Monday, 2015-02-09

*** achanda has quit IRC00:00
*** achanda has joined #openstack-infra00:02
*** sputnik13 has joined #openstack-infra00:07
*** sputnik1_ has quit IRC00:08
*** yamahata has joined #openstack-infra00:10
*** garyh has joined #openstack-infra00:11
*** sputnik13 has quit IRC00:12
*** weshay has quit IRC00:19
*** garyh has quit IRC00:21
*** otter768 has quit IRC00:22
*** mriedem1 has joined #openstack-infra00:23
*** mriedem has quit IRC00:24
*** markvoelker has joined #openstack-infra00:25
*** markvoelker has quit IRC00:30
*** dimsum__ has joined #openstack-infra00:35
*** dims_ has joined #openstack-infra00:36
*** dimsum__ has quit IRC00:39
*** oomichi has joined #openstack-infra00:41
*** talluri has joined #openstack-infra00:44
*** talluri has quit IRC00:48
*** achanda has quit IRC00:52
*** garyh has joined #openstack-infra00:52
*** baoli has joined #openstack-infra00:58
*** garyh has quit IRC01:03
*** zhiwei has joined #openstack-infra01:06
*** garyh has joined #openstack-infra01:09
*** woodster_ has quit IRC01:10
*** otter768 has joined #openstack-infra01:15
*** nelsnelson has joined #openstack-infra01:16
*** sdake_ has quit IRC01:17
*** sdake_ has joined #openstack-infra01:17
*** garyh has quit IRC01:19
*** Ryan_Lane has quit IRC01:20
*** sarob has quit IRC01:26
*** markvoelker has joined #openstack-infra01:26
openstackgerritIan Wienand proposed openstack-infra/nodepool: Add generate-log-config command
*** markvoelker has quit IRC01:31
*** jraim has quit IRC01:38
*** ameade has quit IRC01:39
*** tmcpeak has joined #openstack-infra01:39
*** briancurtin has quit IRC01:40
*** ctracey has quit IRC01:40
*** zhiyan has quit IRC01:40
*** sdake_ has quit IRC01:44
*** sdake_ has joined #openstack-infra01:45
*** ctracey has joined #openstack-infra01:45
*** jraim has joined #openstack-infra01:45
*** tmcpeak has quit IRC01:46
*** ameade has joined #openstack-infra01:46
*** briancurtin has joined #openstack-infra01:46
*** zhiyan has joined #openstack-infra01:46
*** wuhg has joined #openstack-infra01:49
*** woodster_ has joined #openstack-infra01:49
*** achanda has joined #openstack-infra01:52
*** achanda has quit IRC01:57
*** nelsnelson has quit IRC02:01
*** ddieterly has quit IRC02:05
*** ddieterly has joined #openstack-infra02:09
*** yamamot__ has joined #openstack-infra02:11
*** ddieterl_ has joined #openstack-infra02:12
*** ddieterly has quit IRC02:13
*** asettle is now known as asettle-gym02:13
*** garyh has joined #openstack-infra02:13
*** nelsnelson has joined #openstack-infra02:15
*** nelsnelson has quit IRC02:21
*** garyh has quit IRC02:23
*** hdd has joined #openstack-infra02:24
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs for pypi and docs
*** markvoelker has joined #openstack-infra02:27
*** mwagner_lap has joined #openstack-infra02:28
*** r1chardj0n3s is now known as r1chardj0n3s_afk02:30
openstackgerritDi Xu proposed openstack-infra/jenkins-job-builder: fix setuptools version error
*** markvoelker has quit IRC02:32
*** r1chardj0n3s_afk is now known as r1chardj0n3s02:35
openstackgerritGhanshyam Mann proposed openstack-infra/project-config: Add non-voting job for Nova V2.1 API with neutron
openstackgerritThanh Ha proposed openstack-infra/jenkins-job-builder: Add FindBugs plugin support
*** sigmavirus24_awa is now known as sigmavirus2402:42
*** camunoz has joined #openstack-infra02:46
zxiiroI'm stuck on adding unit tests to the reporters module (FindBugs). If anyone's around to take a look and give me some advice on what to do?02:46
*** garyh has joined #openstack-infra02:48
*** garyh has quit IRC02:58
*** radez_g0n3 is now known as radez03:01
*** jerryz has joined #openstack-infra03:03
*** garyh has joined #openstack-infra03:05
*** hdd has quit IRC03:06
*** Longgeek has joined #openstack-infra03:11
*** Longgeek has quit IRC03:14
*** bhunter71 has joined #openstack-infra03:14
*** garyh has quit IRC03:15
*** Ryan_Lane has joined #openstack-infra03:18
*** garyh has joined #openstack-infra03:18
*** Longgeek has joined #openstack-infra03:20
*** shashankhegde has joined #openstack-infra03:22
*** shashankhegde has quit IRC03:23
*** tiswanso has joined #openstack-infra03:28
*** garyh has quit IRC03:28
*** markvoelker has joined #openstack-infra03:28
*** pcrews has quit IRC03:30
*** tiswanso has quit IRC03:30
*** tiswanso has joined #openstack-infra03:30
*** markvoelker has quit IRC03:33
*** bhunter71 has quit IRC03:36
*** koolhead17 has joined #openstack-infra03:38
*** ddieterl_ has quit IRC03:39
*** sigmavirus24 is now known as sigmavirus24_awa03:40
*** mriedem1 has quit IRC03:40
*** dimtruck is now known as zz_dimtruck03:40
*** miqui has quit IRC03:41
*** ddieterly has joined #openstack-infra03:41
*** talluri has joined #openstack-infra03:45
*** baoli has quit IRC03:48
*** talluri has quit IRC03:49
*** coolsvap_ is now known as coolsvap04:02
*** sabeen has quit IRC04:03
*** baoli has joined #openstack-infra04:04
*** lnxnut has quit IRC04:04
*** radez is now known as radez_g0n304:09
*** lnxnut has joined #openstack-infra04:13
*** amotoki has joined #openstack-infra04:13
*** garyh has joined #openstack-infra04:16
*** armax has joined #openstack-infra04:18
*** tiswanso has quit IRC04:18
*** sdake_ has quit IRC04:22
*** dims_ has quit IRC04:23
*** asettle-gym is now known as asettle04:24
*** sdake_ has joined #openstack-infra04:25
*** garyh has quit IRC04:25
*** avozza is now known as zz_avozza04:26
*** armax has quit IRC04:27
*** markvoelker has joined #openstack-infra04:29
*** zz_dimtruck is now known as dimtruck04:31
*** markvoelker has quit IRC04:35
*** baoli has quit IRC04:37
*** armax has joined #openstack-infra04:38
*** achanda has joined #openstack-infra04:39
*** sdake_ has quit IRC04:43
*** dimtruck is now known as zz_dimtruck04:43
*** lnxnut has quit IRC04:43
*** zz_dimtruck is now known as dimtruck04:43
openstackgerritJerry Zhao proposed openstack-infra/project-config: Make IPv4 take precedence over IPv6 when download packages&codes.
*** rmcall has joined #openstack-infra04:45
*** talluri has joined #openstack-infra04:46
*** nikil89_ has joined #openstack-infra04:46
*** nikil89_ is now known as naggappan04:46
*** lnxnut has joined #openstack-infra04:49
*** yfried|afk is now known as yfried04:49
*** ddieterly has quit IRC04:49
*** talluri has quit IRC04:50
*** lnxnut has quit IRC04:53
*** sdake_ has joined #openstack-infra04:55
*** coolsvap is now known as coolsvap_04:55
*** coolsvap_ is now known as coolsvap04:56
*** sdake_ has quit IRC04:59
*** asettle has quit IRC05:04
*** Alexandra_ has joined #openstack-infra05:04
*** Alexandra_ is now known as asettle05:05
*** talluri has joined #openstack-infra05:15
*** garyh has joined #openstack-infra05:16
*** otter768 has quit IRC05:18
*** dimsum__ has joined #openstack-infra05:24
*** garyh has quit IRC05:25
*** dimsum__ has quit IRC05:29
*** markvoelker has joined #openstack-infra05:31
*** markvoelker has quit IRC05:36
*** armax has quit IRC05:37
*** yfried has quit IRC05:38
*** achanda has quit IRC05:38
*** armax has joined #openstack-infra05:39
*** sdake_ has joined #openstack-infra05:40
*** stevemar has quit IRC05:41
*** lnxnut has joined #openstack-infra05:50
*** BharatK has joined #openstack-infra05:50
*** boris-42 has quit IRC05:52
*** r1chardj0n3s is now known as r1chardj0n3s_afk05:55
*** garyh has joined #openstack-infra05:55
*** ildikov has quit IRC05:57
*** achanda has joined #openstack-infra05:57
*** hdd has joined #openstack-infra06:03
*** garyh has quit IRC06:05
*** shashankhegde has joined #openstack-infra06:07
*** sdake_ has quit IRC06:07
*** shashankhegde has quit IRC06:11
*** armax has quit IRC06:15
*** Ryan_Lane has quit IRC06:21
*** Ryan_Lane has joined #openstack-infra06:22
*** achanda has quit IRC06:22
*** doude has quit IRC06:24
*** boris-42 has joined #openstack-infra06:26
*** Ryan_Lane has quit IRC06:26
*** pcaruana has quit IRC06:28
*** markvoelker has joined #openstack-infra06:32
*** markvoelker has quit IRC06:37
*** garyh has joined #openstack-infra06:37
*** lnxnut has quit IRC06:41
*** garyh has quit IRC06:47
*** yfried has joined #openstack-infra06:49
*** ddieterly has joined #openstack-infra06:50
*** asettle has quit IRC06:53
*** yfried_ has joined #openstack-infra06:53
*** yfried has quit IRC06:53
*** ddieterly has quit IRC06:55
*** hdd has quit IRC06:59
*** HeOS has quit IRC07:01
*** vigneshvar has joined #openstack-infra07:02
*** scheuran has joined #openstack-infra07:09
openstackgerritgreghaynes proposed openstack-infra/nodepool: Fail gracefully if disk-image-create is missing
*** achanda has joined #openstack-infra07:15
*** garyh has joined #openstack-infra07:17
*** otter768 has joined #openstack-infra07:19
*** afazekas has joined #openstack-infra07:20
*** rmcall has quit IRC07:22
*** dannywilson has joined #openstack-infra07:22
*** Ryan_Lane has joined #openstack-infra07:23
*** otter768 has quit IRC07:24
*** Ryan_Lane has quit IRC07:27
*** garyh has quit IRC07:28
*** doude has joined #openstack-infra07:28
*** ivar-laz_ has joined #openstack-infra07:29
*** dannywilson has quit IRC07:31
*** dannywilson has joined #openstack-infra07:32
*** ivar-lazzaro has quit IRC07:32
*** markvoelker has joined #openstack-infra07:33
*** ivar-laz_ has quit IRC07:34
*** stevemar has joined #openstack-infra07:35
*** dannywilson has quit IRC07:36
*** ildikov has joined #openstack-infra07:37
*** chlong has quit IRC07:37
*** markvoelker has quit IRC07:38
openstackgerritBharat Kumar Kobagana proposed openstack-infra/project-config: Suffixing GlusterFS experimental job name with '-centos7'
*** achanda has quit IRC07:47
*** ddieterly has joined #openstack-infra07:51
*** achanda has joined #openstack-infra07:51
*** e0ne has joined #openstack-infra07:55
*** ddieterly has quit IRC07:56
*** garyh has joined #openstack-infra07:59
*** doude has quit IRC08:01
*** doude has joined #openstack-infra08:02
*** arxcruz has joined #openstack-infra08:03
*** e0ne has quit IRC08:03
*** Ala has joined #openstack-infra08:09
*** garyh has quit IRC08:09
*** woodster_ has quit IRC08:10
*** dtantsur|afk is now known as dtantsur08:10
*** luqas has joined #openstack-infra08:11
*** Guest90369 has quit IRC08:14
*** naggappan has quit IRC08:16
*** HeOS has joined #openstack-infra08:17
*** shardy has joined #openstack-infra08:17
*** jlibosva has joined #openstack-infra08:18
*** mfisch has joined #openstack-infra08:18
*** doude has quit IRC08:18
*** mfisch is now known as Guest2281708:18
*** mtanino has quit IRC08:20
*** stevemar has quit IRC08:20
*** garyh has joined #openstack-infra08:20
*** achuprin_ has quit IRC08:25
*** doude has joined #openstack-infra08:25
*** bradjones has quit IRC08:31
*** garyh has quit IRC08:31
*** bradjones has joined #openstack-infra08:32
*** markvoelker has joined #openstack-infra08:34
*** achuprin_ has joined #openstack-infra08:37
*** doude has quit IRC08:38
*** ihrachyshka has joined #openstack-infra08:38
*** markvoelker has quit IRC08:39
*** berendt has joined #openstack-infra08:40
*** jcoufal has joined #openstack-infra08:40
*** andreykurilin_ has joined #openstack-infra08:40
*** jerryz has quit IRC08:41
*** tkelsey has joined #openstack-infra08:41
*** jlibosva has quit IRC08:41
openstackgerritAleksey Ripinen proposed openstack-infra/storyboard: Added new paginate query
*** oomichi has quit IRC08:42
*** jlibosva has joined #openstack-infra08:43
*** mrmartin has joined #openstack-infra08:44
*** doude has joined #openstack-infra08:45
*** tkelsey has quit IRC08:45
*** doude has quit IRC08:46
*** berendt has quit IRC08:46
*** koolhead17 has quit IRC08:47
*** markus_z has joined #openstack-infra08:47
*** tkelsey has joined #openstack-infra08:48
*** jistr has joined #openstack-infra08:49
*** dizquierdo has joined #openstack-infra08:49
*** afazekas has quit IRC08:49
*** achanda has quit IRC08:52
*** ddieterly has joined #openstack-infra08:52
*** mpaolino has joined #openstack-infra08:53
openstackgerritAleksey Ripinen proposed openstack-infra/storyboard: Added controller for refresh tokens
*** achanda has joined #openstack-infra08:56
*** ddieterly has quit IRC08:57
*** nikil89_ has joined #openstack-infra08:58
*** nikil89_ is now known as naggappan08:59
*** hashar has joined #openstack-infra08:59
*** hashar has quit IRC09:00
*** garyh has joined #openstack-infra09:02
*** jpich has joined #openstack-infra09:02
*** vigneshvar has quit IRC09:05
*** zz_avozza is now known as avozza09:05
*** sarob has joined #openstack-infra09:06
*** koolhead17 has joined #openstack-infra09:08
*** achanda has quit IRC09:08
*** ghostpl has joined #openstack-infra09:09
*** jlibosva has quit IRC09:09
*** jlibosva has joined #openstack-infra09:10
*** hashar has joined #openstack-infra09:10
*** garyh has quit IRC09:12
*** garyh has joined #openstack-infra09:14
*** MaxV has joined #openstack-infra09:16
*** vigneshvar has joined #openstack-infra09:17
*** achanda has joined #openstack-infra09:18
*** derekh has joined #openstack-infra09:18
*** skolekonov has joined #openstack-infra09:18
*** otter768 has joined #openstack-infra09:20
openstackgerritMehdi Abaakouk proposed openstack-infra/project-config: gnocchi: add swift and ceph functional jobs
*** vponomaryov has joined #openstack-infra09:24
*** otter768 has quit IRC09:25
*** garyh has quit IRC09:25
*** koolhead17 has quit IRC09:26
openstackgerritMerged openstack-infra/project-config: Add Rally job to Mistral
*** yamahata has quit IRC09:31
*** yamamot__ has quit IRC09:32
*** koolhead17 has joined #openstack-infra09:35
*** markvoelker has joined #openstack-infra09:35
*** markvoelker has quit IRC09:39
*** andreykurilin_ has quit IRC09:41
*** doude has joined #openstack-infra09:41
*** jedimike has joined #openstack-infra09:42
*** koolhead17 has quit IRC09:44
*** Hal has joined #openstack-infra09:46
*** Hal is now known as Guest8251109:46
*** achanda has quit IRC09:46
*** marcusvrn has joined #openstack-infra09:49
*** denis_makogon has joined #openstack-infra09:50
*** ddieterly has joined #openstack-infra09:53
*** jgallard__ has joined #openstack-infra09:55
*** dizquierdo has quit IRC09:55
*** ddieterly has quit IRC09:57
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Fix exception in stories db api
*** hichihara has quit IRC09:58
*** garyh has joined #openstack-infra09:58
*** zz_johnthetubagu is now known as johnthetubaguy10:00
*** johnthetubaguy has left #openstack-infra10:00
*** dtantsur is now known as dtantsur|brb10:00
*** e0ne has joined #openstack-infra10:00
*** jp_at_hp has joined #openstack-infra10:02
*** rlandy has joined #openstack-infra10:02
*** koolhead17 has joined #openstack-infra10:02
*** jd__ has quit IRC10:06
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Fix exception in stories db api
*** jd__ has joined #openstack-infra10:07
*** garyh has quit IRC10:08
*** ssam2 has joined #openstack-infra10:08
*** hashar has quit IRC10:08
*** hashar has joined #openstack-infra10:09
*** e0ne has quit IRC10:09
*** e0ne has joined #openstack-infra10:09
*** hashar has quit IRC10:10
*** hashar has joined #openstack-infra10:10
*** zhiwei has quit IRC10:10
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Added controller for refresh tokens
*** e0ne is now known as e0ne_10:19
*** andreaf_ is now known as andreaf10:21
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Revert "Adding pretty-tox"
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Revert "Adding pretty-tox"
*** e0ne_ has quit IRC10:24
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Added controller for refresh tokens
*** e0ne has joined #openstack-infra10:24
*** pblaho has joined #openstack-infra10:25
openstackgerritMerged openstack-infra/storyboard: Added new way to filter stories by tags
*** belmoreira has joined #openstack-infra10:31
*** pelix has joined #openstack-infra10:32
*** yfried_ is now known as yfried|afk10:32
*** dannywilson has joined #openstack-infra10:32
*** yfried|afk is now known as yfried_10:35
*** markvoelker has joined #openstack-infra10:36
*** dannywilson has quit IRC10:37
*** vigneshvar has quit IRC10:37
*** vigneshvar has joined #openstack-infra10:40
*** ldnunes has joined #openstack-infra10:40
*** sarob has quit IRC10:41
*** markvoelker has quit IRC10:41
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Added controller for refresh tokens
*** yfried_ is now known as yfried|afk10:48
*** Guest82511 has quit IRC10:52
*** ssam2 has quit IRC10:52
*** erlon has joined #openstack-infra10:53
*** mika has quit IRC10:56
openstackgerritMerged openstack-infra/storyboard: OAuth checks for remote data access permissions.
*** mikap has joined #openstack-infra10:56
*** mikap is now known as mika10:56
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Explicit Hook Priorities
*** dtantsur|brb is now known as dtantsur10:59
*** rm_work|away has quit IRC11:02
*** garyh has joined #openstack-infra11:02
*** talluri has quit IRC11:03
*** rm_work|away has joined #openstack-infra11:04
*** rm_work|away is now known as rm_work11:04
*** rm_work has joined #openstack-infra11:04
openstackgerritMerged openstack-infra/storyboard: OAuth Validate Token Grant Types
openstackgerritMerged openstack-infra/storyboard: OAuth tests for invalid authorization codes.
*** yfried|afk is now known as yfried_11:08
*** e0ne is now known as e0ne_11:08
openstackgerritMerged openstack-infra/storyboard: OAuth delete refresh tokens when access tokens deleted.
*** garyh has quit IRC11:13
*** e0ne_ has quit IRC11:14
openstackgerritMerged openstack-infra/storyboard: OAuth now validates client ID against a configured list.
