12:48
mpaolinoin particular, cinder and nova are failing12:49
*** JayJ has quit IRC12:49
*** JayJ has joined #openstack-infra12:50
*** dims has quit IRC12:52
*** dims has joined #openstack-infra12:52
*** katyafervent_awa is now known as katyafervent12:53
*** JayJ has quit IRC12:54
jedimikempaolino, I'm not seeing any issues here, what errors are you getting?12:55
*** JayJ has joined #openstack-infra12:55
*** rushiagr is now known as rushiagr_away12:55
mpaolinojedimike, ok thanks I'll recheck the connectivity of my machines12:56
*** miqui has joined #openstack-infra12:56
*** katyafervent is now known as katyafervent_awa12:56
mpaolinoI'm getting a git error, I can't be more specific now (I just rerun ./stack to see if now it works)12:56
*** dizquierdo has quit IRC12:56
*** _nadya_ has quit IRC12:57
*** HenryG has quit IRC12:57
*** _nadya_ has joined #openstack-infra12:58
*** JayJ has quit IRC13:00
*** JayJ has joined #openstack-infra13:00
*** rkukura has quit IRC13:01
*** hashar has quit IRC13:02
*** hashar has joined #openstack-infra13:02
*** e0ne has quit IRC13:03
*** e0ne has joined #openstack-infra13:04
*** JayJ has quit IRC13:05
*** tty57 has quit IRC13:05
*** bknudson has joined #openstack-infra13:05
*** JayJ has joined #openstack-infra13:05
*** katyafervent_awa is now known as katyafervent13:06
*** jcoufal has quit IRC13:07
*** tsg has joined #openstack-infra13:09
*** JayJ has quit IRC13:10
*** _nadya_ has quit IRC13:10
*** JayJ has joined #openstack-infra13:10
*** garyh has joined #openstack-infra13:12
*** julim has joined #openstack-infra13:12
*** eharney has joined #openstack-infra13:13
*** JayJ has quit IRC13:15
*** mfink has joined #openstack-infra13:15
*** JayJ has joined #openstack-infra13:15
*** jistr is now known as jistr|bbl13:16
*** dkranz has quit IRC13:19
*** JayJ has quit IRC13:20
*** mwagner_lap has quit IRC13:20
*** JayJ has joined #openstack-infra13:20
*** MaxV has quit IRC13:21
*** yaguang has quit IRC13:21
*** yaguang has joined #openstack-infra13:22
*** garyh has quit IRC13:23
*** jcoufal has joined #openstack-infra13:25
*** JayJ has quit IRC13:25
*** JayJ has joined #openstack-infra13:25
*** esker has joined #openstack-infra13:28
*** JayJ has quit IRC13:30
*** JayJ has joined #openstack-infra13:30
*** MaxV has joined #openstack-infra13:30
*** kgiusti has quit IRC13:31
*** kgiusti has joined #openstack-infra13:32
*** vhoward has joined #openstack-infra13:33
*** tsg has quit IRC13:35
*** JayJ has quit IRC13:35
*** dkranz has joined #openstack-infra13:35
*** tsg has joined #openstack-infra13:35
*** JayJ has joined #openstack-infra13:36
*** mwagner_lap has joined #openstack-infra13:36
*** bhunter71 has joined #openstack-infra13:39
*** JayJ has quit IRC13:40
*** JayJ has joined #openstack-infra13:41
*** cdent has joined #openstack-infra13:43
*** rushiagr_away is now known as rushiagr13:44
*** JayJ has quit IRC13:45
*** JayJ has joined #openstack-infra13:46
*** homeless has joined #openstack-infra13:47
*** JayJ has quit IRC13:50
*** JayJ has joined #openstack-infra13:51
*** kashyap is now known as kashyap_afk13:51
*** zns has joined #openstack-infra13:55
*** JayJ has quit IRC13:55
*** JayJ has joined #openstack-infra13:56
*** dmsimard_away is now known as dmsimard13:57
*** tty57 has joined #openstack-infra13:58
*** hdd has quit IRC13:59
*** JayJ has quit IRC14:00
*** JayJ has joined #openstack-infra14:01
anteayaso I am here for about 10 minutes and then not here for most of today, Tristan is here today if there are any questions about the election14:02
anteayaand of course you can email us14:03
*** ayoung has quit IRC14:03
anteayareed: thanks, had to get the election started since I am going to be offline a good chunk of today14:03
*** tty57 is now known as jgjl14:04
*** cpowell has joined #openstack-infra14:04
AJaegeranteaya: you were up late last night - and I voted already ;) Thanks for running this so smoothly!14:05
*** JayJ has quit IRC14:05
*** dansmith is now known as superdan14:06
anteayayay voted!!14:06
*** dustins has joined #openstack-infra14:06
anteayathank you14:06
anteayaso far so good14:06
*** JayJ has joined #openstack-infra14:06
*** jgjl has quit IRC14:06
*** dizquierdo has joined #openstack-infra14:06
*** jgjl has joined #openstack-infra14:06
cpowellGood morning everyone. I have recently put in a request for a CI account on third-party-requests, but it has been quite a while and I have not received a response. Is there some way I can follow up on its status?14:07
*** tsg has quit IRC14:07
*** jgjl has quit IRC14:08
anteayaSergeyLukjanov: ^^14:08
*** e0ne has quit IRC14:09
*** jgjl has joined #openstack-infra14:10
*** AJaeger has quit IRC14:10
*** JayJ has quit IRC14:10
*** JayJ has joined #openstack-infra14:11
*** rickerc has quit IRC14:11
*** jgjl has quit IRC14:11
*** jgjl has joined #openstack-infra14:12
*** garyh has joined #openstack-infra14:12
*** tsg has joined #openstack-infra14:13
fungiugh... just a heads up... internet access is basically out across the whole island again. i'm connected via a (laggy, low-bandwidth) wireless modem but don't expect much14:13
jgriffithanteaya: Nope, not intentional (> in front of reponse) :(14:13
fungirestarting the gerritbot now14:13
*** sdake has quit IRC14:15
*** e0ne has joined #openstack-infra14:15
*** JayJ has quit IRC14:15
*** rickerc has joined #openstack-infra14:16
*** openstackgerrit has joined #openstack-infra14:16
*** JayJ has joined #openstack-infra14:16
*** jaypipes is now known as leakypipes14:16
*** vigneshvar has quit IRC14:17
*** e0ne has quit IRC14:17
*** e0ne has joined #openstack-infra14:18
*** zns has quit IRC14:18
*** JayJ has quit IRC14:20
*** JayJ has joined #openstack-infra14:21
*** shashankhegde has joined #openstack-infra14:22
*** sdake has joined #openstack-infra14:22
anteayajgriffith: thanks, I will remove14:22
*** zns has joined #openstack-infra14:22
fungicpowell: normally one of the infrastructure root admins gets around to batching up all new service account requests every week or two. we tend to be very, very busy14:23
*** garyh has quit IRC14:23
mtreinishfungi: is there anything we can do about it's blocking the tempest queue14:23
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’]" [Undecided,Confirmed]14:23
*** hashar has quit IRC14:23
mtreinishit looks like there was a new precise kernel which breaks openvswitch14:23
*** hashar has joined #openstack-infra14:24
fungimtreinish: i talked with clarkb about either rolling back our nodepool images or trying to pin them to an older kernel, but i think we hoped ubuntu would fix this already14:24
jedimikemtreinish, yeah, i emailed the mailing list about it this morning14:24
cpowell@fungi thank you for the response. I just want make sure there isn't something blocking the account.14:24
anteayajgriffith: removed14:25
anteayacpowell: just our lack of time14:25
*** JayJ has quit IRC14:25
fungiright, jedimike was jedi-on-the-spot with a heads up to us for that bug. he had tracked it down before the rest of us even knew it was happening14:25
fungimuch appreciated14:26
*** JayJ has joined #openstack-infra14:26
jedimikefungi, no problem :)14:26
*** rlandy has quit IRC14:27
clarkbfungi or implement a pin and rebuild14:28
*** yfried has quit IRC14:28
mtreinishfungi: ok, yeah I'm thinking we should just rollback the image, who knows when ubuntu will push a fix it...14:28
*** tsg has quit IRC14:28
clarkbI just didnt like the idea of perpetual image deletes14:28
anteayacpowell: just checked your email, you have all the information we require thank you14:28
clarkbbecause it prevents us from fixing newer issues14:29
fungiclarkb: "pin" suggests the older kernel is still in their index14:29
cpowell@anteaya thank you very much, I appreciate the help14:29
*** zz_jgrimm is now known as jgrimm14:29
clarkbjedimike: is your pin change pushed upstream?14:29
clarkbjedimike: I thinj we would take it if pushed14:29
fungiif it is, then awesome. let's do that14:30
jedimikeclarkb, no, but I can start a review for it - it's a little hacky due to the hp core team needing to get patches in today, but it's a start14:30
*** JayJ has quit IRC14:31
*** HenryG has joined #openstack-infra14:31
*** maxbit has joined #openstack-infra14:31
*** markmcclain has joined #openstack-infra14:31
*** JayJ has joined #openstack-infra14:31
anteayacpowell: yeah, hopefully next gerrit upgrade we will have what we need so I can have permissions to create accounts, I don't right now14:31
jedimikeclarkb, fungi, where does openstack_project/files/nodepool/scripts live now?14:32
anteayamy taxi is here, see you later this afternoon14:32
*** skolekonov has quit IRC14:32
clarkbin the project-config repo14:32
*** david-lyle has joined #openstack-infra14:34
*** JayJ has quit IRC14:36
*** JayJ has joined #openstack-infra14:36
openstackgerritJulien Danjou proposed a change to openstack-infra/project-config: Import openstack-monitoring
*** dprince has quit IRC14:38
*** datsun180b has joined #openstack-infra14:39
openstackgerritMike Heald proposed a change to openstack-infra/project-config: Insist on kernel 3.2.0-69 for precise
*** stevemar has joined #openstack-infra14:40
*** dhellmann is now known as dhellmann_14:40
*** dhellmann_ is now known as dhellmann14:41
*** JayJ has quit IRC14:41
*** JayJ has joined #openstack-infra14:41
clarkbjedimike the list of packages in the else block is different14:41
clarkbignoring versions that is14:42
*** pcrews has joined #openstack-infra14:42
jedimikeclarkb, ah, I didn't notice that. Fixing now.14:42
*** ska has joined #openstack-infra14:42
clarkbjedimike we will want to be as close to the old list as possible14:43
*** zz_dimtruck is now known as dimtruck14:43
jedimikeclarkb, yes, the only difference I'm seeing is the precise specific was missing python-software-properties, which was missing from our config14:43
jedimikeclarkb, the reason I explicitly reinstall linux-image is so the grub trigger happens14:44
clarkbya reinstall is fine just wanted packages to line up14:44
openstackgerritMike Heald proposed a change to openstack-infra/project-config: Insist on kernel 3.2.0-69 for precise
jedimikeclarkb, new change pushed up for review14:45
jedimikethanks openstackgerrit :)14:45
*** dustins has quit IRC14:45
*** JayJ has quit IRC14:46
*** JayJ has joined #openstack-infra14:46
*** shashankhegde has quit IRC14:47
skaWe're setting up a dev environment for Openstack. I have a server that runs KVM instances already, and another that is just Debian running nothing special yet. Can I use those to form a dev env without destroying existing KVM instances?14:48
skaIf there is a more appropriate channel for this, please redirect me.14:49
*** hdd has joined #openstack-infra14:50
*** JayJ has quit IRC14:51
fungiska: probably #openstack-dev... we mostly manage the community services in here14:51
*** JayJ has joined #openstack-infra14:51
*** garyh has joined #openstack-infra14:52
skaok..Thanks.14:52 seems unable to accept new pastes. Getting 50014:52
ttxfungi: ^14:52
fungiska: #openstack-infra are the people keeping up the code review system, mailing lists, wiki, etherpad, continuous integration machines14:52
fungioh, and paste14:52
skaThanks again..14:52
*** ska has left #openstack-infra14:53
openstackgerritJulien Danjou proposed a change to openstack-infra/project-config: Import openstack-monitoring
fungittx: go to and then try force-reloading the paste you're getting a 500 on14:53
*** dmellado has quit IRC14:53
fungittx: for some reason lodgeit seems to fail to store session cookies in the trove backend it's using, and that causes your browser to cache an invalid cookie14:53
*** dmellado has joined #openstack-infra14:53
*** ihrachyshka is now known as ihrachyshka|afk14:53
fungifrom what we've been able to tell14:54
ttxfungi: I was trying to paste a new thing though14:54
fungittx: ahh, yeah, go back in your browser and submit again. that's tended to work for me14:54
fungi(back button, then the submit button again, presumably your browser has kept the freeform text input field with your previous content)14:55
fungiclarkb: didn't you have a theory that was rate limiting/throttling in rackspace's mysql configuration?14:55
ttxfungi: I just switched to another paste :P14:56
*** JayJ has quit IRC14:56
ttxthe one that pastebinit uses by default14:56
fungittx: yeah. sorry about that. it's been a hard issue to diagnose14:56
clarkbfungi in the logs were lodgeit connectiin dying14:56
