Monday, 2017-06-26

*** s-shiono has joined #openstack-infra00:04
*** sflanigan has joined #openstack-infra00:05
*** salv-orlando has joined #openstack-infra00:07
*** gmann has quit IRC00:07
*** gmann has joined #openstack-infra00:07
*** salv-orlando has quit IRC00:12
*** felipemonteiro__ has quit IRC00:12
*** tsuzuki has joined #openstack-infra00:13
*** kzaitsev_mb has joined #openstack-infra00:18
*** kzaitsev_mb has quit IRC00:22
*** bobh has joined #openstack-infra00:39
*** jkilpatr has quit IRC00:40
*** zhurong has joined #openstack-infra00:45
*** LindaWang has joined #openstack-infra00:48
*** tuanluong has joined #openstack-infra00:48
openstackgerritMerged openstack-infra/project-config master: Move dib image download jobs to -nv job  https://review.openstack.org/47673700:55
openstackgerritMerged openstack-infra/project-config master: Revert "Pause opensuse image builds"  https://review.openstack.org/47677600:59
*** mriedem has joined #openstack-infra01:02
*** salv-orlando has joined #openstack-infra01:08
*** gildub has quit IRC01:09
*** liujiong has joined #openstack-infra01:11
*** salv-orlando has quit IRC01:13
*** xinliang_ has quit IRC01:17
*** kzaitsev_mb has joined #openstack-infra01:18
*** cshastri has joined #openstack-infra01:20
openstackgerritIan Wienand proposed openstack-infra/project-config master: Check for newlines in project descriptions  https://review.openstack.org/47726101:20
*** gildub has joined #openstack-infra01:21
*** liusheng has quit IRC01:23
*** xinliang_ has joined #openstack-infra01:30
*** gcb has joined #openstack-infra01:32
*** hongbin has joined #openstack-infra01:33
*** MatthewLi has joined #openstack-infra01:33
*** cuongnv has joined #openstack-infra01:37
*** apetrich has quit IRC01:42
*** apetrich has joined #openstack-infra01:42
*** wolverineav has joined #openstack-infra01:46
*** bobh has quit IRC01:48
*** wolverineav has quit IRC01:50
*** zhurong has quit IRC02:01
*** salv-orlando has joined #openstack-infra02:09
*** salv-orlando has quit IRC02:14
*** gouthamr has joined #openstack-infra02:15
*** liusheng has joined #openstack-infra02:24
*** sbezverk has joined #openstack-infra02:34
openstackgerritMerged openstack/diskimage-builder master: Fix mkfs use wrong label option for vfat  https://review.openstack.org/47638202:41
*** zhurong has joined #openstack-infra02:44
*** bobh has joined #openstack-infra02:46
*** yee37924 has joined #openstack-infra02:56
*** wolverineav has joined #openstack-infra02:56
*** zhurong has quit IRC02:56
*** yee379 has quit IRC02:59
*** wolverineav has quit IRC03:00
*** armax has quit IRC03:04
*** armax has joined #openstack-infra03:04
*** tphummel has joined #openstack-infra03:07
*** zhurong has joined #openstack-infra03:07
*** bobh has quit IRC03:13
*** gcb has quit IRC03:14
*** armax has quit IRC03:14
*** gcb has joined #openstack-infra03:15
*** gouthamr has quit IRC03:22
*** zhurong has quit IRC03:23
*** zhurong has joined #openstack-infra03:35
ianwclarkb: dib release, suse images rebuilt & uploaded, seem to boot fine and get network and valid unbound.conf.  so i'm calling that done03:40
*** spzala has quit IRC03:45
*** spzala has joined #openstack-infra03:47
*** spzala has quit IRC03:52
*** kornicameister has joined #openstack-infra03:54
*** yamamoto has joined #openstack-infra03:54
openstackgerritMerged openstack-infra/project-config master: networking-odl: update graphana  https://review.openstack.org/47707603:54
openstackgerritMerged openstack-infra/project-config master: tripleo: move scenario004-multinode-oooq-container to gate  https://review.openstack.org/47701803:55
*** wolverineav has joined #openstack-infra03:57
*** yamahata has joined #openstack-infra04:00
*** wolverineav has quit IRC04:01
*** amotoki_away is now known as amotoki04:02
*** aeng has quit IRC04:03
openstackgerritTony Breeds proposed openstack/diskimage-builder master: Support for Cloud Images on ppc64le for rhel7 and centos7  https://review.openstack.org/45605504:08
openstackgerritTony Breeds proposed openstack/diskimage-builder master: [doc] Add some notes about PowerPC ARchitecture names  https://review.openstack.org/47735904:08
openstackgerritTony Breeds proposed openstack/diskimage-builder master: As far as block-device layout is concerned ppc64le == ppc64el  https://review.openstack.org/47736004:08
*** wolverineav has joined #openstack-infra04:08
*** apetrich has quit IRC04:10
*** salv-orlando has joined #openstack-infra04:10
*** hongbin has quit IRC04:11
*** kornicameister has quit IRC04:11
*** apetrich has joined #openstack-infra04:11
*** salv-orlando has quit IRC04:15
*** wolverineav has quit IRC04:18
*** aeng has joined #openstack-infra04:20
*** shu-mutou-AWAY is now known as shu-mutou04:21
*** adisky__ has joined #openstack-infra04:23
*** wolverineav has joined #openstack-infra04:34
*** aeng has quit IRC04:34
*** wolverineav has quit IRC04:38
*** links has joined #openstack-infra04:40
*** kzaitsev_mb has quit IRC04:40
*** kornicameister has joined #openstack-infra04:41
*** wolverineav has joined #openstack-infra04:47
*** aeng has joined #openstack-infra04:47
*** wolverineav has quit IRC04:51
*** wolverineav has joined #openstack-infra04:55
*** ramishra has joined #openstack-infra04:58
*** wolverineav has quit IRC05:01
*** salv-orlando has joined #openstack-infra05:04
*** tphummel has quit IRC05:28
*** dhajare has joined #openstack-infra05:34
openstackgerritEric Kao proposed openstack-infra/project-config master: enable designate service in congress tests  https://review.openstack.org/47737905:37
*** MasterOfBugs has joined #openstack-infra05:40
openstackgerritMerged openstack-dev/pbr master: rearrange existing documentation using the new standard layout  https://review.openstack.org/47668405:43
*** psachin has joined #openstack-infra05:44
openstackgerritYAMAMOTO Takashi proposed openstack-infra/project-config master: networking-midonet: Disable centos jobs on stable branches  https://review.openstack.org/47738205:47
*** wolverineav has joined #openstack-infra05:50
*** Douhet has joined #openstack-infra05:53
*** wolverineav has quit IRC05:55
openstackgerritIan Wienand proposed openstack-infra/system-config master: Remove pholio.openstack.org  https://review.openstack.org/47738605:55
openstackgerritTony Breeds proposed openstack/diskimage-builder master: Support for Cloud Images on ppc64le for rhel7 and centos7  https://review.openstack.org/45605505:58
openstackgerritTony Breeds proposed openstack/diskimage-builder master: [doc] Add some notes about PowerPC Architecture names  https://review.openstack.org/47735905:58
openstackgerritTony Breeds proposed openstack/diskimage-builder master: As far as block-device layout is concerned ppc64le == ppc64el  https://review.openstack.org/47736005:58
*** wolverineav has joined #openstack-infra05:58
*** rcernin has joined #openstack-infra05:59
*** AJaeger has quit IRC06:07
*** MasterOfBugs has quit IRC06:13
*** MasterOfBugs has joined #openstack-infra06:13
*** wolverineav has quit IRC06:15
openstackgerritMerged openstack-infra/project-config master: Readd "update descriptions of some projects I maintain"  https://review.openstack.org/47681406:17
*** salv-orlando has quit IRC06:19
*** wolverineav has joined #openstack-infra06:19
*** salv-orlando has joined #openstack-infra06:19
*** florianf has joined #openstack-infra06:21
*** tesseract has joined #openstack-infra06:24
*** salv-orlando has quit IRC06:24
*** wolverineav has quit IRC06:25
*** andreas_s has joined #openstack-infra06:27
*** salv-orlando has joined #openstack-infra06:28
*** salv-orl_ has joined #openstack-infra06:37
*** tushar has quit IRC06:38
*** markus_z has joined #openstack-infra06:38
*** tushar has joined #openstack-infra06:39
*** salv-orlando has quit IRC06:39
*** iyamahat has joined #openstack-infra06:41
*** yamahata has quit IRC06:43
*** pcaruana has joined #openstack-infra06:45
*** jpena|off is now known as jpena06:50
*** Hal has joined #openstack-infra06:52
*** shardy has joined #openstack-infra06:54
*** shardy has quit IRC06:54
*** iyamahat has quit IRC06:54
*** shardy has joined #openstack-infra07:07
openstackgerritArx Cruz proposed openstack-infra/system-config master: Adding tripleo check jobs into logstash  https://review.openstack.org/47742107:08
openstackgerritTony Breeds proposed openstack/diskimage-builder master: Support for Cloud Images on ppc64le for rhel7 and centos7  https://review.openstack.org/45605507:09
openstackgerritTony Breeds proposed openstack/diskimage-builder master: [doc] Add some notes about PowerPC Architecture names  https://review.openstack.org/47735907:09
openstackgerritTony Breeds proposed openstack/diskimage-builder master: As far as block-device layout is concerned ppc64le == ppc64el  https://review.openstack.org/47736007:09
*** e0ne has joined #openstack-infra07:15
*** jascott1 has quit IRC07:15
*** jascott1 has joined #openstack-infra07:16
openstackgerritDaniel Mellado proposed openstack-infra/project-config master: Disable Swift on Kuryr-K8s tempest gate  https://review.openstack.org/47742307:19
*** jascott1 has quit IRC07:20
*** ccamacho has joined #openstack-infra07:21
*** tushar has quit IRC07:27
*** gildub has quit IRC07:28
*** amoralej|off is now known as amoralej07:30
*** tushar has joined #openstack-infra07:30
*** Hal has quit IRC07:30
*** martinkopec has joined #openstack-infra07:33
*** jpich has joined #openstack-infra07:35
*** ralonsoh has joined #openstack-infra07:44
*** hashar has joined #openstack-infra07:48
*** Hal has joined #openstack-infra07:48
openstackgerritTuan Luong-Anh proposed openstack/diskimage-builder master: Replace new location of isolinux.bin for Ubuntu 16.04  https://review.openstack.org/47743707:54
openstackgerritTuan Luong-Anh proposed openstack/diskimage-builder master: Replace new location of isolinux.bin for Ubuntu 16.04  https://review.openstack.org/47743707:56
openstackgerritMerged openstack-dev/pbr master: Trivial: Fix docstring  https://review.openstack.org/47723807:56
*** rossella_ has joined #openstack-infra07:57
openstackgerritTuan Luong-Anh proposed openstack/diskimage-builder master: Replace new location of isolinux.bin for Ubuntu 16.04  https://review.openstack.org/47743707:57
*** pcaruana has quit IRC07:58
*** rcernin has quit IRC07:59
*** zzzeek has quit IRC08:00
*** zzzeek has joined #openstack-infra08:00
*** raissa has quit IRC08:03
*** raissa has joined #openstack-infra08:10
*** dhajare has quit IRC08:11
*** rcernin has joined #openstack-infra08:11
*** pcaruana has joined #openstack-infra08:12
*** rossella_ has quit IRC08:18
*** rossella_ has joined #openstack-infra08:20
*** junbo has quit IRC08:21
*** spzala has joined #openstack-infra08:21
openstackgerritChangBo Guo(gcb) proposed openstack-infra/project-config master: convert oslo.reports to the new unified doc build  https://review.openstack.org/47744408:22