*** talluri has joined #openstack-infra11:17
*** pc_m has joined #openstack-infra11:19
*** otter768 has joined #openstack-infra11:21
*** adalbas has joined #openstack-infra11:24
*** otter768 has quit IRC11:25
*** ssam2 has joined #openstack-infra11:28
*** pblaho has quit IRC11:29
*** pblaho has joined #openstack-infra11:29
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Explicit Hook Priorities
*** aysyd has joined #openstack-infra11:30
*** pblaho has quit IRC11:30
*** jgallard__ has quit IRC11:33
*** afazekas has joined #openstack-infra11:34
*** markvoelker has joined #openstack-infra11:37
openstackgerritSwami Reddy proposed openstack/requirements: Ceilometer ceph storage meters need requests-aws
*** afazekas has quit IRC11:39
*** doude has quit IRC11:40
*** markvoelker has quit IRC11:42
*** garyh has joined #openstack-infra11:42
openstackgerritAleksey Ripinen proposed openstack-infra/storyboard: Fix method is_valid in access_token api
*** e0ne has joined #openstack-infra11:51
*** garyh has quit IRC11:52
*** cdent has joined #openstack-infra11:54
*** andreaf_ has joined #openstack-infra11:54
*** ddieterly has joined #openstack-infra11:54
*** jamielennox is now known as jamielennox|away11:56
*** e0ne is now known as e0ne_11:58
*** andreaf_ has quit IRC11:58
*** ddieterly has quit IRC11:59
*** andreaf_ has joined #openstack-infra11:59
*** e0ne_ is now known as e0ne11:59
*** garyh has joined #openstack-infra12:06
*** koolhead17 has quit IRC12:11
*** koolhead17 has joined #openstack-infra12:11
*** Longgeek has quit IRC12:12
*** _nadya_ has joined #openstack-infra12:12
*** garyh has quit IRC12:16
*** koolhead17 has quit IRC12:16
*** afazekas has joined #openstack-infra12:19
*** chlong has joined #openstack-infra12:25
openstackgerritMerged openstack-infra/storyboard: Explicit Hook Priorities
*** rfolco has joined #openstack-infra12:32
*** tmcpeak has joined #openstack-infra12:32
openstackgerritMarton Kiss proposed openstack-infra/system-config: Add puppet askbot and puppet solr modules
*** garyh has joined #openstack-infra12:34
*** chlong has quit IRC12:35
*** WendellM_ has quit IRC12:35
*** chlong has joined #openstack-infra12:36
*** WendellM has joined #openstack-infra12:36
*** rfolco has quit IRC12:36
*** markvoelker has joined #openstack-infra12:38
*** koolhead17 has joined #openstack-infra12:39
*** koolhead17 has joined #openstack-infra12:39
*** tmcpeak1 has joined #openstack-infra12:41
openstackgerritAleksey Ripinen proposed openstack-infra/storyboard: Fix method checks.superuser
*** tmcpeak has quit IRC12:42
*** talluri has quit IRC12:42
*** garyh has quit IRC12:43
*** markvoelker has quit IRC12:43
*** dimsum__ has joined #openstack-infra12:44
*** dims_ has joined #openstack-infra12:45
openstackgerritMerged openstack-infra/project-config: Add maven build and repo upload for monasca-api
openstackgerritMerged openstack-infra/project-config: Add new Tempest job for Manila project
openstackgerritMerged openstack-infra/project-config: Preseed with a number of samples from db instead of aggregate
*** dimsum__ has quit IRC12:49
*** rfolco has joined #openstack-infra12:49
*** ghostpl has quit IRC12:49
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs publishing docs
*** rfolco has quit IRC12:51
*** baoli has joined #openstack-infra12:51
*** lnxnut has joined #openstack-infra12:53
*** ddieterly has joined #openstack-infra12:55
openstackgerritAleksey Ripinen proposed openstack-infra/storyboard: Added controller for refresh tokens
*** doude has joined #openstack-infra12:56
*** dprince has joined #openstack-infra12:57
*** arxcruz has quit IRC12:57
*** arxcruz has joined #openstack-infra12:57
*** lnxnut has quit IRC12:58
*** markvoelker has joined #openstack-infra12:59
*** ddieterly has quit IRC12:59
*** e0ne is now known as e0ne_12:59
*** baoli_ has joined #openstack-infra13:01
openstackgerritThierry Carrez proposed openstack-infra/storyboard: Email Templating Engine
openstackgerritThierry Carrez proposed openstack-infra/storyboard: API events now include resource before/after state.
openstackgerritThierry Carrez proposed openstack-infra/storyboard: Email configuration.
openstackgerritThierry Carrez proposed openstack-infra/storyboard: Pass the old entity values in the NotificationHook
*** _nadya_ has quit IRC13:03
ttxyolanda, NikitaKonovalov: please check out
*** baoli has quit IRC13:04
NikitaKonovalovttx: is that a rebase?13:04
ttxsolves conflict created by renaming priorites to priorities13:04
ttxwill unblock the next 4 patches in the series13:05
*** sdake_ has joined #openstack-infra13:05
*** sdake_ has quit IRC13:05
*** weshay has joined #openstack-infra13:06
*** e0ne_ is now known as e0ne13:06
openstackgerritMarton Kiss proposed openstack-infra/system-config: Initial commit of Ask website
*** sdake_ has joined #openstack-infra13:13
*** ghostpl has joined #openstack-infra13:14
*** garyh has joined #openstack-infra13:14
*** hashar has quit IRC13:15
*** BharatK has quit IRC13:16
*** tkelsey has quit IRC13:18
*** bswartz has quit IRC13:18
yolandasorry, i was lunching13:19
yolandataking a look13:19
*** sdake__ has joined #openstack-infra13:21
*** otter768 has joined #openstack-infra13:22
*** yfried_ is now known as yfried|afk13:23
*** yfried|afk is now known as yfried_13:24
*** garyh has quit IRC13:24
*** sdake_ has quit IRC13:25
openstackgerritMerged openstack-infra/storyboard: Pass the old entity values in the NotificationHook
openstackgerritMerged openstack-infra/storyboard: API events now include resource before/after state.
openstackgerritMerged openstack-infra/storyboard: Email Templating Engine
*** otter768 has quit IRC13:26
*** BobBall_AWOL is now known as BobBall13:27
*** dizquierdo has joined #openstack-infra13:29
openstackgerritMerged openstack-infra/storyboard: Email configuration.
*** e0ne is now known as e0ne_13:30
*** ddieterly has joined #openstack-infra13:31
*** ddieterl_ has joined #openstack-infra13:32
*** mfink_ has joined #openstack-infra13:32
*** ddieterly has quit IRC13:32
*** achuprin_ has quit IRC13:32
*** woodster_ has joined #openstack-infra13:34
*** Guest82511 has joined #openstack-infra13:34
*** chlong has quit IRC13:36
*** bknudson has quit IRC13:36
*** mtanino has joined #openstack-infra13:37
*** rfolco has joined #openstack-infra13:37
*** ddieterl_ has quit IRC13:37
*** e0ne_ is now known as e0ne13:38
*** belmoreira has quit IRC13:40
*** rfolco has quit IRC13:40
*** sarob has joined #openstack-infra13:41
*** bhunter71 has joined #openstack-infra13:42
*** Longgeek has joined #openstack-infra13:44
*** sarob has quit IRC13:46
*** _amrith_ is now known as amrith13:47
*** weshay has quit IRC13:47
*** afazekas has quit IRC13:47
*** achuprin_ has joined #openstack-infra13:48
openstackgerritMerged openstack-dev/pbr: Fixes Windows generated scripts headers
*** dkliban_afk is now known as dkliban13:53
*** weshay has joined #openstack-infra13:54
*** vigneshvar has quit IRC13:55
*** Longgeek has quit IRC13:57
*** bknudson has joined #openstack-infra13:59
*** Longgeek has joined #openstack-infra14:00
*** david-lyle_afk is now known as david-lyle14:01
*** jcoufal has quit IRC14:02
*** bswartz has joined #openstack-infra14:02
*** jcoufal has joined #openstack-infra14:02
*** krtaylor has quit IRC14:03
*** r-daneel has joined #openstack-infra14:04
*** mjturek1 has joined #openstack-infra14:04
*** e0ne has quit IRC14:05
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs publishing docs
*** e0ne has joined #openstack-infra14:05
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs publishing docs
*** hashar has joined #openstack-infra14:06
*** lascii is now known as alaski14:07
*** xyang1 has joined #openstack-infra14:08
*** amotoki has quit IRC14:09
*** jpeeler has joined #openstack-infra14:12
*** jpeeler has quit IRC14:12
*** jpeeler has joined #openstack-infra14:12
*** garyh has joined #openstack-infra14:14
*** jaypipes has joined #openstack-infra14:15
*** krtaylor has joined #openstack-infra14:15
*** dustins has joined #openstack-infra14:16
*** lnxnut has joined #openstack-infra14:17
*** ddieterly has joined #openstack-infra14:19
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Revert "OAuth tests for invalid authorization codes."
*** lnxnut has quit IRC14:21
*** e0ne is now known as e0ne_14:23
*** Longgeek has quit IRC14:23
*** garyh has quit IRC14:24
*** sdake__ is now known as sdake_14:25
openstackgerritDenis M. proposed openstack-dev/pbr: Added extras_require feature
openstackgerritDenis M. proposed openstack-dev/pbr: Added extras_require feature
*** wuhg has quit IRC14:29
*** bcourt has joined #openstack-infra14:29
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Revert "OAuth tests for invalid authorization codes."
*** jcoufal has quit IRC14:34
*** jcoufal has joined #openstack-infra14:34
*** doug-fish has joined #openstack-infra14:38
*** abhirc has joined #openstack-infra14:38
openstackgerritDenis M. proposed openstack-dev/pbr: Added extras_require feature
*** garyh has joined #openstack-infra14:42
*** reed has joined #openstack-infra14:42
*** achuprin_ has quit IRC14:42
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs publishing docs
*** ryanpetrello has quit IRC14:44
*** sandywalsh has quit IRC14:44
mordredfungi: you'll be happy to know that:14:44
mordreda) config-drive network info has the informations (scripting consuming it right now)14:44
*** sandywalsh has joined #openstack-infra14:45
mordredb) doesn't have ipv6 info -so we may still have some follow up work to do - but we can move forward for sure14:45
fungimordred: in all the regions we're using? awesome!14:45
mordredfungi: I have not tested all of the regions yet - but I did spin something up in dfw and it's working14:46
mordredfungi: I figure I'll write the consumption script so I can get it up for review14:46
openstackgerritNikita Konovalov proposed openstack-infra/storyboard: Fix exception in stories db api
*** signed8bit has joined #openstack-infra14:46
openstackgerritMarton Kiss proposed openstack-infra/infra-specs: Add migrate askbot to infra spec
*** ryanpetrello has joined #openstack-infra14:48
*** e0ne_ is now known as e0ne14:49
reedttx, re:, have you considered putting that page in the 'Obsolete:' namespace on the wiki instead of renaming it?14:50
openstackgerritThierry Carrez proposed openstack-infra/infra-manual: Update to reflect recent governance changes
ttxreed: if you teach me I can do it14:52
ttxreed: I considered getting rid of it completely14:52
*** dmsimard_away is now known as dmsimard14:53
reedin the drop-down menu 'new title' go from whatever is default (I believe it's called "Main" in English --my menu is in Italian, testing the translation extension)14:53
*** prad-afk is now known as prad14:54
*** sabeen has joined #openstack-infra14:54
reedso from "main" to "obsolete"14:54
*** afazekas has joined #openstack-infra14:54
reedand add a redirect14:54
ttxok, trying14:54
reedthere is also a macro to add a good looking warning sign that the page is obsolete14:54
*** achuprin_ has joined #openstack-infra14:54
*** kgiusti has joined #openstack-infra14:55
*** talluri has joined #openstack-infra14:55
ttxreed: if you add it to that page I'll learn it14:55
ttxreed: ISTR you had a proposed new main page somewhere, got the link ?14:56
*** koolhead17 has quit IRC14:56
reedI need to review the whole taxonomy now that programs are no more14:56
*** amitgandhinz has joined #openstack-infra14:56
openstackgerritSpencer Krum proposed openstack-infra/project-config: Enable testing of future parser for puppet
*** radez_g0n3 is now known as radez14:58
*** lnxnut has joined #openstack-infra14:58
ttxthe problem with this page is that it doesn't prioritize anything. I think it's good as Index page, not so much as main landing page14:58
ttxLet me propose something else and let me know how that looks14:58
reedthe good thing is that it's automatic though :)14:59
*** elliottk has joined #openstack-infra14:59
ttxLet me try to collapse project team list and main page together15:00
ttxerr. after meetings15:00
*** dimtruck is now known as zz_dimtruck15:00
elliottkgood morning infra folk.  i have a couple of jjb questions.  first...15:01
*** lnxnut_ has joined #openstack-infra15:01
*** mriedem has joined #openstack-infra15:01
elliottki have a global default throttle setting that i am trying to disable in certain job templates but it is not working15:01
*** lnxnut has quit IRC15:02
*** derekh has quit IRC15:02
elliottkthrottleEnabled is still set to true.  however, the number of concurrent build and builds per node are both 0 so, essentially, it IS disabled15:02
*** tiswanso has joined #openstack-infra15:02
elliottkbut i don't even want it to show up as enabled in my job15:02
elliottkthis is what i have in the template that i'm trying to disable it in:15:03
elliottkthrottleEnabled: false15:03
*** derekh has joined #openstack-infra15:03
elliottkwith the appropriate hyphens and such15:03
elliottki don't understand why this isn't working15:04
*** signed8b_ has joined #openstack-infra15:04
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add check/gate jobs for the networking-odl project
elliottkthe properties jjb page clearly says that global defaults can be overridden in a job template15:04
openstackgerritSpencer Krum proposed openstack-infra/project-config: Enable testing of future parser for puppet
*** tiswanso has quit IRC15:05
zxiiropelix: Any advice on what I can do here?
*** tiswanso has joined #openstack-infra15:05
anteayaelliottk: have you considered joining the third party meeting, now in progress in #openstack-meeting?15:06
*** ChuckC has joined #openstack-infra15:06
*** ChuckC has quit IRC15:06
*** signed8bit has quit IRC15:06
elliottksorry, i thought that was this irc.  let me go to that one.  thanks!15:06
*** elliottk has left #openstack-infra15:07
*** ChuckC has joined #openstack-infra15:07
*** carl_baldwin has joined #openstack-infra15:07
*** stevemar has joined #openstack-infra15:08
*** markvan has joined #openstack-infra15:09
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Temporarily bypass expiration date check
*** amotoki has joined #openstack-infra15:10
*** sarob has joined #openstack-infra15:10
*** superdan is now known as dansmith15:10
*** nelsnelson has joined #openstack-infra15:12
*** _nadya_ has joined #openstack-infra15:13
anteayafor the record I wasn't asking him to leave this channel just join the meeting, it appears he either mis-understood or doesn't have a client that can do both15:14
*** _buzz_ has joined #openstack-infra15:15
*** sarob has quit IRC15:15
mordredfungi: have you ever looked at the routing on the rax nodes?15:15
mordredfungi: specifically, there are two routes added to the 10. interface15:15
mordredfungi: any idea what the heck they are for?15:16
openstackgerritamrith proposed openstack-infra/project-config: Making gate-trove-functional-dsvm-mysql a gating test
*** trinaths1 has joined #openstack-infra15:16
pelixzxiiro: added comment, just needs an extra line of yaml code to trigger setting of the project type in the base test class.15:17
trinaths1Hi guys!. Got this doubt. Commit-3 is dependant on Commit-2 and Commit-1. While check out of the of Commit-3, How can I enable git to also checkout Commit-2 and Commit-1 based on dependancy.15:17
mesterysdague: Thanks for the review here! Can you let me know if I correctly set this up to be non-voting? Much appreciated!
fungimordred: looking15:17
*** _buzz has quit IRC15:17
otherwiseguyanteaya: My native-OVSDB patch requires OVS's python bindings...which they don't have in PyPI (and it would be a bit of work to get them in a state to upload to PyPI). Is there a way we can get the python-openvswitch distro packages installed on the test VMs?15:17
fungitrinaths1: commit 2 and commit 1 should be in the git history of commit 3 so if you check it out (not cherry-pick it) then you should have them15:18
trinaths1fungi: thank you will try this out15:18
otherwiseguyI'm going to add skipTest functionality if the library isn't installed, but would like the CI to actually test it.15:18
anteayaotherwiseguy: right now we are suggesting folks install test specific packages within a venv on the node15:18
anteayaotherwiseguy: to prevent clashing with other installed things15:19
anteayaotherwiseguy: after this meeting I will try to dig you up an example15:19
trinaths1fungi: but i observed that "git checkout" gives me the old git code base on a latest code base. then. How to deal with it?15:19
mordredfungi: also - any chance you know a good/smart way to get logical interface name (eth0) given hwmacaddr?15:19
fungitrinaths1: rebase that series on the branch tip15:19
anteayaif you don't want to wait for me look at jhesketh's work for swift logs, I believe he was directed to use a venv15:19
fungimordred: possibly ip link show. i'll test and see15:20
trinaths1fungi: okay. need to try this. Thanks sir15:20
mordredfungi: yah - that seems to work15:20
*** talluri has quit IRC15:22
*** timcline has joined #openstack-infra15:22
*** otter768 has joined #openstack-infra15:23
fungimordred: looking at one in iad i see bound to eth1 and then there are routes through off that interface to and (the interface is in a subnet of the latter destination)15:23
fungimordred: i'm guessing those are the two rfc-1918 cidrs rackspace uses for their backend net15:23
mordredfungi: I have those same things in the node I have in dfw15:23
*** Longgeek has joined #openstack-infra15:24
openstackgerritRicardo Carrillo Cruz proposed openstack-infra/storyboard: Rename map_resource call to get_original_resource
*** signed8b_ is now known as signed8bit15:24
zxiiropelix: thanks that helps15:25
sdaguemestery: - comments included15:25
*** ildikov has quit IRC15:25
openstackgerrityolanda.robla proposed openstack-infra/storyboard: Temporarily bypass expiration date check
mesterysdague: Thanks man!15:26
*** jcoufal_ has joined #openstack-infra15:26
openstackgerritThanh Ha proposed openstack-infra/jenkins-job-builder: Add FindBugs plugin support
mesterysdague: Cool, thanks for the comments. Do I remove the bits from jenkins/jobs/networking-odl.yaml  around voting I had put in there?15:27
*** otter768 has quit IRC15:27
sdaguemestery: yeh15:28
mesterysdague: Thank you15:28
*** jcoufal has quit IRC15:29
*** spzala has joined #openstack-infra15:29
*** koolhead17 has joined #openstack-infra15:30
openstackgerritMerged openstack-infra/devstack-gate: Refactor processing of testr artifacts for reuse
*** zz_dimtruck is now known as dimtruck15:32
openstackgerritMerged openstack-infra/devstack-gate: Enable preseeding testrepository with a number of streams
openstackgerritMerged openstack-infra/devstack-gate: Remove unused variables(PIP_USE_MIRRORS, USE_GET_PIP)
openstackgerritMerged openstack-infra/devstack-gate: Add partial-ironic support
*** afazekas has quit IRC15:32
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add check/gate jobs for the networking-odl project
mordredfungi: cool. I have a script that reproduces /etc/network/interfaces for ubuntu/debian15:33
mordredfungi: next step - centos15:33
*** medberry is now known as med_15:34
vponomaryovwho knows what has been changed to CI that now only one commit in merge queue?15:34
mordredfungi: it's not a sexy script - but it's super simple and doesn't need any non-standard libs15:34
*** enikanorov has joined #openstack-infra15:35
mordrednor any long-lived daemons or xen tools or anything :)15:35
anteayavponomaryov: you seem to be looking at something very specific, care to share what you are seeing?15:35
*** achanda has joined #openstack-infra15:35
*** talluri has joined #openstack-infra15:35
vponomaryovanteya: three commits are approved at once -,n,z and only first of them in queue -
*** enikanorov has quit IRC15:36
*** annegentle has joined #openstack-infra15:36
anteayaQueue lengths: 2146 events, 182 results.15:37
anteayavponomaryov: when events reaches 0 look again15:37
anteayanow the question I would have is has anyone taken a peek at zuul today to find out why we have such a back up in events15:38
*** KanagarajM has joined #openstack-infra15:39
fungianteaya: looks like maybe several changes just merged in the gate15:40
funginote the large spike in the merge-check sparkline15:40
*** yfried_ is now known as yfried|afk15:40
fungii also count 9 changes which have been in the post pipeline for ~1 minute15:41
*** tiswanso has quit IRC15:41
fungiso it was probably 9 changes merging at once15:41
anteayaah cool15:41
fungizuul's queue lengths are already caught back up now15:41
anteayabeen a while since I saw it with over 2000 events15:42
*** tiswanso has joined #openstack-infra15:42
*** achanda has quit IRC15:42
anteayaglad it was just because I didn't take the time to look around15:42
fungiglad it's because zuul is working really, really well15:42
anteayathat too15:42
anteayavponomaryov: do you see your patches?15:42
anteayaand yeah merge-check wow15:43
vponomaryovanteya: yes15:43
anteayavponomaryov: great15:43
*** trinaths1 has left #openstack-infra15:43
*** jcoufal_ has quit IRC15:43
vponomaryovanteya: what was that?15:43
*** jcoufal has joined #openstack-infra15:44
anteayavponomaryov: in the zuul status page the queue lengths tell you if there is a backup anywhere15:44
anteayaso zuul knew there were over 2000 events it hadn't processed at the time you posed your question15:45
anteayaand had 182 results that had not yet posted to gerrit15:45
vponomaryovoh, now I see, thanks15:45
anteayaso it was backed up, which is why you coulnd't see your patches on the page15:45
*** wenlock has joined #openstack-infra15:45
anteayaand fungi just figured out what caused the back up15:45
anteayaand the fast clearance15:45
*** afazekas has joined #openstack-infra15:47
fungihard to call it backed up... it was just prioritizing emptying its event and result queues15:47
anteayaokay prioritizing15:48
anteayayes not really backed up15:48
fungiand two of the three approval events in question were pending in the queue because nine changes merged and there was work to be done to process those related events/results first15:48
anteayaand yay 9 changes merged15:50
*** Longgeek has quit IRC15:50
*** yfried|afk is now known as yfried_15:51
*** zz_jgrimm is now known as jgrimm15:51
*** ChuckC has quit IRC15:51
*** ChuckC has joined #openstack-infra15:52
*** rfolco has joined #openstack-infra15:55
*** joesavak has joined #openstack-infra15:55
*** Guest22817 is now known as mfisch15:56
*** mfisch has joined #openstack-infra15:56
*** dims_ has quit IRC15:56
*** harlowja_at_home has joined #openstack-infra15:56
mesterysdague: Regarding my queries in this review (, I think I can get what I want if I follow what manilla has done here ( and here (