*** JayJ has joined #openstack-infra14:56
clarkbso ya I think one idea was we needed to tweak mysql connection settings14:57
*** xyang0 has joined #openstack-infra14:57
*** JayJ has quit IRC14:57
fungittx: also, is my last comment about tc jurisdiction on bug 1372635 accurate?14:57
uvirtbotLaunchpad bug 1372635 in ossa "MITM vulnerability with EMC VMAX driver" [Undecided,Won't fix]
*** armax has joined #openstack-infra14:57
*** flaviof is now known as flaviof_zzz14:57
* ttx looks14:58
ttxfungi: yeah, mostly14:59
ttxwill comment14:59
fungiappreciated! just trying to make sure the ossg doesn't think the vmt is giving them the blow-off14:59
*** luqas has quit IRC15:00
*** JayJ_ has joined #openstack-infra15:01
timrcjedimike, Thanks for the workaround! As your reward, you can have the weekend off.15:01
fungishhh... don't tell him that!15:02
*** Guest51022 has quit IRC15:02
jedimikecan I just have a day where a failing dsvm-neutron job doesn't block me? :D15:03
ttxfungi: commented15:03
*** IAX has joined #openstack-infra15:04
*** _nadya_ has joined #openstack-infra15:04
*** rwsu has joined #openstack-infra15:05
*** JayJ_ has quit IRC15:05
*** dangers_away is now known as dangers15:05
*** JayJ_ has joined #openstack-infra15:06
*** yaguang has quit IRC15:07
*** dizquierdo has quit IRC15:07
*** tsg has joined #openstack-infra15:08
*** kgiusti has quit IRC15:08
*** kgiusti has joined #openstack-infra15:09
*** otherwiseguy has joined #openstack-infra15:10
*** JayJ_ has quit IRC15:10
*** JayJ_ has joined #openstack-infra15:11
*** rcarrillocruz has quit IRC15:11
*** afazekas has quit IRC15:15
*** rcarrillocruz has joined #openstack-infra15:15
*** JayJ_ has quit IRC15:15
*** JayJ_ has joined #openstack-infra15:16
*** IvanBerezovskiy has left #openstack-infra15:16
openstackgerritJulien Danjou proposed a change to openstack-infra/project-config: Import openstack-monitoring
*** david-lyle has quit IRC15:17
jeblairjedimike: approved, thanks!15:17
*** david-lyle has joined #openstack-infra15:18
*** kmartin has quit IRC15:19
jeblairzul: are you aware of  ?15:20
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’]" [Undecided,Confirmed]15:20
*** JayJ_ has quit IRC15:20
*** JayJ_ has joined #openstack-infra15:21
*** dims is now known as dimsum_15:21
zuljeblair:  nope i was not15:22
jeblairzul: we've just pinned our precise builds to the older -69 kernel because of it15:22
zuljeblair:  ok ill have a look15:23
fungionce that patch merges, i'll get the precise images rebuilding15:23
openstackgerritEvgeniy L proposed a change to openstack-infra/project-config: Fuel-plugins repository for Fuel
jeblairzul: thanks!15:23
*** JayJ_ has quit IRC15:25
*** JayJ_ has joined #openstack-infra15:26
*** andreaf has joined #openstack-infra15:27
*** krotscheck has quit IRC15:28
*** krotscheck has joined #openstack-infra15:29
*** JayJ_ has quit IRC15:30
openstackgerritA change was merged to openstack-infra/project-config: Insist on kernel 3.2.0-69 for precise
*** JayJ_ has joined #openstack-infra15:31
*** Sincler has joined #openstack-infra15:32
*** jgallard has quit IRC15:32
fungigetting going on the image updates for that ^ now15:33
clarkbfungi make sure puppet updates first15:34
clarkbas the nodepool scripts are copied from nodepool.o.o iirc15:34
*** sacharya has joined #openstack-infra15:34
fungiclarkb: yep, hovering on that now15:35
*** JayJ_ has quit IRC15:36
*** JayJ_ has joined #openstack-infra15:36
*** sarob has quit IRC15:38
*** jp_at_hp has quit IRC15:38
*** kgiusti has quit IRC15:41
*** kgiusti has joined #openstack-infra15:43
openstackgerritDavid Kranz proposed a change to openstack-infra/devstack-gate: Pass args to stress tox call
*** flaviof_zzz has quit IRC15:44
*** thedodd has joined #openstack-infra15:45
*** harlowja_at_home has joined #openstack-infra15:45
*** groknix has joined #openstack-infra15:46
*** groknix has quit IRC15:47
*** armax has quit IRC15:47
*** jgjl has quit IRC15:47
*** ZZelle has quit IRC15:48
*** armax has joined #openstack-infra15:49
*** hashar has quit IRC15:50
*** armax has quit IRC15:50
*** jistr|bbl is now known as jistr15:50
*** hashar has joined #openstack-infra15:50
pleia2good morning15:51
*** zaphod42 has joined #openstack-infra15:51
jedimikehi pleia215:51
*** zaphod42 has joined #openstack-infra15:52
*** hashar has quit IRC15:52
*** hashar has joined #openstack-infra15:52
pleia2jedimike: ugh re: -70 kernel issue15:53
*** amuller has joined #openstack-infra15:53
*** ramashri has joined #openstack-infra15:55
*** reed has quit IRC15:55
*** katyafervent is now known as katyafervent_awa15:55
*** kmartin has joined #openstack-infra15:56
*** otherwiseguy has quit IRC15:57
*** otherwiseguy has joined #openstack-infra15:57
*** e0ne has quit IRC15:58
*** markmcclain has quit IRC15:58
*** stevemar has quit IRC15:58
*** stevemar has joined #openstack-infra16:00
*** ZZelle has joined #openstack-infra16:00
*** mfisch has joined #openstack-infra16:02
*** otherwiseguy has quit IRC16:02
*** arxcruz has quit IRC16:02
*** berendt has quit IRC16:02
mfischCan someone point me in the right direction to get a Stackalytics error corrected?16:02
*** otherwiseguy has joined #openstack-infra16:02
fungimfisch: you propose a change to that for review and the stackalytics core reviewers review and approve it16:03
mfischyeah this is beyond that16:03
*** ramashri_ has joined #openstack-infra16:03
mfischsome random person showing up as working for my company, maybe they have a LP bug tracker16:04
*** otherwiseguy has quit IRC16:04
mfischbug filed as an obvious way to request help.. thx16:04
fungimfisch: yeah, in that case, probably you want to reach out to,members or
*** otherwiseguy has joined #openstack-infra16:04
*** andreaf has quit IRC16:05
mfischdone and thanks fungi16:06
*** ramashri has quit IRC16:06
fungigrr... this wireless modem wants to time out idle ssh sessions very, very quickly16:06
*** hashar has quit IRC16:06
jeblairmfisch: i think there's a data file that you may be able to update16:08
mfischthere's nothing in the datafile that should cause this16:08
mfischI added myself16:08
pleia2 explains rationale for how afflilations are determined, if that helps16:09
jeblairmfisch: "domains": [""],16:09
mfischwait I see it16:09
jeblair"company_name": "Time Warner Cable"16:09
jeblairmfisch: could that be it?16:09
mfischnot sure how I missed that16:09
mfischyeah thats it16:09
mfischshe has for her home cable provider16:09
fungiheh. and is using the isp-provided e-mail address16:10
fungiand they mapped it up based on that16:10
mordredISPs give out email?16:10
mfischso the annoying thing will be if you work here and have that for your home email16:10
mfischbut I think we can drop it16:10
mordredmfisch: yah - those folks will just have to list themselves individually in the mapping file16:10
fungimordred: there are probably isps that still have nntp servers too16:10
mfischdiane is making us all look bad anyway so we need to remove her ;)16:11
mordredmfisch: fwiw, 85% of the people who work for me have to manually add themselves to that file because they don't use emails for openstack16:11
mordredfungi: _crazy_16:11
*** derekh has quit IRC16:11
*** yamahata has joined #openstack-infra16:11
*** thedodd has quit IRC16:11
*** MaxV has quit IRC16:11
mfischokay fix in review16:12
*** MarkAtwood has joined #openstack-infra16:12
mfischsee asking people always helps16:12
*** armax has joined #openstack-infra16:13
*** mmaglana has joined #openstack-infra16:13
*** ramashri_ has quit IRC16:13
*** thedodd has joined #openstack-infra16:13
fungiclarkb: devstack-precise-dib came ready before the devstack-precise image in the same privider16:14
jeblairhuh, the data file is not parsable by python-yaml16:14
*** cody-somerville has quit IRC16:16
*** flaviof_zzz has joined #openstack-infra16:19
clarkbfungi: nice! though the -dib image is without jedimike's fix right? not a big deal as I don't quite have tests going there yet16:19
*** eharney_ has joined #openstack-infra16:19
*** sarob has joined #openstack-infra16:19
fungiclarkb: yeah16:19
clarkbfungi: I think the caching of data on nodepool.o.o really helps the build times16:20
clarkbfungi: its like 15 minutes per devstack image16:20
clarkbthen upload time16:20
ekarlsophschwartz: you up dude ?16:20
fungiwell, it was more like 45 minutes start to finish16:20
fungier, no, 2516:20
clarkbalso I am mostly properly awake now. I was going to look at centos7-dib really quickly to catch up from yesterday then look at zaro's gearman-plugin patch16:20
*** eharney has quit IRC16:21
*** eharney_ is now known as eharney16:21
pleia2mostly properly awake \o/16:22
clarkbjeblair: 127371 is the consume dib from master change. I seem to recall you were ok with that yesterday?16:23
clarkbfungi: that is quick :)16:23
*** yamahata has quit IRC16:24
*** patrickeast has joined #openstack-infra16:24
*** jamielennox has quit IRC16:25
*** jpich has quit IRC16:26
sacharyaHi, I am trying to use the upload logs to swift feature in zuul (on a 3 party CI I am working on)… anyone around to answer a couple of questions?16:27
clarkbsacharya: yup.16:28
fungiit's even mostly working now16:28
sacharyaclarkb: I am getting " Environment variable 'SWIFT_logs_LOGSERVER_PREFIX' not found" while its trying to run "/usr/local/jenkins/slave_scripts/ --name logs /tmp/console.txt"16:29
sacharyahow is the env variables passed to the node?16:29
*** eharney has quit IRC16:30
clarkbsacharya: you have to configure the zuul server with swift credentials and info. When you do that, and zuul successfully auths against swift then zuul will provide those env variables to the job runs16:30
clarkbsacharya: is the documentation for that16:31
*** baoli has quit IRC16:31
sacharyaclarkb: This is whats in my zuul.conf
*** ramashri has joined #openstack-infra16:32
*** ChuckC__ has joined #openstack-infra16:32
clarkb ok that looks a lot like ours.16:33
clarkbsacharya: make sure you have python-swiftclient installed and that you can auth using the data you provided. note that rackspace uses a funny tenant id for its swift service16:34
clarkbsacharya: you can double check that value by dumping the keystone catalog for your account16:34
sacharyaclarkb: I dont think I'm even that far.. Right now says "Environment variable 'SWIFT_logs_LOGSERVER_PREFIX' not found"16:35
*** baoli has joined #openstack-infra16:36
clarkbsacharya: yes, because zuul has not provided that info to your job. For zuul to provide that info to your job the swift stuff has to be tested by zuul as working16:36
clarkbsacharya: so I think we should start with double checking that it is working16:37
clarkbonce we have checked that we also need to check that the individual jobs are specified to use swift in the layou.taml16:37
sacharyaclarkb: gotcha…16:37
*** dangers is now known as dangers_away16:37
ekarlsois it possible to disable the gerrit integration if one is using github as atrigger?16:37
clarkb is where we setup swift log data for all of our jobs16:37
* fungi wonders if stackalytics may be using .taml files16:37
clarkbsacharya: the swift: line there is the important bit16:38
clarkber s/line/section/16:38
clarkbekarlso: I don't understand the question16:38
*** ashaeron has quit IRC16:38
clarkbekarlso: can you elaborate on the scope of the question?16:38
clarkbekarlso: eg what project (or is this third party) and is that a jenkins trigger?16:38
ekarlsoclarkb: atm when using zuul-server it tries to connect to gerrit, though if you dont wanna use gerrit aka just using zuul to provide pipelines using github as a trigger it dont need to be triggered ?16:39
ekarlsoehm, s/triggered/started16:39
clarkbekarlso: just remove the gerrit trigger?16:39
ekarlsoclarkb: still when you try to start zuul-server it will start to stream from gerrit (ssh connection)16:40
ekarlsodoesn't it ? ;)16:40
clarkbekarlso: it might, but I would hope it doesn't if it isn't configured to use gerrit. If it does that is probably a bug that needs fixing16:41
mordredwhen did we get a github trigger added to zuul?16:41
*** jcoufal has quit IRC16:41
clarkbmordred: I don't think we have16:41
mordredekarlso: how are you using zuul with github?16:41
*** ChuckC__ has quit IRC16:42
ekarlsomordred: idea is to start using it soon just by exposing a api endpoint..16:42
* fungi gets the impression this is a question about a hypothetical bug in a hypothetical patch to zuul16:42
mordredekarlso: indeed. so I can expect a patch from you then? :)16:42