*** junbo has joined #openstack-infra08:25
*** spzala has quit IRC08:26
*** s-shiono has quit IRC08:31
openstackgerritMarkos Chandras (hwoarang) proposed openstack-infra/project-config master: zuul: bifrost: Promote the openSUSE jobs to non-voting  https://review.openstack.org/47744808:35
*** ivyyy has quit IRC08:36
*** ivyyy has joined #openstack-infra08:37
*** yamamoto has quit IRC08:38
*** bauwser is now known as bauzas08:39
*** yamamoto has joined #openstack-infra08:39
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci master: Fix names of periodic jobs in promotion script  https://review.openstack.org/47745208:40
*** derekh has joined #openstack-infra08:42
*** dhajare has joined #openstack-infra08:43
*** tsuzuki has quit IRC08:43
*** psachin has quit IRC08:43
*** yamamoto has quit IRC08:45
*** MasterOfBugs has quit IRC08:47
*** rwsu has joined #openstack-infra08:47
*** dtantsur|afk is now known as dtantsur08:48
openstackgerritMerged openstack-infra/project-config master: enable designate service in congress tests  https://review.openstack.org/47737908:50
*** shen has joined #openstack-infra08:52
shenHi, I am trying to set up a CI environment, and get some troubles. I follow the guild 'https://docs.openstack.org/infra/openstackci/third_party_ci.html', and in the end of step 'start nodepool' I get a HTTP Error 405, following is the error log in /var/log/nodepool/nodepool.log: ``` 2017-06-22 10:53:08,712 INFO nodepool.NodeLauncher: Node id: 469 is ready 2017-06-22 10:53:08,889 ERROR nodepool.NodeLauncher: HTTPError launching08:52
openstackgerritMerged openstack-infra/project-config master: networking-midonet: Disable centos jobs on stable branches  https://review.openstack.org/47738208:52
openstackgerritMerged openstack-infra/project-config master: Check for newlines in project descriptions  https://review.openstack.org/47726108:53
openstackgerritMerged openstack-infra/project-config master: switch puppet-openstack-guide to unified docs build  https://review.openstack.org/47702808:53
openstackgerritMerged openstack-infra/project-config master: switch os-client-config to the new unified doc jobs  https://review.openstack.org/47667008:54
openstackgerritMerged openstack-infra/project-config master: convert pbr to the new unified documentation jobs  https://review.openstack.org/47668908:57
openstackgerritMerged openstack-infra/project-config master: ceilometer: do not install Gnocchi via devstack in master  https://review.openstack.org/46887608:57
openstackgerritMerged openstack-infra/project-config master: nova-lxd: Add py35 tests, refactor to use templates  https://review.openstack.org/47689008:57
openstackgerritSagi Shnaidman proposed openstack-infra/project-config master: Change time for periodic OVB TripleO CI jobs  https://review.openstack.org/47745608:57
sshnaidmamoralej, ^^08:57
amoralejsshnaidm, thx09:02
*** khappone has quit IRC09:03
*** yamamoto has joined #openstack-infra09:03
*** spzala has joined #openstack-infra09:05
*** khappone has joined #openstack-infra09:05
*** spzala has quit IRC09:09
*** hashar has quit IRC09:13
*** hashar has joined #openstack-infra09:14
*** tosky has joined #openstack-infra09:14
*** jascott1 has joined #openstack-infra09:17
*** electrofelix has joined #openstack-infra09:20
*** jascott1 has quit IRC09:21
electrofelixjamielennox: wonder if you've had a chance to look at https://review.openstack.org/475474 to fix a bug with the Mocker decorator applied to a method called multiple times.09:23
jamielennoxelectrofelix: i hadn't seen it no09:24
jamielennoxbut that seems like a likely problem09:25
electrofelixI took a relatively naive approach in looking at noticing how the class decorator used a copy09:26
*** rwsu has quit IRC09:26
jamielennoxelectrofelix: i don't really like using the class decorator so i haven't come across it yet09:27
jamielennoxelectrofelix: why take the value from self._adapter._case_sensitive rather than self.case_sensitive?09:27
*** rkukura has quit IRC09:27
jamielennoxoh, we don't update it from kwargs09:28
electrofelixjamielennox: because it looks like that is simply a default value used in case case_sensitive isn't passed through as a kwarg to the __init__09:28
electrofelixyep ;-)09:28
*** rkukura has joined #openstack-infra09:28
electrofelixI'm guessing that could be changed in the 2.x release?09:29
jamielennoxyea, so it's an annoying bug that has cause a number of problems - but i don't see why we couldn't update self.case_sensitive at that point09:29
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci master: Fix names of periodic jobs in promotion script  https://review.openstack.org/47745209:30
*** shu-mutou is now known as shu-mutou-AWAY09:30
electrofelixjamielennox: because that would change the value for all instances09:30
jamielennoxthe class value is there because (as mentioned in the comment) you can fix the issue for the whole project09:30
openstackgerritOpenStack Proposal Bot proposed openstack-infra/project-config master: Normalize projects.yaml  https://review.openstack.org/47746809:31
jamielennoxhowever if we do: self.case_sensitive = kwargs.pop('case_sensitive', self.case_sensitive) that shouldn't be an issue09:31
electrofelixjamielennox: would that not change the default behaviour for all subsequent class creations?09:31
jamielennoxthe instance variable will have priority over the class variable, but it doesn't actually change it for other classes09:31
electrofelixjamielennox: I'll have a look, didn't think I could safely shadow a class attribute09:33
jamielennoxelectrofelix: no, the way super() works in python it walks up the hierarchy and takes the first available, so when self.case_sensitive is defined class.case_sensitive is still set to the old value, just the instance taken instead09:33
jamielennoxelectrofelix: this is a really interesting topic that i'm not feeling elequent enough to explain, but there are some great blog posts09:34
electrofelixthat's good to know, guess I'm still learning details about python after many years using it ;-)09:34
electrofelixno, makes sense, I just didn't realise that was how it worked09:34
*** sambetts|afk is now known as sambetts09:35
jamielennoxelectrofelix: but in summary: http://paste.openstack.org/show/613685/09:35
jamielennoxdoh09:36
jamielennoxhttp://paste.openstack.org/show/613686/09:36
jamielennoxso i'd prefer to change that to update self.case_sensitive, and then use self.case_sensitive in the copy()09:37
*** rwsu has joined #openstack-infra09:42
*** jamesden_ has joined #openstack-infra09:46
*** rcernin has quit IRC09:46
*** jamesdenton has quit IRC09:47
*** aarefiev_afk is now known as aarefiev09:47
*** spzala has joined #openstack-infra09:48
*** spzala has quit IRC09:53
*** rossella_ has quit IRC09:54
*** stakeda has quit IRC09:56
*** salv-orl_ has quit IRC09:56
*** salv-orlando has joined #openstack-infra09:57
openstackgerritGergely Csatari proposed openstack-infra/project-config master: Adding Vitrage install guide  https://review.openstack.org/47178609:59
*** salv-orlando has quit IRC10:01
*** rcernin has joined #openstack-infra10:02
*** spzala has joined #openstack-infra10:04
*** kornicameister has quit IRC10:05
*** liujiong has quit IRC10:05
*** spzala has quit IRC10:08
*** rossella_ has joined #openstack-infra10:10
electrofelixjamielennox: done, thanks for the pointers10:16
*** yamamoto has quit IRC10:17
*** cshastri has quit IRC10:19
*** spzala has joined #openstack-infra10:19
*** kornicameister has joined #openstack-infra10:21
*** dizquierdo has joined #openstack-infra10:22
*** spzala has quit IRC10:24
*** jaosorior has joined #openstack-infra10:30
*** florianf has quit IRC10:36
*** LindaWang has quit IRC10:37
*** florianf has joined #openstack-infra10:41
*** mriedem has quit IRC10:41
*** Hal has quit IRC10:43
*** tosky has quit IRC10:49
*** ericyoung has joined #openstack-infra10:52
openstackgerritMerged openstack-infra/project-config master: Normalize projects.yaml  https://review.openstack.org/47746810:53
*** pcaruana has quit IRC10:53
*** LindaWang has joined #openstack-infra10:53
*** yamamoto has joined #openstack-infra10:59
*** rossella_ has quit IRC10:59
*** cuongnv has quit IRC11:01
openstackgerritMonty Taylor proposed openstack-infra/shade master: DNM Test patch to find bug with nics  https://review.openstack.org/47345211:02
*** pblaho has joined #openstack-infra11:02
electrofelixanyone around to review some fixes for git-review? https://review.openstack.org/389701 - fix output with python3, https://review.openstack.org/244083 - add topic field to output, https://review.openstack.org/398353 - refactor display code11:03
*** ccamacho is now known as ccamacho|lunch11:03
electrofelixoh and https://review.openstack.org/312056 - limit list reviews by branch if set11:03
*** pcaruana has joined #openstack-infra11:05
*** tosky has joined #openstack-infra11:06
*** jkilpatr has joined #openstack-infra11:06
openstackgerritMonty Taylor proposed openstack-infra/shade master: WIP Make revision_number go away for testing  https://review.openstack.org/47717511:07
*** rhallisey has joined #openstack-infra11:10
*** martinkopec has quit IRC11:11
*** martinkopec has joined #openstack-infra11:11
*** sdague has joined #openstack-infra11:12
*** gcb has quit IRC11:15
*** rossella_ has joined #openstack-infra11:15
mordredelectrofelix: all +2 from me11:16
mordredmaybe fungi will get a moment when he's up to check them out11:16
electrofelixmordred: thanks, I've been meaning to follow up on them for ages11:17
mordredelectrofelix: our stacks get rather deep don't they?11:20
*** AJaeger has joined #openstack-infra11:21
*** ericyoung has quit IRC11:22
electrofelixmordred: we're all just event driven processes with preemptive handing for critical bugs at heart11:23
*** shen has quit IRC11:24
*** ericyoung has joined #openstack-infra11:25
*** jaypipes has joined #openstack-infra11:26
*** slagle_ is now known as slagle11:30
*** alexchadin has joined #openstack-infra11:31
mordredelectrofelix: ++11:31
*** yamamoto has quit IRC11:37
*** pradk has quit IRC11:38
openstackgerritMerged openstack-infra/nodepool feature/zuulv3: Fix dict key copy operation  https://review.openstack.org/47690211:39
dnikishovAJaeger: hello, does https://review.openstack.org/#/c/475323/ look good? if not, who else I might need to get reviews from on it?11:43
*** ldnunes has joined #openstack-infra11:44
*** dizquierdo has quit IRC11:47
*** witek has joined #openstack-infra11:48
witekhello11:58
witekis it a strong requirement for every OpenStack project to have its own devstack plugin?11:58
*** sshnaidm is now known as sshnaidm|afk11:58
witekin Monasca project, the installation of the system is quite complex11:58
witekand maintaining the devstack plugin has always been a significant effort for our small team11:59