mesterysdague: My question is, how they've done this, is it an accepted way and if I follow it will you folks be happy with it?15:57
*** pcrews has joined #openstack-infra15:59
dhellmanngreghaynes: I'm not sure what you're trying to do. Do you have more than one sphinx doc (so you need multiple or are you just trying to put your doc input files somewhere other than docs/source15:59
*** beekneemech is now known as bnemec15:59
otherwiseguyanteaya: I've been looking for the example you mentioned but haven't found it yet.15:59
*** dangers_away is now known as dangers16:00
krotscheckStoryboard meeting in #openstack-meeting-316:00
anteayaotherwiseguy: thanks for waiting, I will look16:00
*** dimsum__ has joined #openstack-infra16:01
openstackgerritMerged openstack-infra/project-config: Suffixing GlusterFS experimental job name with '-centos7'
*** dalgaaf has joined #openstack-infra16:04
*** shashankhegde has joined #openstack-infra16:05
mordredfungi: <-- there are the two files (I'll make an infra patch in a sec) - I have the python program just printing rather than writing the files at the moment for ease of testing16:06
mordredfungi: the shell script clearly need to run at first boot16:06
mordredfungi: but best I can tell, this produces the same output that's on servers now from nova-agent16:06
*** carl_baldwin has quit IRC16:06
anteayaotherwiseguy: also in this version:
fungimordred: it's amusing how much extra work rh is compared to debian16:07
*** carl_baldwin has joined #openstack-infra16:07
mordredfungi: yah16:07
openstackgerritMerged openstack-infra/storyboard: Rename map_resource call to get_original_resource
mordredfungi: do you think I shoudl add hwaddr= to the network/interfaces version? I'm not really sure it gets us much ...16:08
*** dimsum__ has quit IRC16:08
mordredbut also probably doesn't hurt16:08
mordredpleia2, SergeyLukjanov, clarkb: if you want to make your brain unhappy first thing in the morning ^^ the gist I posted to fungi16:09
fungimordred: was it present in the existing ones? i don't think the hypervisor actually allows us to override that anyway16:09
mordredfungi: nope16:09
mordredbut it WAS set in the rh ones16:09
fungimordred: oh... that may be for anaconda's interface mapping (rather than using udev persistent net rules)?16:09
*** dimsum__ has joined #openstack-infra16:10
fungii've been a little out of touch in the red hat world16:10
*** achuprin_ has quit IRC16:10
mordredyah - me too16:10
ttxreed: have a look at -- I aggressively trimmed stale pages, exploded the top header into a series of links under "Getting started", and collapsed the project team list on the main page (since there was a copy there anyway)16:10
ttxannegentle: ^16:10
annegentlettx: looking16:10
* reed checks16:10
fungimordred: anyway, those scripts lgtm. no hard-coded values for routes or anything, so should be relatively future-proof16:11
anteayattx: will there be a picture?16:11
ttxannegentle: that matches the current list of "project teams"16:11
*** dimsum__ has quit IRC16:11
ttxanteaya: we are trying to move people off the wiki, not trying to make it attractive16:11
anteayathe picture feels like home16:11
anteayawell this is in accordance with stated goals then16:11
*** sarob has joined #openstack-infra16:11
otherwiseguyanteaya: The python-openvswitch packages aren't available to be installed via pip, only yum/apt.16:11
SergeyLukjanovmordred, oh16:11
ttxanteaya: the end goal is to move all the useful content into a more peer-reviewed area16:12
mordredfungi: it will not deal with ipv6 as written - largely because there is no ipv6 info - so I can only guess at how it would go into the config drive data16:12
reedare the 'Demo videos' still valid? or wouldn't it be better to link to www.o.o/start?16:12
annegentlettx: heh, bounce rate increase as a stated goal, not very often :)16:12
SergeyLukjanovmordred, due to my local time (7:12pm) it's already not the first things that makes my brain unhappy, but it's probably the most powerful one16:12
*** dimsum__ has joined #openstack-infra16:12
mordredSergeyLukjanov: :)16:12
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add check/gate jobs for the networking-odl project
mordredSergeyLukjanov: I thought you normally woke up aroud 7pm :)16:13
*** dimsum__ has quit IRC16:13
anteayaotherwiseguy: hmmmmm16:13
anteayaotherwiseguy: well then I think we have a candidate for an infra meeting item16:13
fungimordred: oh, i wonder if v6 is being handled in rackland by dynamic means (dhcp6, radvd, et cetera)16:13
ttxannegentle: "leave the wiki now or we'll open fire using stale information"16:13
*** dimsum__ has joined #openstack-infra16:13
openstackgerritMerged openstack-infra/storyboard: Temporarily bypass expiration date check
SergeyLukjanovmordred, normally at 11am, but today at 5am to flight to Moscow, where I'm now :)16:13
*** funzo has joined #openstack-infra16:13
anteayaotherwiseguy: since I don't know of where else we address such a thing and we might need to think about it as a group16:13
ttxreed: I added a link for the "categorized index of pages in the wiki"16:14
otherwiseguyanteaya: So glad I could make more work for everyone.16:14
anteayattx: does the more peer-reviewed area have a name yet?16:14
*** dimsum__ has quit IRC16:14
*** ildikov has joined #openstack-infra16:14
ttxanteaya: plenty of sites. Think governance.o.o, specs.o.o...16:14
ttxall the places driven by a git repo16:15
anteayaotherwiseguy: not necessarily, just that to get your goal accomplished I think we may need to focus as a group for 5 minutes16:15
fungiotherwiseguy: just catching up, when you say python-openvswitch distro packages you mean they're provided by the distros on which we're testing, or "some random unofficial deb/rpm packages not from the distros themselves"?16:15
anteayaotherwiseguy: don't worry if there is an action item out of the discussion it is all yours16:15
annegentlettx: I think one image would help with the text-heaviness. Perhaps the O logo in a corner. Otherwise it's really a wall.16:15
*** coolsvap is now known as coolsvap_16:15
reedannegentle, good :) it's not a place for 'casual' observer probably :)16:16
ttxannegentle: yes, was thinking adding a logo in front of the main subtitles (OpenSTack and Project Teams)16:16
clarkbmordred: is a config drive always called config-2?16:16
anteayaSergeyLukjanov: safe flight16:16
mordredclarkb: yes16:16
annegentlettx: I also want to de-emphasize Official" for docs... but I do appreciate it being first bullet. That said, probably you should have a link to in that first list16:16
ttxannegentle: OpenStack logo I have. Open to suggestions for the "Project Teams" one16:16
annegentlettx: lowercase Official16:16
mordredclarkb: please don't ask me why ...16:16
otherwiseguyI can add skipTest if the package isn't there. But the gate probably should check it. Ideally, the ovs python bindings could be pushed to PyPI, it's just that they currently don't use setuptools, etc. and also have a file generated by ./configure that hard-codes a path from the build process of the openvswitch package.16:16
anteayattx per review in git repo, yeah that makes sense16:16
annegentlettx: also it looks like ReleaseNotes is a top 3 page based on analytics16:16
otherwiseguyfungi: it's in the distros, just no PyPI.16:16
*** dimsum__ has joined #openstack-infra16:17
reedannegentle, I suspect because those come from at release time16:17
annegentlettx: maybe an image in the lower corner16:17
*** sarob_ has joined #openstack-infra16:17
annegentlereed: yeah that could be, and the date span was around a release date16:17
ttxannegentle: that's because it's listed from the Launchpad pages. You can find them under "the list of openstack releases". Maybe we can add "release notes" there16:17
mordredotherwiseguy: wow. that doesn't make me happy16:17
mordredotherwiseguy: I especially like "hard-codes a path from the build process"16:18
fungiotherwiseguy: you're wanting to use them in unit tests or in a devstack context (dsvm-based functional tests, integration testing, upgrade testing, et cetera)?16:18
annegentlettx: I do think your headings are good16:18
clarkbmordred: and this isn't using cloud init because? let me guess the config drive isn't set up in a way that it understands?16:18
* ttx fixes16:18
otherwiseguyRight now, there are just functional tests covering the feature that uses python-openvswitch.16:18
annegentlettx: and the column/list layout is good16:18
mordredclarkb: nope - has nothign to do with that - it has to do with cloud-init being giant overkill for what we need - and requires installing all of the cloud-init python packages which, on test nodes, is a bit invasive16:19
fungiclarkb: also cloud init distro packages drag in conflicting distro packaged python libs which we try to avoid contaminating our job workers with, right?16:19
mordredclarkb: I mean, _also_ I don't know if upstream cloud-init has support for this yes16:19
clarkbmordred: can we virtualenv it? I just figure using standardish tools to do this is nice16:19
*** sarob has quit IRC16:19
mordredclarkb: I actively want to not have cloud-init on these machiens16:19
clarkbfungi: right virtualenv all the things :)16:19
ttxannegentle: reload16:19
mordredbecause it's giant overkill for what we need16:19
* anteaya imagines annegentle emptying her rifle and adding a new magazine16:20
clarkbmordred: sure but maintaining our own non standard less overkill seems like more work?16:20
fungiotherwiseguy: functional tests in a devstack context, or using some other framework?16:20
* anteaya spent too much time in texas16:20
annegentleheh, I'm so not a texan tho16:20
otherwiseguymordred: yeah, that's why I didn't just jump in and immediately try to get the OVS team to push things to PyPI. It seemed like a "patches welcome" kind of thing, and it was going to be a little tricky.16:20
annegentleMidwesterner through-n-through16:20
mordredclarkb: maybe - I usually find that if I ever have to debug a cloud-init problem the game is over16:20
anteayaannegentle: yeah non-texans in texas throw me16:20
annegentleanteaya: we're everywhere in Austin :)16:20
anteayaannegentle: I saw16:21
otherwiseguyfungi: no devstack, just directly imports neutron.agent.linux.ovs_lib and runs tests based on its methods with ovs-vsctl-based backend and the new ovsdb-based one.16:21
openstackgerritMarton Kiss proposed openstack-infra/infra-specs: Add migrate askbot to infra spec
annegentleanteaya: oh yeah, were you here?16:21
fungiotherwiseguy: in what job? or is this slated to be a separate job?16:21
annegentleanteaya: for midcycle meetups?16:21
anteayayes in san antonio for keystone and in austin for cinder16:21
otherwiseguyI believe it is the dvsm-functional one.16:21
anteayafound some good food16:21
openstackgerritMarton Kiss proposed openstack-infra/infra-specs: Add migrate askbot to infra spec
fungiotherwiseguy: okay, so devstack is handling the dependencies in the dsvm case16:22
anteayawent to the 6th street rackspace office to see iccha16:22
fungiotherwiseguy: in which case you want to get devstack support for installing that package worked out16:22
otherwiseguyeither that or just 'functional' or both.16:22
annegentleanteaya: cool!16:22
clarkbmordred: fair enough16:22
wenlockmordred, agreed...on debug cloud-init.... we love set -x -v and a good logfile function to capture it all during that phase.16:22
anteayawas a last minute thing, I had 2 hours before I had to go to the airport16:22
mordredwenlock: ++16:22
mordredof course, we also only need this on rax because no-dhcp ...16:23
clarkbmordred: though in my experience any cloud init problems are actually nova metadata failing to be reached16:23
fungiotherwiseguy: the dsvm jobs run on "devstack" workers, and we get info on what packages to pre-cache by "asking" devstack16:23
anteayafungi: which is where I had thought the venv might be applied16:23
clarkbmordred: which totally happens on a single router nova network 2 node devstack install with the aiopcpu jobs16:23
anteayabut won't seem to work in this case16:23
anteayaas it isn't via pip16:24
mordredclarkb: have I mentioned that I'm also glad that we don't need nova metadata service in this setup?16:24
clarkbwhich is why the stack at exists16:24
fungianteaya: devstack doesn't (currently anyway) install dependent libraries into a virtualenv, though tempest does16:24
wenlockclarkb we usually see it fail durring mounting of the cloud config volumen, openstack metadata seems to be consistent, so we try to auto gen the config into the user_data16:24
anteayafungi: interesting16:24
wenlockso you are not relying on config folder existing16:25
anteayafungi: might this be a case where the tempest virtualenv approach makes it into devstack?16:25
clarkbwenlock: huh, have had zero problems mounting the config drive iso thing here16:25
otherwiseguyfungi: ok. it'll be a little odd in that it is a configurable option which backend to use, so the package requirement would be dependent on that, but the test is written to test both backends, so we wouldn't want to be configuring devstack to just test the new ovsdb backend.16:25
* otherwiseguy isn't sure how all of that works.16:25
clarkbwenlock: we switched away from it in order to test live migration which does not work with config drive, but other than that is pretty reliable16:25
*** achuprin_ has joined #openstack-infra16:25
fungiotherwiseguy: i think the "let's minimize our test scenarios" discussion over the past year went the way of choosing the most common of each deployment options to build an official test scenario, and then adding less-common options in non-gate jobs (check pipeline pre-flight tests only, periodic jobs on the merged branch state, et cetera)16:27
clarkbmordred: fungi: if I could get reviews on the stack starting at that would be great. Trying to get multinode testing with nova network working again given the metadata madness16:27
fungiotherwiseguy: though sdague can probably state the position more eloquently16:27
*** Guest82511 has quit IRC16:28
*** scottda has left #openstack-infra16:28
mordredwenlock: yeah - we don't really do anything with userdata16:28
otherwiseguyfungi: Perhaps a good compromise would be, if we are installing openvswitch, go ahead and install python-openvswitch too.16:28
mordredwenlock: largely because it's flaky, and doing post-boot commands over ssh is consistent16:28
mordredand easier to debug16:29
wenlockmordred +1 cloud init should be small16:29
mordredclarkb: will do - let me finish dumping my state in to files and then I'll review16:29
fungiotherwiseguy: perhaps. though we do have to be fairly careful with distro packages of python libraries since they tend to drag in distro packages of other python libraries they depend on, some of which can conflict with the dependencies we're installing from pypi16:29
clarkbmordred: thank you16:30
otherwiseguyfungi: python-openvswitch was written by C developers. They even re-implemented their own json parser. :p16:30
fungiotherwiseguy: so care should be exercised to only install it if it's going to be tested, in an effort to avoid contaminating the environment of other jobs16:30
jrollclarkb: mordred: reading back, we have patches and builders for cloud-init to handle this stuff, see e.g. and similar repos on JayF's github there16:31
otherwiseguySo I don't think it has any external dependencies.16:31
dteselkinHi, I found an issue in my nodepool installation, could anyone help?16:31
fungiotherwiseguy: confirmed, for ubuntu trusty at least, just depends on the python interpreter16:31
clarkbmordred: an amazing thing just happened. I can pay $20/month for ESPN and friends without getting cble16:31
jrollclarkb: mordred: "this stuff" being "parsing network metadata and applying it"16:31
clarkbmordred: now I just have to figure out if I can stream on linux16:32
anteayadteselkin: we really like to see stacktraces and error messages, could you put something on
fungiotherwiseguy: anyway, to get back to the point, if it's going to be used in a devstack-based (dsvm) job, the dependency needs to be added in devstack not directly in our node build configuration16:32
*** derekh has quit IRC16:32
otherwiseguyfungi: ok, I'll write up a devstack patch. Thanks for the help.16:32
fungiotherwiseguy: so you probably want to re-hash the discussion with the devstack developers in #openstack-qa16:32
*** afazekas has quit IRC16:33
*** gyee has joined #openstack-infra16:33
dteselkinThe problem is that under a 'high load' (I have about 10 of jobs in a queue, because all workers are busy) when one job finishes, it immediately assigned a new job, and after a minute or so it is removed by nodepool. In that case the second job fails with Channel$OrderlyShutdown error from Jenkins16:33
pc_mHi folks! Could use some guidance getting coverage tests to run in the VPN repo. Can anyone help/advise?16:33
mordredclarkb: nice!16:33
*** ayoung has joined #openstack-infra16:33
openstackgerritMerged openstack-infra/storyboard: Prevent duplicate register_options call
dteselkinHere is the part of debug.log
mordredjroll: yah - it's honestly largely that I don't want to install all of cloud-init into the images when we use none of the features except for setting network info16:34
fungidteselkin: you need to set the offline when complete parameter from zuul16:34
*** naggappan has quit IRC16:34
jrollmordred: right, I'm not saying cloud-init is the right thing to do, just pointing out that it can do this :)16:34
mordredjroll: I think that's the right choice for the generic provider thign16:34
mordredjroll: woot16:34
fungidteselkin: that will make the jenkins-gearman plugin immediately offline the slave before it can grab a second job16:34
*** rmcall has joined #openstack-infra16:35
dteselkinAh, and then it'll be removed in background ?16:35
*** flashgordon is now known as jogo16:35
dteselkinThis option should be modified in, right?16:35
clarkbmordred: the really quick answer is no, looks like you have to install a thing on windows or OSX (android, ios, roku also currentl supported)16:35
fungidteselkin: nodepoold will still unregister it from jenkins (as long as the jenkins zeromq publisher plugin is working and nodepool is connected to it)16:35
fungidteselkin: it can be set via a parameter function that way, yes16:36
clarkbI wonderif my old roku will run the thing16:36
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add check/gate jobs for the networking-odl project
anteayapc_m: what have you done so far?16:36
dteselkinfungi, thanks a lot! I'll try now.16:36
fungidteselkin: the way we do it is perhaps slightly overcomplicated because we have some jobs which run on persistent workers where we don't want that set, and others which run on ephemeral workers where we do want that set16:36
pc_manteaya: Well, I just tried to run "tox -e cover" in the neutron-vpnaas repo.16:36
pc_manteaya: When it runs, I get a warning that says "Module neutron-vpnaas was never imported." and no data is collected.16:37
pc_manteaya: This warning is from
*** shashankhegde has quit IRC16:38
dteselkinfungi, it looks like I'll have to set that option based on, so doesn't look too complicated16:38
pc_manteaya: HenryG suggested trying to pull off the lockutils-wrapper on the command, so it is the same as Neutron's, but that makes no difference.16:38
anteayaah okay, let's start by looking at the tox file for neutron-vpnaas16:38
clarkbpc_m: I think you need to specify on setup.cfg what the name of your module under coverage is16:39
*** rmcall has quit IRC16:39
clarkbor maybe thats is in the config16:39
anteayapc_m: I don't know if I have seen a bigger tox file:
clarkbeither way its looking for neutron-vpnaas to be covered and that doesn't happen16:39
*** dizquierdo has quit IRC16:39
pc_manteaya: I also changed .coveragerc16:40
*** rmcall has joined #openstack-infra16:40
clarkbpc_m: you updated the source = neutron line?16:40
pc_mclarkb: Yeah, tried neutron-vpnaas, neutron_vpnaas, and /opt/stackl/neutron-vpnaas16:41
*** derekh has joined #openstack-infra16:41
clarkbI think it should be neutron_vpnaas16:41
pc_mclarkb: I'll try that again, with a omit line of neutron_fpnaas/tests/*16:42
*** scheuran has quit IRC16:42
pc_mclarkb: Does that sound right?16:42
pc_manteaya: Neutron's tox file is bigger :)16:42
*** talluri has quit IRC16:43
anteayapc_m: I believe it16:43
clarkbI don't think you need the omit line to get this working initially (and I personally prefer to see that tests are being covered too and don't understand why openstack as a whole has mostly decided to omit that coverage data)16:43
anteayapc_m: I've just never viewed it16:43
pc_mclarkb: I thought the intent was to run the tests and see if the tests cover all the production code.16:44
*** ashleighfarnham has joined #openstack-infra16:44
*** rwsu has joined #openstack-infra16:44
clarkbpc_m: sure, but shouldn't you also be checking that the tests are being exercised too?16:44
*** ashleighfarnham has quit IRC16:44
clarkbpc_m: you can still check that the "production" code is covered...16:44
*** ashleighfarnham has joined #openstack-infra16:44
pc_mclarkb: true16:44
pc_mclarkb: So I ran with neutron_vpnaas as source (and had the omit line).16:45
fungiright, if your tests are completely run as part of running your tests, they get 100% coverage16:45
pc_mIt still says...16:45 warning: Module neutron-vpnaas was never imported.16:45
* pc_m note the dash vs underscore16:45
anteayapc_m: can you paste your .coveragerc file for us?16:45
clarkbits possible that and pbr are overriding us here16:46
*** yfried_ has quit IRC16:46
pc_manteaya: sure:
*** unicell has quit IRC16:46
clarkbpc_m: in your tox.ini command add --coverage-package-name=neutron_vpnaas16:47
clarkbI think pbr may be overriding the default value16:47
*** Longgeek has joined #openstack-infra16:47
*** nicodemos has joined #openstack-infra16:47
pc_mclarkb: trying... Here is the tox.ini now
*** afazekas has joined #openstack-infra16:50
clarkbthat looks right16:51
pc_mclarkb: Ah! No warning, and I have coverage files. That was it. Thanks!16:52
pc_mSo I should remove the omit line too.16:52
clarkbThats up tp you16:52
clarkbI am in the minority with my opinion tests should be covered16:52
pc_mclarkb: I could comment it out and then, if someone doesn't want it they can uncomment.16:53
anteayapc_m: glad you are on your way16:53
pc_mclarkb: I did have one other question...16:53
pc_mclarkb: I tried the docs command, and it seems like I get an index.html file that is pretty much empty.16:54
pc_mclarkb: Should this have something in it, like API definitions or something?16:54
clarkbdo you have an index.rst?16:54
clarkbyou have to fill that in16:54