*** krtaylor has quit IRC16:42
fungii recommend, when you implement that feature in zuul, don't include the bug you're predicting16:42
* mordred looks forward to this16:43
ekarlsomordred: guess so .. :p16:43
* mordred throws wet cat at ekarlso16:43
ekarlsowhy a cat ? :p16:43
ekarlsocoding cat ?  p16:43
*** slagle has quit IRC16:44
fungimordred: seems to prefer cat-hurling as an alternative to fish-slapping16:44
*** mpaolino has quit IRC16:45
*** e0ne has joined #openstack-infra16:46
clarkbmordred: thoughts on 127371?16:46
clarkbI want to approve but missed much of yesterday and don't want to step on any decisions made yesterday around this16:46
mordredI did not make any decisions yesterday16:47
mordreduhm - wait a sec16:47
mordredwhy are we installing from git?16:47
* mordred missing something16:48
mordredSpamapS: ^^ is a safe thing to do?16:48
clarkbmordred: because old pip cannot centos716:49
clarkber not old pip16:49
clarkbold dib16:49
mordredthen what we need is for SpamapS to make a dib release?16:49
fungirereading that change, ianw could have included a bit more detail on his reasons in the commit message16:49
mordredyes. I have no idea why we want to do that16:49
mordredI'm -1 pending follow up questions16:50
fungimordred: ianw: clarkb: 0.1.33 seems to have been tagged just yesterday16:51
clarkbwithout my dib change in it :(16:52
* clarkb wonders what the point of a release is if you don't flush the review queue first16:52
*** kgiusti has quit IRC16:52
*** esker has quit IRC16:53
mordredmaybe I should be poking greghaynes instead of SpamapS ...16:53
*** kgiusti has joined #openstack-infra16:53
mordredgreghaynes: we need things. please fix. kthxbai16:53
fungi"now the shoe is on the other foot," he says16:54
*** doude has quit IRC16:54
greghaynesYou just need a dib release cut?16:55
clarkbno, my dib change isn't merged yet16:55
fungia proposed bugfix approved, merged, then a release cut16:55
mordredthen we'll let you live16:55
clarkbso I think we can switch to consuming releases of dib via pip16:56
clarkbusing ensure => latest16:56
clarkbfungi: mordred: is that what you had in mind?16:56
fungii agree that would be reasonable16:56
*** MarkAtwood has quit IRC16:56
fungiinstead of ensure=>present16:56
*** cpowell has quit IRC16:56
*** MaxV has joined #openstack-infra16:57
greghaynesgot a review link for bugfix?16:57
*** achanda has joined #openstack-infra16:57
clarkbfungi: is bugfix related to centos thing?16:57
clarkbmy change is a new feature not bug16:57
fungiit fixes our bug ;)16:57
fungibut yes, new feature really16:57
greghayneswell, I am the only +2 on that patch so I cant really help :(16:58
clarkbright. my gripe was tagging a release on a project with open reviews that are primed to merge16:58
*** esker has joined #openstack-infra16:58
greghaynesThere are lots of open reviews, we tag releases this time of week-ish16:59
greghaynesor whenever someone asks nicely16:59
fungibut yeah, it's arguably a bug in glance which needs to be worked around in dib17:00
fungithat can go onto hogepodge's interop wishlist17:00
*** ivar-lazzaro has joined #openstack-infra17:01
*** bhunter71 has quit IRC17:01
*** harlowja_at_home has quit IRC17:04
*** krtaylor has joined #openstack-infra17:05
*** zns has quit IRC17:06
*** viktors is now known as viktors|afk17:08
*** thedodd has quit IRC17:08
*** eharney has joined #openstack-infra17:08
*** jamielennox has joined #openstack-infra17:10
*** _nadya_ has quit IRC17:10
*** Longgeek has quit IRC17:12
*** kgiusti has quit IRC17:12
clarkbfungi: mordred: if I update ianw's change will you be around to rereview in the near future?17:13
fungiclarkb: yes, i'm around all day, modulo heading to lunch in a little while and dealing with flaking wireless modem connection since my broadband isp still hasn't gotten their act together today17:14
*** jheroux has joined #openstack-infra17:16
*** kgiusti has joined #openstack-infra17:16
openstackgerritClark Boylan proposed a change to openstack-infra/config: Install latest diskimage-builder on nodepool
clarkbmordred: fungi ^ how about that?17:17
fungiglad we no longer have to care about pointless rocketship alignment17:18
mordredclarkb: +217:18
clarkboh wait17:19
clarkbfungi: will that fail? should I fix it?17:19
fungiclarkb: i think we ditched the lint job months ago17:19
clarkbgah let me fix it, that way this code doens't need fixing later if we go back to enforcing that check17:19
clarkbfungi: ya you are right17:19
openstackgerritClark Boylan proposed a change to openstack-infra/config: Install latest diskimage-builder on nodepool
*** arosen has joined #openstack-infra17:23
fungidid we stop including the template vm name for the image used in job logs? for example
*** torandu has quit IRC17:24
clarkbfungi: I don't think so, maybe we only did it for non d-g jobs?17:25
fungiohh... i think you're right17:25
fungiyeah, it's in the slave scripts17:25
*** markmcclain has joined #openstack-infra17:25
*** markmcclain has quit IRC17:25
fungimaybe echoing that should become a builder17:25
*** ryanpetrello has quit IRC17:25
stevemarho boy, that one bug sure exploded in rechecks since last night17:26
fungistevemar: yeah17:26
*** ryanpetrello has joined #openstack-infra17:26
fungistevemar: we're hoping has fixed it, but not all images are done updating yet17:26
stevemarfungi, ah that explains why my latest recheck still came back with the same error17:27
fungistevemar: yeah, i'm hanging my hopes on since the provider it's running in has had an updated devstack-precise image for half an hour now17:29
fungier, not that one... needs to be neutron17:29
stevemarfungi, hehe, yeah not that one, it's well into the tempest job now, the bug fails on devstack setup17:30
*** chandankumar has joined #openstack-infra17:30
mordredOk. My friend's band just released a video of their new Star Trek based love song:
fungigrr, the neutron version for that change is running in a provider which only got an updated image a few minutes ago, so still too soon17:30
mordredyou're welcome17:30
*** kashyap_afk is now known as kashyap17:32
*** HeOS has quit IRC17:33
clarkbnibalizer: you should watch ^17:33
fungimordred: that is _sick_17:33
*** ramashri_ has joined #openstack-infra17:34
fungiwe're holding off rewatching season 7 until they get the blu-ray set out for it (soon!)17:34
fungimaking do with ds9 in the meantime17:34
*** groknix has joined #openstack-infra17:34
mordredthey were one of my favorite bands anyway - but that just takes the cake17:35
clarkbnot sure that watching ds9 instead of TNG is making do17:36
*** ramashri has quit IRC17:36
clarkbmordred: 127371 please :)17:37
fungiwell, we went through all of tos on blu-ray and got about halfway through tng before we started having to wait for them to release new seasons17:37
*** johnthetubaguy is now known as zz_johnthetubagu17:37
clarkbit was approved nevermind. I gues sfungi decided that your previous +2 was sufficient17:37
adam_gis there something that pegs stable/icehouse neutron jobs to precise nodes?17:37
clarkbadam_g: yes, all icehouse jobs run on precise17:37
clarkb(excluding python2.6 jobs which run on centos6)17:38
fungiclarkb: yes, that was indeed my reasoning. no need to waste a perfectly good existing +2 on whitespace17:38
fungiadam_g: it's done in a zuul parameter function called from the layout.yaml17:38
clarkbfungi: I tried rewatching TNG recently started with season one. promptly gave up17:38
mordreddon't watch season one17:39
clarkbmordred: yes I know17:39
mordredMAYBE the last episode17:39
clarkbbut I am OCD and must start at the beginning!17:39
adam_gclarkb, i guess thats intentional, huh?  new kernel update to precise introduced a nice bug that breaks OVS module builds,
fungiclarkb: yes, it takes some stomach to get through the first couple seasons17:39
clarkbadam_g: yes and we are currently working to pin the kernel to older version17:39
adam_gfungi, thanks!17:39
adam_gclarkb, oh, me too slow17:40
fungiadam_g: i think it's probably already fixed in some of our providers, just looking for confirmation so we can be sure. if so, the rest are also minuts from being fixed17:40
fungiminutes too17:41
*** chandankumar has quit IRC17:42
fungiclarkb: i can't get my wife to watch babylon 5 with me because i made the mistake of trying to get her to start at season 1. after about two episodes she patently refused to watch any more. at least with st:tng she had seen enough of later seasons in the past to know it would get better without having to take my word for it17:42
adam_gfungi, cool. for my own edification, where does packaging version pinning happen? puppet?17:42
fungiadam_g: in this case, nodepool prep scripts17:42
adam_gfungi, thanks17:43
*** chuckC_ has quit IRC17:43
*** wenlock has joined #openstack-infra17:43
jedimikefungi, my wife insists on watching everything from the beginning. I told her we could skip the first star trek movie and go straight to Khan... but nooooo... so I made her sit through the directors cut, with the 16 minute starfield credits sequence at the start :)17:43
fungijedimike: _so_ _much_ _nebula_ _footage_17:44
jedimikeshe kept on "can we skip this bit?" "No! You must watch it ALL"17:44
openstackgerritA change was merged to openstack-infra/config: Install latest diskimage-builder on nodepool
jedimikefungi, yeah... more like "the slow motion picture"17:45
fungiadam_g: i was about to suggest that you prod any old friends you might have to pick up bug 1379201, but looks like zul assigned it to himself a couple hours ago17:45
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’]" [Undecided,Confirmed]
*** dalgaaf__ has quit IRC17:47
adam_gfungi, that looks like the type of bug that will move slow.  its going to probably require an SRU to precise which, at best, takes a couple of weeks17:47
clarkbadam_g: thats ... unfortunate17:48
fungiadam_g: ick!17:48
fungiadam_g: let's just hope nobody uses ubuntu precise to run neutron (or anything else) with openvswitch17:48
clarkbzaro: I have reviewed the gearman-plugin change17:48
adam_gactually maybe not, it may be addressed with new kernel17:49
fungitwo-week sru to fix a serious regression introduced by a stable update seems slow17:49
*** david-lyle has quit IRC17:50
*** david-lyle has joined #openstack-infra17:50
*** ChuckC has quit IRC17:50
clarkbwaking up early feels so productive. It isn't lunch time and the two things I needed to get done this morning are both happening what is with that17:50
*** vhoward has left #openstack-infra17:51
clarkbin other news if a horseback rider raises their arm it means something important that you will have to answer a question on your drivers test about. But I have already forgotten what the answer was17:51
jedimikeif they're wearing chain mail, that means you have been challenged to a duel, I think17:52
adam_gclarkb, haha i failed the same question when i transferred my license to OR17:53
fungiclarkb: it means lower your lance and charge?17:53
clarkbI think the real answer was something like stop beause the horse is spooked17:53
adam_gclarkb, yeah17:54
clarkbI wanted it to be a turn signal17:54
*** cpowell has joined #openstack-infra17:54
fungiclarkb: no, they mount blinkers on the horses for that17:54
*** harlowja has joined #openstack-infra17:54
fungialso halogen headlamps for night-time riding17:54
*** groknix has quit IRC17:55
openstackgerritDavid Kranz proposed a change to openstack-infra/devstack-gate: Pass args to stress tox call
clarkbI am glad I didn't hit the downtown DMV too because we had to deq for vehicle registration17:56
clarkbit is times like this that I remember the states were basically little countries that pretended to get along17:57
mordredclarkb, fungi, jeblair: so - you may be interested in knowing about SpamapS's recently released tool flamel (pip install flamel)17:57
*** groknix has joined #openstack-infra17:58
mordredI just added a patch to preserve order - but it's a yaml|json|yaml converter - which also can be used to easily reformat json or yaml cleanly17:58
fungithat does look hella useful17:58
* clarkb remembers a thing about rails derping that so hard17:59
fungialso makes me think of nicolas flamel17:59
SpamapSfungi: that is intentional :)17:59
*** markmcclain has joined #openstack-infra17:59
*** baoli has quit IRC17:59
fungiSpamapS: knowing you, i figured it probably was17:59
SpamapSit was supposed to be yaml2json17:59
clarkboh cool you can netflix on linux now if you intsall the chromes17:59