witekin the last months we have done a lot of work to install Monasca using Docker containers11:59
witekwould it be possible to use such environment instead of devstack to run tempest tests in OpenStack CI?11:59
openstackgerritMerged openstack-infra/project-config master: Add fuel-plugin-detach-netnode  https://review.openstack.org/47532312:00
*** dprince has joined #openstack-infra12:00
AJaegerwitek: better ask the QA team on #openstack-qa12:00
*** rossella_ has quit IRC12:01
AJaegerwitek: (on the requirement)12:01
arxcruzmtreinish: hey, around? Can you take a look at https://review.openstack.org/#/c/475788/ ?12:01
arxcruzthe stackviz on tripleo jobs are being overwritted because of the process_stackviz12:02
*** rcernin has quit IRC12:02
*** amoralej is now known as amoralej|lunch12:02
witekAJaeger: but technically it shouldn't be a problem, right?12:02
*** rcernin has joined #openstack-infra12:04
AJaegerwitek: something for others to answer - My expectation: Far too early for that - we start VMs for each test...12:05
openstackgerritAttila Darazs proposed openstack-infra/project-config master: tripleo: move containerized scenario002 to gate  https://review.openstack.org/47750512:05
* AJaeger does not know how exactly kolla and tripleo use container for testing12:05
witekAJaeger: thanks anyway12:06
dnikishovAJaeger: thanks!12:08
mtreinisharxcruz: is the problem that the tripleo jobs generate there own stackviz data?12:08
mtreinishit feels weird to me that tripleo would use devstack gate at all...12:08
arxcruzmtreinish: yes, that's the problem, and we use to call our internal scripts...12:12
arxcruzwould be nice to at least have the option to not call the process_stackviz12:12
openstackgerritAttila Darazs proposed openstack-infra/project-config master: tripleo: move containerized scenario002 to gate  https://review.openstack.org/47750512:13
*** yamamoto has joined #openstack-infra12:13
*** rossella_ has joined #openstack-infra12:13
*** jpena is now known as jpena|lunch12:14
*** ldnunes has quit IRC12:15
*** gordc has joined #openstack-infra12:17
*** tushar has quit IRC12:20
dnikishovAJaeger: can I also be added to fuel-plugin-detach-netnode-core and fuel-plugin-detach-netnode-release groups?12:21
*** rcernin has quit IRC12:21
*** rcernin has joined #openstack-infra12:21
*** Hal has joined #openstack-infra12:23
*** tushar has joined #openstack-infra12:23
*** ccamacho|lunch is now known as ccamacho12:25
*** ldnunes has joined #openstack-infra12:28
*** salv-orlando has joined #openstack-infra12:29
*** ralonsoh has quit IRC12:30
*** baoli has joined #openstack-infra12:31
*** rlandy has joined #openstack-infra12:31
openstackgerritMerged openstack-infra/project-config master: Disable Swift on Kuryr-K8s tempest gate  https://review.openstack.org/47742312:32
*** jcoufal has joined #openstack-infra12:32
*** salv-orlando has quit IRC12:33
*** rfolco has joined #openstack-infra12:34
*** baoli has quit IRC12:35
*** martinkopec has quit IRC12:36
*** trown|outtypewww is now known as trown12:40
*** bobh has joined #openstack-infra12:40
*** mdrabe has joined #openstack-infra12:41
*** jamesdenton has joined #openstack-infra12:42
*** jamesden_ has quit IRC12:43
openstackgerritTristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Abstract Nodepool request handling code  https://review.openstack.org/46874812:43
openstackgerritTristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Abstract Nodepool provider management code  https://review.openstack.org/46874912:43
openstackgerritTristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Collect request handling implementation in an OpenStack driver  https://review.openstack.org/46875012:43
openstackgerritTristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Extend Nodepool configuration syntax to support multiple drivers  https://review.openstack.org/46875112:43
openstackgerritTristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement a static driver for Nodepool  https://review.openstack.org/46862412:43
*** spzala has joined #openstack-infra12:48
*** bobh has quit IRC12:50
*** kgiusti has joined #openstack-infra12:50
*** tuanluong has quit IRC12:50
*** camunoz has joined #openstack-infra12:52
*** spzala has quit IRC12:53
*** jkilpatr has quit IRC12:54
*** hrybacki is now known as hrybacki|trainin12:55
*** edmondsw_ has joined #openstack-infra12:56
*** wolverineav has joined #openstack-infra12:57
*** edmondsw_ has quit IRC12:59
openstackgerritDavid Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Add tox py35 environment  https://review.openstack.org/47752512:59
*** rossella_ has quit IRC13:04
*** isaacb has joined #openstack-infra13:04
*** jkilpatr has joined #openstack-infra13:08
*** florianf has quit IRC13:16
*** gouthamr has joined #openstack-infra13:17
*** florianf has joined #openstack-infra13:18
*** amoralej|lunch is now known as amoralej13:19
*** jheroux has joined #openstack-infra13:19
openstackgerritGabriele Cerami proposed openstack-infra/tripleo-ci master: WIP: containers periodic test  https://review.openstack.org/47574713:19
*** ralonsoh has joined #openstack-infra13:21
*** ivyyy has quit IRC13:23
*** electrofelix has quit IRC13:23
*** jpena|lunch is now known as jpena13:24
*** rossella_ has joined #openstack-infra13:25
*** gouthamr has quit IRC13:26
*** esberglu has joined #openstack-infra13:26
*** felipemonteiro has joined #openstack-infra13:28
*** salv-orlando has joined #openstack-infra13:30
*** felipemonteiro_ has joined #openstack-infra13:30
*** edmondsw_ has joined #openstack-infra13:31
*** thorst has joined #openstack-infra13:31
*** yland has joined #openstack-infra13:32
*** edmondsw_ has quit IRC13:32
*** rossella_ has quit IRC13:32
*** eharney has joined #openstack-infra13:33
*** spzala has joined #openstack-infra13:34
*** salv-orlando has quit IRC13:34
openstackgerritGabriele Cerami proposed openstack-infra/tripleo-ci master: WIP: containers periodic test  https://review.openstack.org/47574713:34
*** felipemonteiro has quit IRC13:34
*** dhajare has quit IRC13:34
*** felipemonteiro_ has quit IRC13:37
*** felipemonteiro has joined #openstack-infra13:38
AJaegerdnikishov: I don't have those permissions13:38
AJaegerdnikishov: give full details (Reference link to change) and ask later again, hope somebody picks it up13:38
*** spzala has quit IRC13:39
*** ijw has joined #openstack-infra13:41
*** gcb has joined #openstack-infra13:42
*** spzala has joined #openstack-infra13:42
*** dhajare has joined #openstack-infra13:43
*** xyang1 has joined #openstack-infra13:43
openstackgerritDong Ma proposed openstack-infra/jenkins-job-builder master: Update dependency_check plugin  https://review.openstack.org/47753713:44
*** tphummel has joined #openstack-infra13:45
*** bobh has joined #openstack-infra13:45
*** rossella_ has joined #openstack-infra13:48
*** camunoz has quit IRC13:49
*** srwilkers has joined #openstack-infra13:50
*** superdan is now known as dansmith13:50
srwilkersclarkb, fungi: what do i need to do to get some cores added to openstack-helm-infra?13:51
*** chlong has joined #openstack-infra13:52
openstackgerritAttila Darazs proposed openstack-infra/tripleo-ci master: Add script parsing status multinode periodic jobs  https://review.openstack.org/47519713:53
AJaegersrwilkers: is the initial person added?13:55
AJaegerIf not, link to the change that merged and created them and ask for initial person to be added.13:55
AJaegersrwilkers: btw. too early for clarkb - and fungi is not around today. They might read backscroll13:55
srwilkersoh nice, thanks AJaeger. https://review.openstack.org/#/c/469079/, lamt would be the one that needs added13:57
AJaegerinfra-root, could somebody set up the ACLs, please ? ^13:57
*** links has quit IRC13:58
*** esberglu has quit IRC13:59
*** esberglu has joined #openstack-infra14:00
*** mriedem has joined #openstack-infra14:00
*** chlong has quit IRC14:01
jheskethAJaeger: I can take a look14:03
jheskethsrwilkers: excuse my ignorance, who is the ptl/should be the first member?14:04
*** esberglu has quit IRC14:04
srwilkersjhesketh: lamt (Tin Lam) should be added first i think, since he's the one who submitted the patchset to bring it up14:04
*** camunoz has joined #openstack-infra14:04
*** rbrndt has joined #openstack-infra14:05
*** edmondsw_ has joined #openstack-infra14:05
jheskethsrwilkers: yep, I've added him to both the core and infra groups14:05
*** salv-orlando has joined #openstack-infra14:05
srwilkersawesome, thank you :)14:05
jheskethno worries :-)14:06
*** edmondsw_ has quit IRC14:09
*** felipemonteiro_ has joined #openstack-infra14:09
*** felipemonteiro has quit IRC14:12
*** chlong has joined #openstack-infra14:14
*** Hal has quit IRC14:16
*** yland has quit IRC14:17
*** tphummel has quit IRC14:22
*** alexchadin has quit IRC14:22
*** hongbin has joined #openstack-infra14:24
openstackgerritCarlos Camacho proposed openstack-infra/tripleo-ci master: Enable the HA deployment by default  https://review.openstack.org/41517914:24
*** sshnaidm|afk is now known as sshnaidm14:24
sshnaidmyolanda, pabelanger can you please review in your time? "Set up new logstash index files for TripleO" https://review.openstack.org/#/c/476524/14:25
*** ijw has quit IRC14:29
*** ijw has joined #openstack-infra14:30
*** eeiden has quit IRC14:30
*** markmcclain has quit IRC14:32
*** ijw has quit IRC14:32
*** jtomasek is now known as jtomasek|afk14:32
*** esberglu has joined #openstack-infra14:32
*** ijw has joined #openstack-infra14:32
*** markmcclain has joined #openstack-infra14:33
*** rossella_ has quit IRC14:38
*** srwilkers has quit IRC14:39
*** nibalizer has quit IRC14:40
*** xarses has joined #openstack-infra14:41
*** yamahata has joined #openstack-infra14:41
*** trown is now known as trown|brb14:41
*** Guest16704 is now known as rook14:42
*** dizquierdo has joined #openstack-infra14:43
*** electrofelix has joined #openstack-infra14:44
*** trown|brb is now known as trown14:45
*** pradk has joined #openstack-infra14:47
*** pradk has quit IRC14:47
*** camunoz has quit IRC14:48
clarkbsshnaidm: how is progress on cleaning up the logs? looks like docs are still being built in every job and etc/ is still full of items (just looking at recent tripleo-common change that merged)14:50
sshnaidmclarkb, we'll have docs in periodic jobs, so we can remove them from gates: https://review.openstack.org/47548014:52
sshnaidmclarkb, about /etc folders in containers, I don't think we have something in pipeline when we reduced the size significantly last time14:52
sshnaidmclarkb, how much does it take now in comparison to other jobs?14:53
*** LindaWang has quit IRC14:53
*** rossella_ has joined #openstack-infra14:53
clarkbsshnaidm: I haven't run a check on disk usage yet but it was much lower last I checked. I'd just like to see the other items we identified for cleanup cleaned up (docs, etc/ etc)14:54
