Tuesday, 2015-06-16

« Monday, 2015-06-15 Index Wednesday, 2015-06-17 »
*** javeriak has joined #openstack-ansible00:00
*** Mudpuppy has joined #openstack-ansible00:05
*** Mudpuppy has quit IRC00:11
*** Mudpuppy has joined #openstack-ansible00:11
*** stevemar has joined #openstack-ansible00:18
*** javeriak has quit IRC00:40
*** Mudpuppy is now known as Cloudpuppy00:55
*** britthou_ has joined #openstack-ansible00:58
*** britthouser has quit IRC01:01
*** Cloudpuppy has quit IRC01:02
*** dkalleg has quit IRC01:05
*** abitha has quit IRC01:08
*** sigmavirus24 is now known as sigmavirus24_awa01:20
*** Mudpuppy has joined #openstack-ansible01:28
*** stevemar has quit IRC03:06
*** tlian has joined #openstack-ansible03:14
*** sura8257 has quit IRC03:17
*** mfisch has quit IRC03:32
*** sdake has quit IRC03:32
*** mfisch has joined #openstack-ansible03:36
*** mfisch is now known as Guest6866003:36
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Move swift_allow_all_users to role defaults  https://review.openstack.org/19098803:41
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Set permissions on user_secrets.yml to 0600  https://review.openstack.org/19185103:44
*** abitha has joined #openstack-ansible03:49
*** fangfenghua has joined #openstack-ansible03:55
*** abitha has quit IRC04:12
*** radek_ has joined #openstack-ansible04:14
*** radek__ has quit IRC04:15
*** sdake has joined #openstack-ansible04:17
*** meshok0 has joined #openstack-ansible04:21
openstackgerritMerged stackforge/os-ansible-deployment: Configurable lxc cache and rootfs path  https://review.openstack.org/19004804:28
openstackgerritMerged stackforge/os-ansible-deployment: Fix spelling of runlevel in init scripts  https://review.openstack.org/19149504:49
*** meshok0 has quit IRC05:11
openstackgerritMerged stackforge/os-ansible-deployment: Configure DB addresses for each service  https://review.openstack.org/19027105:23
*** javeriak has joined #openstack-ansible05:24
*** tlian has quit IRC05:28
*** stevemar has joined #openstack-ansible05:30
*** meshok0 has joined #openstack-ansible06:09
*** sdake has quit IRC06:14
*** fawadkhaliq has joined #openstack-ansible06:17
*** fangfenghua has quit IRC06:28
openstackgerritJaveria Khan proposed stackforge/os-ansible-deployment: Adding PLUMgrid plugin option to neutron  https://review.openstack.org/18896406:54
*** fawadkhaliq has quit IRC07:05
*** jmccrory has quit IRC07:08
*** jmccrory has joined #openstack-ansible07:12
*** JRobinson__ has quit IRC07:16
*** meshok0 has quit IRC07:19
*** fangfenghua has joined #openstack-ansible07:23
*** meshok0 has joined #openstack-ansible07:57
*** meshok0 has quit IRC08:09
*** meshok0 has joined #openstack-ansible08:09
*** javeriak has quit IRC08:26
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Split environment file into component parts  https://review.openstack.org/19210208:29
*** stevemar has quit IRC08:52
*** fawadkhaliq has joined #openstack-ansible08:52
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Set default pass in aio scripts to be random  https://review.openstack.org/19026608:54
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Move swift_allow_all_users to role defaults  https://review.openstack.org/19098808:54
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Allow protocol to be set per endpoint-type  https://review.openstack.org/19073608:57
*** meshok0 has quit IRC08:58
openstackgerritMerged stackforge/os-ansible-deployment: Set permissions on user_secrets.yml to 0600  https://review.openstack.org/19185109:44
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152810:15
openstackgerritMerged stackforge/os-ansible-deployment: Allow protocol to be set per endpoint-type  https://review.openstack.org/19073610:34