SpamapS <-- but it got squatted18:00
pleia2clarkb: yeah, it even works :)18:00
clarkbpleia2: without user agent switching now too18:00
fungiSpamapS: you could ask r1chardj0n3s_afk for special dispensation18:00
pleia2clarkb: yep!18:00
*** amuller has quit IRC18:00
SpamapSGood job Netflix, just in time for me to never try to watch netflix on anything other than chromecast or roku .. ;)18:00
SpamapSfungi: no, now I want it to be flamel :)18:00
pleia2heh, yeah tbh I just use my android tablet18:00
pleia2and my tv18:00
*** groknix has quit IRC18:00
SpamapSbecause its an awesome name. :)18:00
fungiSpamapS: oh, looks like yaml2json has an actual upload anyway18:01
clarkbSpamapS: I mostly roku myself, but there are so many things that work better with a real computer so I will gladly switch when I can18:01
SpamapSfungi: contents are nothing18:01
clarkbamazon video is pretty derpy still18:01
clarkbok we should have new dib on nodepool real soon now18:01
SpamapSclarkb: netflix doesn't strike me as one of the things that I want to use a computer for.18:01
clarkbgoing to kick off centos7 build when that updates18:01
fungiSpamapS: but yes, flamel is way better. the whole alchemy theme and making json lead into yaml gold ;)18:01
SpamapSfungi: haha I was going to say the same thing. Glad I'm not the only one who goes cross eyed when somebody says "edit this json"18:02
clarkbSpamapS: there is a computer under my television. It runs VMs during the work day and hulu/netflix/hbogo/et al in the evening18:02
fungiSpamapS: i want it to be a c program. it doesn't look like data to me18:02
*** MarkAtwood has joined #openstack-infra18:02
clarkbat some point it will likely become a steam box18:03
clarkbthen I can stream to it from desktop18:03
*** coolsvap|afk has quit IRC18:03
*** baoli has joined #openstack-infra18:04
fungiclarkb: yep, i too have a beefy debian machine in a home theater looking atx case so it blends in with the rest of my a/v components. hdmi to the teevee and optical spdif to the surround receiver18:04
fungiworks out great18:04
SpamapSI did that so long ago .. like.. first version of mythtv long ago..18:04
mtreinishfungi: heh, ditto18:04
SpamapSand then it broke18:04
SpamapSand I was like oh.. I can buy a roku and never care about this again. :)18:04
mtreinishfungi: although s/debian/archlinux18:05
clarkbexcept my roku doesn't do subtitles or hbogo18:05
fungigot an awesome rechargeable wireless bluetooth keyboard sitting on a charger on the end table next to the sofa. very convenient18:05
clarkbit is a bit annoying that upstream basically killed all updates to my roku18:05
clarkb\o/ throw away consumer devices that I cannot haxor18:06
fungiwell, i _mostly_ have it so that i can play linux-based games on the hdtv as if it were a game console18:06
jeblairfungi: this is what i use:
jeblairfungi: (i do not have a tivo; i use it with my myth box)18:06
fungijeblair: neat-o18:07
jeblairthe 'premiere or mini' version comes with a usb dongle receiver that is a standard usb hid input dev18:07
fungithat's conveniently flexible18:08
clarkbcentos7 dib image rebuilding now18:08
mtreinishjeblair, fungi: I use this:
*** mwagner_lap has quit IRC18:08
fungirax-iad still hasn't gotten a devstack-precise update to "stick"18:08
jeblairmtreinish: i mostly use it as a remote; it's just nice to have a keyboard pop out for searching for titles, etc18:09
jeblair(i use it with my roku too via the roko http api)18:09
clarkbI have first gen of
mtreinishjeblair: well I've mostly stopped using the keyboard, I use an android xbmc remote which works well18:10
adam_gmtreinish, same here18:10
clarkbits terrible. So I have a first gen HHKB too that I hook up for real keybaording :)18:10
*** _nadya_ has joined #openstack-infra18:11
*** ZZelle_ has joined #openstack-infra18:11
clarkbtar: unrecognized option '--xattrs'18:12
clarkbgoing to guess that is a peril of consuming newer dib18:12
*** jistr has quit IRC18:13
boris-42jeblair hi there18:13
boris-42jeblair I have some idea maybe crazy=)18:13
clarkbSpamapS: ^18:13
jeblairmtreinish: i'm a fan of buttons i can find in the dark for tv watching.  i have an android app for most of this stuff but i rarely use it because of the time it takes to unlock/find correct buttons18:13
clarkbSpamapS: does this mean you don't intend on supporting debian or precise?18:13
fungicurrently using as the keyboard for my htpc, and then an irda interface (the htpc case i'm using had one built in) to interface from a normal universal remote (phillips pronto touchscreen)18:13
boris-42jeblair we have issue with gerrit that we are not able to have cores in some subset of repo18:13
jeblairboris-42: who is "we"?18:14
boris-42jeblair I think every project =)18:14
boris-42jeblair in openstack,stackforge18:14
boris-42jeblair like it was discussion before about nova virt drivers splitting18:14
jeblairboris-42: are you a nova core?18:14
boris-42jeblair nope =)18:14
boris-42jeblair I am just saying sample*18:15
*** bdpayne has joined #openstack-infra18:15
mtreinishfungi: damn, that's a shiny looking keyboard18:15
clarkboh fun this is a rhel7 thing18:15
boris-42jeblair like one of reason why splitting of repo is required is to scale review process18:15
fungimtreinish: it's nice for computer games which work better from a full keyboard rather than a game pad18:15
*** tonytan4ever has joined #openstack-infra18:15
clarkbmordred: we should just run nixos18:16
boris-42jeblair and the reason is that it's not possible to allow people to be core in part of repo*18:16
mordredclarkb: ++18:16
clarkbseriously this is ridiculous18:16
jeblairboris-42: it is not required.  i think you misunderstand the issues.  anyway, it may be faster for you to simply state your crazy idea.18:16
clarkbif we want centos7 dib images we need a super new version of tar to support their selinux crap which we don't need on slaves18:16
boris-42jeblair yep sure, just to have people that has ability to +2 but not to merge stuff18:16
boris-42jeblair I mean another acl group18:16
mordredoh - like a set of people with +2 but not +A ?18:17
boris-42mordred ya18:17
jeblairi believe that has been suggested previously18:17
boris-42mordred so they can review their parts18:17
fungithat's what openstack/governance does currently18:17
boris-42mordred and if real core found +218:17
mordredfungi: not really - we have a set of people who only have +1 and one person with +2/+A18:17
boris-42mordred from that reviewer it can just workflow +1 it18:17
fungier, i guess they do +1 restricted instead, yeah18:17
boris-42mordred so this can help to scale process18:17
boris-42mordred like in rally we have benchmarks for different projects (that can be maintained by different people)18:18
boris-42mordred so and we will have scale issues with that18:18
boris-42mordred I believe tempest guys have the same issue (but already)18:18
boris-42mtreinish ^18:18
fungioh, i know what i was remembering... one of the programs was talking about doing that for their specs repo so that they could have a subset of approvers smaller than the set of core reviewers18:18
*** ivar-lazzaro has quit IRC18:19
jeblairboris-42: here's an alternative suggestion: give people +2 with the understanding that they should only review relevant work, and if they violate the trust, kick them out.18:19
jeblairfungi: that makes sense for specs18:19
mtreinishboris-42: not really, if we need subject matter experts on a specific project api we just get a core from the project to review it18:19
boris-42jeblair that can work, but for some reason doesn't work=)18:19
mtreinishbefore we approve a patch18:20
boris-42mtreinish so you have core from project + one tempest core?18:20
*** packet has quit IRC18:20
boris-42mtreinish I mean no need in 2 cores from project?18:20
mtreinishno still needs 2 tempest cores to +2 it18:20
jeblairboris-42: afaik it's never been tried :)18:20
fungiboris-42: why do you say that doesn't work?18:20
boris-42mtreinish ah18:20
*** pelix has quit IRC18:20
boris-42fungi I didn't see that18:20
mtreinishboris-42: but if there is a question about an api intricacy we just ask someone from that project to review it18:21
boris-42mtreinish ya sure, I was thinking about landing faster patches in master18:21
*** e0ne has quit IRC18:21
jeblairboris-42: we can also scale out projects by adopting the kernel subsystem model using branches in gerrit18:21
boris-42mtreinish and approaches to scale team by reducing amount of work of core team18:21
jeblairboris-42: however, i believe nova has chosen to split out modules because it makes sense technically18:21
notmynameare core team so big that they don't regularly communicate or know what each are good at?18:22
*** zns has joined #openstack-infra18:22
boris-42notmyname it's not about what I am saying, it's about making from regular developers halfcores18:22
fungithat sounds like a good reason to start breaking components out into separate projects anyway18:22
SpamapSclarkb: No, it means we don't _test_ on debian or precise. ;)18:23
fungihaving lots of people hacking independently on different parts of one piece of software is a recipe for bugs18:23
boris-42jeblair heh it's becomes hard for end user to use end product18:23
clarkbmordred: so this is a flavor of the systemd and kernel thing I was worried about18:23
clarkbmordred: you cannot build centos7 images on precise or wheezy18:23
boris-42jeblair like if it was one repo for project + one repo for client -> simple18:23
boris-42jeblair over 9k repos -> hard=18:24
SpamapSclarkb: you can't?18:24
SpamapSclarkb: we've been doing fedora images for quite a while...18:24
clarkbSpamapS: nope, you need tar --xattrs18:24
SpamapSOh that18:24
clarkbSpamapS: other wise selinux breaks simple things like ping18:24
clarkbbecause broken secure linux is better than useable linux18:24
SpamapSclarkb: isn't that only for -t ?18:24
boris-42jeblair and in any case big duplication of code=)18:25
boris-42jeblair even with libs=)18:25
mordredclarkb: gross18:25
clarkbSpamapS: I am reading edd74778912db5d4ae261f17d2ea8695d9e9ee0a in dib18:25
SpamapSclarkb: anyway, sounds like we can't support -t on old versions of tar18:25
mordredclarkb: will my docker-dib patch help this problem?18:25
mordredSpamapS: who cares about old versions of tar?18:25
*** andreykurilin_ has joined #openstack-infra18:25
clarkbmordred: I do18:25
SpamapSmordred: philistines18:25
clarkbyes I cannot build iamges right now18:26
SpamapScase in point..18:26
mordredclarkb: I'm not trolling - why?18:26
boris-42jeblair but it's another big topic=)18:26
clarkbmordred: because nodepool runs on precise18:26
mordredOH. and precise has an old tar18:26
clarkbdib is now broken18:26
SpamapShonestly, there's nothing you can do about it18:26
SpamapScentos7 is now broken you mean. ;)18:26
clarkbSpamapS: no I think everything18:26
SpamapSwe could try to use tar from inside the chroot.18:26
clarkbits a common extraction function18:26
clarkboh its a common redhat function nevermind just centos7 is broken18:27
fungiwe could, in theory, install a trusty chroot on nodepool.o.o and then run dib from that... another turtle on the pile18:27
clarkband potentially fedora18:27
SpamapSno Ubuntu wants the xattrs to18:27
SpamapSwe just worked around it18:27
SpamapSmmmm turtlepile18:27
fungior we could work on upgrading nodepool.o.o to trusty18:27
fungiwhich needs to eventually happen anyway18:28
jesusaurusis it not possible to backport new tar to precise? has that not already been done by someone?18:28
clarkbmost of my frustration is that this is specifically a thing I was told not to worry aout18:28
clarkband its happening18:28
*** vigneshvar has joined #openstack-infra18:28
greghaynesdoes the selinux attrs issue matter if you run in permissive? I think we currently require that right now...18:28
greghayneser, run selinux in permissive18:28
clarkbgreghaynes: looks like other xattrs are at play too18:29
jeblairclarkb: the problem is you are worrying about it, contrary to the instructions you were given!18:29
*** melwitt has joined #openstack-infra18:29
clarkbdib is not a sufficient abstraction to ensure builds work cross distro18:29
*** dangers_away is now known as dangers18:29
clarkbor across releases of a distro18:29
clarkbif you are homogenous its fine18:29
*** otherwiseguy has quit IRC18:29
openstackgerritChristian Fetzer proposed a change to openstack-infra/jenkins-job-builder: Add support for the Jenkins Clang Scan-Build publisher plugin.