clarkbianw: thanks for getting suse images building again14:55
*** andreww has joined #openstack-infra14:56
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Add multinode jobs to promotion  https://review.openstack.org/47520014:56
*** felipemonteiro_ has quit IRC14:57
*** xarses_ has quit IRC14:58
*** mriedem has quit IRC14:58
*** Julien-zte has joined #openstack-infra15:00
*** camunoz has joined #openstack-infra15:00
*** xarses_ has joined #openstack-infra15:00
*** andreww has quit IRC15:00
*** rcernin has quit IRC15:04
*** mriedem has joined #openstack-infra15:04
*** wolverineav has quit IRC15:06
*** wolverineav has joined #openstack-infra15:06
*** rbrndt has quit IRC15:07
*** rbrndt_ has joined #openstack-infra15:07
*** mriedem has quit IRC15:08
clarkbrally rally jobs hit http://www.openstack.org in a HttpRequests.check_random_request test15:10
*** Julien-zte has quit IRC15:12
*** rossella_ has quit IRC15:12
*** Julien-zte has joined #openstack-infra15:13
*** ijw has quit IRC15:16
*** wolverineav has quit IRC15:16
*** ijw has joined #openstack-infra15:16
*** marst has joined #openstack-infra15:17
*** tphummel has joined #openstack-infra15:17
*** links has joined #openstack-infra15:18
*** Julien-zte has quit IRC15:20
*** jascott1 has joined #openstack-infra15:20
*** aarefiev is now known as aarefiev_afk15:20
*** yamamoto has quit IRC15:21
*** yamahata has quit IRC15:21
clarkbhttp://tarballs.openstack.org/ci/gerrit/plugins/its-storyboard/its-storyboard-805f9ac.jar and http://tarballs.openstack.org/ci/gerrit/plugins/javamelody/javamelody-v2.13.3.e4233d6.jar built implying those build are working \o/ the version on melody file won't work with our current puppetry to install gerrit plugins though. Will have to sort that out15:22
clarkbfungi: we are very close to being able to try another review-dev upgrade ^15:22
*** Julien-zte has joined #openstack-infra15:22
*** jascott1 has quit IRC15:24
*** yamamoto has joined #openstack-infra15:24
*** links has quit IRC15:25
*** felipemonteiro has joined #openstack-infra15:25
*** rossella_ has joined #openstack-infra15:25
*** ramishra has quit IRC15:27
*** felipemonteiro_ has joined #openstack-infra15:27
*** chlong has quit IRC15:27
openstackgerritGabriele Cerami proposed openstack-infra/tripleo-ci master: WIP: containers periodic test  https://review.openstack.org/47574715:29
*** yamamoto has quit IRC15:29
*** felipemonteiro has quit IRC15:30
*** e0ne has quit IRC15:31
*** ijw has quit IRC15:34
*** andreas_s has quit IRC15:35
*** dizquierdo has quit IRC15:39
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Add configuration documentation  https://review.openstack.org/46332815:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Reorganize docs into user/admin guide  https://review.openstack.org/47592815:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use oslosphinx theme  https://review.openstack.org/47758515:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move status_expiry to webapp section  https://review.openstack.org/47758615:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move tenant_config option to scheduler section  https://review.openstack.org/47758715:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct sample zuul.conf  https://review.openstack.org/47758815:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use executor section of zuul.conf for executor dirs  https://review.openstack.org/47758915:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use scheduler-specific log config and pidfile  https://review.openstack.org/47759015:40
*** Julien-zte has quit IRC15:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move zookeeper_hosts to zookeeper section  https://review.openstack.org/47759115:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Add some information about canonical_hostname  https://review.openstack.org/47759215:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix some inconsistent indentation in docs  https://review.openstack.org/47759315:40
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Rename git_host to server in github driver  https://review.openstack.org/47759415:40
*** chlong has joined #openstack-infra15:41
*** Julien-zte has joined #openstack-infra15:41
*** Julien-zte has quit IRC15:41
*** gcb has quit IRC15:41
*** Julien-zte has joined #openstack-infra15:41
openstackgerritDaniel Mellado proposed openstack-infra/project-config master: Use export tempest plugins  https://review.openstack.org/47759515:47
*** tosky has quit IRC15:49
*** Julien-zte has quit IRC15:50
*** markus_z has quit IRC15:50
*** Julien-zte has joined #openstack-infra15:50
*** rossella_ has quit IRC15:51
*** Julien-zte has quit IRC15:55
*** nich0s has joined #openstack-infra15:55
*** Julien-zte has joined #openstack-infra15:56
*** isaacb has quit IRC15:57
*** annegentle has joined #openstack-infra15:58
*** rossella_ has joined #openstack-infra16:00
*** rbrndt_ has quit IRC16:00
*** rbrndt_ has joined #openstack-infra16:00
*** rbrndt_ has quit IRC16:01
openstackgerritYaron de Leeuw proposed openstack-infra/jenkins-job-builder master: Add groovy-script options to extended parameters  https://review.openstack.org/47700316:01
*** pblaho has quit IRC16:02
inc0good morning everyone, so quick question about publishing to dockerhub. https://review.openstack.org/#/c/476989/3 when this merges and related changes to kolla merges, we'll tested tarball in tarballs.o.o16:03
*** pcaruana has quit IRC16:03
inc0so we will be able to run job to pull from these tarballs and upload to registries16:03
inc0which means access to secrets16:03
inc0how do we go about that?16:04
*** jamesmcarthur has joined #openstack-infra16:05
*** amotoki is now known as amotoki_away16:05
clarkbinc0: you have to set up a job that runs on a static test node with access to those secrets. This job can only run code within in the project-config repo to control access to those secrets16:05
*** ijw has joined #openstack-infra16:06
inc0so, static node which will run cronjob? how do I define it?16:06
*** iyamahat has joined #openstack-infra16:06
inc0and where do I write code to do execute?16:06
clarkbinc0: in openstack-infra/system-config/manifests/site.pp look at proposal.slave.openstack.org for examples on how we set up the static node and hand it secrets16:07
*** ijw has quit IRC16:08
clarkbinc0: for code to execute I think we have been keeping those scripts in project-config/jenkins/scripts, but if it is a small amout of code you might just put it in the jjb directly16:08
*** cody-somerville has joined #openstack-infra16:08
*** cody-somerville has quit IRC16:08
*** cody-somerville has joined #openstack-infra16:08
inc0well, I wrote some python to do the job16:08
clarkbthe project-config/jenkins/scripts should be fine16:08
*** Julien-zte has quit IRC16:09
inc0ok I'll clean it up and make single workflow, then get back to you16:09
inc0one thing tho, this process will require docker deamon running16:09
*** Apoorva has joined #openstack-infra16:10
*** gyee has joined #openstack-infra16:11
*** tosky has joined #openstack-infra16:11
*** jpich has quit IRC16:14
*** pcaruana has joined #openstack-infra16:14
*** isaacb has joined #openstack-infra16:15
*** pblaho has joined #openstack-infra16:16
*** jtomasek|afk is now known as jtomasek16:24
*** jpena is now known as jpena|off16:24
openstackgerritDaniel Mellado proposed openstack-infra/project-config master: Use export tempest plugins  https://review.openstack.org/47759516:26
*** yamamoto has joined #openstack-infra16:27
*** yamamoto has quit IRC16:32
clarkbinfra-root can we get a second review on https://review.openstack.org/#/c/476929/ ? expected to correct the problem with cgit config we ran into thursday/friday16:34
jeblairclarkb: o/16:38
clarkbmorning16:40
clarkbpabelanger: https://review.openstack.org/#/c/476289/1 is one that may interest you. Poking at alternatives to the afs cache across atlantic problem16:41
*** iyamahat has quit IRC16:41
jeblairclarkb, sdague: did anyone suggest ">" to accomplish multiline descriptions in the yaml file instead of "|" ?16:41
clarkbjeblair: no, I think the concern is more that | is valid and if not caught in review it breaks cgit16:42
clarkbso we should handle it gracefully16:42
jeblairclarkb: oh yeah, i'm down with that.  :)16:42
sdaguejeblair: it was a question I had, however the safety concern seemed valid there16:42
jeblairclarkb: just *also* saying that > probably aligns with what was intended16:42
sdagueso that we can't break cgit by accident16:42
jeblairwe should definitely make it so we can't break by accident :)16:43
jeblair"aligns"  see what i did there? :)16:43
sdaguejeblair: yeh, it would, though unfortunately the mass reformatter that processes that yaml script actually doesn't preserve any of this16:43
jeblairsdague: oh neat.  :)16:44
*** dizquierdo has joined #openstack-infra16:44
*** rossella_ has quit IRC16:47
*** SumitNaiksatam has quit IRC16:47
*** iyamahat has joined #openstack-infra16:47
openstackgerritMerged openstack-infra/jeepyb master: strip out newlines when writing repo.desc  https://review.openstack.org/47692916:48
*** iyamahat has quit IRC16:48
clarkblogs.o.o utilization seems under control now. Trimming job size and then reducing retention seems to have worked (for now)16:49
*** iyamahat has joined #openstack-infra16:49
*** tosky has quit IRC16:49
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Fix undercloud-upgrade job for Newton to Ocata  https://review.openstack.org/47760616:49
*** annegentle has quit IRC16:55
*** dhajare has quit IRC16:55
*** jascott1 has joined #openstack-infra16:56
*** iyamahat has quit IRC16:57
*** dprince has quit IRC16:58
*** derekh has quit IRC16:59
*** dprince has joined #openstack-infra17:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix some inconsistent indentation in docs  https://review.openstack.org/47759317:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Add some information about canonical_hostname  https://review.openstack.org/47759217:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Rename git_host to server in github driver  https://review.openstack.org/47759417:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move tenant_config option to scheduler section  https://review.openstack.org/47758717:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use executor section of zuul.conf for executor dirs  https://review.openstack.org/47758917:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct sample zuul.conf  https://review.openstack.org/47758817:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move zookeeper_hosts to zookeeper section  https://review.openstack.org/47759117:00
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use scheduler-specific log config and pidfile  https://review.openstack.org/47759017:00