openstackgerritMerged stackforge/os-ansible-deployment: Set default pass in aio scripts to be random  https://review.openstack.org/19026610:39
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Use default service_region var for all roles  https://review.openstack.org/18966410:45
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152810:45
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152810:48
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Implement service_region user_group_var  https://review.openstack.org/18966410:51
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152811:08
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Allow protocol to be set per endpoint-type  https://review.openstack.org/19214811:15
openstackgerritDarren Birkett proposed stackforge/os-ansible-deployment: Set default pass in aio scripts to be random  https://review.openstack.org/19216211:51
openstackgerritJesse Pretorius proposed stackforge/os-ansible-deployment: Configurable lxc cache and rootfs path  https://review.openstack.org/19216511:52
mancdazhttps://review.openstack.org/192162 cores this is a straight cherry-pick of a simple patch from master-kilo. Appreciate if you could do the needful11:53
*** fangfenghua has quit IRC11:58
*** fawadkhaliq has quit IRC12:10
*** tlian has joined #openstack-ansible12:54
openstackgerritMerged stackforge/os-ansible-deployment: Move swift_allow_all_users to role defaults  https://review.openstack.org/19098812:56
*** sdake has joined #openstack-ansible13:32
*** sdake has quit IRC13:32
*** sdake has joined #openstack-ansible13:33
*** Bjoern__ has joined #openstack-ansible13:41
*** rrrobbb has joined #openstack-ansible13:59
*** Mudpuppy has joined #openstack-ansible14:04
*** Bjoern__ is now known as BjoernT14:08
*** rromans has quit IRC14:11
sigmavirus24_awaCool story: the problem I'm having with the fernet review is fixed by cryptography 1.0 (which is unreleased) thanks to a cffi 1.0 release14:12
palendaeBetter go help them release14:12
sigmavirus24_awaSo here's the question, do we point to master of cryptography for a while and wait for them to release 1.0?14:12
sigmavirus24_awapalendae: okay14:12
sigmavirus24_awawill do14:12
Sam-I-Amsigmavirus24_awa: become core?14:12
odyssey4mesigmavirus24_awa what's the issue exactly?14:13
sigmavirus24_awaodyssey4me: even with single-threaded apache, cffi reloads the bindings for openssl which causes some rather nasty exceptions that are pretty much https://github.com/pyca/cryptography/issues/177614:13
sigmavirus24_awaWe're using a version with the commit (that fixed that issue) but it still happens anyway14:14
*** rromans has joined #openstack-ansible14:14
*** jwagner_away is now known as jwagner14:14
sigmavirus24_awaAt least on my AIOs and it prevents me from completeing a full AIO build using fernet tokens14:14
odyssey4mesigmavirus24_awa so what are our options? do we drop the apache front-end? have you tested any updates beyond to see if it is fixed by the next proposed release?14:14
odyssey4mesigmavirus24_awa also, it would seem that the openstack-ci build worked?14:15
sigmavirus24_awaodyssey4me: openstack CI build isn't using fernet14:15
*** jwagner is now known as jwagner_ceph_tra14:15
odyssey4mesigmavirus24_awa ah, of course - it wasn't something you enabled for the AIO so it would build successfully... fair enough14:16
* odyssey4me needs coffee or something14:16
sigmavirus24_awaodyssey4me: or something14:16
*** sigmavirus24_awa is now known as sigmavirus2414:16
palendaeodyssey4me: Come on, you should be wide awake by now :)14:17
*** Mudpuppy has quit IRC14:17
odyssey4mepalendae the afternoon slump14:17
palendaeAh, yes14:17
palendaeGo do your 10k steps, re-energize ;)14:17
*** Mudpuppy has joined #openstack-ansible14:18
*** yaya has joined #openstack-ansible14:19
odyssey4mesigmavirus24 bah, it seems that was the last commit to the 0.8.x branch too14:23