*** vigneshvar has joined #openstack-infra16:55
pc_mclarkb: Where is that located?16:55
pc_mclarkb: I've never run this before, but did once in Neutron and there were several web pages created.16:56
openstackgerritamrith proposed openstack-infra/project-config: Making gate-trove-functional-dsvm-mysql a gating test
*** timcline has quit IRC16:57
*** jasonsb has quit IRC16:58
anteayapc_m: here is the index file from neutron:
anteayapc_m: it has a bit more in it, which is added via a patch using gerrit16:59
pc_mclarkb: anteaya: Thanks. There is one, and it looks like the Neutron one that clarkb posted.16:59
*** tqtran has joined #openstack-infra17:00
anteayapc_m: clarkb posted the neutron-vpnaas one17:00
pc_mmakes sense why it's the same :)17:00
anteayaso if you wanted it to have more content, all you have to do is add more content17:00
*** marun has joined #openstack-infra17:00
anteayayou are a man for details17:00
*** sarob_ is now known as sarob17:01
pc_mclarkb: anteaya thanks, I'll look at the neutron one and see if there is stuff that makes sense to add here.17:02
*** hdd has joined #openstack-infra17:02
*** coolsvap_ is now known as coolsvap17:02
anteayasounds good17:02
*** joesavak has quit IRC17:02
*** amotoki has quit IRC17:03
*** timcline has joined #openstack-infra17:03
*** annegentle has quit IRC17:03
*** MaxV has quit IRC17:03
pc_mclarkb: anteaya Thanks for all the help!17:03
anteayaof course17:04
SpamapScrinkle: Hi, I saw your pull request to solve the rspec issue. Is there anything you can think of that I can try in my chroot that shows the bad behavior to maybe explain why it's breaking for me and not others?17:04
SpamapScrinkle: oh and thanks for working on the patch. :)17:04
*** enikanorov has joined #openstack-infra17:05
*** ihrachyshka has quit IRC17:06
fungiSpamapS: i wouldn't say it's not breaking for others, you're just the first to notice and start looking into it17:07
SpamapSfungi: mordred asserts it is not breaking for him.17:07
openstackgerritMatthew Treinish proposed openstack-infra/devstack-gate: DO NOT MERGE COPY ALL OF .TESTREPOSITORY to log server
fungiSpamapS: he said in scrollback he was able to reproduce it, just came later in the build than he expected17:07
SpamapSThats not something he said to me. :-/17:08
*** skolekonov has quit IRC17:08
fungiSpamapS: or for another potential example, we haven't successfully built any new bare-.* worker snapshot images from in ~170 hours, but i don't think any of us have looked into an underlying cause yet17:08
SpamapSSo is the best workaround to just add a patch to  gem install --version ~>2.99.0  before the puppet call?17:09
fungiSpamapS: anyway, from scrollback, "20:36 mordred SpamapS, nibalizer: I have hit your problem - it comes way later in the process than I was expecting"17:09
* SpamapS curses irssi17:10
*** Longgeek has quit IRC17:10
SpamapSyou know maybe its time to start using a bouncer17:10
clarkbfungi: wait really? seems like I started last week fixing those builds, they must've run once then promptly broken again17:10
*** jistr has quit IRC17:10
clarkbfungi: I will look into that now17:10
*** rwsu has quit IRC17:10
fungiclarkb: tie to place bets on rspec version matching errord for puppetlabs_spec_helper17:11
fungier, time17:11
clarkbits a reasonably good bet17:11
*** rwsu has joined #openstack-infra17:11
*** harlowja_at_home has quit IRC17:11
*** armax has joined #openstack-infra17:12
fungialso, i have yet to be able to reproduce the git timeouts on centos6. i switched to git fetch --all on nova and have been looping that most of the weekend17:12
SpamapShopefully next time I run into this I'll actually have enough context not to just scratch my head and be confused for 4 days17:13
clarkbfungi: maybe you need to do updates for many repos like zuul cloner does?17:13
*** isviridov_away is now known as isviridov_vacati17:13
SpamapSWhy does Ubuntu have to have new cloud images every damn day?17:13
*** isviridov_vacati is now known as isviridov_vction17:14
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add elements for Infra servers
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add static network config element for Rackspace
mordredSpamapS: this is one of the reasons I'm looking forward to that ^^17:14
mordredSpamapS: and ceasing to consume other people's base images17:14
openstackgerritClint 'SpamapS' Byrum proposed openstack-infra/project-config: Add element for bare nodes
openstackgerritClint 'SpamapS' Byrum proposed openstack-infra/project-config: Work around rspec dependency issues
mordredSpamapS: also - yeah - I did hit your issue17:14
clarkbfungi: Error: Could not update: Execution of '/usr/bin/gem install --no-rdoc --no-ri puppetlabs_spec_helper' returned 117:15
clarkbfungi: we have a winner17:15
SpamapSmordred: yeah I didn't see that because I left irssi in this window and forgot to /lastlog17:15
*** EmilienM is now known as EmilienM|afk17:15
fungiclarkb: SpamapS wins17:15
SpamapS^^ 154133 is working for me17:15
SpamapSbut I'm chasing the patch before it17:15
SpamapSso dunno if that helps you all :)17:15
clarkbSpamapS: not really, we need it in nodepool/scripts too17:16
mordredSpamapS: also - since you know upstart stuff, I'm going to assume you also know systemd stuff ... has both in it17:16
fungiSpamapS: clarkb: on a positive note, we can surmise that puppetlabs_spec_helper broke themselves <170 hours ago17:16
mordredSpamapS: at least it wasn't just you - there actually was a broken thing17:17
fungisince we know it to have been working at that point17:17
SpamapSOpenStack: a project created to find bugs in _every single library on the planet_17:17
clarkbSpamapS: if you add that same thing to nodepool/scripts/ and rebase it, we can merge17:17
fungishould totally be our new project slogan17:17
SpamapSmordred: so your is interesting. I wonder if we should just add that to os-net-config.17:18
SpamapSprobably not similar enough to be a quick win17:18
clarkbmordred: status update on espn without cable, I need a newer roku. I think the lesson here is I should give up on technology working the way I expect it to17:18
SpamapSclarkb: Roku 3's are on sale right now17:18
SpamapSand are fantastic17:19
*** jpich has quit IRC17:19
clarkbSpamapS: the refurb special?17:19
SpamapSheadphone jack in the remote is a nice feature for those of us who watch zombies devour people whilst children sleep in the next room. ;)17:19
mordredSpamapS: I'm glad you think it's interesting - it kinda makes me sad - but at least it doesn't require installing libraries on the system17:20
clarkbI just want to be able to watch sportsball17:20
mordredclarkb: yah17:21
SpamapSclarkb: youre team might score a goal unit point without you!17:21
SpamapSmordred: oh wow static-network-waiter is getting into the deepest evil of upstart17:22
*** ghostpl has quit IRC17:23
mordredSpamapS: I got that logic from a post that you made on a forum, btw17:23
SpamapSmordred: yes it looks quite familiar and is probably right. ;)17:24
*** jerryz has joined #openstack-infra17:24
*** otter768 has joined #openstack-infra17:24
*** btran has joined #openstack-infra17:24
mordredbut I think that will do exactly what we want- which is "please run this script before doing anything with the network including starting dhcp or anything"17:24
openstackgerritRicardo Carrillo Cruz proposed openstack-infra/storyboard: Jsonify the result from get_original_resource
*** btran has quit IRC17:24
*** btran has joined #openstack-infra17:25
*** hashar has quit IRC17:26
clarkbBTW I -1'd 153921 last night. Quick sanity check on my message is probably worthwhile17:26
clarkbfungi: ^ its ipv6 related you are likely interested17:26
SpamapSmordred: yes, before I got sucked into the Juju vortex I was trying to rally people around adding a few hook-points like this so that people don't have to know upstart's internals to get obvious stuff done.17:26
fungiclarkb: thanks, looking17:26
SpamapSmordred: moot now. You seem to have cargo culted correctly. Let us hope that the cargo cult has served us well. :-P17:26
nibalizerSpamapS: so are you still busted?17:26
nibalizeri think we decided the issue was an rspec dependency in p_s_h17:27
openstackgerritDouglas Mendizábal proposed openstack-infra/project-config: Use python-jobs template for barbicanclient
*** crc32 has joined #openstack-infra17:27
SpamapSnibalizer: no, installing rspec --version '~>2.99.0' before running puppet fixes me.17:28
*** mpaolino has quit IRC17:28
*** otter768 has quit IRC17:28
mordredSpamapS: my next step is to make an image and upload it and see if it boots17:28
*** zigo has quit IRC17:29
mordredSpamapS: and then to also upload it to HP and verify that it does not break the existing config17:29
anteayamorning zaro17:31
clarkbzaro: is this storm hitting you up north too?17:31
clarkbzaro: winter is very spring like this year17:32
zaroclarkb: rain, rain, and rain.17:32
*** ddieterly has quit IRC17:33
zaroyeah, good thing is not too cold17:33
*** ddieterly has joined #openstack-infra17:33
*** KanagarajM has quit IRC17:33
*** HeOS has quit IRC17:35
*** ddieterly has quit IRC17:36
*** ddieterly has joined #openstack-infra17:36
* mordred looks outside at the snow17:36
* mordred doesn't look too hard, lest anteaya show him up ...17:37
anteayasnow is good17:38
anteayacold is good too17:38
*** unicell has joined #openstack-infra17:38
SpamapSmordred: +1'd with a few thoughts on long term improvements.17:38
anteayawar stories available if you want em17:38
* SpamapS looks outside at the .. well.. lets just say, at the LA..17:39
mordredSpamapS: yay!17:39
openstackgerritMerged openstack-infra/project-config: gnocchi: add swift and ceph functional jobs
clarkbmordred: question, how does your config drive stuff avoid the default rax is gonna write stuff to your files method of configuring networks?17:40
clarkbmordred: is config drive XOR'd with that method or do we need to handle potential races there?17:41
*** krtaylor has quit IRC17:41
mordredthe "they write stuff to your files" thing is nova-agent17:41
mordredso with this, we will not install nova-agent17:41
fungiit replaces what rax is doing17:42
mordredand then we'll be happy17:42
fungishouldn't race17:42
*** jlibosva has quit IRC17:43
zarofungi, clarkb, jeblair : i'm seeing connections issues between review-dev.o.o and the trove db.  it's basically this issue!msg/repo-discuss/rR_ouEeAQzU/bA70CYLT0t0J17:44
SpamapSOh rax stopped doing super crazy evil file injection?17:44
zarowondering if you guys see on review.o.o?17:44
openstackgerritSteve Martinelli proposed openstack/requirements: Add oslo.policy to projects.txt
fungiSpamapS: rax does super crazy evil file injection by preinstalling a custom userspace daemon which writes what they tell it at boot17:45
clarkbzaro: no we haven't seen it with review.o.o or at least I haven't noticed ithappening. What should we be looking for to detect it?17:45
zaroit's probably not hapenning on review.o.o because it's so busy.17:45
clarkbzaro: that could be17:45
jeblairzaro: weren't you looking into that happening every few minutes on review-dev?17:45
zaroSo review-dev.o.o is still running Gerrit 2.8.4 with same settings to db as review.o.o17:46
jeblairzaro: with 2.917:46
zarowe'll it doesn't look like a 2.9 issue at all since it's happening with 2.8 as well17:46
*** Ryan_Lane has joined #openstack-infra17:47
SpamapSfungi: thats just crazy evil ... super crazy evil was mounting the backing file and writing /etc/network/interfaces17:47
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for grenade bug 1419913
openstackbug 1419913 in grenade "Grenade fails with ContextualVersionConflict between oslo.config and tempest-lib on juno side" [Undecided,New]
*** mfink__ has joined #openstack-infra17:47
zarojeblair: the only way i was able to fix last time was to set connectionpool=false in gerrit.config17:47
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add static network config element for Rackspace
clarkbSpamapS: thats what I thought they were doing17:47
clarkbSpamapS: which is why I asked the question.17:48
mordredSpamapS: I tried your less-bad method of detecting upstart/systemd above ^^17:48
fungiSpamapS: i think they decided 1) to do that you need your host to understand the filesystem of your guests, and 2) with a daemon you can also change it while running (which we'17:48
fungive seen them do to us before)17:48
jeblairzaro: yeah.  so if it times out after 12 hours, that's because it's not busy and fine.  if it times out after 5 minutes like you saw with 2.9 on review-dev, that's something we need to understand before we go into production17:48
mordredif only there was a dynamic host config protocol17:48
mordredwith a standard daemon that shipped with the distros that could be used!17:49
fungimordred: you should propose an rfc17:49
mordredfungi: I really should17:49
clarkbif only17:49
mordredfungi: otoh - Tom Preston-Warner thinks that the yaml spec is too long to read which means it's sane to write an ini-style replacement for it17:49
SpamapSmordred: and a 30+ year history of working and scaling to tens of thousands of nodes. Yes. that would be helpful.17:49
jeblairmordred: you should call it "RFC 2131".17:49
mordredfungi: so it probably wouldn't stop much17:49
mordredmaybe if the host config protocol used TOML for communication it would be more acceptable ...17:50
jeblairor maybe even "RFC 1531"17:50
SpamapSoh durn, DHCP is only 23 years old. My guesstimate was way off.17:50
*** mfink_ has quit IRC17:50
jeblairSpamapS: however, it is older than rackspace17:51
zarofungi: yeah, it's much sooner, it's about every 3 mins.17:51
openstackgerritDan Prince proposed openstack-infra/project-config: Add puppet tripleo check tests
SpamapSWell if it's only 23 years old.. it can't possible be useful or proven.17:51
*** mfink_ has joined #openstack-infra17:51
openstackgerritMaru Newby proposed openstack-infra/project-config: Set neutron jobs to use devstack-gate testr processing
fungimordred's rule of specifications: "the specification for any moderately complex system will be sufficiently long as to make at least some less clueful individuals to think they can replace it in less time than it will take to fully read and understand it (see also dunning-kruger effect)"17:52
zarojeblair: we are currently using the mysql driver from ubuntu package repo.  i'll see if using the specific one that gerrit downloads makes any difference.  although i think i tested that last time and it didn't.  will try again just to make sure.17:53
openstackgerritMaru Newby proposed openstack-infra/project-config: Set neutron jobs to use devstack-gate testr processing
*** mfink__ has quit IRC17:54
jeblairzaro: it's not worth addressing the problem if you are only seeing the 12 hour timeouts17:54
*** krtaylor has joined #openstack-infra17:54
mordredfungi: *sob*17:54
jeblairzaro: only debug this if you are seeing < 5 minute ones17:54
zarojeblair: ohh, yeah, it's like 3 mins.17:54
clarkbsdague: 153811 is functional testing for ec2api and your change was for tempest?17:54
clarkbsdague: trying to understand if there is overlap here or not17:55
jeblairzaro: so you are seeing 3 minute timeouts with 2.8?17:55
zarojeblair: i think this problem is already in production just that we don't see it17:55
jeblairzaro: (same version as production?)17:55
sdagueclarkb: so eventually the tempest job goes away17:55
zarojeblair: yes,17:55
sdaguebut right now we do have ec2 code in nova, and ec2 tests in tempest17:55
zarojeblair: review-dev still running same version as prod17:55
jeblairzaro: hrm.  i suppose that's possible17:55
sdagueso that's a good backstop17:55
otherwiseguysdague: responded via the python-openvswitch review, but this won't break anything because neutron's dvsm-functional tests already run with sitepackages=True.17:55
clarkbsdague: gotcha17:56
otherwiseguyAlso, the test will do skipTest if the package isn't installed, just in case.17:56
jeblairzaro: we might be able to check very early saturday morning logs and see if there are any requests > 3 mins apart.  though we have people running gertty, so that seems unlikely.17:56
sdagueotherwiseguy: that gets dragged in for devstack in general17:56
*** jcoufal has quit IRC17:57
jeblairzaro: i accept your hypothesis that production never sees requests more than 3 minutes apart. :)17:57
otherwiseguysdague: yes, but it is a package that has no other dependencies and is very small. It doesn't hurt anything.17:57
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for busted requirements on juno bug 1419919
openstackbug 1419919 in Keystone "gate-requirements-integration-dsvm fails on stable/juno with ValueError for missing 2014.2.3" [Undecided,New]
fungijeblair: zaro: i too find that entirely believable17:57
otherwiseguyAnd only being installed when openvswitch is otherwise being installed.17:57
zarojeblair: yeah, not sure how to debug on prod.17:57
sdagueotherwiseguy: how does it impact a world where stable/juno openstack with neutron is installed in a venv ?17:57
*** maxbit has joined #openstack-infra17:58
sdagueand all the services are run out of that17:58
zaroi can set connectionpool=true on review-dev and just continue to use it for 2.8 and migration to 2.917:58
zaronot sure what the adverse affect would be though.17:59
otherwiseguysdague: I don't see how it would affect it at all, other than a small extra package being installed. The OVSDB native version is optional and non-default, and tests would pass because of the skipTest.17:59
*** dmorita has joined #openstack-infra17:59
*** timcline has quit IRC18:00
clarkbSpamapS: were you planning to update 154133 to update nodepool/scripts/ as well?18:00
otherwiseguysdague: It is basically impossible to even test the patch via CI without this change.18:00
sdagueotherwiseguy: so neutron outside of functional tests *never* uses this code path?18:00
clarkbotherwiseguy: neutron functional tests could do the install themselves18:01
zarojeblair: oops i mean connectionpool=false since it's currently set to true now.18:01
otherwiseguysdague: unless it is specifically configured to do so with new option :ovsdb_interface=native.18:01
otherwiseguyThen it is never even imported.18:01
*** sarob is now known as sarob_18:01
*** dtantsur is now known as dtantsur|afk18:01
*** sarob_ is now known as sarob18:01
fungisdague: also it would at the moment be impossible to run a service using this from a virtualenv (at least without turning on sitepackages) since it doesn't use setuptools at all18:02
otherwiseguyclarkb: I'm not sure that trying to re-implement devstack's handling of installing dependencies for multiple distros is really the right thing for a functional test to do.18:02
jeblairzaro, fungi: do we have old review-dev around still?18:02
jeblairzaro, fungi: the precise one?18:02
sdagueotherwiseguy: so I'm with clarkb, I think the neutron functional tests should just do the install. Because the implications for how this interacts with normal devstack is kind of big18:02
fungijeblair: yes, i have not deleted it yet18:02
fungijeblair: lemme get you the ip address18:02
jeblairzaro, fungi: can we put 2.8 back on that and see if this problem appears?18:02
sdaguefungi: yeh, exactly18:03
sdaguewhich is a direction we're headed down18:03
*** yamahata has joined #openstack-infra18:03
jeblairzaro, fungi: i could be convinced that this affects prod now and is harmless because we are busy; but also, i don't remember it happening with 2.8+precise+dev (only the 12 hour timeout).  so i'm still a bit worried18:04
otherwiseguysdague: I just don't see how that would work other than doing some kind of testing of what distro we were using in the test itself and then running yum/apt during the test. Sounds crazy.18:04
*** annegentle has joined #openstack-infra18:04
*** harlowja has joined #openstack-infra18:04
sdagueotherwiseguy: that's why d-g has a hooks interface that lets you do this at different places18:05
SpamapSclarkb: I was not planning on that no, was kind of just throwing it out there as a "this works for me" .. if you want me to circle back to it I will do so before EOD today (context switched at the moment)18:05
*** derekh has quit IRC18:05
fungijeblair: zaro: old review-dev is 2001:4800:780d:509:3bc3:d7f6:ff04:395418:06
clarkbSpamapS: no thats fine, do you care if I end up rebasing and pushing a new patchset to your change?18:06
clarkbotherwiseguy: devstack and devstack-gate both have support for figuring out the package manager18:06
fungizaro: obviously you'll want to stop the gerrit service on new review-dev before starting it on old review-dev18:06
clarkbotherwiseguy: it basically boils down to if apt-get then apt-get update && apt-get install foo elif yum then yum install foo18:07
*** afazekas has quit IRC18:07
openstackgerritMerged openstack-infra/elastic-recheck: Add query for grenade bug 1419913
openstackbug 1419913 in grenade "Grenade fails with ContextualVersionConflict between oslo.config and tempest-lib on juno side" [Undecided,Confirmed]
zarofungi: i'm on it18:08
otherwiseguyclarkb: ok, so then what do I need to do to get python-openvswitch installed when the neutron-dvsm-functional test runs?18:09
*** e0ne has quit IRC18:09
openstackgerritDavanum Srinivas (dims) proposed openstack-infra/project-config: Update magnum jobs publishing docs
openstackgerritMerged openstack-infra/project-config: Add functional test jobs for EC2 project
clarkbopenstackgerrit: in that job, just install the package before anything that depends on it runs (probably in the pre test hook)18:11
clarkbotherwiseguy: ^18:11
*** timcline has joined #openstack-infra18:13
openstackgerritMerged openstack-infra/project-config: Use python-jobs template for barbicanclient
*** carl_baldwin has quit IRC18:14
marunotherwiseguy: link to review?18:14
otherwiseguyclarkb: So I have infra telling me to modify devstack, and devstack telling me to modify the hooks for tests?18:14
*** armax has quit IRC18:14
yjiang5_Hi, all, my multi-host devstack environment failed because will die devstack if I have my controller node in another host. I remember devstack should support multiple host enviornment, right?18:14
*** melwitt has joined #openstack-infra18:14
marunotherwiseguy: I'd like to ensure that what you're proposing won't break
clarkbsdague: I am beginning to wonder if all the neutron driver project things will need the same sort of job as the odl project and if we should make a single template for that job?18:14
clarkbmestery: ^18:14
clarkbotherwiseguy: ?18:15
openstackgerritMerged openstack-infra/elastic-recheck: Add query for busted requirements on juno bug 1419919
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: Email base plugin.