jeblairclarkb: pabelanger is out for 2 weeks17:01
*** chlong has quit IRC17:02
openstackgerritBrianna Poulos proposed openstack-infra/project-config master: Change Barbican Tempest Plugin Gate to Voting  https://review.openstack.org/47282117:02
*** pcaruana has quit IRC17:03
*** isaacb has quit IRC17:04
*** kzaitsev_mb has joined #openstack-infra17:04
*** trown is now known as trown|lunch17:05
jeblairclarkb: re 476289, don't you think if we're going to switch to http caching, we should take the opportunity to stop using afs?17:09
*** pblaho has quit IRC17:10
*** dizquierdo has quit IRC17:11
clarkbjeblair: I'm not sure. AFS gives us a few useful features, particularly with the distro repos. We can avoid publishing mirrors until they are ready very easily, people can anonymously browse them locally, and we get redundant volumes17:11
*** dizquierdo has joined #openstack-infra17:11
clarkbjeblair: I mostly haven't gotten that far bceuase it is easy to test the reverse caching proxy against the afs mirrors whereas deciding to not use afs would require a lot more implementation and thinking17:11
jeblairclarkb: okay; i worry that 2 layers of caching is unecessarily complex....17:13
*** chlong has joined #openstack-infra17:14
clarkbto start I just want to see if this is viable for speeding up jobs under real conditions. And for that pointing at mirror.dfw. shouldn't be a big difference to pointing at something with the data on eg ext417:14
*** Swami has joined #openstack-infra17:14
jeblairclarkb: i'm happy to +2 that change for experimentation... i'm just not sure i agree that if it works well, that should be the end state rather than just apache->pypi.  but if you want to proceed with what's there and kick that coversation down the road for when it matters, i'm happy to do so.17:14
clarkbjeblair: well the problem as ianw points out is less pypi and more ubunut/debina/fedora/centos etc17:15
clarkband those all really need the checkpointed releases17:15
jeblairclarkb: yeah (though -- people keep saying that maybe got better?  is it worth verifying we still need it?)17:15
clarkbjeblair: it gets better on >= xenial if you enable fancy apt features. We might want to just consider trusty slightly less reliable and enable those features17:16
clarkbnot sure about fedora/centos17:16
jeblairclarkb: another thing to consider, if we do want to keep afs, is to shard volumes more.17:16
jeblairclarkb: afs can handle hundreds of thousands of volumes easily.  if we sharded the pypi mirror into *many* volumes, then each bandersnatch run may result in only a small number which need to be released.17:17
jeblairclarkb: however, that's somewhat counter to the goal of atomic-release, so we have to be careful there.17:17
*** SumitNaiksatam has joined #openstack-infra17:17
jeblair(perhaps a process like: release all of the package volumes, then release the index volume would be sufficient)17:18
*** _bmjen is now known as bmjen17:18
clarkbya, we could also run cache warmers after release to reduce the chance a job has to warm the cache itself17:18
jeblairclarkb: or, of course, your suggestion of warm..17:18
jeblairyeah that :)17:19
clarkbI think we can test ^ at the same time as testing the reverse caching proxy stuff17:19
clarkbsince dfw mirror cold vs warm cache difference should be minimal if any17:19
jeblairclarkb: ++17:20
jeblairclarkb: i just really want to avoid the day where we have to figure out whether we got an old package because apache was out of sync, or afs was, or....:/17:20
*** pblaho has joined #openstack-infra17:23
clarkbjeblair: ya, though my understanding of the afs cache recently has been that you should never get an out of date file from afs because anytime you release (even on a noop) the afs client is forced to go check if it needs a new version17:23
clarkbgranted there may be bugs or that understanding could be incomplete/wrong17:24
*** jamesmcarthur has quit IRC17:28
openstackgerritGabriele Cerami proposed openstack-infra/tripleo-ci master: WIP: containers periodic test  https://review.openstack.org/47574717:34
*** jcoufal has quit IRC17:35
*** ZZelle has joined #openstack-infra17:39
*** jcoufal has joined #openstack-infra17:39
*** HeOS has quit IRC17:40
*** ralonsoh has quit IRC17:41
*** Hal has joined #openstack-infra17:42
*** shardy is now known as shardy_afk17:44
openstackgerritMark Hamzy proposed openstack/diskimage-builder master: Support CentOS7 in 01-override-yum-arch  https://review.openstack.org/47761617:46
*** dizquierdo has quit IRC17:48
*** felipemonteiro_ has quit IRC17:49
*** felipemonteiro has joined #openstack-infra17:49
*** felipemonteiro_ has joined #openstack-infra17:50
electrofelixanyone to review some changes for git-review?17:52
electrofelixhttps://review.openstack.org/389701 - string format fix for py3, https://review.openstack.org/244083 - adds topic field if using '-ll', https://review.openstack.org/398353 - refactor of review display code, https://review.openstack.org/312056 - list reviews for given branch only17:52
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move tenant_config option to scheduler section  https://review.openstack.org/47758717:52
electrofelixhopefully they all look good and I'll check in later in case someone had a question on them17:53
*** electrofelix has quit IRC17:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix some inconsistent indentation in docs  https://review.openstack.org/47759317:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Add some information about canonical_hostname  https://review.openstack.org/47759217:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Rename git_host to server in github driver  https://review.openstack.org/47759417:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use executor section of zuul.conf for executor dirs  https://review.openstack.org/47758917:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct sample zuul.conf  https://review.openstack.org/47758817:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move zookeeper_hosts to zookeeper section  https://review.openstack.org/47759117:53
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use scheduler-specific log config and pidfile  https://review.openstack.org/47759017:53
*** felipemonteiro has quit IRC17:53
*** Rockyg has joined #openstack-infra18:00
*** Apoorva_ has joined #openstack-infra18:01
*** Apoorva has quit IRC18:04
*** dtantsur is now known as dtantsur|afk18:07
*** salv-orlando has quit IRC18:07
*** salv-orlando has joined #openstack-infra18:08
*** kzaitsev_mb has quit IRC18:10
*** salv-orlando has quit IRC18:12
*** amoralej is now known as amoralej|off18:13
*** tosky has joined #openstack-infra18:16
*** camunoz has quit IRC18:19
*** trown|lunch is now known as trown18:22
*** felipemonteiro_ has quit IRC18:23
clarkbactually java melody can be installed with existing puppet we just have to pass that entier version string to it in puppet rather than just the short sha18:24
clarkbI think that means we have have all of the artifacts necessary to attempt an upgrade to 2.13.8 now. The only remaining question is which changes do we want to backport to openstack/2.13.8 maybe all of them and rather than base on tag we base on tip of branch?18:24
openstackgerritMerged openstack-infra/tripleo-ci master: Fix names of periodic jobs in promotion script  https://review.openstack.org/47745218:25
clarkbodyssey4me: are you around? did you catch my comments about possible reason for dns weirdness last week?18:26
*** felipemonteiro has joined #openstack-infra18:28
*** florianf has quit IRC18:29
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use executor section of zuul.conf for executor dirs  https://review.openstack.org/47758918:29
*** felipemonteiro_ has joined #openstack-infra18:29
*** JudeC has joined #openstack-infra18:29
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Use scheduler-specific log config and pidfile  https://review.openstack.org/47759018:30
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix some inconsistent indentation in docs  https://review.openstack.org/47759318:31
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Add some information about canonical_hostname  https://review.openstack.org/47759218:31
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Rename git_host to server in github driver  https://review.openstack.org/47759418:31
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Move zookeeper_hosts to zookeeper section  https://review.openstack.org/47759118:31
*** felipemonteiro has quit IRC18:33
*** kzaitsev_mb has joined #openstack-infra18:33
jeblairclarkb: i missed the reason; what was it?18:35
*** markmcclain has quit IRC18:35
clarkbjeblair: lxc runs dnsmasq for both dhcp and dns for the containers it starts. The openstack-ansible jobs also disable all iptables firwall rules on the test nodes. My theory was that we were seeing rogue dns and dhcp servers all fighting each other18:36
clarkbjeblair: in part because on osic the IP addr that it binds to is within the 10/8 range assigned to our cloud ipv4 interface18:36
clarkbjeblair: lxc uses 10.0.3.1/24 which is routable in osic (unfortunately)18:37
clarkbjeblair: on the flip side our default iptables rules prevent dnsmasq from working with lxc proeprly so some editing to end up in the middle likely necessary18:37
*** annegentle has joined #openstack-infra18:38
*** markmcclain has joined #openstack-infra18:38
*** kzaitsev_mb has quit IRC18:38
cloudnullclarkb: jeblair: are you all running lxc things? in OSA we run lxc w/ dnsmaq maybe maybe you could pull from there18:43
*** kzaitsev_mb has joined #openstack-infra18:44
cloudnullhttps://github.com/openstack/openstack-ansible-tests/blob/master/test-vars.yml#L59-L7218:44
clarkbcloudnull: ya its yall running lxc that is the problem :) your tests blow up on dns resolution18:44
clarkbcloudnull: well some subset of them do aiui18:44
cloudnullis that still happening ?18:45
clarkbcloudnull: and I think it is possibly because dnsmasq is using an ip range for containers and dnsmasq that is within the ipv4 range assigned by osic18:45
cloudnulli thought that was fixed recently18:45
cloudnullodyssey4me: ^18:45
clarkbcloudnull: I think odyssey4me worked around it but not sure if it got fixed?18:45
odyssey4mewe worked around it by hard coding the troublesome names into /etc/hosts18:46
cloudnullah.18:46
openstackgerritJames E. Blair proposed openstack-infra/zuul feature/zuulv3: Make sure playbook and role repos are updated  https://review.openstack.org/47762718:46
odyssey4mehttps://review.openstack.org/#/q/topic:lnxcnt-org-workaround18:46
odyssey4mewe were just hitting too many failures and couldn't merge patches reliably for a week18:47
odyssey4meso that's there for a workaround18:47