openstackgerritHugh Saunders proposed stackforge/os-ansible-deployment: Generate consistent length passwords in AIO  https://review.openstack.org/19222014:23
sigmavirus24odyssey4me: yep14:23
*** Guest68660 is now known as mfisch14:30
*** mfisch is now known as Guest8122614:31
sigmavirus24so odyssey4me thoughts on this problem?14:32
sigmavirus24odyssey4me: fwiw, I tested out cryptography 1.0 from master in the keystone containers and it works perfectly14:32
sigmavirus24Didn't fail once. I'm even thinking of turning up the thread values to see if it continues to work perfectly14:32
odyssey4mesigmavirus24 give that a bash, for interest's sake...14:33
odyssey4meI'm curious about what's different between 0.8.2 and 1.0 then14:33
sigmavirus24Should I use Mjolnar?14:33
sigmavirus24odyssey4me: cffi pre-1.0 and cffi 1.014:33
sigmavirus24cffi 1.0 was a huge effort partially funded by the Python Software Foundation14:33
sigmavirus24It has a huge list of improvements14:33
sigmavirus24cffi 1.0 required some breaking changes though which are addressed in cryptography master14:34
* sigmavirus24 hangs out with the cryptography developers which is the only reason he knows this14:34
sigmavirus24*"hangs out"14:34
sigmavirus24(a.k.a, lurks)14:34
odyssey4mesigmavirus24 https://github.com/pyca/cryptography/issues/186814:35
odyssey4mesigmavirus24 is there some sort of timeline for the 1.0 release?14:36
sigmavirus24I jumped off IRC last night and no one's around this morning. I think half the team is on vacation at the moment14:36
sigmavirus24I've also offered to help get cryptography 1.0 out the door however possible14:37
sigmavirus24Which of course means staying up past 1AM instead of staying up til 1AM to keep my other stuff moving too14:37
*** Guest81226 has quit IRC14:38
odyssey4mesigmavirus24 did you try 0.91 ?14:46
odyssey4me*0.9.114:46
*** rrrobbb has quit IRC14:46
sigmavirus24odyssey4me: that's what's installed by default14:46
sigmavirus24That's the version I was seeing the issue with initially14:46
sigmavirus24sorry14:46
odyssey4mebah :/14:47
sigmavirus24odyssey4me: I was working on this until 1900 local time last night14:47
sigmavirus24or there-abous14:47
sigmavirus24*abouts14:47
odyssey4memaybe a solution then is to fix to a current sha in master?14:47
odyssey4meonce 1.0 releases we can revise14:47
*** stevemar has joined #openstack-ansible14:50
*** galstrom_zzz is now known as galstrom14:50
odyssey4mesigmavirus24 yeah, I'd suggest doing that - just explain why you're doing it in the review14:51
sigmavirus24Yep14:51
sigmavirus24And fwiw, with the default thread count, it works still14:51
sigmavirus24This is definitely a cffi problem14:51
sigmavirus24What I've learned from tailing the logs, though, is that we need to make sure the key_repository directory is not world readable14:52
sigmavirus24keystone issues warnings when it sees that14:52
odyssey4mecool - perhaps also make a note in the two issues indicating that it's resolved in the current master14:52
sigmavirus24Which two issues?14:53
odyssey4mesigmavirus24 https://github.com/pyca/cryptography/issues/1776 and https://github.com/pyca/cryptography/issues/186814:55
sigmavirus24Oh on cryptography's issues, yeah sure14:56
sigmavirus24There appears to be two features left before cryptography wanted to release 1.014:56
*** mfisch has joined #openstack-ansible14:56
sigmavirus24So I'm going to see how I can help accelerate those14:56
*** mfisch is now known as Guest2567614:56
odyssey4mesigmavirus24 sweet, that'll be helpful I'm sure :)14:59
sigmavirus24Paul's been trying to trick me into working on Cryptography for a while now15:00
sigmavirus24So15:00
sigmavirus24He's welcome15:00
palendaeJust make a bug that affects you and boom15:01