openstackbug 1419919 in Keystone "gate-requirements-integration-dsvm fails on stable/juno with ValueError for missing 2014.2.3" [Undecided,New]
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: Email SMTP Sender
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: [WIP] Email Send Cron Worker
*** bhuvan has joined #openstack-infra18:15
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: Email Working Directory Utility Method
anteayaI'd add marun to the list of folks to consult on the creation of a template18:15
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: Email Outbox
marunsdague, clarkb, dtroyer: btw, your eyes would be appreciated on my work to automate host config for functional testing:
fungiyjiang5_: you might want to ask the devstack developers in #openstack-qa about that change18:15
otherwiseguymarun: the devstack review is here: I don't care how it gets done, all I care about is that python-openvswitch gets installed before the test runs. :)18:16
*** hdd has quit IRC18:16
* otherwiseguy is not that familiar with the various test hooks, etc.18:16
openstackgerritMerged openstack/requirements: Bump tempest-lib min version to 0.2.0
yjiang5_fungi: oh, sorry for wrong channel. I thought this is for devstack. Will do that.18:16
fungiotherwiseguy: the hooks do modify devstack by way of being run within devstack-gate18:16
mesteryclarkb: ++18:16
mesteryclarkb: Although, not the vendor ones.18:16
fungiotherwiseguy: you were originally asking why can't nodepool preinstall python-openvswitch on all our job workers for you18:16
mesteryclarkb: The vendor ones require proprietary crap to run, the open source ones like ODL do not18:16
mesteryclarkb: Make sense?18:17
marunotherwiseguy: I've proposed a change that would break you -
fungiotherwiseguy: and i didn't tell you to submit a patch to devstack, i suggested you go ask the devstack developers how to go about it18:17
mesteryclarkb: In other words, you're never going to run NSX in the gate, or Cisco Nexus, or Arista, etc.18:17
otherwiseguyfungi: fair enough.18:17
marunsdague: What if we were to make the addition #testonly?18:18
marunotherwiseguy: this is only required for testing, right?18:18
clarkbmestery: sure, so maybe when the second open source one comes along we can templatize your proposed job a bit more18:18
otherwiseguyfungi: this is all stuff that I unfortunately have very little experience with. I just have patches I've been working for months on that have a bit of a wrinkle that are holding them up. :)18:18
*** _nadya_ has quit IRC18:18
mesteryclarkb: Ack18:18
otherwiseguymarun: yes. just for testing.18:19
*** armax has joined #openstack-infra18:19
marunotherwiseguy: ok, mark it #testonly.  there are examples of how this should be done in the same file.18:19
sdaguemarun: it still impacts the whole system, the fact that it's a python package which is not on pypi is a giant red flag18:19
fungiotherwiseguy: the last 1% is always the hardest ;)18:19
marunsdague: hmmm18:19
sdagueand breaks other assumptions we are about to make18:19
clarkbsdague: mestery fwiw it would be nice if the enable_plugin machinery could update the enabled services list18:19
otherwiseguyfungi: so unfortunately, even when people tell me what to do I don't always completely understand.18:19
marunsdague: what assumptions do you mean?18:19
otherwiseguyThere's a lot of...infrastructure.18:20
mesteryclarkb: In devstack-gate?18:20
clarkbmestery: no in devstack18:20
sdaguemarun: that we can run the services in a venv, for instance18:20
marunsdague: ...18:20
* anteaya comiserates with otherwiseguy18:20
marunsdague: with site_packages=True or False?18:20
clarkbmestery: I assume that if you enable the odl plugin in devstack that should imply the odl services are enabled18:20
mesteryclarkb: Ah, got it. It could, but ofr ODL, it's complicated as we have 3 services and we don't want to enable htem by default because it depends if you node is a control node or not.18:20
otherwiseguyanteaya: thanks. :)18:20
mesteryclarkb: Not always, but for gate testing, yes.18:20
sdagueclarkb: it's expected that plugins do that in their settings file18:21
anteayaotherwiseguy: you are not alone18:21
sdaguemarun: with site_packages=False18:21
marunsdague: though tbh I'm not sure I recall why we're doing that :/18:21
clarkbsdague: ok, in mestery's change its part of the job def18:21
marunsdague: I guess we'll need to verify that our functional testing can work with that.18:21
sdagueso that we can actually test new clients with old servers18:21
fungioh, but this is a case of a test using python-openvswitch, not neutron using python-openvswitch right? (as in neutron just interacts with openvswitch cli/configs directly not via this python lib?)18:21
mesterysdague clarkb: We can't do it in the settings file. ODL has 3 different options to enable, it depends on the type of node which is running.18:21
sdagueand have the old servers use old versions of clients18:21
mesteryclarkb: Thus, we're doing it in the job.18:21
sdaguemestery: .... ?18:21
marunotherwiseguy: let's work together to figure this out18:21
sdaguemestery: show me your devstack plugin18:21
marunanteaya: which template are you referring to?18:22
*** sarob has quit IRC18:22
sdaguemarun: so for this, please just make it a d-g pre hook or something, it shouldn't be bad to do it there18:22
*** luqas has quit IRC18:22
mesterysdague: odl-server means "this node is goign to run ODL in devstack", odl-compute means "this node is a compute node controlled by ODL", and odl-neutron means 'this node is running neutron but ODL is not running on the node"18:22
mestery3 cases, so we can't enable one by default18:22
marunsdague: Ok18:23
marunsdague: speaking of pre-hook, as per a previous comment, I would appreciate your thoughts on what I'm doing here:
*** hdd has joined #openstack-infra18:24
sdaguemestery: so... this plugin should change a bit18:24
marunsdague: I'm not sure that this approach is good long term, but I think it represents a possible starting point for something like a devstack plugin?18:24
anteayamarun: the template discussion clarkb proposed, the chat you are having now18:24
sdaguethe settings should turn on one of these modes by default18:24
mesterysdague: But hat's the point, we can't do that.18:24
marunsdague: admittedly I'll need to review the docs you've recently posted18:24
mesterysdague: Because we can't assume what to turn on.18:25
mesteryWe'd break people if we did that.18:25
sdaguemestery: I 100% disagree18:25
sdagueyou have to have a sane default18:25
mesterysdague: What? Why?18:25
sdaguewhich isn't "don't do anything"18:25
mesteryBut how do we allow for overrides?18:25
mesteryIf we enable odl-server for example, how do I let people override that?18:25
sdagueODL_MODE= ...18:25
sdagueor something18:25
sdagueanything people put in localrc is available here18:25
marcusvrnclarkb: anteaya: hello!! just to you guys take aware about my problem with CI optimization, I uninstalled the devstack requirements as you suggested and now it's working well :) (letting devstack install what it needs). Each build runs in about an hour... I think it's ok, isn't it?18:25
openstackgerritMerged openstack-infra/project-config: Set neutron jobs to use devstack-gate testr processing
mesterysdague: These are services, but I see your point.18:26
anteayamarcusvrn: sounds good to me18:26
mesterysdague: I think the problem is in the way we migrated this, I'll dig a bit and see what I can do.18:26
sdaguesure, but there should be something that works out of the box with no extra config18:26
anteayamarcusvrn: thank you for letting us know, glad builds are happening for you18:26
mesterysdague clarkb: thanks for your itme18:26
sdaguethen ways to tweak it18:26
mesterysdague: It will work out of the box for a single node deployment, but that's all it could work for.18:26
mesterysdague: If htat's ok, we'll settle on that.18:26
clarkbsounds perfect for a dev stack :)18:26
anteayamarcusvrn: the next optimization you might look at is using disk image builder to create yourself an image18:26
sdaguemestery: sure, but that's going to be 90% of the use case :)18:27
*** krtaylor has quit IRC18:27
*** shashankhegde has joined #openstack-infra18:27
*** talluri has joined #openstack-infra18:27
mesterysdague: lol :)18:27
anteayamarcusvrn: but keep your current workflow while you investigate disk image builder, as that has it's own hazzards18:27
mesterysdague: Reformulating now.18:27
sdaguemestery: maybe that's the fundamental impedance that hit us in the neutron devstack support ...18:27
clarkbwell SpamapS didn't say if he cared about me rebasing his change so I am going to assume its ok and do that now so that we can build iamges again :)18:28
mesterysdague: I think you're conflating two different things together here.18:28
sdaguefrom a devstack perspective the zero config case should be valid and produce a working system18:28
clarkbhrm actually wait maybe rebase isn't necessary since no dib on bare-* yet. just a nwe change. I will do that18:28
mesterysdague: In a model with a central controller, for a multi-node devstack you don't want the controller running on all the nodes.18:28
mesteryI had previously done this by having services you enable.18:28
sdaguemestery: sure18:28
mesteryBut clearly, this won't work with pluggable devstack.18:28
sdaguebut that's an advanced case18:28
mesterySo it just needs to be changed to a variable, as you say, and we go from there.18:28
mesteryIt's not advanced for hte users I'm trying to serve ;)18:29
mesteryBut we can make the single node work easily with this too18:29
sdagueit's an advanced case for most devstack deployments18:29
mesterySo it's a win-win18:29
sdaguewhich is how people evaluate a lot of openstack18:29
sdagueagree, I think there's a plan now18:29
marcusvrnanteaya: nice! I will take a look on it18:29
*** yamahata has quit IRC18:29
mesterysdague: Agreed, thanks for the talk through here.18:29
*** patrickeast has joined #openstack-infra18:30
openstackgerritMerged openstack-infra/project-config: Making gate-trove-functional-dsvm-mysql a gating test
mesterysdague: I'll reformulate those patches and hopefully we can come to a solid conclusion this week on this. Thanks again for the reviews!18:30
*** yamahata has joined #openstack-infra18:30
sdagueyep, no prob18:30
anteayamestery: give yourself time to learn it, so keep your builds happening while you experiement18:30
anteayasorry that was for marcusvrn18:30
*** doude has quit IRC18:32
marcusvrnanteaya: yep, I will do that18:32
*** markus_z has quit IRC18:33
zarojeblair, fungi : running review-dev on precise @ and seeing the same connection error in the log.18:33
anteayamarcusvrn: great18:33
crinkleSpamapS: sorry, was in meetings18:33
openstackgerritClark Boylan proposed openstack-infra/project-config: Fix rspec installs during image builds
clarkbfungi: crinkle SpamapS ^ that should fix it for the non DIB case18:34
clarkbI think18:34
crinkleclarkb: where/why is puppetlabs_spec_helper getting installed?18:34
clarkbcrinkle: its part of the puppet manifest for the "thick" nodes iirc. Let me get a link18:35
crinkleI know it is bundle installed for the puppet module tests but I don't see why it should ever get installed directly18:35
*** annegentle has quit IRC18:36
clarkbheads up the wind is picking up, I may end up semi useless today if the power goes out18:36
crinkleI would just take that out and let the individual tests install it18:38
crinkleunless it needs to be there ahead of time for a different reason18:38
clarkbya I am digging through git logs now to find out why it was added18:39
*** patrickeast has quit IRC18:39
clarkbIf6e06bf632deb0680869ed963c934ff8ee8a783f added it in 201218:40
mordredclarkb: I believe I agree with crinkle18:40
clarkblooks like it was added for the lint job18:40
mordredI think we installed it back when we still had long-lived test slaves18:40
clarkbmordred: ya 2012 would be back then18:40
*** jp_at_hp has quit IRC18:41
clarkbmordred: does our lint job not depend on it anymore?18:41
*** avozza is now known as zz_avozza18:42
clarkblooks likeit just runs rake18:42
*** patrickeast has joined #openstack-infra18:42
*** edwarnicke has joined #openstack-infra18:42
clarkbso we would likely need to update that job to install spec_helper first? then remove the install on all nodes?18:42
edwarnickeI am having difficulties with the CLA app as reached from the gerrit18:42
edwarnickeI fill out all the fields, and click the 'Submit' button18:42
edwarnickeAnd nothing happens18:43
edwarnickeThe button greys out18:43
*** achanda has joined #openstack-infra18:43
edwarnickeBut there is *no* traffic to the network18:43
edwarnickeAnd I have now tried with Safari *and* Chrome18:43
clarkbedwarnicke: did you fill in the I AGREE portion?18:43
edwarnickeDoes anyone know the state of the CLA app?18:43
edwarnickeclarkb: I did :)18:43
crinkleclarkb: each job should 'bundle install' first which should install puppetlabs_spec_helper if it's in the Gemfile18:43
edwarnickeI even *typed* it :)18:43
clarkbcrinkle: you assume we have gemfiles18:43
edwarnickeIf I type AGREE, I can click the submit button.18:43
crinkleclarkb: you should make gemfiles18:43
edwarnickeClicking the Submit button clears the AGREE field, and regreys the button18:43
clarkbcrinkle: I have no idea how to do that :)18:44
clarkbfwiw is the repo in question18:44
anteayaedwarnicke: did you type 'I AGREE'18:44
clarkbanteaya: yes confirmed. But it really does sound like the behavior when you leave that out18:44
mordredclarkb, crinkle: I also do not know how to make a gemfile - but it seems the other puppet tests do this bundle install step ...18:45
mordredso I could imagine that we should do somethign similar18:45
anteayamordred: bundler reads the gemfile18:45
edwarnickeanteaya: Ah... no, I typed 'AGREE'18:45
edwarnickeLet me try 'I AGREE'18:45
clarkbmordred: yup not saying we shouldn't do it, just trying to understand what is required18:45
*** coolsvap is now known as coolsvap_18:45
clarkbedwarnicke: itshould give you a verbatim string to enter if you do agree18:45
mordredclarkb: me too - I'm just saying it strangely18:45
openstackgerritMichael Krotscheck proposed openstack-infra/storyboard: Jsonify the result from get_original_resource
anteayathe gemfile is a list of gem dependencies18:45
clarkbedwarnicke: use the string that it asks for18:45
*** yfried_ has joined #openstack-infra18:45
anteayamordred: bundler installes the gems listed in the gemfile18:46
mordredso we'd just ahve a Gemfile with "puppetlabs_spec_helper" in it18:46
mordredand then bundle install18:46
mordredwould install puppetlabs_spec_helper?18:46
anteayaonly if puppetlabs_spec_helper is a gem18:46
mordredI think that it is18:46
crinkleit is18:46
anteayahosted someplace that bundler knows to look18:46
*** annegentle has joined #openstack-infra18:46
*** coolsvap_ is now known as coolsvap18:46
crinkleand it has puppet-lint and pupet-syntax as dependencies18:46
mordredbut we'd need to put one of those Gemfiles in to every one of our puppet repos I guess?18:47
anteayayou can also set a source= value somewhere18:47
*** devlaps has joined #openstack-infra18:47
anteayamordred: yes18:47
clarkbmordred: yes and make sure they all include puppetlabs_spec_helper18:47
*** vigneshvar has quit IRC18:47
mordredis there a reason to not just make a jjb macro that does "gem install puppetlabs_spec_helper" ?18:47
*** coolsvap is now known as coolsvap_18:47
crinkleit's hosted on the rubygems forge so you don't need a source=18:47
anteayamordred: every project needs a gemfile for its dependencies18:47
*** achanda has quit IRC18:47
anteayacrinkle: cool18:47
clarkbmordred: probably because its not very rubyy18:47
anteayawell it isn't very ruby, that is true18:48
fungicrinkle: any suggestions for how we can pre-cache that locally on the filesystem to avoid having it hit the network when the job runs?18:48
* mordred defers in all ruby judgement to the rubie18:48
* mordred defers in all ruby judgement to the rubies18:48
anteayabut how many gems do the repos need now?18:48
mordredor maybe 318:48
anteayafungi: set source= to the cache18:48
crinklefungi: I have no idea, I'm not sure ruby is that smart18:48
crinklemaybe that ^18:48
fungianteaya: oh, cool!18:48
edwarnickeanteaya: 'I AGREE' gets stuff to the server, but then I get an error back "Code Review Error: Server Error Cannot Store Contact Information"18:48
anteayafungi: I'd try it anyway18:48
edwarnickeanteaya: Suggestions ?18:48
anteayaedwarnicke: have you filled out a foundation individual membership?18:49
crinklefwiw the stackforge modules don't do any caching that I'm aware of and they run reasonably fast18:49
edwarnickeanteaya: I filled in the forms on the page presented here: for the ICLA18:49
edwarnickeanteaya: Including all blanks (even fax.. who has a fax these days ? ;) )18:49
anteayaedwarnicke: yeah no kidding18:50
*** patrickeast has quit IRC18:50
fungiedwarnicke: what instructions did you find telling you to do that? all the instructions we've written tell you to first get your foundation individual membership set up18:50
anteayamordred: yeah so I would go the jjb route for now with a comment to create gemfiles for all the puppet repos as we have over 100 of them18:50
jeblairedwarnicke: (we'd love to find those docs and fix them for people in the future)18:50
anteayabut it isn't very ruby18:51
edwarnickefungi: That looks about right18:51
edwarnickejeblair: I feel ya... we have similar issues at ODL... docs are never done ;)18:51
fungicrinkle: it's less of a speed/performance concern, more of a stability concern. every network access is one more place a job can unnecessarily fail when running in clouds with less-than-awesome network functionality18:51
openstackgerritMerged openstack-infra/storyboard: Fix method checks.superuser
edwarnickeanteaya: fungi jeblair What is the local tradition on 'Unaffiliated'... I am employed to work on ODL... but not OpenStack... I just happen to need to push a patch to OpenStack18:52
openstackgerritMerged openstack-infra/storyboard: Fix method is_valid in access_token api
crinklefungi: oh i see18:52
anteayaand I have no data on how often the ruby gems servers are down18:52
anteayaedwarnicke: affiliated means who pays your paycheque18:52
edwarnickeGot it :)18:53
anteayamostly for stackalytics tracking and such18:53
edwarnickeMakes sense :)18:53
anteayaedwarnicke: feel free to fill it in as you see fit18:53
*** ChuckC has quit IRC18:53
anteayait shouldn't block any contributions18:53
jeblairanteaya: there are important bylaws things regarding affiliation18:53
edwarnickeThe final line about 'If you are not being paid to work on OpenStack' left the question open in my mind :)18:53
*** achanda has joined #openstack-infra18:53
*** BobBall is now known as BobBall_AWOL18:53
fungiedwarnicke: specifically if you're contributing in your spare time and no employer can claim to be contracting/employing you to do the work, you're unaffiliated18:53
anteayajeblair: that is true, for membership of certain elected bodies18:53
*** maxbit has left #openstack-infra18:54
mordredbut if you are making a patch to openstack in service of your employer paying you to work on ODL, then I'd imagine your work would be related to your employer. however - IANAL18:54
*** ChuckC has joined #openstack-infra18:54
jeblairyou know what we really need to do?18:54
clarkbmordred: you are an honorary fake lawyer :)18:55
jeblairnot require individual membership for contributions18:55
mordredjeblair: ++18:55
fungianteaya: the rubygems site can be up 100% of the time, but if the cloud provider arbitrarily ditches 1% of your outbound tcp sessions, it's still beneficial to avoid making network connections to the rubygems site during the job18:55
greghaynesdhellmann: I basically want a bunch of doc source files to be outside of doc/source and id like to avoid having to maintain an explicit list of docs because theres a lot and they each for a self-contained component. is the best way ive come up with (note the tox.ini)18:55
*** Longgeek has joined #openstack-infra18:55
edwarnickefungi: anteaya mordred Victory!  Many thanks :)18:55
anteayajeblair: works for me18:56
anteayaedwarnicke: yay18:56
anteayaedwarnicke: welcome to openstack18:56
zarojeblair, fungi : i've also verified that the connection issue between gerrit and the db occurs when using the specific mysql driver downloaded by gerrit.18:56
anteayaedwarnicke: might I suggest starting here:
anteayafungi: I agree18:56
fungiedwarnicke: what instructions were you following which jumped right into setting up a gerrit account but didn't tell you to join the foundation first? we've been trying to eliminate any documentation about that which isn't the official documentation at so definitely want to track it down18:56
zarojeblair, fungi : just wondering if you want to do anything about this atm? or just leave it alone?18:57
*** tmcpeak1 has quit IRC18:57
*** carl_baldwin has joined #openstack-infra18:57
fungizaro: jeblair: we probably want to figure out what's causing it (for example, this might be the same issue which was impacting before i moved it back off trove) but i think it's likely safe to move forward with the upgrade plan since this doesn't seem to be a regression18:58
annegentlegreghaynes: I'm very interested in generated TOCs across repos18:59
greghaynesoh? :)18:59
jeblairfungi: i agree.  i can't think of anything else to do to prove that we aren't already affected, other than to take a 30 minute downtime over the weekend and firewall off production gerrit for testing18:59
annegentlegreghaynes: I was thinking it would be a good way to collect say "user-guide" related topics18:59
openstackgerritMORITA Kazutaka proposed openstack-infra/project-config: Add devstack-plugin-sheepdog project into stackforge
clarkbjeblair: we'll do it live!19:00
annegentlegreghaynes: so you'd add some sort of tag in your RST file and then run a build job that looks across named repos for that tag, creates an index.rst from that collection, then builds the user-guide from that index.rst19:00
*** arxcruz has quit IRC19:00
*** dizquierdo has joined #openstack-infra19:01
fungizaro: also thanks for double-checking19:01
greghaynesannegentle: yep, makes sense. My sphinx-fu is fairly weak and im unfortunately having issues generating toc's within a single repo19:02
greghaynesI am definitely a fan of decentralising that though19:02
*** HeOS has joined #openstack-infra19:02
zarofungi: np.19:02
openstackgerrityolanda.robla proposed openstack-infra/infra-specs: StoryBoard integration tests specification
clarkbmrmartin: you around?19:03
clarkbmrmartin: looking at puppet-solr. was it the case that you said existing solr modules weren't good for some reason?19:03
funginikhil_k: i see you just updated glance-specs-core in launchpad... i think we're also overdue for similar treatment on glance-coresec (looks like markwash is still the admin there). need help fixing that?19:03
zarofungi: just wondering, connectionpool=false seems to fix it but i guess that's something we don't want to do?19:03
*** zigo has joined #openstack-infra19:03
jeblairzaro: not in production -- we definitely want a connection pool19:04
mrmartinclarkb: the situation, that the puppetlabs-solr module exists, and the one that was looking good was broken, but the missing pull request was merged during this weekend, and the maintainer made a new puppet module release too, so my puppet-solr module can be abandoned now19:04
mrmartinI did the integration today, and works well with askbot module.19:05
zaroi'm real not knowledgable so i'll continue to look for another work around during upgrade testing.  the connection timeout is real annoying.19:05
clarkbmrmartin: cool, can you abandon your change so taht we don't create a new project that we need?19:05
sdaguemestery: ok... can you walk be back on something
jeblairzaro: feel free to set it on dev19:05
mrmartinclarkb, yes, doing now19:05
clarkbmrmartin: thank you19:05
*** _nadya_ has joined #openstack-infra19:05
*** devlaps has quit IRC19:06
sdaguebecause it's unclear to me when / how you'd set anything of those in the existing setup19:06
mesterysdague: Hold on, I'm reformulating the patch to use ODL_MODE instead, and defualt it to a single-mode install.19:07
*** dizquierdo has quit IRC19:07
sdagueyeh, I saw a new round of the patch, but... I think it's changing too much19:07
*** vigneshvar has joined #openstack-infra19:07
sdagueI think this entirely can be encapsulated in the settings file19:07
sdagueI'm happy to draft a version of it, but I need some more info about what these flags mean in your current env19:08
mesterysdague: Ack.19:08
*** jerryz has quit IRC19:09
*** ssam2 has quit IRC19:09
mesterysdague: OK, so look here for what they mean: (line 36)19:09
*** dimtruck is now known as zz_dimtruck19:09
mesterysdague: Does that clear things up?19:09
mesterysdague: The piece I was missing was, if we enalbe something by default in the settings file, how can the user override that?19:10
sdagueah, yeh, I think so. let me propose something.19:10
mesterysdague: Sweet!19:10
openstackgerritJames E. Blair proposed openstack-infra/zuul: Fix independent pipeline CRD
openstackgerritJames E. Blair proposed openstack-infra/zuul: Avoid leaking dynamic change queues
openstackgerritJames E. Blair proposed openstack-infra/zuul: Check liveness of changes before removing
fungimrmartin: on a related note, i'm trying to replace the cert on the ask site with a new one (the current one expires in a couple weeks) and having trouble working out where it goes on the existing server. it's not in /etc/ssl like your new puppet manifest is doing in 14004319:10
mrmartinfungi: it can be19:11
nikhil_kfungi: that would be great, thanks. I was just waiting for him to give me admin rights on these repos (that would have been more convenient) however, he seems too busy atm. Please let me know if you need any steps from my side!19:11
jeblairfungi: ^ i have revised those 3 patches, but the _only_ changes i made were to add tests19:11
*** Longgeek has quit IRC19:11
clarkbfungi: shouldn't the vhost or other apache config point at a location?19:11
fungiclarkb: vhost? ;)19:11
jeblairfungi: (and i changed the commit message on the last one to reflect that i could not, after all, find any code path that exercised the supposed problem, but still think we should "fix" it anyway)19:12
clarkbfungi: isn't it running as a wsgi app in apache?19:12
clarkbfungi: clearly I shouldn't even look19:12
mrmartinfungi: give me a minute, I'll try to help find it, the config of that server is a bit interesting.19:12
*** EmilienM|afk is now known as EmilienM19:12
fungiclarkb: dpkg -l|grep apache19:12
fungiaha! it's running nginx19:13
mrmartinfungi: nginx19:13
*** melwitt has quit IRC19:13
*** melwitt has joined #openstack-infra19:13
mrmartin   ssl_certificate /etc/nginx/conf/ask_openstack_org.full.crt;19:13
mrmartin    ssl_certificate_key /etc/nginx/conf/ask_openstack_org.key;19:13
fungimrmartin: yep, found in /etc/nginx/conf/ thanks!19:13
mrmartinand I guess an nginx restart is required19:13
*** _nadya_ has quit IRC19:14
fungii'll backup the old files to alternate names before i replace them too19:14
*** melwitt has quit IRC19:15
mrmartinand double check the file permissions, specially for the private key19:15
*** melwitt has joined #openstack-infra19:15
fungimrmartin: wow, yeah it's world-readable. odd19:16
zxiiroDo you folks plan on getting interns for GSoC?19:16
mrmartinwe could speed up this migration ask.o.o, I could sleep better.19:16
clarkbif at first you don't succeed recheck, if you still don't succeed then chmod -R 77719:16
clarkbzxiiro: we did last year, but I haven't seen as much organization around it this year yet19:16
dhellmanngreghaynes: we do this in the specs repositories by checking the symlink into git19:16
fungiclarkb: well, 0644 but still...19:17
anteayazxiiro: for us to have an intern for GSoC we need a mentor19:17
*** zz_avozza is now known as avozza19:17
zxiiroah ok, I'm looking into options on my side as well to see if we can get a summer intern to work on JJB19:17
*** pelix has quit IRC19:17
zxiirolet me know if you guys need help with mentoring students for JJB related stuff19:17
greghaynesdhellmann: ok, thanks! So my symlink deal is a sane approach :)19:17
*** zz_dimtruck is now known as dimtruck19:17
anteayazxiiro: well we certainly will do our best to support you and your intern if you get one19:18
*** BharatK has joined #openstack-infra19:18
dhellmanngreghaynes: yeah, but just check it in, don't script creating it19:18
sdague - mestery, how about that19:18
anteayaas working on jjb stuff would be great19:18
zxiiroanteaya: good to know thanks :)19:18
mesterysdague: Looking19:18
anteayazxiiro: sure19:18
*** Ryan_Lane has quit IRC19:18
greghaynesdhellmann: yep, a lot easier19:18
mesterysdague: Even better, I like this. :)19:18
jeblairdhellmann, greghaynes: yeah, we don't want to add anything to the venv section of tox.ini -- doc builds should work with standard tooling19:19
dhellmannfungi, clarkb, jeblair :  could you put the patch to import oslo.versionedobjects on your review list for this week, please? We're ready to move ahead.