odyssey4meyou think it might be an IP conflict? we can do an audit for that - I thought we had resolved it but it's possible there are stragglers18:48
odyssey4meespecially in the older branches18:48
clarkbodyssey4me: well 10/8 is the ip range we get on every ipv4 interface in osic and 10.0.3.1/24 is what dnsmasq uses18:48
*** rbrndt_ has joined #openstack-infra18:48
*** salv-orlando has joined #openstack-infra18:49
clarkband containers can route to the outside world as a result too. I think the removal of the iptables rules is potentially causing that to be problematic18:49
clarkbgranted you have to relax the rules because otherwise dns/dhcp flat out doesn't work18:49
odyssey4meyeah, but we arranged exclusions - cloudnull do you still have access to check what those were?18:49
clarkbso ya I think it has to do with routing and firewalls just based on my local testing18:49
cloudnullodyssey4me: I can get them18:49
cloudnullwe use the following config in test18:50
cloudnullhttps://github.com/openstack/openstack-ansible-tests/blob/master/test-vars.yml#L59-L7218:50
cloudnulli'll go pull the exclusions in the subnet18:50
clarkbcloudnull: odyssey4me I would be curious what happens if you moved to something outside of 10/818:51
clarkbsince 10.100.100/24 is still within 10/8 the behavior between it and 10.0.3/24 should be roughly the same18:51
odyssey4meclarkb I'm just trying to find our reference range18:51
odyssey4mewe tried to use the same ranges as devstack where possible18:51
clarkbodyssey4me: the difference is the devstack run (actually neutron run) dnsmasq is firwalled off by neutron18:52
clarkbodyssey4me: whereas lxc is far more yolo18:52
odyssey4mewell, the 10.x range is actually a host-only interface too18:52
odyssey4mebut that will possibly still have issues if the host knows other routes18:52
clarkbodyssey4me: except it isn't because there is another interface with 10/8 and routing is enabled18:52
clarkb(Ideally osic would use a smaller range that we could just avoid in testing, but ya)18:53
odyssey4meah, we use 10.100.100.x/24 as the override default in tests18:53
odyssey4meyeah, maybe osic could do that now that there are less people using it18:53
odyssey4mewe did do a dhcp exclusion to block some parts of the range being used18:54
odyssey4menote that the dns issue we were seeing were not just osic though18:54
clarkbodyssey4me: yes I think you may have similar problems potentially anywhere else we share l218:55
clarkbodyssey4me: because no firewall18:55
clarkbthe firewall is pretty important to test isolation here and it is being removed18:55
odyssey4meyeah18:55
cloudnullodyssey4me: https://pasteboard.co/1UiAhvc46.png18:56
*** rbrndt_ has quit IRC18:57
*** adisky__ has quit IRC18:58
*** shardy_afk is now known as shardy18:59
*** chlong has quit IRC19:02
clarkbodyssey4me: cloudnull I think the problem is less liekly osic handing out a 10.100.100/24 addr to some VM and more that 10/8 is routable19:05
clarkband directly attached19:05
clarkbwhcih affects how arp will be used for example19:05
odyssey4meyeah, that's probable19:05
clarkbso having a dhcp exclusion is good it won't fix all the problems19:06
odyssey4meso I have a suggestion19:06
clarkb*potential problems19:06
odyssey4mecan we put a page up somewhere that defines a set of CIDR's for use in testing19:06
odyssey4meobviously use devstack as the reference, but we may wish to add a few more to it19:06
odyssey4mewe can then model our deployment on that set19:07
clarkbodyssey4me: this is another case where devstack-gate aims to address this problem. We define a range that is expected to be good with it19:07
clarkb(though it also falls into the /8)19:07
clarkbso this particular corner case may not be solved by it19:07
odyssey4mewe're wanting to move towards using devstack-gate to do the host bootstrapping, but that will take time and is more likely to happen next cycle than this19:07
clarkbthe problem with statically defining it in eg docs is that we add and remove clouds so this changes19:09
*** rbrndt has joined #openstack-infra19:09
clarkbsome clouds allow us to change it via custom defined networks otehrs don't19:09
clarkb(it is a hard problem and there hasn't been a great answer so far)19:10
odyssey4meyeah, I remember that being part of the previous discussion19:10
clarkbI think it would be good to update the firewall rules to only allow traffic from/to that range if it originates on the test host19:14
clarkbI think you may be able to do that simply by having a rule that only allows the traffic on the lxcbr0 interface19:15
odyssey4mehmm, but that doesn't help the 172.x interfaces which are bridges19:15
clarkbis that where neutron is attaching floating IPs?19:16
clarkbI think those are less problematic because they aren't bootstrapped by lxc19:16
clarkb(neutron properly firewalls stuff I think)19:16
*** chlong has joined #openstack-infra19:16
odyssey4mewe bridge the containers onto host bridges19:16
odyssey4methe lxcbr0 interface is only used for fetching things from the internet19:17
clarkband dns and dhcp19:17
*** sdague has quit IRC19:17
*** jcoufal has quit IRC19:17
odyssey4methe containers should be default routing to lxcbr0, then that uses dnsmasq on the host for dhcp and dns resolution19:18
clarkbodyssey4me: sort of19:18
odyssey4mednsmasq should just use the host's resolvers19:18
clarkbodyssey4me: you are confusing host vs container here when that is sort of orthogonal. The issue is more ip range and routes and firwall19:18
clarkbdnsmasq attaches itself to lxcbr0 with ip 10.0.3.1 by default. It runs in the hosts ip namespace but that doesn't actually matter too much19:19
clarkbwhat matters is can the packets originating from/headed to 10.0.3.0/24 get to their intended destination and not go anywhere else19:19
clarkbwhen the host VM wants to send packets from itself to a container say at 10.0.3.22/24 it will source them as originating from 10.0.3.1/24 because of the dnsmasq setup19:21
clarkbso I think you can firewall off traffic to 10.0.3.0/24 and only allow it to originate on lxcbr0 then allow established connections created outbound for internet connectivity19:21
clarkbthat should isolate you from other test envs out there19:22
clarkbyou still potentially run the risk of internet bound traffic going to the wrong desitnation due to a conflit but 10.0.3.1 is  going to be a consistent conflict19:23
*** nicosuave has joined #openstack-infra19:28
openstackgerritAmrith Kumar proposed openstack-infra/project-config master: promote all trove experimental scenario jobs to check as non-voting  https://review.openstack.org/47764419:28
clarkbthe problem with our default iptables rules is that it blocks the dnsmasq to unbound resolution it also prevents dhcp from happening to dnsmasq from lxcbr0 attached containers19:29
*** nich0s has quit IRC19:29
*** nicosuave has quit IRC19:29
clarkbso you do need to open things up, I'm just concerned that opening everything up completetly is what is causing the weirdness here19:29
*** nich0s has joined #openstack-infra19:29
*** salv-orlando has quit IRC19:29
*** nich0s_ has joined #openstack-infra19:30
openstackgerritDavid Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Add web-based console log streaming  https://review.openstack.org/46335319:30
*** salv-orlando has joined #openstack-infra19:30
*** ChanServ has quit IRC19:30
*** nich0s_ has left #openstack-infra19:30
*** Rockyg has quit IRC19:31
*** nich0s has quit IRC19:32
*** salv-orlando has quit IRC19:35
*** ChanServ has joined #openstack-infra19:35
*** card.freenode.net sets mode: +o ChanServ19:35
*** nich0s has joined #openstack-infra19:37
*** nich0s has quit IRC19:37
*** nich0s has joined #openstack-infra19:37
*** sdague has joined #openstack-infra19:37
clarkbanyways I *think* that if you were to use something out of 192.168.x.y/24 instead that would break the overlap with 10/8 and confirm if that is related19:38
clarkbI would expect the dns issue to go away in that case19:39
odyssey4meclarkb hmm, interesting - as a data note - we're only clearing the iptables rules in our integrated gate19:39
odyssey4mein our role tests we don't do that19:39
*** spzala has quit IRC19:39
odyssey4medo you have job names which are prolematic?19:39
clarkbodyssey4me: I should be able to get them from logstash again. Give me a few to look that up19:40
*** spzala has joined #openstack-infra19:40
*** spzala has quit IRC19:40
*** spzala has joined #openstack-infra19:40
odyssey4methat said, lxc's deployment and our container deployment does implement more rules which may be conflicting19:40
odyssey4memy logstash-foo is not good, I can't seem to get queries that work well for me19:41
clarkb gate-openstack-ansible-os_neutron-ansible-upgrade-ubuntu-xenial is one of them19:41
clarkbthe nova and glance upgrade tests also affected19:41
odyssey4meany particular branch?19:42
clarkblooks like master19:42
clarkb(message:"Temporary failure in name resolution." OR message:"Temporary failure resolving" OR message:"wget: unable to resolve" OR  message:"Could not resolve host") AND NOT message:"Could not resolve host: fake" AND NOT message:"somehost" AND NOT message:" epel.vpsnet.lt" AND tags:"console"19:42
clarkbthat is my query19:42
clarkbthen I extended that to 7 days19:42
clarkbtries to filter out known bad names like fake, somehost, and epel.vpsnet.lt (this one stopped resolving)19:43
odyssey4meI'm seeing a ton of keystone jobs, and yeah - seems osic is the worst for it19:46
*** yamahata has joined #openstack-infra19:47
*** Apoorva_ has quit IRC19:47
*** Apoorva has joined #openstack-infra19:48
*** nich0s has quit IRC19:48
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Fix undercloud-upgrade job for Newton to Ocata  https://review.openstack.org/47760619:51
clarkbwe are definitely still seeing some flakyness outside of osic though the osic stuff seems to have largely gone away by only using ipv6 resolvers there19:51
*** sshnaidm is now known as sshnaidm|afk19:52
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Fix undercloud-upgrade job for Newton to Ocata  https://review.openstack.org/47760619:52
*** xyang1 has quit IRC19:55
odyssey4meclarkb ok, we'll try to fit in some time to look at using devstack gate for host bootstrapping sooner rather than later... but for now we may have to hobble along19:55
clarkbodyssey4me: ya I'd mostly just be curious if it was happier with 192.168 range and removing the /etc/hosts workaround19:56
clarkbjust to confirm if ip range overlap is problematic19:56
*** e0ne has joined #openstack-infra19:57
*** spzala has quit IRC19:57
*** spzala has joined #openstack-infra19:58
odyssey4meclarkb hmm, well I could fire up a review for that which we can recheck ad nauseum19:58
*** xyang1 has joined #openstack-infra19:59
*** jkilpatr has quit IRC20:00
openstackgerritTobias Henkel proposed openstack-infra/zuul feature/zuulv3: Fix encrypt_secret for python3  https://review.openstack.org/47716820:01