fungithough as a future design idea, building on the idea to farm out dib tasks to separate workers, we could in theory have nodepool use base images of various platforms to run native dib builds18:31
*** ChuckC has joined #openstack-infra18:31
SpamapSclarkb: honestly, we can revert it18:32
clarkbI don't think a revert is the right answer. THis is just a limitation of dib18:32
clarkbits one that most people are happ with because they deploy a homogenous environment18:32
mordredclarkb: to be fair - it's actually a limitation of dib on precise18:32
SpamapSclarkb: or provide a flag to disable it.18:33
SpamapSclarkb: but it would still break centos718:33
mordredclarkb: at this point, there are plenty of things that dont' work well on precise18:33
SpamapSThe only want to make centos7 builds work is to do them from an OS that has a working tar.18:33
*** groknix has joined #openstack-infra18:33
clarkbmordred: it is a specific manifestation of a generic bug18:33
arosenfungi:  I think you were maybe mentioning in issue around using git fetch from zuul directly a while back? I'm seeing this weird issue in our ci that it doesn't seem to get the right patch using: $ZUUL_URL/$ZUUL_PROJECT $ZUUL_REF in jenkins_jobs/config18:33
mordredwe could change dib to use libguestfs, for instance, which uses kvm vms to do its work18:33
clarkbs/bug/limitation/ whatever term is appropriate18:34
mordredbut you'd STILL need a modern host os to do that18:34
SpamapSdib isn't going to spin up an os, just to build a chroot for another OS.18:34
arosenfungi: is this a thing?18:34
fungiarosen: i don't specifically _remember_ saying something about that18:34
SpamapSlibguestfs has come up multiple times.18:34
SpamapSprobably worth looking into it for the next backward incompatible breakage OS vendors embrace. ;)18:35
fungiarosen: do you have apache configured on the machine where your merger is running so that it serves those repositories?18:35
mordredSpamapS: I actually like the docker patch better, tbh18:35
mordredSpamapS: and then just using libguestfs to do the tarball-to-image conversion18:35
*** chuckC_ has joined #openstack-infra18:35
SpamapSyeah docker wouldn't solve this though, would it? We still do root.d from outside the docker or chroot.18:35
mordredSpamapS: what do you think happens in root.d in the docker patch?18:36
arosenfungi: i do. let me show make a quick paste with the snips:18:36
mordredSpamapS: because let me tell you, extracting a cloud tarball is not one of the things18:36
*** ChuckC has quit IRC18:36
SpamapSI haven't looked closely at what root.d changes there are.18:36
*** ChuckC has joined #openstack-infra18:36
mordredSpamapS: in the docker patch, images start from the docker base images18:37
*** achanda has quit IRC18:37
fungistevemar: this succeeded...
openstackgerritKhai Do proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests
fungistevemar: check-tempest-dsvm-neutron-heat-slow-icehouse was one of the affected jobs, right?18:38
SpamapSmordred: so side-steps the issue. And would also not work on precise most likely. ;)18:38
mordredSpamapS: I do not care about supporting running on precise18:38
clarkbarosen: we would be very surprised if the zuul ref did not point to the right patch18:38
mordredthis is new software18:38
clarkbarosen: how are you checking that?18:38
SpamapSmordred: nor I, but I do care about not breaking backward compatibility without good reason.18:38
arosenclarkb: ^18:39
SpamapShaving docker == good reason for dib 2.0 ;)18:39
nkinderfungi: I was seeing it in check-tempest-dsvm-neutron-icehouse18:39
mordredSpamapS: it would be ENTIRELY possible to write a root.d that pulled a cloud image, extracted it and imported it as a docker base image18:39
nkinderfungi: I'm not sure if check-tempest-dsvm-neutron-heat-slow was affected18:39
mordredI just think that would be a giant waste of time and be much less pleasant for everyone18:39
mordredsince the cloud images are all bloated and large and contain tons of stuff that people may not actually want18:40
SpamapSmordred: it's necessary to support other people whose root.d elements are not in-tree and who may not ship docker images *cough*hlinux*cough*18:40
mordredSpamapS: I do not care18:40
clarkbarosen: did you check out fetch_head?18:40
mordredSpamapS: they can ship a docker image18:40
mordredit's REALLY easy to do18:40
clarkbarosen: it looks like you haven't checked out the code you fetched18:40
*** achanda_ has joined #openstack-infra18:40
mordredSpamapS: I do not believe tools need to take in to account the random workflows of people who chose to not engage or participate in the process18:41
SpamapSmordred: what you're describing is a complete backward compatibility break. I am sure when you're on the other side of those, you get all Naruto-ragey on the perpetrator. :)18:41
arosenclarkb:  i did: git fetch is that not right?18:41
clarkbmordred: so my gripe with just saying "upgrade" is this process has been 6 months in the making and it should've been a couple weeks of work.18:41
mordredSpamapS: I would be suggesting the patch be a 2.018:41
clarkbarosen: no git fetch only retrives the refs, it does not check them out18:41
fungiarosen: right, git checkout FETCH_HEAD18:41
mordredclarkb: I agree. I'm not saying I don't think that this is annoying18:42
clarkbmordred: so something is clearly broken18:42
clarkbit may not be dib18:42
mordredclarkb: only that, for a dib 2.0 I don't particulary think that supporting precise needs to be a goal18:42
clarkbwe are talking about 1.33 though18:42
funginkinder: ahh, i thought it was all neutron-based tempest jobs on icehouse (because of trying to use ovswitch on ubuntu precise)18:43
*** wenlock has quit IRC18:43
arosenshould FETCH_HEAD be some combination of  $ZUUL_URL/$ZUUL_PROJECT $ZUUL_REF?18:43
*** stevemar has quit IRC18:43
clarkbarosen: no FETCH_HEAD is the last ref that you fetched18:43
clarkbso `git fetch something && git checkout FETCH_HEAD` checks out whatever something was18:43
*** stevemar has joined #openstack-infra18:44
SpamapSclarkb: I'm pretty sure dib is failing fast, where it used to fail slow by booting centos7 without a working ping. :-P18:44
nkinderfungi: it's consistently failed in just the one job for stevemar and I -
clarkbSpamapS: it didn't even boot centos7 before18:44
clarkbSpamapS: but that is apparently because the base image was broken18:44
jeblairmordred, clarkb: there are some folks (including inside of hp) that would like to have windows nodes with nodepool; that may be possible using windows-diskimage-builder, but that needs to run on a windows host18:44
arosenclarkb:  ah okay I see I need to add && git checkout FETCH_HEAD exactly what you were saying18:44
arosenthanks guys!18:45
funginkinder: okay, so in theory this one is running with the kernel downgrade applied...