sdaguemestery: the comments probably need some cleaning up19:19
sdaguebut you can run with it from there19:19
mesterysdague: Cool! I'll post comments, want me to respin the next version or you ok doing that?19:19
sdaguemestery: honestly, why don't you just take over the patch and modify it to your heart's content19:19
mesterysdague: lol19:19
sdagueI figured it would probably just be easier to show you what I was thinking than keep saying "no, that's not quite it"19:20
openstackgerritJan Klare proposed openstack-infra/project-config: move from macro install-chefdk to chef-bootstrap
jeblairsdague: patch development by evolutionary algorithm?19:21
mordredjeblair: with enough cloud servers ...19:22
*** joesavak has joined #openstack-infra19:22
sdaguewe would need a lot more cloud servers19:23
mrmartinsdague: planet-size cloud!19:23
clarkbdhellmann: so I can go ahead and approve that if it all looks good?19:23
*** Ryan_Lane has joined #openstack-infra19:24
*** otter768 has joined #openstack-infra19:24
*** gokrokve has joined #openstack-infra19:25
*** achanda has quit IRC19:25
*** talluri has quit IRC19:26
sdagueclarkb: ????19:26
sdagueis that just hpcloud network falling over?19:27
*** signed8bit has quit IRC19:27
sdagueoh, it's a rax node19:27
*** achanda has joined #openstack-infra19:27
clarkbsdague: we have caches for a raisin19:27
*** ChuckC_ has joined #openstack-infra19:28
clarkbdhellmann: I am approving the versionedobjects change now19:29
openstackgerritDoug Hellmann proposed openstack-infra/project-config: Remove stable-compat-jobs from Oslo libraries
*** otter768 has quit IRC19:29
*** ChuckC has quit IRC19:29
jeblairwhat is that job doing with neutron?19:29
clarkbjeblair: installing it as a dependency19:29
clarkbjeblair: via git bypassing any of our git caches19:29
*** e0ne has joined #openstack-infra19:30
clarkbI do not know the answer to that question19:30
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add check/gate jobs for the networking-odl project
sdagueclarkb: more importantly, how?19:30
clarkbsdague: looks like the git string is in its requirements?19:30
sdagueoh.... it's in the requirements.txt19:30
sdague-e git://
jeblairand i'm guessing it's not running a requirements-check job19:30
clarkbjeblair: no its a stackfoge project19:31
clarkbso meh19:31
sdagueclarkb: well... this is how all the neutron drivers are being split out19:31
jeblairsdague: to unofficial projects?19:31
sdaguejeblair: ask mestery ?19:32
* mestery reads backscroll19:32
mesteryjeblair: ++, this is how we're doing. If you folks have feedback on this, please advise :)19:33
*** AJaeger has joined #openstack-infra19:33
*** sileht has quit IRC19:33
mesteryjeblair: Are we doing the requirements pieces wrong here?19:33
jeblairmestery: is networking-odl part of neutron?19:33
*** Adri2000 has quit IRC19:34
mesteryjeblair: No, it's on stackforge19:34
dhellmannclarkb: thanks!19:34
*** Adri2000 has joined #openstack-infra19:34
*** ZZelle has joined #openstack-infra19:34
jeblairmestery: but it used to be and you have split it out?19:34
openstackgerritgreghaynes proposed openstack-infra/nodepool: Fail gracefully if disk-image-create is missing
mesteryjeblair: Yes, it was an ML2 driver, now it's out on it's own in the world :)19:35
jeblairmestery: are you going to test changes to neutron with it?19:35
mesteryjeblair: I think that would be a good idea, yes. Once we stabilize it, fix some things on the ODL side, it may be good to do that.19:36
asselin_I'm running into this issue setting up nodepool images. anyone seen it before? "sudo: unknown user: jenkins"
mesteryjeblair: But we're about a month out from that I would expect.19:36
jeblairmestery: but it's not the default open source implementation that neutron uses in its own tests?19:36
asselin_my images are about 161 hours old19:36
mesteryjeblair: Correct, it's not.19:37
jeblairasselin_: ours too, search backscroll for puppet_spec_helper or something like that :)19:37
*** stevemar has quit IRC19:37
clarkbasselin_: can you look back in your logs to see if puppet is failing?19:37
asselin_jeblair, thanks19:37
*** ivar-lazzaro has joined #openstack-infra19:37
clarkbasselin_: I think you don't have a jenkins user because puppet fails on the spec_helper thing19:37
*** stevemar has joined #openstack-infra19:38
openstackgerritMerged openstack-infra/project-config: Add openstack/oslo.versionedobjects
asselin_clarkb, ok will look19:38
*** andreykurilin_ has joined #openstack-infra19:38
openstackgerritBharat Kumar Kobagana proposed openstack-infra/project-config: Pointing to correct devstack-plugin-glusterfs repository location
*** ivar-lazzaro has quit IRC19:39
*** sarob has joined #openstack-infra19:39
*** ivar-lazzaro has joined #openstack-infra19:39
*** sileht has joined #openstack-infra19:40
clarkbasselin_: 154188 is a possible work around for the issue. We may end up solving this differently though19:40
*** afazekas has joined #openstack-infra19:41
*** openstackgerrit has quit IRC19:41
jeblairmestery: in that case, the only feedback i have is: don't put git repos in requirements.txt.  install from pypi or use zuul-cloner and python install if you want to install from git (so it uses the local cache)19:41
*** openstackgerrit has joined #openstack-infra19:41
clarkbdhellmann: considering 154211, does that also mean we can stop running python26 tests against oslo?19:41
mesteryjeblair: Good feedback, thank you!19:41
clarkbdhellmann: oh probably not because oslo is used by the clients?19:41
dhellmannclarkb: right, no, we need to keep the 26 stuff for a while longer19:42
jeblairmestery: i realize installing neutron from pypi is not really on the table atm, but i think people want to fix that.  (hi mordred)19:42
clarkbjeblair: mestery maybe we should consider that the dependencies are backwards too19:42
openstackgerritMORITA Kazutaka proposed openstack-infra/project-config: Add devstack-plugin-sheepdog project into stackforge
clarkbnetwork-odl doesn't depend on neutron, neutron depends on network-odl19:42
jeblairclarkb: agreed.19:42
openstackgerritgreghaynes proposed openstack-infra/project-config: Add docs jobs for diskimage-builder
mesteryclarkb: We didn't want to do that, armax can explain why :)19:43
mesteryclarkb: We debated that in the spec, I think we should have included you, jeblair and friends in that discussion it appears.19:43
armaxmestery: can someone fill me in?19:43
mesteryarmax: This is around the dependencies from the stackforge repos and neutron19:43
mesteryHow the stackforge repos depend on neutron, not hte other way around19:43
mesterySee clarkb's comments above armax :)19:44
mordredyah - it sounds like we should take a look at the larger design of that system - git repos refs in requirements.txt is going to lead to a place of disaster19:45
*** thedodd has joined #openstack-infra19:46
mesterymordred: In the networking-odl case, I think the git reference in requirements was a short-term thing. /cc armax19:46
clarkbjeblair: I just approved 153334 and realized you may still be debugging that race. Should I remove my approval?19:46
armaxmestery: sorry I am struggling to follow up, what is it that you’re asking of me?19:46
mordredmestery: nod.19:47
mesteryarmax: Just pulling you into the discussion, I'll ping you once htis call I'm on is done in 13 minutes and do a high bandwidth catch-up. :)19:47
jeblairclarkb: nah, it's unrelated19:48
armaxok, please do: I am already having 3 other converssations in as many channels :)19:48
AJaegerclarkb: I think is not needed at all and your comment confused the author, could you have a look, please? Nothing urgent...19:49
morganfainbergWe need a project called "Vince" in OpenStack to complement Zuul.19:50
mesterymordred: lol19:50
mesterymorganfainberg: lol19:50
clarkbAJaeger: :(19:50
morganfainbergmestery: I made19:50
clarkbAJaeger: but ya shouldn't be needed19:50
openstackgerritMerged openstack-infra/zuul: Add extra liveness check to queue processor
morganfainbergComment to that effect in keystone channel.19:50
morganfainbergMost people didn't get it :(19:50
morganfainbergmestery: I felt pained that they obviously didn't see enough ghostbusters.19:51
mordredmorganfainberg: it's spelled vinz19:51
stevemarmorganfainberg, google says it's "Vinz", just sayin19:51
morganfainbergmordred: auto-correct :(19:51
morganfainbergstevemar: ^^19:51
mesterymorganfainberg: You're ahead of your time :)19:51
mesterymorganfainberg: Or behind it, I'm not sure which ;)19:52
morganfainbergkilling my referred to bad 80s movies. Stupid iPhone.19:52
mordredmorganfainberg: steve jobs does not want you to like 80s movies anymore19:52
*** abhirc has quit IRC19:52
morganfainbergIt also autocorrects to "vine".19:52
morganfainbergI gave up fighting it.19:52
openstackgerritgreghaynes proposed openstack-infra/nodepool: Fail gracefully if disk-image-create is missing
asselin_clarkb, jeblair ok error's on my side. I need to update a bit:
openstackgerritSergey Kulanov proposed openstack-infra/project-config: Add Fuel LMA collector Plugin to Stackforge
*** Rockyg has joined #openstack-infra19:53
morganfainbergThree autocorrects and I figure the technology wins "vinz" no not "vins" no not "vine" no not "Vince" oh whatever close enough.19:53
morganfainbergmordred: Steve wouldn't have allowed 80s movies to happen /meme19:54
*** melwitt has quit IRC19:55
*** melwitt has joined #openstack-infra19:56
*** afazekas has quit IRC19:56
BharatKHi clarkb AJaeger19:56
mordredmorganfainberg: of course he wouldn't have - all of the cool computer movies from the 80s feature an IBM PC19:56
BharatKglusterfs job failing because of repository issue.19:57
BharatKPlease have a look at the log:
AJaegerclarkb: this is about
clarkbya I think sdague has figured it all out if you look at the ec2api job it should give you a good template to work off of19:58
*** sdake is now known as sdake__19:58
clarkbbut basically you add your project to the PROJECTS list in the job then refer to the repo normally19:58
morganfainbergmordred: sneakers, best hacker movie ever (yay mostly social engineering with minor deus ex machnina) or ... Still has pcs in it.19:58
*** afazekas has joined #openstack-infra19:58
sdagueBharatK: please see my openstack-dev ML post from Friday19:58
AJaegerBharatK: check jenkins/jobs/ec2-api.yaml in project-config19:58
sdagueI explained it there, with lots of links19:59
*** krtaylor has joined #openstack-infra19:59
*** sdake_ is now known as sdake19:59
BharatKsdague, okay I will see thank you19:59
openstackgerritMerged openstack-infra/project-config: Remove stable-compat-jobs from Oslo libraries
fungimorganfainberg: i thought it was terrible on a technical basis, but aykroyd's character made up for that19:59
morganfainbergfungi: yes. But that it was mostly social engineering made me happy.20:00
*** andreykurilin_ has quit IRC20:00
morganfainbergAnd aykroyd's character was great.20:00
openstackgerritMerged openstack-infra/zuul: Don't show jobs for non-live changes
*** mrmartin has quit IRC20:00
BharatKsdague, Thank you Sean, I have seen "enable_plugin ec2-api git://"20:03
BharatKsdague, but clarkb commented that not to use git URL in localrc.20:03
*** annegentle has quit IRC20:03
sdagueBharatK: read the whole email20:03
*** bswartz has quit IRC20:03
*** hashar has joined #openstack-infra20:03
BharatKsdague, okay20:03
clarkbBharatK: well I specifically said don't use github20:03
sdagueyou need the PROJECTS line as well20:04
*** tmcpeak has joined #openstack-infra20:04
BharatKsdague, I think this patch adds PROJECT line I guess
BharatKsdague, please review it and let me know if it is wrong.20:05
*** abhirc has joined #openstack-infra20:05
clarkbBharatK: it is, that is the wrong place to make that change20:05
sdagueBharatK: you should not be patching devstack-gate20:05
BharatKclarkb, sdague Okay thank you. Sorry for wasting your time. I will correct my mistake.20:06
*** afazekas has quit IRC20:07
*** abhirc has quit IRC20:07
*** afazekas has joined #openstack-infra20:07
*** gyee has quit IRC20:07
*** changbl has quit IRC20:09
*** aysyd has quit IRC20:10
*** dprince has quit IRC20:10
*** signed8bit has joined #openstack-infra20:12
*** r1chardj0n3s_afk is now known as r1chardj0n3s20:13
*** ashleighfarnham has quit IRC20:15
*** ldnunes has quit IRC20:16
*** changbl has joined #openstack-infra20:17
*** koolhead17 has quit IRC20:20
*** _nadya_ has joined #openstack-infra20:21
SpamapSmordred: oh btw, I got the script to run. So now how do I make it fail because of mysql in diskimage-builder?20:22
clarkbSpamapS: mordred I was just looking at that again and I think you need to unset the in chroot flag20:23
*** ghostpl has joined #openstack-infra20:24
*** shashankhegde has quit IRC20:24
mesterysdague: When you get a moment, a quick eyeball on the changes I made to your ODL_MODE patch would be good before we merge it:
SpamapSclarkb: ty20:26
openstackgerritgreghaynes proposed openstack-infra/project-config: Add docs jobs for diskimage-builder
* SpamapS is so puppet stupid he isn't entirely sure how to do that. :-P20:26
*** dizquierdo has joined #openstack-infra20:27
mordredclarkb: so - just in case you were wondering ... when you design a thing to get networking info from config-drive20:27
mordredclarkb: it REALLY doesn't work if you boot with config-drive not on20:27
SpamapSmordred: you lie20:27
* mordred drops truth bombs20:28
clarkbmordred: its a good thing that works as expected :)20:28
lifelessmordred: so this is the non-cloud-init thing, custom config-drive reader?20:28
*** ghostpl has quit IRC20:28
mordredlifeless: yah - although it's actually the non-nova-agent thing20:28
mordredstock cloud-init also no workie on these20:29
lifelessmordred: oh because rackspace?20:29
sdaguemestery: yeh, that should be fine20:29
*** talluri has joined #openstack-infra20:29
*** timcline has quit IRC20:29
*** timcline has joined #openstack-infra20:30
*** amitgandhinz has quit IRC20:32
mesterysdague: Cool, I respun with enhanced documentation per your suggestion. Thanks again, I think this has the makings of being pretty awesome. :)20:32
clarkbjeblair these are some large zuul changes :) will get back to reviews after lunch. Any chance I can interest you in the stack that starts at ?20:32
* mestery steps out for a bit20:32
*** AJaeger has quit IRC20:33
openstackgerritClaudiu Belu proposed openstack-infra/project-config: Adds networking-hyperv in StackForge
*** _buzz_ has quit IRC20:33
*** talluri has quit IRC20:33
*** timcline has quit IRC20:34
openstackgerritBharat Kumar Kobagana proposed openstack-infra/project-config: Exporting PROJECTS variable for glusterfs job
*** timcline has joined #openstack-infra20:37
*** Guest46399 has joined #openstack-infra20:37
*** kmartin has quit IRC20:38
*** jedimike has quit IRC20:39
jeblairclarkb: well, they are big tests, hopefully the changes themselves aren't that big :)20:40
*** Rockyg has quit IRC20:40
*** afazekas has quit IRC20:44
*** salv-orlando has joined #openstack-infra20:44
*** tqtran is now known as tqtran_afk20:44
*** radez is now known as radez_g0n320:46
*** shashankhegde has joined #openstack-infra20:47
*** maxbit has joined #openstack-infra20:48
*** maxbit has left #openstack-infra20:48
*** kmartin has joined #openstack-infra20:50
*** _nadya_ has quit IRC20:50
fungiokay, i need to duck out for a few hours. it's that time of year again where i go talk to a professional about how much money i've been giving my various governments, and then go drown my sorrows in booze20:50
*** afazekas has joined #openstack-infra20:50
jeblairfungi: do you want to re-review the zuul changes you reviewed over the weekend?20:50
jeblairfungi: (i only added tests)20:50
jeblairfungi: (if not, i can just mentally carry-over your +2)20:51
*** andreaf_ has quit IRC20:51
fungijeblair: i was going to, but feel free. it'll be later this evening before i'm back to a computer20:51
*** baoli_ has quit IRC20:51
jeblairfungi: okay, we'll just see what happens with them then :)20:51
fungimoar tests == moar better20:51
* fungi wanders away until later20:52
*** baoli has joined #openstack-infra20:52
*** BharatK has quit IRC20:55
*** Longgeek has joined #openstack-infra20:56
*** hamzy has quit IRC20:58
*** hamzy has joined #openstack-infra20:58
hamzyhuh crashed xchat20:59
openstackgerritJames E. Blair proposed openstack-infra/system-config: Zuul status: Use a grey icon for non-live changes
*** carl_baldwin has quit IRC21:00
clarkbjeblair: 153813 has a comment but still +221:02
*** unicell has quit IRC21:03
*** unicell has joined #openstack-infra21:03
jeblairclarkb: responded (agreed).21:04
*** weshay has quit IRC21:04
*** maxbit has joined #openstack-infra21:07
*** maxbit has quit IRC21:08
*** Longgeek has quit IRC21:10
*** andreaf_ has joined #openstack-infra21:12
*** ashleighfarnham has joined #openstack-infra21:12
*** bswartz has joined #openstack-infra21:12
*** andreaf has quit IRC21:12
*** andreaf_ is now known as andreaf21:12
*** andreaf_ has joined #openstack-infra21:13
*** gyee has joined #openstack-infra21:14
*** afazekas has quit IRC21:14
mordredsyntax error in upstart file...21:15
jeblairmordred: couldn't you just run it before you... oh, right.21:15
*** dmellado has quit IRC21:16
*** dmellado has joined #openstack-infra21:16
*** carl_baldwin has joined #openstack-infra21:16
*** maxbit has joined #openstack-infra21:16
mordredjeblair: I suppose I could learn how to run VMs on my laptop21:16
*** achanda has quit IRC21:20
*** ChuckC_ has quit IRC21:20
*** ChuckC_ has joined #openstack-infra21:21
jeblairi'm restarting zuul21:21
*** ChuckC_ is now known as ChuckC21:21
*** tqtran_afk is now known as tqtran21:22
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add static network config element for Rackspace
mordredgreghaynes: ^^ you actually don't need nodepool to play with those though - I'm mostly doing manual things to verify on rax21:23
nibalizerjeblair: mordred do you know if hpcloud overwrites /etc/hosts?21:24
greghaynesmordred: good deal21:24
mordrednibalizer: if it does I will kill it dead with a stick21:24
clarkbmordred: nibalizer: the answer is sort of, iirc dhcp works there and it does/can21:25
mordrednibalizer: bt it probably does21:25
*** dmellado has quit IRC21:25
*** otter768 has joined #openstack-infra21:25
nibalizergozer tried to make dns be better with /etc/hosts21:26
nibalizerand on inspection, didn't work21:26
nibalizerim investigating21:26
greghaynesmordred: that patch looks eerily similar to what os-collect-config does21:27
*** nicodemos has quit IRC21:27
mordredgreghaynes: I would not doubt it21:27
clarkbjeblair: commented on 15381421:29
mordredgreghaynes: I imagine there will be several ways in which this could be improved on21:29
clarkbjhesketh: good morning21:29
mordredgreghaynes: wow. os-collect-config depends on eventlet21:30
*** otter768 has quit IRC21:30
mordredgreghaynes: it's ok - I'm pretty sure that they're different enough that using o-c-c isn't still a bit much21:31
greghayneseh, os-c-c is basically grab data from $place and then run $thing with data21:32
greghaynesbut a config drive collector would totally make sense for it21:32
mordredthat part woudl be fine - but the (very correct) choice by o-c-c to just depend on all the things it might want to poll ... means that it's 8 times more heavyweight than is needed here21:32
greghaynesoh, right, that problem21:33
*** cdent has quit IRC21:33
greghaynesthere are many bad ways to solve that problem, and we chose one ;)21:33
mordredgreghaynes: there are almost no _good_ ways21:34
mordredalthough I am partial to dhcp ...21:34
jeblairclarkb: responded21:34
clarkbjeblair: it might even help to just flip around the args getChangeQueue(whatwasdefault, thingthatIreadasdefault)21:36
clarkbjeblair: then if not whatwasdefault return whatwasdefault else do stuff with thingsthatIreadasdefault21:36
ianwdo we have any preference for people cloning from via https:// v git:// ?21:37
jeblairrestarting zuul again21:37
clarkbianw: git:// is nice for the builds farms. I mostly prefer https elsewhere21:37
*** sarob has quit IRC21:37
*** ChuckC has quit IRC21:38
mordredI like archie:// myself21:38