*** marst has quit IRC20:01
*** marst has joined #openstack-infra20:01
*** spzala has quit IRC20:03
*** Odd_Bloke has quit IRC20:03
*** Odd_Bloke has joined #openstack-infra20:03
*** e0ne has quit IRC20:06
*** e0ne has joined #openstack-infra20:09
openstackgerritTobias Henkel proposed openstack-infra/zuul feature/zuulv3: Fix encrypt_secret for python3  https://review.openstack.org/47716820:10
*** ericyoung has quit IRC20:10
odyssey4meclarkb I'm tied up in a VPN at the moment and unfortunately can't git-review. I'll pick it up in the morning and push up some test patches.20:11
clarkbodyssey4me: ok, sounds good20:12
cmurphyclarkb: I have a devstack-gate patch for you when you have a minute https://review.openstack.org/#/c/47644020:13
*** jamesmcarthur has joined #openstack-infra20:14
clarkbcmurphy: I've got https://review.openstack.org/474293 to address that. But maybe I should just merge yours to avoid review conflict :)20:15
*** e0ne has quit IRC20:15
openstackgerritEric Kao proposed openstack-infra/project-config master: add designate-tempest-plugin in congress job  https://review.openstack.org/47766620:16
cmurphyclarkb: yours won't fix my problem20:16
clarkboh I see20:17
clarkbwe actually want a combo of both things I think20:17
clarkbcmurphy: if using horizon then include django auth else if using django auth then include horizon?20:17
clarkbs/else if/ or/20:17
cmurphyclarkb: yeah something like that20:18
*** yamahata has quit IRC20:18
*** tesseract has quit IRC20:21
*** jcoufal has joined #openstack-infra20:21
*** shardy has quit IRC20:22
*** jamesmcarthur has quit IRC20:23
*** jamesmcarthur has joined #openstack-infra20:23
clarkbcmurphy: do you want to squash my changes into your change and I can approve?20:26
clarkbor I can rebase on yours and approve your thing20:26
*** rderose has joined #openstack-infra20:27
clarkband wait for someone to review mine20:27
cmurphyclarkb: oh i just commented on yours, but yeah let me squash your changes into mine20:27
cmurphythat will be faster20:27
*** salv-orlando has joined #openstack-infra20:28
*** kgiusti has quit IRC20:29
*** annegentle has quit IRC20:33
*** felipemonteiro has joined #openstack-infra20:33
*** felipemonteiro_ has quit IRC20:34
*** tinwood has quit IRC20:35
*** tinwood has joined #openstack-infra20:36
*** rbrndt has quit IRC20:38
openstackgerritColleen Murphy proposed openstack-infra/devstack-gate master: Group all horizon-related projects together  https://review.openstack.org/47644020:38
cmurphyclarkb: ^20:38
openstackgerritMerged openstack-infra/zuul feature/zuulv3: Fix parsing of github PR url  https://review.openstack.org/47628620:38
*** felipemonteiro_ has joined #openstack-infra20:38
clarkbsdague: maybe you can review 466440 real quick? I'll probably approve it after lunch if no second reviewer takes a look20:40
*** felipemonteiro has quit IRC20:41
clarkbcmurphy: oh it exploded20:43
*** rbrndt has joined #openstack-infra20:43
clarkbcmurphy: need a space before the ]] I think20:43
cmurphyoops20:43
openstackgerritColleen Murphy proposed openstack-infra/devstack-gate master: Group all horizon-related projects together  https://review.openstack.org/47644020:44
clarkbyup tested locally that is fix20:44
sdagueclarkb: that's a tempest change?20:44
openstackgerritDoug Hellmann proposed openstack-dev/pbr master: switch from oslosphinx to openstackdocstheme  https://review.openstack.org/47668620:44
clarkbsdague: devstack-gate20:44
clarkboh I typed it out wrong20:45
sdagueurls are friends :)20:45
clarkbsdague: https://review.openstack.org/476440 that one20:45
clarkbsdague: thanks20:46
openstackgerritSamuel de Medeiros Queiroz proposed openstack-infra/shade master: Project update to change enabled only when provided  https://review.openstack.org/47691420:46
cmurphythanks guys20:46
*** jheroux has quit IRC20:49
*** lamt has joined #openstack-infra20:51
openstackgerritDoug Hellmann proposed openstack-dev/pbr master: switch from oslosphinx to openstackdocstheme  https://review.openstack.org/47668620:51
jeblairdhellmann: can you point me at a tldr on oslosphinx/openstackdocstheme if one exists?20:52
clarkbcmurphy: I have abandoned my old change20:53
dhellmannjeblair : do you mean how to migrate or why to migrate?20:53
jeblairdhellmann: why20:54
cmurphyclarkb: okie20:54
dhellmannjeblair: the docs team is going to maintain the theme to be consistent with the rest of openstack.org and the oslo team is going to deprecate oslosphinx20:54
jeblairdhellmann: differences between the two, etc.  audience would be a hypothetical person who suddenly realized openstackdocstheme existed.  :)20:54
openstackgerritTobias Henkel proposed openstack-infra/zuul feature/zuulv3: wip: support custom playbook locations  https://review.openstack.org/47767220:55
dhellmannaha20:55
dhellmannthe openstackdocstheme has some features for linking pages to bug trackers (though that only works with launchpad right now)20:55
dhellmannotherwise, they should have comparable features20:55
dhellmannwell, the openstackdocstheme doesn't force itself on you, so there's a little less magic20:55
dhellmannbasically, we learned better how to write themes and did that work in openstackdocstheme instead of oslosphinx20:56
jeblairdhellmann: ah cool20:56
jeblairi just wrote a change to switch zuul's docs to oslosphinx... i guess i'll write another one :)20:56
dhellmannannegentle is working on bringing some of the other features from oslosphinx into the docs theme, so we can stop maintaining multiple themes20:56
dhellmannsorry, we should probably do an announcement20:56
dhellmannwe sort of rolled it into the doc-migration spec, since we had to touch every repo at least once for that anyway20:56
jeblairdhellmann: i confess, i can't find openstackdocstheme's own docs after about 10 seconds of looking20:57
dhellmannhrm20:58
jeblair(i mean, i found the pypi entry, which has docs, but i'm sure there must be an entry on docs.openstack.org with it self-rendering...)20:58
dhellmannI wonder if there's a docs job20:58
*** annegentle has joined #openstack-infra21:02
dhellmannjeblair : it appears to have its own special job21:03
*** annegentle has quit IRC21:03
dhellmannoh, no, not quite21:03
*** jamesmcarthur has quit IRC21:03
dhellmannit has the check and gate jobs, but no post job21:04
*** chlong has quit IRC21:04
jeblairyay i helped find a bug!21:05
*** HeOS has joined #openstack-infra21:05
*** rockyg has joined #openstack-infra21:05
openstackgerritDoug Hellmann proposed openstack-infra/project-config master: add doc publishing job for openstackdocstheme  https://review.openstack.org/47767421:05
dhellmannjeblair : ^^21:05
jeblair+221:07
dhellmannty21:07
*** felipemonteiro_ has quit IRC21:08
*** wolverineav has joined #openstack-infra21:09
*** kevinbenton has quit IRC21:12
*** ldnunes has quit IRC21:15
*** kevinbenton has joined #openstack-infra21:15
openstackgerritMerged openstack-infra/zuul feature/zuulv3: Implement pipeline requirement of github labels  https://review.openstack.org/47396221:15
*** annegentle has joined #openstack-infra21:19
openstackgerritMerged openstack-infra/zuul feature/zuulv3: config: refactor config get default  https://review.openstack.org/47448421:20
*** dprince has quit IRC21:21
openstackgerritMerged openstack-infra/zuul feature/zuulv3: bubblewrap: adds --die-with-parent option  https://review.openstack.org/47316421:25
*** aeng_ has joined #openstack-infra21:25
openstackgerritMerged openstack-infra/zuul feature/zuulv3: executor: run trusted playbook in a bubblewrap  https://review.openstack.org/47446021:25
openstackgerritSlawek Kaplonski proposed openstack-infra/project-config master: Add Neutron fullstack python35 tests  https://review.openstack.org/47767921:28
*** spzala has joined #openstack-infra21:28
*** esberglu has quit IRC21:32
*** esberglu has joined #openstack-infra21:32
*** rbrndt has quit IRC21:33
*** esberglu has quit IRC21:37
*** spzala has quit IRC21:39
*** xyang1 has quit IRC21:41
*** thorst has quit IRC21:44
*** rbrndt has joined #openstack-infra21:45
*** thorst has joined #openstack-infra21:46
*** thorst has quit IRC21:50
clarkbas a followup on the dns things we have what appears to be a single dns failure in the last 6 hours that is valid21:51
clarkbhappened in http://logs.openstack.org/17/475617/4/gate/gate-kolla-dsvm-build-oraclelinux-source-centos-7/354126a/console.html so possibly a fail in container networking there too?21:52
clarkbinc0: just fyi ^21:53
openstackgerritMonty Taylor proposed openstack-infra/shade master: Deal with revision_number in neutron update_port tests  https://review.openstack.org/47717521:53
mordredclarkb: woot!21:53
*** kzaitsev_mb has quit IRC21:54
clarkb(this based on my logstash query which may be incomplete)21:55
*** esberglu has joined #openstack-infra21:57
openstackgerritMichael Johnson proposed openstack-infra/project-config master: Make neutron-lbaas tempest jobs voting  https://review.openstack.org/47768921:57
*** gordc has quit IRC21:58
*** esberglu has quit IRC21:58
*** esberglu has joined #openstack-infra21:58
openstackgerritSlawek Kaplonski proposed openstack-infra/project-config master: Add Neutron fullstack python35 tests  https://review.openstack.org/47767921:59
jeblairzuul meeting time in #openstack-meeting-alt22:00
*** jascott1 has quit IRC22:02
*** jascott1 has joined #openstack-infra22:03
*** thorst has joined #openstack-infra22:05
*** rockyg has quit IRC22:05
*** eharney has quit IRC22:06
*** jascott1 has quit IRC22:07
*** thorst has quit IRC22:09
*** rwsu has quit IRC22:11
openstackgerritMerged openstack-infra/jenkins-job-builder master: Update stash plugin  https://review.openstack.org/47718922:13
openstackgerritMerged openstack-infra/jenkins-job-builder master: Update dependency_check plugin  https://review.openstack.org/47753722:13
*** xarses has quit IRC22:13
openstackgerritMerged openstack-infra/jenkins-job-builder master: Update build_publisher plugin  https://review.openstack.org/47707722:14
*** bobh has quit IRC22:14
*** kzaitsev_mb has joined #openstack-infra22:15
openstackgerritMerged openstack-infra/jenkins-job-builder master: Update git plugin test cases  https://review.openstack.org/47692022:16
*** yamahata has joined #openstack-infra22:18
*** kzaitsev_mb has quit IRC22:20
*** mrhillsman is now known as codebauss22:22
*** ericyoung has joined #openstack-infra22:22
*** codebauss is now known as mrhillsman22:22
inc0clarkb: not sure, possibly22:30
*** xarses has joined #openstack-infra22:30
inc0if this keeps break, tomorrow EU time our local oraclelinux expert will wake up22:31
*** yamahata has quit IRC22:31
openstackgerritMichael Johnson proposed openstack-infra/project-config master: Make neutron-lbaas tempest jobs voting  https://review.openstack.org/47768922:31