palendaeDone15:01
*** Guest25676 is now known as mfisch15:01
*** mfisch has quit IRC15:02
*** mfisch has joined #openstack-ansible15:02
sigmavirus24I'm a scourge on projects15:03
sigmavirus24You don't want me15:03
palendaeI know I don't15:03
palendaeI've seen you contributions15:03
sigmavirus24Urban Dictionary definition of "sigmavirus24": Brings death and pestilence to all he touches15:03
*** Mudpuppy has quit IRC15:03
palendaesigmavirus24: Please go work on io and node then15:04
*** Mudpuppy has joined #openstack-ansible15:04
sigmavirus24I'm working on rack right now15:04
sigmavirus24palendae: https://github.com/rack/rack/issues/726 https://github.com/rack/rack/pull/89215:05
palendaeWell that one's already dead15:05
palendaeAlso not openstack, stop.15:05
palendaeThough you're helping tenderlove, so carry on15:05
sigmavirus24Also https://github.com/kennethreitz/requests/issues/211715:06
*** jaypipes has quit IRC15:06
*** yaya has quit IRC15:07
*** jwagner_ceph_tra is now known as jwagner_away15:24
*** jwagner_away is now known as jwagner15:28
openstackgerritMerged stackforge/os-ansible-deployment: Set default pass in aio scripts to be random  https://review.openstack.org/19216215:55
*** yaya has joined #openstack-ansible16:03
Sam-I-Ammmmmeeting?16:04
*** gtt116_ has joined #openstack-ansible16:05
odyssey4meo/ yeah, who else is around?16:07
palendae<-16:08
palendaeb3rnard0: ?16:08
*** gtt116__ has quit IRC16:08
odyssey4mebug list: https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.status=New16:08
*** javeriak has joined #openstack-ansible16:09
odyssey4mehttps://bugs.launchpad.net/openstack-ansible/+bug/146395616:09
openstackLaunchpad bug 1463956 in openstack-ansible "slow_query_log should be off by default and relocated" [Undecided,New]16:09
b3rnard0did cloudnull assign a backup?16:09
Sam-I-Amb3rnard0: looks like you volunteered16:09
odyssey4meb3rnard0 he's just asked me to cover for him today and thu16:10
b3rnard0k16:10
Sam-I-Amb3rnard0: denied16:10
odyssey4methis info looks wrong for master/kilo - the slow log is located in /var/log/mysql_logs/ for master/kilo as I recall, so it would appear that this bug info is for juno/icehouse16:11
odyssey4meBjoernT are you around?16:11
BjoernTyes16:11
*** daneyon has joined #openstack-ansible16:11
odyssey4meok, confirmed - /var/log/mysql/mysql-slow.log is for juno; /var/log/mysql_logs/mysql-slow.log is for kilo/master16:13
*** javeriak has quit IRC16:13
odyssey4meas I recall log rotation is enabled for master/kilo too16:13
*** daneyon_ has joined #openstack-ansible16:14
odyssey4meah - no log rotation for anything16:16
BjoernTyeah ii should not go into the container at all16:16
BjoernTbecause our log rotation is still incomplete, or mostly missing16:16
*** daneyon has quit IRC16:17
odyssey4meBjoernT not sure I agree with that, the slow log is meant to be a useful tool for diagnosis - so I'd prefer to ensure that log rotation is set, but the slow log is left enabled... but that would be up to an implementer I guess16:17
BjoernTwell as I said no one is looking into it so why even logging in the first place unless you actually need it16:18
odyssey4mewe should allow that to be enabled/disabled though instead of hard setting it16:18
*** vdo has quit IRC16:18
BjoernTyeah thats fine too16:18
*** javeriak has joined #openstack-ansible16:20
odyssey4meany thoughts about importance?16:20
BjoernTnormal16:21
BjoernTnothing high16:21
odyssey4mealright, set to low16:22
odyssey4menext: https://bugs.launchpad.net/openstack-ansible/+bug/146437216:23
openstackLaunchpad bug 1464372 in openstack-ansible "Environment variable setup does not work in container creation" [Undecided,New]16:23
odyssey4mefor now that looks pertinent to kilo/master - any thoughts on whether this matters for juno?16:25