jeblairclarkb: but it's an optional argument21:39
clarkbjeblair: oh I see ya21:39
*** ChuckC has joined #openstack-infra21:39
clarkbjeblair: I think thats why I got confused, its optional but overrides everything else if present, which is different than dict.get(foo,bar)21:39
clarkbjeblair: it is probably fine as is, just requires me to read more carefully :)21:40
Shrewsmordred: lol. i must be old if i understand that joke21:41
jeblairclarkb: it's designed to reduce the chances of a programming error; it's otherwise apparently not necessary.  so if it's failing to do that, we might want to consider other options21:41
*** afazekas has joined #openstack-infra21:41
clarkbI did not get the joke other than as a reference to a comic book and its protagonist21:41
ianwclarkb: ok, in reference to -- (Use 'https' protocol by default as easier for proxies) .  i'm not sure i really care, other than my usual "if it ain't broke..." (nobody has complained for over a year)21:41
Shrewsclarkb: darn your youth21:41
jeblairclarkb: you got the beginning of the joke, but not the middle :)21:42
*** ChuckC_ has joined #openstack-infra21:42
clarkbianw: that should be fine at least for the test botnet, we update all th git repos independent of stack.sh21:42
clarkbianw: so if it makes devstack users happy its probably a good thing21:42
*** carl_baldwin_ has joined #openstack-infra21:43
clarkbianw: note that the android gerrit server is a bad thing to compare against simply because gerrit does not use the git:// protocol for acl reasons iirc21:43
clarkbianw: (has nothing to do with proxies or any of that)21:44
*** carl_baldwin has quit IRC21:44
*** carl_baldwin_ is now known as carl_baldwin21:44
ianwclarkb: right, just didn't want everyone to start hitting https and sending encryption overhead through the roof21:44
clarkbianw: it should be ok, our biggest bottleneck today is outbound bandwidth which we should hit our cap on before we run out of cpu time for ssl21:44
ianwi know occasionally some git servers have struggled21:45
*** ChuckC has quit IRC21:46
clarkbjeblair: what if we kept the context manager bits but more explicitly grab the manager before using it with the with block21:46
clarkbjeblair: so if last_head.queue then change_queue = new Manager(last_head.queue) else new Manager() fi with change_queue:21:47
jeblairclarkb, fungi: i stopped zuul because i'm not entirely sure how i feel about commit dependencies in the merge-check pipeline (if we think it's long now, it could get very, very long with them).21:47
clarkbjeblair: oh hrm good point especially since merge failures aren't a cross repo concern21:48
jeblairclarkb: that works, but does not force the use of the context manager21:48
*** ashleighfarnham has quit IRC21:49
*** bswartz has quit IRC21:50
clarkbjeblair: actually maybe something simple is all we need s/default/override/21:50
clarkbjeblair: then it doesn't read as a fallback default but an explicit, this things wins if valid21:50
jeblairclarkb: wfm21:51
mordredSpamapS: ping21:51
mordredSpamapS: on that upstart patch that I copied from your forum post21:51
*** rlandy has quit IRC21:51
mordredSpamapS: when I try to run that job, I get "Unknown parameter: UPSTART_JOB"21:51
openstackgerritMatthew Treinish proposed openstack-infra/project-config: Don't run thirdparty tests on icehouse
*** afazekas has quit IRC21:53
mordredSpamapS: is it possible that in your post you intended $UPSTART_JOB to be something I fill in directly in the description?21:53
*** achanda has joined #openstack-infra21:54
*** dkliban is now known as dkliban_brb21:55
openstackgerritMatthew Treinish proposed openstack-infra/project-config: Don't run thirdparty tests on icehouse
dhellmannclarkb, fungi, jeblair: mtreinish's patch ^^ is needed to unwedge stable/juno work21:56
*** ashleighfarnham has joined #openstack-infra21:57
openstackgerritMerged openstack-infra/infra-manual: Explain changes and patch sets
jeblairmtreinish: isn't the solution to that problem the tempest venv?  (i ask because this issue blocking tempest-venv sounds circular)21:59
mtreinishjeblair: see -qa, sdague just asked the same thing21:59
stevemarfungi, i've got a weird issue with a build that i can't figure out - can i bug you :D21:59
*** dmellado has joined #openstack-infra22:00
stevemarI seem to be getting: Couldn't find remote ref refs/zuul/stable/icehouse/ errors in the devstack setup logs - but i think that's a generic warning22:01
clarkbreed: can you rebase 145971 then I can merge it?22:01
openstackgerritMerged openstack-infra/infra-manual: Clarify that Change-Id should not be modified
jeblairclarkb: so do we need a pipeline flag to ignore dependencies that we can set for merge-check ?22:01
reedclarkb, let me try22:02
clarkbjeblair: ya I think we need a no cross repo deps flag22:02
jeblairclarkb: well, it would be a 'no deps' flag since they are the same now :)22:02
clarkboh right hrm22:02
sdakeHey folks have a hopefully easy question22:03
clarkbyes, I think for merge-check we need it22:03
jeblairclarkb: but still, the merge-check would do the right thing with no deps at all22:03
*** daneyon has joined #openstack-infra22:03
sdakethe title of our repository in stackforge needs a change, because we are dropping kubernetes support22:03
clarkbit is a bit of a corner case but an important one for UI imo22:03
*** ddieterly has quit IRC22:03
sdakeis that done via a infra review?22:03
clarkbstevemar: can you link to the jobs that are failing with that?22:03
jeblairsdake: yes, the same projects.yaml where the project was created22:04
SpamapSmordred: $UPSTART_JOB should be in the environment for every upstart job.22:04
*** afazekas has joined #openstack-infra22:04
mordredSpamapS: I now understand that part22:04
stevemarclarkb, sure, it's here:
sdakejeblair cool thanks22:04
mordredSpamapS: my upstart job does not seem to be getting tirggered and I cannot figure out why22:04
mordredSpamapS: any clues on debugging such a thing?22:04
SpamapSmordred: hm, the use of it in 'instance' is superfluous though22:04
SpamapSmordred: remove that I think.22:04
jeblairclarkb: yeah, i also think that there _might_ be implications if a needed dependency failed to merge (for that matter, it would probably fail check too, but i think that's desired)22:04
stevemarthis seems suspicious 'oslo.context' is not a global requirement but it should be,something went wrong22:04
jeblairclarkb: (failed to be mergable, i guess i should say)22:05
*** hashar has quit IRC22:05
dhellmannstevemar: ?22:05
mordredSpamapS: in the waiter script?22:05
SpamapSmordred: yes, I just updated the askubuntu answer. :-P22:06
stevemardhellmann, back to the issue of whether of not to add oslo.context to global req for icehouse22:06
SpamapSgah and I see another error22:06
jeblairclarkb: i'll do s/default/existing/.  sound good?22:06
dhellmannstevemar: it shouldn't be needed if we're capping requirements22:06
SpamapS$INTERFACE won't be exported by 'starting networking'22:06
*** oomichi has joined #openstack-infra22:06
clarkb jeblair yup thank you22:06
stevemardhellmann, the issue is it's not working in pycadf: but it's working for keystonemiddleware:
SpamapSmordred: so $UPSTART_JOB is available in exec/pre-start/post-start/pre-start/post-start but not in instance22:07
clarkbdhellmann: stevemar: ya I think thats a matter of either backporting the requirement or not adding the requirement to pycadf22:08
stevemarboth patches are adding oslo.context to requirements, and both test against tempest-icehouse22:08
dhellmannstevemar: I just dropped all of the stable branch jobs for oslo libs because we're capping them. I didn't include any libraries outside of oslo in that patch, though. You could do the same, if it's safe to assume those libs will be capped.22:08
*** ddieterly has joined #openstack-infra22:08
*** afazekas has quit IRC22:09
mordredSpamapS: INTERFACE probably won't be exported by networking either22:09
mordredSpamapS: in my case here - what's the chance that I'm going to get one vs. the other?22:09
dhellmannstevemar: those tests are running the master branches of those libs under the icehouse versions of the servers. we don't care about that configuration any more.22:09
SpamapSmordred: right I fixed them now22:09
mordredSpamapS: is there a defined order at all for "networking" or "network-interface"?22:09
mordredah - cool22:09
* mordred goes to lookg22:09
clarkbI much prefer sdague's change to the d-g regex change. I am going to -1 the d-g change now22:10
SpamapSmordred: no22:10
SpamapSmordred: network-interface happens whenever udev ifup's stuff22:10
SpamapSmordred: networking happens at various points22:11
SpamapSstart on (local-filesystems and (stopped udevtrigger or container)) or runlevel [2345] or stopped networking RESULT=failed PROCESS=post-stop EXIT_STATUS=10022:11
SpamapSPerhaps one of the best examples of the reason event-based is fail for psychological reasons.22:11
*** roshan has joined #openstack-infra22:11
clarkbjeblair: any opposition to me approving those d-g changes now? (you +2'd but did not approve)22:12
stevemardhellmann, ah so you mean we can remove the icehouse/juno related jobs from pycadf?22:13
jeblairclarkb: nope -- but note that while i grokked most of that, i did not carefully read the network commands because i am unlikely to spot an error there.22:13
mordredSpamapS: I was just chatting with jeblair about the fact that this would be super simple with sysvinit - I'd just run the read-from-config-drive with a very low number22:14
mordredSpamapS: ok - your changes there  make sense - I am trying them22:14
dhellmannstevemar: yes. Also, the icehouse version of devstack doesn't seem to be installing keystonemiddleware from source anyway.22:14
*** abhirc has joined #openstack-infra22:15
*** afazekas has joined #openstack-infra22:15
mordredSpamapS: also - it turns out "auth eth0" is not a valid line in /etc/network/interfaces - but "auto eth0" is22:15
dhellmannI see the function defined, but I'm not seeing it called in
reedclarkb, I screwed up my repo :)22:15
mordredSpamapS: you've poked at some of these things more recently than I ... do I need to care about the eth0/en0 stuff?22:16
reedclarkb, tried to do three things at once, after 11pm22:16
stevemardhellmann, yeah you *just* removed the stable-compat jobs, okay, i will do the same to pycadf (at least), and look into ksm22:16
stevemarthank you sir22:16
clarkbjeblair: the network setup is the bit I am most confident in :)22:16
SpamapSmordred: I just commented with the right answer in the review. :-P22:17
clarkbjeblair: its all of the openstack stuff that I rely on jogo for :)22:17
dhellmannstevemar: it looks like when the middleware was added to devstack's stable/icehouse branch the call to install_keystonemiddleware was left out of stack.sh22:17
mordredSpamapS: woot22:17
SpamapSmordred: biosdevname you mean (em1, em2, etc?) I don't recall what en0 would be22:17
mordredSpamapS: of course, "trying that out" means "build a new image; upload the image to swift; import image in to glance; set image metadata; boot nova instance based on that image"22:17
mordredSpamapS: yah22:17
openstackgerritDaneyon Hansen proposed openstack-infra/project-config: Updates Kolla Project Description
jeblairclarkb: yeah, consider my approval an "I don't see this causing problems for future d-g use" :)22:17
mordredthe biosdevname stuff22:17
mordredSpamapS: do I need to care about that?22:18
stevemardhellmann, yeah i noticed something weird was going on since that icehouse job didn't actuall install oslo.context either, so i was very confused22:18
SpamapSmordred: I don't think so for the cloud. For baremetal you would.22:18
stevemarbut if it's installing from pypi, then it makes sense22:18
clarkbSpamapS: I just ordered a roku 3, with my luck they will refresh all models next week22:18
dhellmannstevemar: I'm not sure I ever saw the middleware installed at all, but I'd have to look back at the log again22:19
mordredSpamapS: awesome. well, I will not care about it until I do then22:19
SpamapSclarkb: You can then claim you liked roku 3 before it was cool.22:19
sdakeclarkb roku 3 rocks - super fast22:19
greghaynesclarkb: maybe we can move convo in here22:19
sdakeI can't wait for the roku 4k22:19
sdakethen I get to upgrade all the tvs in the house to 4k - yay !22:19
clarkbsdake: and move to a place with google fiber22:20
sdakecox is rolling gige to my hood in the next 6 mo22:20
clarkbgreghaynes points out that oslo.versionedobjects isn't on the mirrors yet22:20
clarkbI am looking into it22:20
*** dmellado has quit IRC22:20
greghaynesYep, its making my openstack-repos element... not happy22:20
mordredoh -that's not good22:21
*** roshan has quit IRC22:21
dhellmannclarkb: I'm getting an empty repository when I try to clone it :-(22:21
clarkbdhellmann: ya I am going to check if the empty repo is on all mirrors then retrigger replication if so22:22
clarkbgit is really derpy on centos6 now22:22
dhellmannclarkb: cool22:22
* dhellmann goes to make sure that git repo URL he provided is right22:22
mordredclarkb: I really don't like statements like "git is derpy"22:22
clarkbmordred: well its true :)22:23
*** afazekas has quit IRC22:23
mordredclarkb: sure. I mean, EVERYTHING is derpy - but I don't like it :)22:23
clarkb confirms the repo is empty and present on all the mirror nodes so kicking replication now22:23
*** dustins has quit IRC22:23
dhellmannclarkb: verified that is good22:23
dhellmannclarkb: not oslo.version -- oslo.versionedobjects22:23
clarkbdhellmann: ya I checked the source repo before I approved too22:23
*** dkliban_brb is now known as dkliban22:24
clarkbdhellmann: I think its in gerrit just not replicated to the mirrors22:24
dhellmannclarkb: ok, you had the wrong name in that url ^^22:24
clarkbdhellmann: ya cgit takes a prefix22:24
clarkbdhellmann: its a super handy feature for when I am lazy :)22:24
greghaynesapparently they know their target audience22:24
openstackgerritSteve Martinelli proposed openstack-infra/project-config: Remove stable-compat jobs from pycadf
SpamapSwhen I turn off in_chroot I get this22:25
SpamapSUpdating cache of git:// in /home/clint/.cache/image-create/source-repositories/oslo_versionedobjects_572b63e19f5f89ed731451a2ca6c9ee558fc22e6 with ref master22:25
SpamapSfatal: Couldn't find remote ref master22:25
greghaynesSpamapS: yep, thats exactly what were dealing with :)22:25
clarkbok replication kicked, now to look at logs to see how that happened22:25
*** vigneshvar has quit IRC22:25
SpamapSI'm not even reading what you guys are saying22:26
* SpamapS will context switch back to $other22:26
*** unicell has quit IRC22:29
*** gokrokve has quit IRC22:29
*** unicell has joined #openstack-infra22:29
*** gokrokve has joined #openstack-infra22:29
clarkbfwiw I have confirmed that you can clone from gerrit and the gerrit local replica just fine22:30
clarkbstill looks like an empty repo after kicking replication though22:30
clarkb*empty repo on the mirror nodes22:30
clarkbalso really needs timestamps22:30
*** carl_baldwin has quit IRC22:30
clarkbmordred: ^ any way to convince ansible to timestamp?22:30
reedclarkb, I don't understand why   ! [remote rejected] HEAD -> refs/publish/master/clear-cla-section (you are not allowed to upload merges)22:31
mordredclarkb: yeah - prolly so22:31
mordredclarkb: I'll put it on my plate22:31
clarkbreed because you ended up with a merge commit that you are trying to push to gerrit22:32
openstackgerritSteve Martinelli proposed openstack-infra/project-config: Remove stable-compat jobs from pycadf
clarkbshould use rebase not merge22:32
*** dhp has quit IRC22:32
reedi screwed up :)22:32
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for ironic bug 1418833
openstackbug 1418833 in Ironic "check-tempest-dsvm-ironic-agent_ssh fails due to ConfigDriveWriteError" [Critical,Fix committed] - Assigned to Jay Faulkner (jason-oldos)22:33
*** pc_m has quit IRC22:34
*** timcline has quit IRC22:34
greghayneswow, cursory googling says ansible playbooks require you to write a plugin to get timestamps22:35
*** timcline has joined #openstack-infra22:35
greghaynesyou can always, ya know, ansible | while read -r line do echo $(date) $line done ;)22:36
*** thedodd has quit IRC22:36
*** timcline has quit IRC22:37
openstackgerritMerged openstack/requirements: Add oslo.policy to projects.txt
*** timcline has joined #openstack-infra22:38
*** maxbit has quit IRC22:38
SpamapSgreghaynes: probably a feature of Tower. ;)22:39
greghaynesSpamapS: no joke, it is22:39
greghayneser, sorry, ansible commander22:39
greghayneswhich I assume is part of that22:39
clarkbthats cool cgit has segfaulted on git01 at least once22:40
clarkb(not related just noticed it in the logs)22:40
jlkgreghaynes: timestamps of what?22:40
greghaynesjlk: playbook runs22:40
jlkgreghaynes: if you write out the log file, you get timestamps for all the plays and tasks22:40
greghaynesjlk: nice!22:40
clarkband stdout doesn't count as a logfile?22:40
*** unicell has quit IRC22:40
clarkbits a file descriptor22:40
*** unicell has joined #openstack-infra22:41
jlkactually I'm seeing it in the stdout too22:41
jlkthat shows time of day, time of play, time of whole playbook.22:42
jlkunless this is a plugin we're using22:42
*** juice has quit IRC22:42
*** vipul has quit IRC22:42
jlkpoop, i't sa plugin22:42
greghaynesheh, /me was really confused there for a minute22:43
clarkbwhat is the centos equivalent to an auth.log?22:43
*** ashleighfarnham has quit IRC22:44
jlkhere's the log file version without a plugin22:44
notmynameclarkb: mordred: jeblair: so I want to have a new feature branch in swift, but I don't remember the process for doing so. can you help?22:44
clarkbya so probably what we need to do is stop redirecting stdout/stderr and just log directly to the file with ansible22:44
*** juice has joined #openstack-infra22:45
greghaynesclarkb: Im assuming this is in system-config?22:45
*** vipul has joined #openstack-infra22:45
mordredI think there was a reason I didn't do that originally, but I do not remember what that reason actually was22:45
clarkbgreghaynes: yes in modules/openstack_project somewhere22:45
*** dmorita has quit IRC22:45
openstackgerritJames E. Blair proposed openstack-infra/zuul: Avoid leaking dynamic change queues
openstackgerritJames E. Blair proposed openstack-infra/zuul: Add ignore-dependencies option
jeblairclarkb: ^22:46
*** crc32 has quit IRC22:46
clarkbjeblair: thanks22:46
clarkbok /me is going to ignore stuff until gerrit replication is figured out22:46
*** dmorita has joined #openstack-infra22:46
*** ashleighfarnham has joined #openstack-infra22:46
reedclarkb, I give up, feel free to take over the patch or write it again... it's a trivial set of ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~22:46
*** asettle has joined #openstack-infra22:47
*** crc32 has joined #openstack-infra22:47
jeblairnotmyname: let us know what's missing from that :)22:47
clarkbreed: ok22:47
clarkb/var/log/secure appears to be the red hat log file for auth22:47
*** dizquierdo has quit IRC22:48
*** miqui has joined #openstack-infra22:48
*** rmcall has quit IRC22:49
clarkbthere is a gerrit error when trying to create the local mirror path but then it succeeds in creating it shortly after22:49
*** amitgandhinz has joined #openstack-infra22:49
*** mriedem has quit IRC22:51
*** rfolco has quit IRC22:51
openstackgerritJames E. Blair proposed openstack-infra/project-config: Zuul: Ignore dependencies in merge-check
*** dmellado has joined #openstack-infra22:52
*** dprince has joined #openstack-infra22:52
*** e0ne has quit IRC22:53
*** Alexandra_ has joined #openstack-infra22:54
clarkbso pushing a change got the change to replicate22:54
clarkbam I just not triggering replication properly?22:54
*** Longgeek has joined #openstack-infra22:55
clarkbzaro: any idea where the replication docs are now?22:56
clarkb has no obvious doc dir22:56
*** dmellado has quit IRC22:56
mordredSpamapS: poke again re: upstart - that time it worked - the script ran, /etc/network/interfaces was TOTALLY created as expected22:56
clarkboh of course because its in the source repo22:56