*** marst has quit IRC22:32
openstackgerritClark Boylan proposed openstack-infra/devstack-gate master: Use 3/4 of CPUs by default instead of 1/2  https://review.openstack.org/47770122:33
clarkbinc0: I'm not sure its specifically an oraclelinux problem. More likely a generic routing/firewall/dns resolver problem22:34
*** marst has joined #openstack-infra22:35
inc0well docker will use this stuff from host22:35
inc0so if node vm can resolve it, so will container22:36
clarkbinc0: thats not true as we've discovered with lxc at least22:36
*** spzala has joined #openstack-infra22:36
clarkbbecause of firewalls routing and such22:36
inc0docker is different than lxc22:36
clarkbinc0: yes but its the same fundamental problem22:36
inc0yeah, but docker deamon will put these confs22:36
inc0into container22:36
clarkbif process X with ip foo wants to talk to resolver proxess Y with ip bar you have to get hte packets back and forth22:37
inc0also, other tests seems to pass22:37
clarkbif iptables says no, then too bad packets get dropped22:37
clarkbeven if you are running all on the same host22:37
clarkb(also I thought docker just hardcoded google dns, but I guess that changed)22:38
inc0https://review.openstack.org/#/c/454743/2 other gate jobs seems to work fine22:38
inc0clarkb: let me check for sure, but I'm almost certain it copies resolv.conf from host22:38
inc0I know it copies /etc/hosts22:38
inc0yup, it copies resolv too22:39
clarkbya at least with lxc things I think it could be a cloud interface ip range interaction so you wouldn't see it everywhere22:39
*** jascott1 has joined #openstack-infra22:40
clarkbinc0: so if our resolv.conf says nameserver 127.0.0.1 how do lookups in a container find the host's 127.0.0.1?22:40
clarkbinc0: are you running in the hosts network namespace?22:40
inc0not during build22:40
clarkb(our iamges all use 127.0.0.1 as the resolver)22:40
inc0is it going to say 127.0.0.1? That might break too22:40
inc0ahh22:40
inc0hmm22:40
*** spzala has quit IRC22:40
inc0yeah, that might be an issue22:41
inc0can we put regular IP to resolv too?22:41
clarkbthat is where the firewall becomes a problem because we don't allow packets to it22:41
clarkbon the public interface(S)22:41
*** hashar has quit IRC22:41
inc0https://review.openstack.org/#/c/454743/2 ubuntu and centos passed tho, but both of these uses mirrors in infra22:42
inc0oracle doesn't'22:42
inc0so that might be affecting this22:42
*** ericyoung has quit IRC22:44
* clarkb wonders how docker generally deals with localhost is resolver22:45
*** annegentle has quit IRC22:47
*** eeiden has joined #openstack-infra22:48
inc0probably it doesn't22:48
*** xarses has quit IRC22:48
*** jascott1 has quit IRC22:51
*** rbrndt has quit IRC22:51
*** rbrndt has joined #openstack-infra22:51
*** rbrndt has quit IRC22:51
*** wolverineav has quit IRC22:54
inc0clarkb: https://review.openstack.org/#/c/475617/ on the other hand these gates passed22:55
inc0and they were run after one that failed22:55
inc0so it's not consistent22:56
*** sdague has quit IRC22:56
clarkbinc0: were they on the same cloud? perhaps there is a cloud/region weirdness there22:56
clarkbor perhaps it legitimetely just had the lookup fail upstream of it22:56
inc0where can I check nodepool?22:57
*** tosky has quit IRC22:57
clarkbinc0: the beginning of each console log should record the host it ran on which includes the cloud and region22:58
inc0osic failed22:58
inc0seems like22:58
clarkbya osic is where we have the 10/8 interface22:59
clarkbwhich could potentially make routing and firwalling funny for other things using 10.whatever23:00
clarkbI wonder if this is happening there23:00
clarkbinc0: I suggested to osa folks that they try using 192.168.x.y/24 ranges instead of 10.x.y.z/24 ranges to see if that avoided the problem in osic23:02
clarkbthough in the docker case you aren't bridging directly on the host are you? so maybe thats less of an issue with docker23:03
inc0we don't give network addresses to our containers23:03
clarkbya that23:03
inc0that being said, docker indeed uses 10.* ips for container23:03
inc0maybe it's configurable23:03
*** annegentle has joined #openstack-infra23:08
cloudnullall of these dns failures all seem very odd. While I can't prove it, it semes like we're seeing this https://bbs.archlinux.org/viewtopic.php?id=20411923:09
cloudnullis systemd-resolved running on some of these nodes?23:10
clarkbcloudnull: no systemd-resolved is not enabled on ubuntu xenial23:11
clarkbit is enabled in 17.04 apparently though23:11
*** iyamahat has joined #openstack-infra23:11
cloudnullit looks like 17+ is shipping it by default23:11
cloudnulldo we have test images on 17 ?23:11
clarkbno23:11
clarkbwe stopped testing the non lts releases when they went to 9 months of support23:11
cloudnullmaybe some of these jobs are installing packages which bring in systemd-networkd/resolved ?23:12
clarkbcloudnull: also switching to ipv6 only forwarders for unbound fixed 99% of the problems looks like23:12
cloudnullwhich ubuntu then starts the service by default ?23:12
cloudnullok23:12
clarkbcloudnull: that is a possibility, maybe lxc/docker/etc pull that in?23:12
*** marst has quit IRC23:13
cloudnullmaybe ?23:13
*** iyamahat has quit IRC23:13
clarkbdevstack-gate dumps the dpkg package listing but I don't think osa or kolla do that in their logs23:13
* cloudnull spitballing23:13
inc0well we won't have resolved running in container23:14
inc0because well, we run single pid23:14
clarkbinc0: right but it may be running on the host if however you install docker says it needs it23:14
clarkbinc0: not sure that it does but it is a possibility23:14
cloudnullmaybe we need to have a larger conversation with projects to get folks using experemental class networks (class E) instead of the noraml class ABC?23:14
inc0also container images are normally cut deeply23:14
inc0from anything that can be useful23:15
inc0like ping23:15
clarkbcloudnull: well I think the major problem is the overlap with the class A23:15
clarkbcloudnull: if osic handed out say a /20 then it would be a non issue23:15
clarkbinc0: yes thats actually one of my major grupes with containers. You add a ton of complixty and remove all the tools necessary to debug that complexity :)23:16
clarkbparticularly when ti comes to networking23:16
inc0you trade one complexity to another23:16
cloudnullexcept for the places where folks overlaped with the /20 :)23:16
clarkbcloudnull: right but at least they now have the option to stop overlapping23:16
inc0fwiw, running kolla container is no issue because we run net=host23:16
clarkbcloudnull: with the 10/8 as is you don't really have that choice if using any of 10/823:16
cloudnullif we all used class E there'd never be any overlap with any provider23:17
cloudnullwe do have a few blocks we dont assign23:17
cloudnullwhich was made to accomodate devstack23:17
clarkbcloudnull: you don't assign ips out of them but you still overlap in the range23:17
clarkbcloudnull: they are two related problems but fixing one doesn't fix the other23:18
cloudnull++23:18
clarkbinc0: I've learned that because all these containers have python installed to run openstack that the most reliable way to test if dns is working is python import socket; socket.gethostbyname() because none of the other tools are present :)23:19
inc0yeah, that;) or running a container with all the tools23:19
clarkbinc0: but once you do that you are no longer replicating necessarily23:20
clarkbcloudnull: aiui some ip implementations nuke class E hard?23:20
inc0container images are glorified packages, but I hear ya23:20
clarkbinc0: but they aren't because of things like networkign :)23:20
*** kzaitsev_mb has joined #openstack-infra23:20
*** hongbin has quit IRC23:20
inc0depends how you use it23:20
cloudnullclarkb: that may be .23:20
inc0again, we run net=host23:21
cloudnullI've only done it locally .23:21
openstackgerritRosario Di Somma proposed openstack-infra/shade master: Don't remove top-container element for flavors and clusters  https://review.openstack.org/47770823:21
clarkbinc0: deb or rpm packages don't need a linux bridge with a dns server and dhcp server running in order to use networking23:21
cloudnullI suspect it may be more headache than it's worth23:21
cloudnull^ getting everyone on board with a experimental class23:21
inc0hence glorified23:21
clarkbcloudnull: ya I imagine we'll spend more time debugging weird corner cases related to that then just tracking down non overlapping addrs23:21
clarkbclarly the answer is to ipv6 harder23:22
clarkbcan anyone bug github about getting ipv6?23:22
cloudnullnet=host makes containers more like an over complicated chroot :)23:22
cloudnullI can poke them again23:23
*** rderose has quit IRC23:23
cloudnullI'm all for making more things ipv6 -all the things23:23
clarkbgithub seems to remain our biggest ipv4 target23:24
cloudnullnot that github will listen to little old me . but I'll keep at it23:24
clarkbbut there are others too. I blame consumer ISPs not embracing ipv623:26
clarkbmy isp won't do native ipv6 :(23:26
cloudnull:'(23:26
cloudnullI blame the ISPs too.23:26
cloudnullthey should simply start with IPv6, and then give IPv4 access to folks who need it (which they can pay more for).23:27
*** wolverineav has joined #openstack-infra23:29
clarkband since facebook has ipv6 AAAA records that will just work for the vast majority of people too :)23:29
cloudnullhahaha :)23:29
*** SumitNaiksatam has quit IRC23:32
*** tsuzuki has joined #openstack-infra23:32
*** tsuzuki_ has joined #openstack-infra23:33
openstackgerritRosario Di Somma proposed openstack-infra/shade master: Don't remove top-container element for flavors and clusters  https://review.openstack.org/47770823:34
*** s-shiono has joined #openstack-infra23:35
*** tsuzuki has quit IRC23:36
*** wolverineav has quit IRC23:38
*** wolverineav has joined #openstack-infra23:40
*** Swami has quit IRC23:41
*** wolverineav has quit IRC23:44
openstackgerritTony Breeds proposed openstack/diskimage-builder master: dib-lint: Ignore editor temp files for linting run  https://review.openstack.org/47771123:44
*** marst has joined #openstack-infra23:44
clarkbdirk: fyi the suse devstack-gate tempest jobs are failing. http://logs.openstack.org/40/476440/3/check/gate-tempest-dsvm-neutron-full-opensuse-422-nv/7e83bef/console.html appears to be related to libguestfs?23:47
*** wolverineav has joined #openstack-infra23:49
openstackgerritMerged openstack-infra/devstack-gate master: Group all horizon-related projects together  https://review.openstack.org/47644023:51
openstackgerritYolande Amate proposed openstack-infra/jenkins-job-builder master: Update hipchat to use convert_mapping_to_xml()  https://review.openstack.org/47722223:51
*** JudeC has quit IRC23:58
*** wolverineav has quit IRC23:58

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!