*** daneyon_ has quit IRC16:25
*** daneyon has joined #openstack-ansible16:25
palendaeI would say leave juno out unless we have an urgent need16:26
odyssey4mepalendae agreed, any thoughts on importance here?16:27
palendaeLow, I'd say. Also sounds like we may have a commit incoming from the reporter16:27
odyssey4meI suppose for the proxy stuff this is pretty important?16:27
palendaeHrm, yeah...I haven't worked with them enough to know overall impact though16:28
palendaeNot sure how many deployments are behind them16:28
openstackgerritIan Cordasco proposed stackforge/os-ansible-deployment: Add support for deploying Keystone with Fernet  https://review.openstack.org/18999816:28
odyssey4meI'm surprised that it wasn't resolved as part of the proxy work: https://review.openstack.org/#/c/181603/16:28
odyssey4memaybe this issue is actually resolved for master/kilo and only is an issue for juno?16:29
sigmavirus24Did juno get that included? I don't recall if I backported that patch16:29
*** rrrobbb has joined #openstack-ansible16:30
odyssey4mesigmavirus24 only master/kilo here: https://review.openstack.org/#/q/I2a397abad6557a48bc737dcd9cca787fc5afa728,n,z16:31
sigmavirus24palendae: you're on bugs now right?16:31
palendaeI suppose16:31
sigmavirus24(not on the multi-region stuff)16:31
palendaeYeah16:31
sigmavirus24Mind taking over that backport, you'll probably need https://gist.github.com/cloudnull/3015a4cff9573866ce5516:31
palendaeSure16:32
sigmavirus24juno/icehouse were going to be a pain compared to master/kilo16:32
palendaeRe-reading that bug16:33
palendaeIt sounds like this might actually be juno16:33
odyssey4mepalendae if you can confirm that'd be great - shall I assign it to you?16:33
palendaeSure16:33
palendaehrm, no, that role doesn't exist in Juno. Still something that'll need to be done though16:34
odyssey4methanks palendae16:34
odyssey4menext: https://bugs.launchpad.net/openstack-ansible/+bug/146572316:34
openstackLaunchpad bug 1465723 in openstack-ansible "Please add bash completion for the openstack clients" [Undecided,New]16:34
odyssey4methis seems like something that would be useful for us to do16:35
sigmavirus24BjoernT: how do you propose we add bash completion for the clients? Do all of them have bash completion for their individual packages?16:35
odyssey4meit would seem that https://github.com/openstack/python-novaclient/blob/master/tools/nova.bash_completion has something to do with it?16:35
BjoernTyes16:36
odyssey4me(for nova)16:36
* sigmavirus24 is pretty sure nova's the only one he's seen with bash completion16:36
BjoernTeach project has the own config, at least nova and neutron does16:36
odyssey4mekeystone: https://github.com/openstack/python-keystoneclient/blob/master/tools/keystone.bash_completion16:36
odyssey4mewe'll have to figure out how that's implemented16:37
odyssey4meI think this is a good candidate for master/kilo16:37
sigmavirus24Anyway, as long as it's installed  properly (which may not be the case anymore, depending on how pbr implements this) we should be able to automate that16:37
BjoernTyeah16:37
* sigmavirus24 agrees with odyssey4me 16:37
* sigmavirus24 has a concern with pbr + pip 7.1 doing things the right way16:38
sigmavirus24but I'll have to experiment with that before I comment further =P16:38
odyssey4meis that a volunteer I hear sigmavirus24 ?16:38
sigmavirus24not for the next few weeks =P16:38
odyssey4meok, I think this is a wishlist item personally - any thoughts?16:39
sigmavirus24My concern is only with how pbr tells setuptools to consider those files16:39
sigmavirus24odyssey4me: I agree16:39
sigmavirus24there's a bug with setuptools + wheel + data_files and if pbr is pointing at those files calling them data_files then they're already broken and there's going to be a lot of fun getting those bash completions in the right place for use16:39