*** dmellado_ has joined #openstack-infra22:56
mordredSpamapS: BUT22:56
*** dmellado_ is now known as dmellado22:57
clarkbyay! I figured it out22:57
clarkbalso the --wait flag is handy22:57
clarkbnow to go back to trying to understand why this broke at all22:58
mordredSpamapS: the network interfaces are down22:58
mordredSpamapS: (If I run "ifdown -a; ifup -a" it's all good)22:58
mordredSpamapS: this makes me think that upstart ran network-interface INTERFACE=eth0 and it ran normally and stuff22:58
*** bcourt has quit IRC22:59
openstackgerritGuido Günther proposed stackforge/python-jenkins: Don't return None on unhandled http errors
openstackgerritGuido Günther proposed stackforge/python-jenkins: Fix incorrect indentation
openstackgerritGuido Günther proposed stackforge/python-jenkins: Use empty crumb on empty response
openstackgerritGuido Günther proposed stackforge/python-jenkins: Don't assume we get a valid response
clarkbok 19:45 UTC is when replication was kicked off in gerrit23:00
*** reed has quit IRC23:00
openstackgerritgreghaynes proposed openstack-infra/system-config: Use ansible logging durring puppet
clarkb20:19 appears to be when mirrors had the files written to them23:01
clarkb thats 20:1923:01
clarkb thats 19:4523:01
clarkbmordred: so how the heck did ansible screw this up?23:01
jlkgreghaynes: so Ansible will keep writing to that file every time it runs23:02
greghaynesjlk: thats the current behavior right?23:02
jlkgreghaynes: is that file going to stay around between runs, do you want it truncated between runs, or do you want it to just grow forever?23:02
clarkbit should append23:02
jlkyeah, actually you're right23:02
nibalizerfor once it isn't puppet's fault \o/23:03
*** andreykurilin_ has joined #openstack-infra23:03
*** timcline has quit IRC23:03
clarkbnibalizer: well it might be :)23:03
greghaynesnibalizer: good point. how can we make this puppets fault?23:03
clarkbnibalizer: I think if puppet returns 0 when it shouldn't then ansible will continue23:03
nibalizerim sure puppet will be blamed before the debugging is over23:03
nibalizerthere it is ^^23:03
jlkyeah, puppet has fun exit codes23:04
clarkbthat is where puppet should've updated but didn't23:04
clarkbthis is going to be lolzy23:04
clarkbmordred: I am 99% sure I know what the problem is. project-config is cloned on each node from when puppet runs on the node23:05
clarkbmordred: so if thta change merged between puppet running on git01 and puppet running on review.o.o we lose23:05
mordredclarkb: yes23:05
mordredthat is it23:05
clarkbI am confirming merge timestamp now23:05
mordredclarkb: I have an idea ...23:05
jlkoh my23:05
clarkb1938UTC perfect23:05
*** jamielennox|away is now known as jamielennox23:05
clarkbso ya thats the bug23:05
mordredthat's totally the bug23:05
mordredclarkb: I have an idea for a fix in mind - I'll code up a POC in just a bit23:06
mordredclarkb: that's a good bug - nice tracking it down, btw23:06
clarkbmordred: probably just supply a ref via a fact?23:06
*** Ng has quit IRC23:06
clarkbmordred: then have ansible be like "this is the ref for this run on all nodes"23:07
*** hichihara has joined #openstack-infra23:07
mordredand since we want to get to passing in params from ansible to puppet anyway to support just using apply, this is actually potentially a nice first-stab at something like that23:07
*** rackertom has quit IRC23:07
*** amitgandhinz has quit IRC23:08
SpamapSmordred: I'm not sure I follow the timeline of what you're saying23:08
mordredSpamapS: let me start again23:08
mordredSpamapS: those upstart scripts we were working on - they write out /etc/network/interfaces23:09
*** enikanorov_ has quit IRC23:09
mordredSpamapS: so they _Want_ to run early enough that when the interfaces get brought up, the config file will be accurate, yeah?23:09
mordredSpamapS: what I have observed in this latest attempt23:09
mordredSpamapS: is that the script ran as expected (This is good, that was not true before)23:10
mordredand the file had the right contents23:10
*** thedodd has joined #openstack-infra23:10
mordredbut the network interfaces were not configured with teh right values (they were in down state)23:10
*** nelsnelson has quit IRC23:10
*** thedodd has quit IRC23:10
nibalizerbefore you write any ruby23:10
*** mjturek1 has quit IRC23:10
mordrednibalizer: yup23:10
mordrednibalizer: exactly what I was thinking23:11
openstackgerritClark Boylan proposed openstack-infra/infra-manual: Provide a subsection for CLA information
*** ddieterly has quit IRC23:11
openstackgerritJames E. Blair proposed openstack-infra/infra-manual: Clarify feature branch creation
SpamapSmordred: Hm..23:12
clarkbjeblair: I am just going to approve 145971 now. It should've been a trivial rebase but git didn't see it that way23:12
jeblairnotmyname, mordred, ttx: ^ I think that should clarify things23:12
mordredjeblair: ++23:12
*** Longgeek has quit IRC23:12
jeblairclarkb: i have re-reviewed it regardless and it has my +2 :)23:12
SpamapSmordred: OH, btw, there's a 'init_is_upstart' function in /lib/lsb/init-functions .. FYI23:12
clarkbjeblair: even better thanks23:12
SpamapSmordred: almost the same as mine :)23:13
SpamapS   if [ -x /sbin/initctl ] && /sbin/initctl version 2>/dev/null | /bin/grep -q upstart; then23:13
mordredSpamapS: nice23:13
SpamapSmordred: ok so here's an interesting problem we'll have23:13
mordredSpamapS: goood ...23:13
SpamapSmordred: eth0 is in the ubuntu cloud image, setup for DHCP23:13
mordredSpamapS: I'm not using the ubuntu cloud image23:14
SpamapSmordred: ok, that's good then23:14
SpamapSmordred: so do you have no /etc/network/interfaces file?23:14
mordredSpamapS: oh wait ... I _do_ have a dhcp interfaces file though23:14
mordredis that my problem then?23:14
*** joesavak has quit IRC23:14
SpamapSmordred: well here's the thing.. what you really want is to not have one.23:14
SpamapSmordred: or rather, auto lo iface lo ...23:14
clarkbjogo: I know you have been busy with other things but I just approved the d-g work to make nova net and multinode/multihost work23:14
SpamapSmordred: then, udev won't run any network-interface jobs23:15
clarkbjogo: hopefully that gets us moving much closer to having tempest full working23:15
mordredSpamapS: so then I'd really just be dealing with networking and not network-interface, yeah?23:15
SpamapSmordred: right23:15
SpamapSmordred: which will allow you to generate it, and then ifup -a takes over23:15
openstackgerritJames E. Blair proposed openstack-infra/zuul: Add ignore-dependencies option
jeblairnow with all needed files!23:16
SpamapSmordred: which is what i'd expect to happen anyway right now23:16
SpamapSjeblair: but wait, there's more23:16
SpamapScomment on this review now and you'll get not one, but +223:17
mordredSpamapS: hangon - what runs ifup -a?23:17
SpamapSmordred: /etc/init/networking.conf23:17
jeblairSpamapS: lol23:17
mordredok. cool23:17
mordredso just "start on starting networking"23:17
SpamapSmordred: it is the catchall for "not already setup or has no udev presence"23:17
SpamapSmordred: your laptop wouldn't work so incredibly well without it23:18
SpamapSYou might wait _SECONDS_23:18
mordredthe idea that the laptop and the server need to be the same ...23:18
SpamapSwell truth be told they don't need to be the same because networkmanager handles laptop connections23:18
mordredyah - and let me tell you for sure what I'm not running on a server ...23:19
mordredSpamapS: do I need the waiter pattern at this point?23:19
mordredif I'm just start on starting networking?23:19
SpamapSmordred: you don't since you won't get an 'ifup' for those interfaces if they're not configured.23:19
clarkbmordred: SpamapS: what if and this may be bonghits we just make the ifup and ifdown scripts look in config drive directly23:20
clarkbmordred: SpamapS that way you don't care about upstart23:20
SpamapSclarkb: no that is not bonghits at all23:20
mordredclarkb: we'd also have to modify them for rh - but maybe so23:20
SpamapSRH makes that relatively easy too23:20
mordredclarkb: ifup is a binary on ubuntu23:21
nibalizernm-tool and nmcli have given me brain damage, but they work23:21
clarkbmordred: but ou can replace them with shell23:21
mordredclarkb: I'll try that next - I'm almost done with this iteration23:22
clarkbmordred: it just seems like a whole hell of a lot less work to just have the calling thing do the right thing rather than setting up all this env stuff so that when it is called it does the right thing23:22
clarkbalso its binary? wtf23:22
mordredclarkb: the EASY thing would be for sysvinit to work23:23
mordredI'd like to argue that sysvinit is actuall more better for cloud23:23
fungiso, so much with the scrollback23:24
clarkbI wonder, can we put a rogue dhcp server on a neutron network we control then just use that for everything? iirc that isn't quite supported in rax with public IPs23:24
clarkbfungi: tl;dr found a bung in ansible puppeting, and dynamic host configuration protocol without using DHCP is bonghits23:25
clarkbjeblair: did you see jesusaurus' comment on 153025?23:25
*** otter768 has joined #openstack-infra23:26
SpamapSclarkb: ifupdown's wtf is even more powerful than you can possibly imagine23:27
clarkbSpamapS: I have resisted looking on my local machine23:28
*** EmilienM is now known as EmilienM|afk23:28
SpamapSclarkb: it is written in one of the most amazingly o_O languages on the planet23:28
clarkbSpamapS: but maybe I should23:28
*** Alexandra_ has quit IRC23:28
*** hdd has quit IRC23:28
StevenKLiterate programming!23:28
StevenKBecause WCPGW.23:28
StevenKOh wait.23:28
SpamapSNoWebM ___generated by perl___23:28
StevenKSpamapS: I have previously asked aj if he lost a bet WRT ifupdown's language of choice.23:29
SpamapSLooking at its code thinking you'll just patch it real quick is the programmer equivalent of mistaking wasabi for guacamole.23:29
*** carl_baldwin has joined #openstack-infra23:30
jlkSpamapS: golden.23:30
mordredSpamapS: ok. here we go again ...23:31
*** otter768 has quit IRC23:31
*** bknudson has quit IRC23:31
*** hdd has joined #openstack-infra23:33
*** wenlock has quit IRC23:33
*** jgrimm is now known as zz_jgrimm23:35
*** wenlock has joined #openstack-infra23:35
*** mfink_ has quit IRC23:36
clarkbjeblair: I am +2 on the remaining CRD changes. Should we get an ignore-dependencies layout.yaml update change pushed too?23:36
openstackgerritMerged openstack-infra/zuul: Fix independent pipeline CRD
*** dmellado has quit IRC23:36
clarkboh you did that already /me review23:36
*** tmcpeak has quit IRC23:36
clarkbso is 154188 a thing we should move forward on until the other work to use bundler and gemfiles is done?23:38
clarkbmordred: ^23:38
openstackgerritMerged openstack-infra/infra-manual: Clarify feature branch creation
*** dmellado has joined #openstack-infra23:38
openstackgerritBoris Pavlovic proposed openstack-infra/project-config: Add rally performance job to nova experimental queue
*** elliottk_ has joined #openstack-infra23:39
fungiclarkb: saw the ansible chatter. awesome stuffs23:40
*** ddieterly has joined #openstack-infra23:40
fungiSpamapS: clarkb: mordred: ifupdown supports hook scripts though, right?23:41
fungilike you can say "on up run this" and "on down run this other"23:42
clarkbya /etc/network/ifup.d/ iirc23:42
clarkbbut in this case we need to set the specific addrs and things23:42
clarkbthough there are pre scripts too iirc23:42
zaroclarkb: latest replication doc are here:
fungisure, i thought you could do that with the hooks calling ip address cli subcommands23:42
*** mriedem has joined #openstack-infra23:43
SpamapSfungi: yes it does23:43
SpamapSand RH's syconfig thing does too23:43
*** elliottk_ has left #openstack-infra23:43
SpamapSHonestly though23:43
clarkbfungi: I half expect it to be like but over here I was told to do X so X wins23:43
*** elliottk has joined #openstack-infra23:43
SpamapSmordred does not want to do things per interface23:43
SpamapSmordred wants to do things _before any interfaces_23:43
SpamapSupstart's implementation in debian and ubuntu don't provide that as a single hook point23:44
SpamapSsystemd does in theory23:44
fungiSpamapS: good point23:44
*** mriedem has quit IRC23:44
clarkbSpamapS: right but you could set it up to work per interface23:44
clarkbSpamapS: if your scripts in ifup.d/ directly looked into the config drive properly23:44
mordredSpamapS: the systemd hook is a good one23:45
SpamapSclarkb: yes you can set it up per interface, but you _ALSO_ need to set it up to go when there are no interfaces.23:45
fungiifupdown provides the interface name as an envvar i think, though it's been ages since i've written hook scripts for it so don't recall the details now23:45
SpamapSmordred: yes systemd's "state driven" paradigm is how our minds think23:45
clarkbSpamapS: to go when there are no interfaces?23:46
clarkbSpamapS: I don't parse that23:46
mordredalso - yes - there are MANY things that could be done - but I'm about 5 seconds from having this approach finished23:46
*** mriedem has joined #openstack-infra23:46
fungii might also be confusing them with rc on *bsd, it's been _that_ long23:46
SpamapSclarkb: well there is also the problem of /etc/network/interfaces not having any interfaces in it that actually exist via udev.23:46
clarkbI have an idea23:47
SpamapSclarkb: if that is the case... you have the catchall "ifup -a" that /etc/init/networking.conf runs .. and that is the time to generate /etc/network/interfaces with the config drive data.23:47
clarkbat the vancouver summit we get booth space and use it solely for signing up petitioners to require all clouds to dhcp23:47
SpamapSwhich is why you need two upstart jobs. ;)23:47
mordredclarkb: +1000023:47
mordredI mean, honestly, I cannot believe this is a problem we're having to solve23:47
fungiclarkb: i'll take gladly a 4-hour booth shift23:47
clarkbmordred: did you submit that talk proposal yet?23:47
SpamapSclarkb: I'm in23:47
SpamapSclarkb: let's have t-shirts23:48
clarkbmordred: the "Things that your cloud should give me as a user" talk23:48
mordredI ALSO cannot believe that crazypants have caused us to have a completely unintelligible init system23:48
fungithough we'd do better wandering with clipboards... would you consider signing this petition to save dhcp?23:48
mordredclarkb: no - I did not23:48
clarkbmordred: today is the last day? do we still have time23:48
clarkbmordred: lets write a proposal really quickly23:48
mordredwe should do a talk on how to configure network services in a scalable fashion ...23:48
mordredand also how to do service discovery by a distributed database of names23:48
clarkbmordred: :)23:49
mordredand seriously just talk about dhcp and dns like they're new concepts23:49
fungiclarkb: the 9th is the last day if you want to present something in front of people who aren't in design sessions, yes23:49
SpamapSmordred: systemd is a lot more intellgibile. And to be fair, it's not all upstart's fault. It is also that such a beast had never been in existence before, and so the implementation of the boot itself is... obtuse.23:49
mordredSpamapS: sure - and I dont' blame upstart the same was I blame systemd23:49
clarkbfungi: I think that is where we have to propose this23:49
mordredbut MAN - I mean, I had to write a "unit" today23:49
fungithere are precisely 10 minutes remaining utc in february 923:50
mordredand I did not like doing it23:50
clarkboh wait is it utc'd?23:50
clarkbnot central'd?23:50
fungino idea23:50
fungieverything should be utc23:50
*** wenlock has left #openstack-infra23:50
fungifile a bug23:50
mordredI cannot imagine it's not central time23:50
clarkbmordred: is this something you want to do?23:50
fungiaustin is the new greenwich23:50
mordredclarkb: guh ... in theory, yes. in practice ... instead we could go sit at a bar in vancouver23:50
fungiclarkb: are you _certain_ the summit is _not_ a bar in vancouver?23:51
*** sarob has joined #openstack-infra23:51
clarkbI just do not know how to say it any louder than I already have other than taking up conference time to be like "DHCP please use it also dns is handy and put services on port 443"23:51
*** andreykurilin_ has quit IRC23:51
mordredSpamapS: so now I have the element installing blank files, then the "write network scripts" portion writing the dhcp config - then the "oh, and also there's config drive" overwriting that23:51
mordredclarkb: ++23:51
fungiclarkb: ultra-lightning talks23:51
openstackgerritBoris Pavlovic proposed openstack-infra/project-config: Nicer order of jobs in Rally pipelines
mordredclarkb: we should just lightning talk23:52
clarkboh and image names are part of the damn api23:52
clarkbmordred: wfm23:52
SpamapSmordred: Ok. You're putting a loopback in /etc/network/interfaces, right?23:52
mordredthat way less prep23:52
mordredSpamapS: yah23:52
fungiclarkb: totally separate lightning talk too23:52
SpamapSmordred: ok, it gets ... weird.. without it. :)23:52
fungiyou already have two proposals there23:52
* SpamapS goes another summit w/o any good ideas for talks23:52
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add static network config element for Rackspace
mordredSpamapS: ^^ you can see the latest there23:53
clarkbSpamapS: you can steal my talk if you want to shame cloud deployers into fixing these problems23:53
mordredfungi: you might want to cover your eyes before looking at that ^^23:53
*** dimtruck is now known as zz_dimtruck23:53
* fungi prepares to shudder23:53
mordredSpamapS: yah - it's not a "feature" to update the image content AND the current image name AND the old image name23:53
*** sdake__ is now known as sdake_23:53
clarkbmordred: now that you pushed that please review 154188 :)23:54
*** dimsum__ has quit IRC23:54
*** dprince has quit IRC23:54
SpamapSmordred: so I think this will work. One thing to note: you will end up having to wait 5 minutes for eth0/eth1 to not get DHCP if they exist.23:54
mordredSpamapS: why?23:55
SpamapShm maybe not actually23:55
SpamapSmordred: static-networking-up23:55
SpamapSmordred: it's a feature :)23:55
*** dimsum__ has joined #openstack-infra23:55
SpamapSmordred: ok so actually no you won't23:55
openstackgerritJeremy Stanley proposed openstack-infra/infra-manual: Provide a subsection for CLA information
mordredSpamapS: it should be an empty file when udev dets to it23:55
mordredI was about to give up on thinking I understood linux23:55
openstackgerritMaru Newby proposed openstack-infra/devstack-gate: Use sudo to move hook script logs
fungiclarkb: 145971 needed fixing--was failing the docs job, hence not merging23:55
SpamapSmordred: as soon as local-filesystems and stopped udevtrigger happen, your script will go, which will write the thing out.23:56
*** ddieterly has quit IRC23:56
SpamapSmordred: then ifup -a will fire, and eth0 and eth1 will come up, and static-networking-up will be emitted23:56
SpamapSmordred: well really you don't want it to be empty. You want it to wait. :)23:56
SpamapSmordred: unless you can handle being in runlevel 2 without any interfaces.23:57
*** dimsum__ is now known as dims23:57
SpamapS(which 99.9% of things can handle, but not things that explicitly bind to ips, for instance)23:57
*** dims is now known as Guest7024723:58
clarkbfungi: wrong character was used?23:58
fungiclarkb: yep23:58
*** Guest70247 has quit IRC23:58
openstackgerritMerged openstack-infra/infra-manual: Provide a subsection for CLA information
SpamapSbtw I just built an image with is_chroot=='false'23:58
fungiclarkb: other parts of the document used ^ for that level rather than ~23:58
mordredSpamapS: and it worked and stuff?23:58
mordredclarkb: it's possible we were just high when we thought it wouldn't work?23:59
mordredSpamapS: OH23:59
clarkbSpamapS: nice!23:59
*** dmsimard is now known as dmsimard_away23:59
SpamapSyes it worked23:59
mordredwait - one more wrinkle23:59
SpamapSI mean23:59
SpamapSit built23:59
SpamapSno 'sploding23:59
mordreddo you have mysql running on your machine?23:59
mordredthat's it23:59
mordredsorry - I totally forgot the issue until just now23:59
SpamapSit just skips the things if no 3306?23:59
mordredwe run mysql on the nodepool node23:59

Generated by 2.14.0 by Marius Gedminas - find it at!