jeblairmordred, clarkb: the suggestion from clarkb of farming out image creation may solve both our current problem and the windows problem18:45
SpamapSright the waters are murky. My point is, the xattrs stuff is isolated to the bits that it fixes from what I can tell18:45
*** kgiusti has quit IRC18:45
SpamapSmmm API for dib18:45
clarkbjeblair: yes I think that is the correct long term thing to do18:46
nkinderfungi: cool, I'll keep an eye on it and cross my fingers18:46
*** HeOS has joined #openstack-infra18:46
clarkbjeblair: but it is hard for me as the person doing a bunch of the work to keep tripping over all these things and not want to go back to what we currently have18:46
*** wenlock has joined #openstack-infra18:46
clarkbit works, reliably, with minimal fuss beyond publication of images18:46
*** jogo is now known as flashgordon18:46
*** stevemar2 has joined #openstack-infra18:46
jeblairpublication of images is not the primary reason we're doing this, it's just a nice side benefit18:47
*** stevemar has quit IRC18:47
*** shashankhegde has joined #openstack-infra18:47
jeblairconsistency and efficiency are i think the primary drivers18:47
*** thedodd has joined #openstack-infra18:47
jeblair(rax != hp)18:47
clarkbwe can get that too with the current system18:47
jeblairclarkb: we can get consistency?18:47
*** openstackgerrit has quit IRC18:48
clarkbwe could build a single snapshot image, download it, convert it to whatever other formats we need then upload it to all of the others18:48
clarkbthat snapshot image could be based on base images we control not images provided by the cloud providers18:48
clarkband we would have consistent images across all providers18:48
*** zns has quit IRC18:49
*** kgiusti has joined #openstack-infra18:49
*** __TheDodd__ has joined #openstack-infra18:50
*** Sukhdev has joined #openstack-infra18:50
jesusauruswhy does openstackgerrit timeout so often?18:51
clarkb(if we don't want nodepool to be growing features like this packer does support most of that workflow)18:51
*** kmartin has quit IRC18:52
jeblairjesusaurus: because it doesn't actively ping the server and reconnect on ping timeouts like uvirtbot18:52
*** thedodd has quit IRC18:52
jeblairmordred: ?18:52
jesusaurusjeblair: ahh18:52
mordredeverything about how packer works grinds my brain like broken glass18:53
mordredI want no part in use touching it18:53
clarkbmordred: it basically does what nodepool does today18:53
clarkbthey work almost identically18:53
mordredit just does it obtusely and completely unintelligently18:53
fungirestarted the gerritbot just now18:54
*** openstackgerrit has joined #openstack-infra18:54
fungiand it's back18:55
fungiout to grab a late lunch, then back as soon as i can be18:56
mordredclarkb: ranting aside, can you point me towards an example of a packer config file that does something?18:56
mordredclarkb: because every time I've tried to read their docs, their examples are so mind-bendingly stupid that I cannot tell how I'd actually perform a real task18:57
mordredclarkb: this may come from the fact that something about the way the packer/vagrant world thinks of things and the way I think of them are at odds18:57
mordredconsidering I find vagrant useless as well18:57
clarkbyou have a builder section and shell provisioners18:58
clarkbits almost identical to what nodepool config looks like18:58
*** maxbit has quit IRC18:58
mordredclarkb: what would we gain from doing that?18:58
clarkbmordred: nodepool wouldn't need to have the code in it18:59
clarkbwe get image built in similar way that should always work because it runs on the resulting platform18:59
clarkbthen nodepool can copy resulting image to all the other regions18:59
clarkbor just have nodepool do it18:59
clarkbit already does most of it18:59
*** MarkAtwood has quit IRC19:00
mordredI don't think we gain anything by adding packer to the mix19:00
mordredbeing able to build locally without needing a cloud is one of the nice things about dib, as is the local caching of objects and the fact that the tool is easy to set up and reproduce locally19:00
clarkbmordred: except we have just said that we cannot do it without a cloud19:00
mordredif we stay with using cloud instances to do things, then I think just adding features to nodepool is fine19:00
mordredclarkb: that is a thing that has been said19:01
mordredI believe that there is currentlya  bug in dib19:01
*** maxbit has joined #openstack-infra19:01
mordredsince being able to do this is a core feature of the tool19:01
mordredand I dont think that ditching because we've uncovered a bug is the right choice19:01
clarkbmordred: I am more worried that we are ditching a known good for a new thing that has known bads19:02
mordredwe are ditching a thing with some known good and some known bad for a new thing with some known good and some known bad19:03
*** jedimike has quit IRC19:03
mordredthis is always the case in a tech switch19:03
clarkb(and I have been happy to patch both nodepool and dib to accomodate this, but the idea that we need a new server farm just to build images with dib which is already a thing we can do without dib seems silly)19:04
mordredI do not think we need a new server farm19:04
jeblairi think we should consider a new server farm :)19:05
jeblairso how about this:
jeblairi'm attempting to solve the current problem, the windows problem, and just plain make the system more flexible19:05
mordredjeblair: on 5 - what do you mean by providers?19:06
jeblairi'm thinking split things out via gearman workers like we do with zuul19:06
jeblairmordred: cloud providers: hp/rax19:06
mordredjeblair: k. I think I might want to suggest a slightly different thing then ...19:06
jeblairclarkb, mordred: if we decide we don't want any new servers for this, we can still run it all on one host19:07
*** homeless has quit IRC19:07
jeblairclarkb, mordred: if we decide we want to split nodepool from the image builder (2 servers) we can do that19:07
jeblairclarkb, mordred: if we decide we need to build centos7 images on a separate host (3 servers) we can do that19:07
mordredhrm. wait - let me float the idea here first19:07
jeblairclarkb, mordred: and the windows folks can run windows-dib on a separate server19:08
mordredjeblair: I think having a content producer job and an uploader job would be helpful19:08
*** homeless has joined #openstack-infra19:08
mordredjeblair: because the content producer can produce a base thing like a raw or a tar.gz19:08
jeblairclarkb, mordred: and we can swap out builders if we decide we want to use dib/docker/packer/whatever19:08
mordredjeblair: and different providers need that in different formats19:08
jeblairmordred: i was thinking the upload should be part of the build job so that it's guaranteed to run on the same host and the resulting image does not need to be copied around19:09
mordredyou know - if we start with your thing, we can always split it in twain later if need be19:10
mordredI think in the docker case there is an optimization that can be gained by splitting - but we're so many steps away from that ...19:10
mordredso I'll drop it for now19:10
jeblairyeah; i mean those are solvable problems (we can include routing information so the job goes to the right worker; we do that with zuul today)19:10
jeblair(i do think copying the image around is silly though)19:11
*** achanda_ has quit IRC19:11
jeblairmordred: i sketched out how do do a two-phase thing19:13
mordredyah. I like that as a possible-step-219:13
mordredbut not needed right now19:13
jeblairi guess that gets us a separation of the "convert image to format needed by provider and upload" bit from the "know how to build an image" bit.  that makes sense for more plugability/flexibility.19:14
mordredand like zuul, if you want a single-node thing, the workers can just co-locate19:14
*** ivar-lazzaro has joined #openstack-infra19:15
*** ivar-lazzaro has quit IRC19:15
*** ivar-lazzaro has joined #openstack-infra19:16
*** rushiagr is now known as rushiagr_away19:18
*** cdent has quit IRC19:21
*** eharney has quit IRC19:21
stevemar2fungi, do you think it's worth setting the neutron-icehouse job as non-voting? cc nkinder i'm wondering how long it'll take that bug to be fixed19:23
*** packet has joined #openstack-infra19:24
*** hashar has joined #openstack-infra19:25
openstackgerritEdward Raigosa (wenlock) proposed a change to openstack-infra/config: provide an option in plugin to pre-update a plugin with puppet
*** rainya has joined #openstack-infra19:29
clarkbstevemar2 didnt we just fix the kernel thing?19:29
stevemar2clarkb, let me do a recheck... last i tried it wasn't19:30
clarkbI dont want to make something like that nonvoting when we have a fix19:30
stevemar2clarkb, of course not19:31
*** kashyap has quit IRC19:34
*** kmartin has joined #openstack-infra19:34
*** cpowell has quit IRC19:36
*** achanda has joined #openstack-infra19:36
*** liusheng has quit IRC19:39
*** liusheng has joined #openstack-infra19:39
*** dkehn has quit IRC19:41
*** e0ne has joined #openstack-infra19:41
*** dkehn has joined #openstack-infra19:43
*** ChuckC__ has joined #openstack-infra19:43
*** kgiusti has quit IRC19:43
*** kgiusti has joined #openstack-infra19:44
*** mjturek has quit IRC19:45
*** ChuckC__ has quit IRC19:47
*** achanda has quit IRC19:48
*** achanda has joined #openstack-infra19:48
*** tonytan4ever has quit IRC19:49
*** hdd has quit IRC19:49
*** hdd has joined #openstack-infra19:50
*** __TheDodd__ has quit IRC19:53
*** unicell has quit IRC19:53
*** imcsk8 has quit IRC19:55
*** imcsk8 has joined #openstack-infra19:55
*** ociuhandu has quit IRC19:56
*** stevemar2 is now known as stevemar19:57
nkinderfungi: looks like that neutron-icehouse job succeeded!19:59
*** otherwiseguy has joined #openstack-infra19:59
clarkbSpamapS: mordred because I just noticed after tests on my dib change failed. we do test dib on precise...19:59
*** Ryan_Lane has joined #openstack-infra19:59
nkinderfungi: I'll kick one of the failing reviews in a bit and see if it makes it through20:00
*** groknix has quit IRC20:01
*** baoli has quit IRC20:03
ZZelle_Hi everyone!20:04
*** kgiusti has quit IRC20:05
*** imcsk8 has quit IRC20:05
*** imcsk8 has joined #openstack-infra20:05
*** kgiusti has joined #openstack-infra20:06
*** alexpilotti has quit IRC20:06
openstackgerritA change was merged to openstack/requirements: Add tempest-lib as a global requirement and a tracked project
ZZelle_Does someone know how many times it takes to ubuntu to publish as packages OpenStack last release?20:06
ZZelle_(typically 2014.1.3 :))20:07
*** wenlock has quit IRC20:07
*** mjturek has joined #openstack-infra20:08
*** ihrachyshka|afk has quit IRC20:09
*** wenlock has joined #openstack-infra20:10
*** e0ne has quit IRC20:10
*** stevemar has quit IRC20:12
*** Longgeek has joined #openstack-infra20:13
*** andreaf has joined #openstack-infra20:15
*** Longgeek has quit IRC20:17
funginkinder: thanks--excellent news!20:17
*** jheroux has quit IRC20:18
*** _nadya_ has quit IRC20:19
fungii should have gone to lunch sooner. broadband was back in service when i returned home20:21
pleia2fungi: ssshhh, don't say that outloud, it might go away again20:22
pleia2lunch is a good idea20:22
fungii'm just glad to not be forwarding all my packets through cell towers again20:24
fungisuch is island life20:24
*** bknudson has quit IRC20:24
*** kgiusti has quit IRC20:25
*** yolanda has quit IRC20:25
fungimrmartin: if you're still around, when do you think would be a good time for me to rebuild/replace with a 4gb server?20:26
*** zz_gondoi is now known as gondoi20:26
*** denis_makogon has joined #openstack-infra20:26
*** david-lyle has quit IRC20:26
*** kgiusti has joined #openstack-infra20:26
fungimrmartin: judging from it very quickly spiked back up to using all its memory after i restarted apache20:27
*** krtaylor has quit IRC20:28
fungimrmartin: by comparison, groups-dev seems to use right at 1.5gb of ram so maybe this is a tuning choice?
*** tonytan4ever has joined #openstack-infra20:29
fungii would hate for us to throw extra capacity at a system just because we hadn't bothered to adjust it from its default settings20:30
*** kgiusti has left #openstack-infra20:32
*** koolhead_ has joined #openstack-infra20:34
*** koolhead17 has quit IRC20:34
*** achanda has quit IRC20:36
*** Ryan_Lane has quit IRC20:36
*** e0ne has joined #openstack-infra20:38
mrmartinfungi, I'm here now20:39
mrmartinif you don't want to recreate the instance now, what you think about adding an 1GB of swap?20:40
fungimrmartin: it already has 1gb of swap, but rackspace seems to freak out when an instance starts swapping heavily. apparently their storage backend can't set quotas around i/o ops20:41
fungiso heavy swap usage creates a denial-of-service situation for other instances sharing the same backend20:41
mrmartinok. I also want to add a varnish to this stack, and it needs at least 1G memory20:42
fungimrmartin: no problem. i'm happy to upgrade us to a bigger instance if we actually need it, just curious when you think would be good timing20:43
mrmartincould we recreate this instance with 4G? Yesterday I asked a guy who is running a larger Drupal 7 site, and they have servers with 12G and 20G of memory.20:43
fungimrmartin: i can easily build a 4g replacement for it20:43
mrmartincould this survive until monday, or we need to do it now?20:43
fungimrmartin: it can wait to next week, no problem20:43
*** harlowja is now known as harlowja_away20:43
fungimrmartin: my biggest concern is whether bringing up a replacement in parallel with the running instance will create database locking issues since they would attempt to share the same database server20:44
fungimrmartin: which determines whether we need to shut down the current machine when bringing up the replacement, or can do a faster dns cut-over with both of them up and running in parallel20:45
mrmartinyou are right, if you bring up a new instance, it will kill the db :)20:45
mrmartinthe actual drupal puppet was not planned as a cluster solution20:45
fungimrmartin: we can also mitigate that in a few different ways, but just want to figure out if it's worth the effort, or if taking it down for an hour on a weekend is acceptable20:46
fungino point in making things hard for ourselves for minimal gain, after all20:46
mrmartinI think it is, we don't have an SLA20:46
mrmartindo we have db backups for this database?20:47
*** harlowja_away is now known as harlowja20:47
*** e0ne has quit IRC20:48
mrmartinanyway I think, if you recreate the instance, we won't lost any data, because now everything comes from third party data sources.20:48
fungimrmartin: doesn't look like we have automated backups of it, though we can create on-demand backups fairly easily20:48
jeblairwe should use the typical local backup -> backup server pattern20:49
fungimrmartin: and if we want, we can set up automated backups, yes20:49
jeblairthe mysql_backups puppet module20:49
mrmartinso don't even need to do a backup now. but this will change if we can finish this openstackid integration and end users and the users will produce content.20:49
*** Sukhdev has quit IRC20:50
fungijeblair: yeah, that's what i was checking for. we can add it fairly easily though20:50
mrmartinin the future it is easy to setup at least a 2 node cluster, so it can survive this type of upgrade. the only criteria is that it requires a shared fs for user content files.20:51
*** jgrimm is now known as zz_jgrimm20:52
*** e0ne has joined #openstack-infra20:52
fungimrmartin: right, the shared fs might be hard to come by unless we provide some sort of virtual shared filesystem... (there's one in particular which comes to mind)20:52
mrmartinok, so now, you can even recreate the instance, we won't lost any data.20:52
clarkbzaro: are you around? would really like to see the comments on gearman-plugin change addressed so that we can update that plugin in the near future20:53
mrmartinfungi, we used nfs in a production system before, but I can imagine better solutions now.20:53
*** e0ne has quit IRC20:53
mrmartinif the varnish is caching up the static assets it can be right.20:53
mordredshared filesystem is easy20:54
mordredit's in our plans already20:54
fungimordred: swift fuse plugin? cinder mount with takeover signaling?20:54
fungimordred: ohhhh20:54
* mordred throws wet cat at fungi20:54
fungi(i think you mean the same one i meant in my "can think of one" offhand comment)20:55
mordredjeblair: ^^ another use case / need for our next project20:55
mrmartinso, what's plan with this rebuild? Postpone to monday?20:55
clarkbshould we remove the centos7-dib image from nodepool until we are able to build it reliably? otherwise we will continue to boot broken nodes20:55
mordredclarkb: yes. I think that's probably very right20:55
fungimrmartin: yeah, it's fine for the weekend. i just wanted to find out what sort of process i should plan to follow20:55
fungimrmartin: we can revisit on minday20:56
fungimonday even20:56
mrmartinok, write on me on monday.20:56
fungiwill do20:56
fungithanks and have a great evening!20:56
mrmartinone more thing20:56
* fungi nods20:56
*** dhellmann has quit IRC20:56
mrmartinI had a discussion with Stefano today about this openstackid, and we have two options here, a, put cloudflare in front the service20:57
*** dhellmann has joined #openstack-infra20:57
mrmartinor b, add the mod_security plugin to apache.20:57
*** miqui has quit IRC20:57
*** mbacchi has quit IRC20:57
jeblairmrmartin: why?20:57
mrmartinwe are a waiting for a decision in this topic also.20:57
mrmartinjeblair: because this app is handling the authentication, and have access to user accounts.20:58
mrmartinit is also a decision, to not do anything :)20:58
fungii gather there are concerns with unknown vulnerabilities lingering in the php code which is the openstackid service20:58
openstackgerritClark Boylan proposed a change to openstack-infra/config: Set devstack-centos7-dib min ready to 0
mrmartinbut I sleep better if I know we did everything reasonable.20:59
fungiin the interim, i built the server but disabled http/https access from the internet via iptables until we have the all-clear to make it reachable20:59
fungii've whitelisted my own ip address and performed some rudimentary tests against the interface which show it seems to be working as designed, at least21:00
*** julim has quit IRC21:00
*** wenlock has quit IRC21:01
mrmartinallright, I need to go now, let's do this upgrade on Monday.21:01
jeblairmod_security is reasonable, if quite a bit of work to set up correctly, and really only blocks the most basic of attacks (eg, sql injection through URLs)21:01
fungii have a suspicion cloudflare doesn't block much more than that, but is also not free software21:01
mrmartinjeblair: in both case we need to retest the application again, because mod_security also can brake things :)21:01
mordredwhat's a cloud flare?21:01
nibalizermordred: its like if you took a plane to the clouds and told them to be faster21:02
*** dhellmann-phone has joined #openstack-infra21:02
jeblairtbh, if we're so afraid that the app has that kind of vulnerability, perhaps we should use a different app.21:02
mordredwhy do we need a CDN for security?21:02
nibalizerjeblair: ++ for that philosophy21:02
fungimordred: the non-free caching proxy service in front of the site which likes to periodically break fro anyone connecting from the summit conference network21:02
mordredoh. so, I'll say21:02
mordredthat's done21:02
mordrednext question?21:02
*** dhellmann-phone has quit IRC21:02
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Filter stories by project_id, status, project_group_id, more.