sigmavirus24(or even determining where they're installed)16:40
palendaeOops - we removed all the environment variable stuff: https://github.com/stackforge/os-ansible-deployment/commit/86b6e8e6409f76c46551becdac783a76bb853980#diff-9724ba1a4e7c37915c868ad54dd4655216:41
sigmavirus24palendae: https://github.com/stackforge/os-ansible-deployment/commit/86b6e8e6409f76c46551becdac783a76bb853980#diff-9724ba1a4e7c37915c868ad54dd46552R20416:42
sigmavirus24palendae: split out (it seems) to a newtask16:42
palendaeAh, you're right16:43
odyssey4meyep16:43
palendaethough that's a poor name...16:43
palendaeOur case was proxies16:43
* odyssey4me agrees16:43
palendaeBut it's doing all env vars16:43
palendaew/e16:43
sigmavirus24palendae: patches welcome16:43
palendaeThanks16:43
sigmavirus24You're welcome16:43
sigmavirus24palendae: EVERYONE IS WELCOME16:43
palendaeThat was thanks for the actual help, not the snark16:44
* sigmavirus24 really should have slept in this morning16:44
odyssey4meok, new bug triage done - any specific bugs that anyone wants to look through or get help with?16:44
* Apsu should have slept in this week16:44
sigmavirus24I'd like to file a bug about my caffeine leves16:45
sigmavirus24*levels16:45
ApsuDrove my caffeine to the levee but the levee was dry..16:45
sigmavirus24A+ reference Apsu16:45
sigmavirus24mmmmm pie16:46
Apsu'merica16:46
odyssey4melol, alright guys - thanks for all the fish :)16:46
sigmavirus24fish is an okay shell16:46
* sigmavirus24 prefers zsh though16:47
odyssey4mego to sleep sigmavirus24  and Apsu16:47
sigmavirus24nou16:47
Apsukk16:47
ApsuFish is a hilarious shell16:47
ApsuThough, I think sea creatures with shells are by definition not fish.16:47
*** javeriak has quit IRC16:48
sigmavirus24they're rustaceans, no?16:49
sigmavirus24WE SHOULD WRITE A SHELL IN RUST!16:50
sigmavirus24anyone else seeing build failures due to a ppa problem with rsyslog?16:51
sigmavirus24(http://paste.openstack.org/show/5MaLRETdnp4GWCNyNAdu/)16:52
*** yaya has quit IRC16:58
odyssey4mesigmavirus24 that has been seen to happen from time to time16:59
sigmavirus24gate's seeing a different error16:59
sigmavirus24maybe it's something transient on my AIO16:59
odyssey4mewe should actually just log a bug for it and remove it - the ppa isn't actually needed for anything we're doing in master/kilo16:59
*** Mudpuppy has quit IRC17:00
sigmavirus24So yaprt doesn't seem to like https://review.openstack.org/#/c/189998/14/playbooks/defaults/repo_packages/openstack_other.yml it seems17:00
sigmavirus24That17:00
sigmavirus24Or ansible auto-fails when something's printed to stderr17:01
sigmavirus24http://logs.openstack.org/98/189998/14/check/os-ansible-deployment-dsvm-check-commit/fd54977/console.html#_2015-06-16_16_52_27_96217:01
odyssey4mesigmavirus24 check the full ansible log: http://logs.openstack.org/98/189998/14/check/os-ansible-deployment-dsvm-check-commit/fd54977/logs/ansible-logging/ansible.log17:05
sigmavirus24I did17:05
sigmavirus24it doesn't give much more info17:05
sigmavirus24the error is at the very bottm17:05
*** yaya has joined #openstack-ansible17:05
*** javeriak has joined #openstack-ansible17:05
*** abitha has joined #openstack-ansible17:11
*** rrrobbb has quit IRC17:12
*** javeriak has quit IRC17:16
*** javeriak has joined #openstack-ansible17:16
*** rrrobbb has joined #openstack-ansible17:16
*** gtt116_ has quit IRC17:16
*** gtt116_ has joined #openstack-ansible17:17
*** yaya has quit IRC17:19
*** sdake has quit IRC17:19
*** sdake_ has joined #openstack-ansible17:19
odyssey4mesigmavirus24 sorry, was talking with mancdaz - that should work, dammit17:24
sigmavirus24lol17:25
*** dkalleg has joined #openstack-ansible17:25
sigmavirus24I scrapped my AIO for a new one17:25