mrmartinso which one is tho no, the cloudflare?21:03
mordredno to cloudflare21:03
mordrednot an optino21:03
jeblairyes: no.  :)21:03
mrmartinok, I'll prepare a patch for mod_security on monday21:03
fungii'd like to say they're both no and we should just make sure we have confidence in the security of the software we're running instead21:03
* krotscheck finally figured out SQLAlchemy magic.21:03
mrmartinand try to do some app level test, also.21:03
krotscheckBut he’s NOT happy with the size of the join that the search story by project group is creating.21:04
jeblairfungi: mod_false_sense_of_security21:04
mordredkrotscheck: woot!21:04
fungithe one thing mod_security *does* have going for it is that it's moderately easier to add your own patterns to block if you end up under an active attack and can capture a sample of the payload21:04
mordredthat's a nice feature21:04
mrmartinlast time I deployed mod_security, half of the app was broken due some rules.21:04
fungibut that's a bit of a corner case21:04
fungiand really, you *can* do the same thing directly in your apache config21:05
*** aysyd has quit IRC21:05
krotscheckmordred: You know SQL things, mind looking through this review chain? I promise it won’t take more than a day.
fungimod_rewrite to the bitbucket21:05
mrmartinok, guys, have a nice day there, really going now.21:06
*** mrmartin has quit IRC21:06
fungihaving managed my share of different network intrusion detection services, some free software, some commercial, some hardware-based devices, the one thing they all have in common is that they break you at random for no good reason21:06
*** achanda has joined #openstack-infra21:06
*** whoops has joined #openstack-infra21:07
*** sacharya has quit IRC21:07
fungibecause, really, the line between unusual but legitimate traffic vs unknown new malicious attack is basically indistinguishable21:07
clarkbthe one feature that I Have used in the past that workedreliably was not returning 500 responses to clients because security21:08
*** mjturek has quit IRC21:08
clarkbit just confused people but it did so reliably21:08
fungiand i can say i never recommended one to anybody, but managed plenty because a customer (or more often their $tandard$ compliance auditors) insisted they had to have one even if we couldn't guarantee against it causing random breakage in the middle of their black friday gonzo sale spectacular21:10
*** achanda_ has joined #openstack-infra21:10
*** MaxV has quit IRC21:10
*** achanda has quit IRC21:12
krotscheckIt’s not like our code isn’t visible to the world. Anyone who cares can pull our server profiles from puppetboard, the code that’s being run from git, and look for intrusion measures to their eharts content.21:12
clarkbSpamapS: so why is tar involved at all? are distros not handing out qcows that can be mounted direclty?21:13
flashgordonmordred: do you know why nova has this
flashgordonauth plugins in the client21:15
flashgordonI think RAX uses it21:15
mordredclarkb: because it needs to be in a chroot21:15
*** achanda_ has quit IRC21:15
mordredclarkb: most of dib does chroot, not nbd-bind-mount21:15
clarkbmordred: but you could mount and chroot into the mount right?21:16
*** homeless has quit IRC21:16
mordredclarkb: this is why this particularly thing is fixed by the docker-dib patch, because that just pulls base images from docker21:16
mordredclarkb: maybe so? dunno, I haven't tried that21:16
*** homeless has joined #openstack-infra21:17
clarkbapparentl my dib patch is now broken too but I can't reproduce locally21:17
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Subscription events now appear in UI
*** maxbit has quit IRC21:19
*** zaphod42_ has joined #openstack-infra21:26
*** zaphod42 has quit IRC21:29
mordredclarkb: ok. I've just officially done something on reddit21:30
mordredclarkb: does that mean either I or reddit are so last year?21:31
*** Sukhdev has joined #openstack-infra21:31
*** zaphod42_ has quit IRC21:31
clarkbmordred: reddit is last year. they are going to start paying users and charging people to link to their own stuff21:31
clarkbwhich is probably a great thing for some people21:31
mordredclarkb: ewww21:33
clarkbI am running a centos7 build on trusty rael quick to see if that works21:34
clarkbmordred: SpamapS: so I think "Repacking base image as tarball." is a really weird step21:35
clarkbits slow21:35
clarkband breaks builds21:35
*** homeless has quit IRC21:36
clarkbregardless of xattr and tar and any other potential issues. I am still having a hard time understanding why that step is necessary21:36
fungimordred: reddit is the new 4chan21:37
JayFfungi: HN is the new reddit?21:37
fungiJayF: usenet is the new hn21:37
JayFfungi: and nobody will tell us where the new HN is because when it gets popular everyone gets pushed a step down the line?21:38
mordredfungi: yes!21:38
* fungi blows your mind21:38
JayFhiding in plain sight. I like it.21:38
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: New search parameters.
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: New search parameters.
*** sacharya has joined #openstack-infra21:40
mordredflashgordon: it was an old thing21:40
flashgordonmordred: I think its a case of special code for racksapce21:40
flashgordonwhich you should submit a patch to remove21:41
mordredright. well, it was added when we originally removed direct support for rax auth in python-novaclient21:41
mordredflashgordon: I think we should not bother until novaclient is using keystoneclient21:42
mordredbecause keystoneclient does have an auth plugin system that is for things other than vendor compat21:42
mordred(oauth comes to mind)21:42
mordredand any monkeying around with novaclient auth support that isn't porting to keystoneclient is a waste of time, imho21:42
*** packet has quit IRC21:44
*** achanda has joined #openstack-infra21:44
flashgordonmordred: we are close to using keystone client for everything21:45
flashgordonoh I didn't realize keystoneclient has what we need already21:45
flashgordon I suspected that was the case though21:45
*** jedimike has joined #openstack-infra21:48
*** achanda has quit IRC21:48
*** xyang0 has quit IRC21:49
mordredflashgordon: well, you'll LOVE the thing deva and I just leaned about tenant_name vs. tenant_id21:50
*** gondoi is now known as zz_gondoi21:51
mmedvedeHi there, I am having issues with nodepoold startng up. It runs for 10 seconds, and dies. Logs are empty. Anybody has an advise to diagnose the issue?21:51
* fungi is afraid to find out21:51
flashgordonmordred: oh do tell21:51
fungimmedvede: check that there isn't a stale pidfile in /var/run/nodepool, and that nodepool has appropriate permissions to create a pidfile there21:52
mordredflashgordon: well, against hp and rackspace public cloud, project_id and project_name are interchangable, it does not matter which one you use21:52
mordredflashgordon: but against devstack, it does matter21:52
mmedvedefungi, I made sure that the pidfile is empty and is ownded by nodepool user21:52
flashgordonmordred: guh21:52
flashgordonmordred: well project_id and project_name are different things right?21:53
fungimmedvede: it has permission to write its logfiles too?21:53
mordredflashgordon: yes - except unless you're talking to rax or hp public cloud21:53
flashgordonand is that all with keystone v2 or v3?21:53
flashgordonso rax and HP both have the same hack?21:53
mordredflashgordon: you can pass a tenant name in as project_id against those and it works21:53
fungimmedvede: you might have more luck using su to switch to your nodepool user and trying to start nodepoold interactively without disassociating from the controlling terminal21:53
mmedvedefungi, yes, nodepool user is 666 on the logs, double checking to make sure...21:54
mordredflashgordon: also, news flash - as a user, I do not know or care if I'm using keystone v2 or v3 - I just want to authenticate21:54
mtreinishmordred: sure you do, on one you have to worry about a domain and what that means21:54
mmedvedefungi, I think the interactive node is what I am looking for21:54
mtreinishand the other works the way you know21:54
* flashgordon gasps in disbelief21:55
mordredmtreinish: let me rephrase that for you:21:55
fungimmedvede: nodepoold -d21:55
fungimmedvede: nodepoold --help tells you what command-line options it accepts too21:55
mtreinishmordred: I'm not saying it should be that way21:55
mordredmtreinish: def get_keystone_token(auth_url, username, password, tenant_name, region=None, domain=none):21:56
flashgordondevananda: lolz21:56
mordrednow, as a user, I do not have to care ^^21:56
mmedvedefungi, great! thanks for advise, I should be good from here21:56
mordredbecause if I provide a domain, something will happen, and if I don't, a different thing will happen21:56
mtreinishmordred: fair enough, is it that way now?21:56
*** doug-fish has left #openstack-infra21:57
fungimmedvede: well, if you can't diagnose it from that, i still have other possible suggestions (depending on what you end up seeing)21:57
mordredI mean, it almost is21:57
mordredexcept when it's not21:57
mtreinishheh, well that's always how it seems to be21:58
mmedvedefungi, your suggestion about the log permissions was correct. I just missed one log file21:58
fungimmedvede: also, i highly recommend reading the friendly documentation at
fungimmedvede: er, i meant
fungi(though that other link is potentially useful for examples of how we use it)21:59
*** achanda has joined #openstack-infra22:01
mmedvedefungi, I have skimmed through the docs earlier, and there is actually no pointers to the nodepoold options.22:02
mmedvedeThere is a good reference for nodepool cli, not the daemon22:03
fungimmedvede: oh, indeed, skimming it now i see that22:03
*** datsun180b has quit IRC22:04
*** JayJ_ has quit IRC22:04
*** hashar has quit IRC22:04
fungimmedvede: care to file a very quick bug at requesting that? i might make a good low-hanging-fruit task for someone22:04
fungier, it might22:04
*** denis_makogon has quit IRC22:05
mmedvedefungi, sure, can do. Thanks again for the help, you saved me some friday night time :)22:05
fungiany time, and have a great weekend!22:05
mmedvedetks, you too22:05
*** krtaylor has joined #openstack-infra22:07
ianwclarkb / mordred : yeah, hindsight is 20/20 but my testing till now has been done on RH platforms22:08
*** andreykurilin_ has quit IRC22:08
*** sacharya has quit IRC22:09
*** tonytan4ever has quit IRC22:10
pleia2clarkb: can you have a look at the Pootle server? It's gone sideways :\22:11
pleia2or anyone with access to translate-dev22:11
*** jedimike has quit IRC22:12
clarkbI am currently transitingto meet hogepodge22:13
fungipleia2: i'll take a peek22:13
clarkbfungi it should be running out of screen :/22:13
pleia2clarkb: have fun, tell him I put a debian pin on my red hat :D22:13
hogepodgepleia2 \o/22:13
pleia2hey hogepodge!22:13
fungipleia2: server's up 12 days since last restart... when did you last see it work?22:13
hogepodgeI'll ask here in infra22:13
pleia2fungi: I have been working on the zanata server lately, so I don't remember22:14
clarkboh it would have died in the restarts22:14
fungiclarkb: definitely no screen session22:14
hogepodgeIf anyone has a base trove guest image for me, I'd love to have a link to it.22:14
fungiclarkb: quick hints on restarting it then?22:14
clarkbfungi ya so it will need a new screen as pootle user then run the service22:14
hogepodgeOne that I can pry open, change for my server, then start demoing against.22:14
fungiclarkb: bash history in a pootle user account maybe?22:14
clarkbthe command should be in bash history22:14
hogepodgeclarkb there's a contingent from eugene going to baileys22:15
clarkbis this a good thing ;)22:16
hogepodgeyup, I'm going to start heading that way22:16
clarkbawesome. my eta is about 30 minutes22:16
fungipleia2: any better?22:17
fungiclarkb: there is no pootle user, only root22:17
*** gyee has quit IRC22:18
clarkboh we have two servers because I didnt want to kill jeblairs22:18
pleia2that's the old one22:18
clarkband the second is not in dns22:18
clarkb:/ I should probably kill old one and fix dns22:18
fungii should pull the not-dns-one out of nova list?22:19
*** hdd has quit IRC22:19
fungiand start it on there?22:19
pleia2oh, right is the new one22:20
pleia2forgot that we never switched dns over22:20
fungipleia2: thanks, confirmed22:20
pleia2that one is down too :)22:20
fungipleia2: figures... fixing tho22:20
pleia2heh, my email22:20
pleia2" (we'll be updating the DNS over from the old22:20 server momentarily, but I wanted to make22:20
pleia2sure you got to the right server)"22:20
pleia2"or, you know, never" :)22:21
*** bcourt has quit IRC22:21
clarkbthat is my fault22:21
pleia2nah, I should have followed up22:21
fungiwow, there is no ~pootle/.bash_history either22:21
fungioic, pootle account has no shell22:22
pleia2fungi: this should help:
clarkboh yes try root again probably has sudo -u commands22:22
clarkbit does run as pootle but like a system user22:22
pleia2should just need: sudo -H -u pootle /home/pootle/venv/bin/pootle start22:23
pleia2I hope :)22:23
*** jergerber has joined #openstack-infra22:23
pleia2(in screen)22:24
fungiit wants a subcommand22:25
fungiUsage: pootle subcommand [options] [args]22:25
pleia2that would be "start"22:25
fungiaha, yup. just found it in the etherpad22:25
fungii see, your command wrapped and i missed the word start on the next line22:26
fungirunning nopw22:26
fungier, now22:26
pleia2thanks fungi22:27
*** erlon_ has quit IRC22:29
*** alexpilotti has joined #openstack-infra22:30
*** otherwiseguy has quit IRC22:31
*** alexpilotti has quit IRC22:31
*** rainya has quit IRC22:40
*** ayoung has joined #openstack-infra22:40
*** dimsum_ has quit IRC22:43
mordredfungi, clarkb: did you see tchaypo's email to the list about "prep-source-repos"22:44
*** flaper87 is now known as flaper87|afk22:46
mtreinishis there a delay for bandersnatch to pick up new projects from pypi?22:49
*** whoops has quit IRC22:50
fungimtreinish: ~5 minutes normally22:50
fungimtreinish: what are you seeing?22:50
mtreinishthe tempest python unit tests are failing trying to pip install tempest-lib22:50
mtreinishbut I pushed the release yesterday22:50
fungi2014-10-09 21:15:02,342 INFO: Downloading:
fungiso at least on it's fine22:52
mtreinish2014-10-10 22:37:34.989 | Could not find any downloads that satisfy the requirement tempest-lib (from tempest==2.dev1241.ge8fb428)22:52
fungichecking that one22:52
fungilooks like it hasn't run bandersnatch on that mirror since 2014-10-09 14:35 utc22:53
*** koolhead_ has quit IRC22:53
fungithe logs just end there22:53
mtreinishok, that would explain it22:53
clarkbhogepodge I am here. dont see you22:54
fungiaha, stray lockfile22:54
hogepodgeclarkb I haven't left my apt yet22:54
*** ociuhandu has joined #openstack-infra22:54
clarkbboo :P22:54
*** openstackgerrit has quit IRC22:54
hogepodgeI'm on my way. Been struggling with this stupid trove image problem, plus no help from anyone on the openstack-trove room22:54
hogepodgesorry I'm late22:55
jeblairclarkb, fungi, mordred:
fungimtreinish: it's updating on that mirror now. i think maybe i wrapped timeout around lockfile when i should have wrapped lockfile around timeout22:55
*** koolhead_ has joined #openstack-infra22:55
jeblairthat's a spec for nodepool image build workers22:55
fungier, no it's flock calling timeout22:55
mtreinishfungi: heh ok, thanks22:55
funginot sure how it left a stale lockfile there in that case22:55
*** dmsimard is now known as dmsimard_away22:56
clarkbjeblair reading now22:56
clarkbjeblair I prefer having a tool like dib do the image conversion22:59
*** mfink has quit IRC22:59
*** rwsu has quit IRC22:59
clarkbI dont think nodepool builders should need to grok qemu-img or other tools22:59
jeblairclarkb: how does that work?22:59
jeblairclarkb: how do we convert image formats now?23:00
mordredjeblair: you can give dib a list of formats to produce23:00
clarkbjeblair no current conversions so I wrote and
*** esker has quit IRC23:01
mtreinishfungi: hmm, I'm seeing the tempest jobs hit the same mirror too, but devstack should be installing it from source already...23:01
mtreinishyet another thing to dig into23:01
clarkbapparently I broke ramdisk builds which do a save_image in the elements which should be verboten imo23:01
clarkbbut we'll see "ramdisks are special"23:02
mordredclarkb: if there is an _option_ for having the two be separate operations23:02
*** marcusvrn_ has quit IRC23:02
jeblairwhat are our actual needs?23:02
clarkbmordred which two?23:02
mordredclarkb: then a solution that provides image building could use two different tools for those two tasks23:02
jeblaireg, hp needs X, rax needs Y...?23:02
mordredjeblair: hp needs qcow, rax needs xen server thing23:02
clarkbya rax is vpc/vhd23:03
*** andreaf has quit IRC23:03
*** andreaf has joined #openstack-infra23:03
jeblairso i think we have 2 reasonable solutions:23:03
jeblair1) use qemu-img convert in all builders as necessary23:04
jeblair2) have the builder supply its own image conversion method (in a subclass-plugable way) so the dib builder will use dib to do the conversion23:04
*** sacharya has joined #openstack-infra23:04
*** mfink has joined #openstack-infra23:05
jeblairclarkb: it doesn't look like your patch is for a standalone tool though, yeah?23:05
jeblairclarkb: it looks like it just says "when you create this, output multiple formats"?23:05
clarkbno its for the current dib tooling23:05
fungi3) glance should eventually grow the ability to convert between image formats for users23:05
jeblairfungi: 3++23:05
clarkbjeblair yes23:05
clarkbI was trying to get this done last week...23:05
jeblairclarkb: understood :)23:06
jeblairclarkb: ideally, i'd like to keep it separate, because the format for upload is tied to the provider, not to the image being built23:06
fungiin between fits of zomg openstack is not portable/interoperable23:06
jeblairclarkb: this way you can add an image to a provider (which may require a new format) and not have to rebuild23:06
clarkbmy concern with that is nodepool is trying to stop knowing about image internals23:07
jeblairclarkb: that makes me lean toward option #123:07
jeblairclarkb: it's more of an 'external' :)23:07
clarkbbut that is a good point about mixing up providers without full rebuilds23:08
jeblairclarkb: i mean, it already knows it has a file, it knows it needs to upload it somewhere, and it knows what format the somewhere needs the file in23:08
*** andreaf has quit IRC23:08
clarkbfungi ya that was my first thought on the matter23:09
jeblairmine too, i just feel like if i voice it at this point i'm just repeating myself :(23:09
clarkbif glance exists for any reason its so I dont have to think about this stuff right?23:09
clarkbbut we wont get those features soon so forward23:10
*** koolhead_ has quit IRC23:10
clarkbjeblair should I abandon my changes then?23:10
clarkbneither currently work so I am meh about then23:10
*** melwitt has quit IRC23:11
*** chuckC_ has quit IRC23:13
jeblairclarkb: i hesitate to say abandon because i'm still catching up to what's going on, and i don't know how long it will take to implement the spec i wrote.  however, if we end up on the trajectory the spec predicts, i think we won't end up using them in the long run.23:14
jeblairi have to run now23:14
*** Sincler has quit IRC23:15
*** achanda has quit IRC23:16
mordredclarkb: also, one could argue that the image-conversion routines in dib shoudl also be standalone23:16
mordredclarkb: so taht you could run dib with a list and have it do everyting23:16
mordredclarkb: but also later decide you just wanted to run dib-image-convert --to qemu my.tarball.file.tgz23:17
mordredwhich would be a handy thing to encapsulate for the purposes of manual hacking23:17
mordredand the fun part is - since dib is in shell - making the other two additional shell programs would not make disk-image-create any  more complex :)23:18
mordredclarkb: maybe I'll make a patch on top of your patch which does that23:18
clarkbI just have zero interest debugging thei ramdisk element :)23:21
*** ChuckC has quit IRC23:22
clarkbwill talk to greghaynes about it23:24
*** dmsimard_away is now known as dmsimard23:29
*** Sukhdev has quit IRC23:31
*** dimsum_ has joined #openstack-infra23:44
*** mmedvede has quit IRC23:44
*** mmaglana has quit IRC23:46
clarkbmordred: ok greghaynes has educated me a bit I think I know how to debug this now23:48
mordredclarkb: woot23:48
*** dimsum_ has quit IRC23:49
*** dmsimard is now known as dmsimard_away23:49
*** adalbas has quit IRC23:53
* jeblair wonders if greghaynes transferred the knowledge to clarkb via words or beer23:55
clarkbjeblair: both :)23:55
greghaynesjeblair: Howd you guess!?23:55
clarkbbeer and cider23:55
jeblairgreghaynes: it seemed particularly efficient :)23:55
mordredstill don't like vagrant23:56
mordredI really wish it worked23:56
* mordred just tried to use sdague's devstack-vagrant23:57
mordredgot this23:57
mordredwhich is annoying because there IS no vm setting in any of the config files I'm aware of23:58
*** rfolco has quit IRC23:58
clarkbmordred: jeblair: latest patchset of my dib patch has a bit more -x to track what is happening wiht traps23:59
clarkbbut greghaynes thinks there is a bug around exit codes and says he is going to fix it23:59

Generated by 2.14.0 by Marius Gedminas - find it at!