sigmavirus24because it wasn't doin gthe build packages step and I'm running stuff on a fresh AIO to repro17:25
odyssey4mesigmavirus24 I'm out - if it's still an issue tomorrow then I'll build an AIO to see if I can figure it out.17:29
sigmavirus24No problem17:29
sigmavirus24I'm working on it17:29
odyssey4mepalendae if you could peek at https://review.openstack.org/191528 it'd be great - this patch will help overcome some of the gating failures we've been seeing, specifically hpcloud apt failures.17:35
odyssey4mecloudnull I did a revision of https://review.openstack.org/189664 which you may prefer. Please review.17:36
*** jaypipes has joined #openstack-ansible17:37
*** yaya has joined #openstack-ansible17:56
*** daneyon_ has joined #openstack-ansible17:57
*** yaya has quit IRC17:59
*** daneyon has quit IRC18:00
palendaeodyssey4me: Reviewed, think I found a very small problem18:00
*** fawadkhaliq has joined #openstack-ansible18:03
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Implement service_region user_group_var  https://review.openstack.org/18966418:08
*** Mudpuppy has joined #openstack-ansible18:11
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152818:14
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Implement service_region user_group_var  https://review.openstack.org/18966418:14
*** BjoernT has left #openstack-ansible18:18
*** Mudpuppy has quit IRC18:21
*** fawadkhaliq has quit IRC18:22
*** jwagner is now known as jwagner_away18:36
*** daneyon has joined #openstack-ansible18:42
*** daneyon_ has quit IRC18:45
openstackgerritMiguel Grinberg proposed stackforge/os-ansible-deployment: Implement service_region user_group_var  https://review.openstack.org/18966418:52
openstackgerritMiguel Grinberg proposed stackforge/os-ansible-deployment: Support an externally hosted keystone  https://review.openstack.org/19201518:52
openstackgerritMiguel Grinberg proposed stackforge/os-ansible-deployment: Support an externally hosted keystone  https://review.openstack.org/19201519:06
*** dkalleg has quit IRC19:06
*** sdake_ is now known as sdake19:13
*** sdake has quit IRC19:15
*** sdake has joined #openstack-ansible19:15
openstackgerritMiguel Grinberg proposed stackforge/os-ansible-deployment: Support an externally hosted keystone  https://review.openstack.org/19201519:16
*** yaya has joined #openstack-ansible19:20
*** gtt116 has joined #openstack-ansible19:29
* sigmavirus24 is very uncertain why yaprt is barfing on cryptography19:32
*** gtt116_ has quit IRC19:32
Sam-I-Amsigmavirus24: cryptic errors?19:32
cloudnullsigmavirus24: maybe i can help :)19:33
sigmavirus24cloudnull: out of training already?19:33
cloudnullwe were sent home , because of storm .19:33
Sam-I-Amlol storm19:34
Sam-I-Amthis aint no storm19:34
sigmavirus24lol19:34
sigmavirus24cloudnull: so http://logs.openstack.org/98/189998/14/check/os-ansible-deployment-dsvm-check-commit/fd54977/console.html#_2015-06-16_16_52_27_962 is from https://review.openstack.org/#/c/189998/14/playbooks/defaults/repo_packages/openstack_other.yml19:34
cloudnullidk, trainer bounced thus here i am19:34
sigmavirus24which is necessary for the fernet work19:34
sigmavirus24so my report generated on my AIO is showing no requirements for cryptography19:36
sigmavirus24which I'm pretty sure it has19:36
palendaeCeph can't stand up to storms19:36
stevelleso don't put your datacenter in Oklahoma?19:38
cloudnullsigmavirus24:  looking now.19:39
sigmavirus24cloudnull: I can give you access to this AIO boxen too if you want19:39
*** rrrobbb has quit IRC19:41
openstackgerritMerged stackforge/os-ansible-deployment: Adding PLUMgrid plugin option to neutron  https://review.openstack.org/18896419:55
vincent_vdkI suddenly understand Dutch...19:55
svgvincent_vdk: ?19:56
sigmavirus24vincent_vdk: do you know kung fu too?19:56
vincent_vdksvg: just went through the history :)19:56
vincent_vdksigmavirus24: not yet :)19:57
*** jwagner_away is now known as jwagner19:57
sigmavirus24cloudnull: ideas?19:58
*** dkalleg has joined #openstack-ansible19:58
cloudnulli believe its the "-e ., -e vectors" because the dep processor is looking at the items in the requirement files as literal requirements.19:59
cloudnullim testing something with that now.19:59
sigmavirus24ah19:59
sigmavirus24that explains that error message20:00
openstackgerritKen Johnston proposed stackforge/os-ansible-deployment-specs: Spec for multiregion ansible deployment  https://review.openstack.org/19242120:02
cloudnullyup thats the issue.20:06
* cloudnull going to go fix20:06
sigmavirus24cloudnull: can probably just s/-e //g20:07
*** rrrobbb has joined #openstack-ansible20:08
*** rrrobbb has quit IRC20:08
cloudnulli thought about that. however in the case of cryptography they have a dev requirement of vector which is a local sub package.20:08
*** rrrobbb has joined #openstack-ansible20:08
sigmavirus24cloudnull: the other thing is that you're picking up the dev-requirements.txt and there is a slightly different set of requirements in https://review.openstack.org/#/dashboard/?foreach=%28project%3Aopenstack%2Fglance+OR+project%3Aopenstack%2Fpython%252Dglanceclient+OR+project%3Aopenstack%2Fglance_store+OR+project%3Aopenstack%2Fglance%252Dspecs%29+status%3Aopen+NOT+label%3ACode%252DReview%3C%3D%252D2&title=Glance+Review+Dashboar20:09
sigmavirus24d&With+Negative+Feedback=NOT+label%3AWorkflow%3C%3D%252D1+NOT+label%3AVerified%3C%3D%252D1%252cjenkins+NOT+owner%3Aself+label%3ACode%252DReview%252D1+limit%3A20+branch%3Amaster&Needs+Reverify=label%3AVerified%3C%3D%252D1%252cjenkins+branch%3Amaster+NOT+label%3ACode%252DReview%3C%3D%252D1+NOT+label%3AWorkflow%3C%3D%252D1&Work+In+Progress=NOT+label%3AWorkflow%3E%3D1+NOT+owner%3Aself+label%3AWorkflow%3C%3D%252D1+limit%3A20+branch%20:09
sigmavirus243Amaster&Needs+Approval=NOT+label%3AWorkflow%3E%3D1+NOT+label%3AWorkflow%3C%3D%252D1+label%3AVerified%3E%3D1%252cjenkins+NOT+owner%3Aself+label%3ACode%252DReview%3E%3D2+NOT+label%3ACode%252DReview%252D1+branch%3Amaster&Without+Negative+Feedback=NOT+label%3AWorkflow%3E%3D1+NOT+label%3AWorkflow%3C%3D%252D1+label%3AVerified%3E%3D1%252cjenkins+NOT+owner%3Aself+NOT+label%3ACode%252DReview%3C%3D%252D1+NOT+label%3ACode%252DReview%3E%320:09
sigmavirus24D2+limit%3A50+branch%3Amaster&5+Days+Without+Feedback=NOT+label%3AWorkflow%3E%3D1+NOT+label%3AWorkflow%3C%3D%252D1+label%3AVerified%3E%3D1%252cjenkins+NOT+owner%3Aself+NOT+label%3ACode%252DReview%3C%3D2+age%3A5d+branch%3Amaster&Own+Patches=owner%3Aself+status%3Aopen20:09
sigmavirus24ugh not that20:09
sigmavirus24https://github.com/pyca/cryptography/blob/master/setup.py#L3420:10
cloudnullits the processing of all the requirements which includes dev20:13
*** sdake_ has joined #openstack-ansible20:16
*** sdake has quit IRC20:20
openstackgerritMerged stackforge/os-ansible-deployment: Added apt update tasks to everything using apt  https://review.openstack.org/19152820:28
*** radek_ has quit IRC20:30
*** yaya has left #openstack-ansible20:51
sigmavirus24https://review.openstack.org/#/c/153057/ was finally approved20:52
sigmavirus24cloudnull: I see you pushed some stuff to github21:02
sigmavirus24should I test that out?21:02
cloudnullhttps://github.com/cloudnull/yaprt/commit/c0a91e38842d71aa22a5b83c5f53056d5acfbb2d21:02
cloudnullthat makes it all happy21:02
sigmavirus24is that up on github already?21:03
sigmavirus24*pypi21:03
cloudnullno21:03
sigmavirus24(or does it not need to be?)21:03
cloudnullit would need to be uploaded for osad to use it21:03
cloudnullim building now.21:03
cloudnullto see if it goes. and then ill push21:04
cloudnullif you wouldnt mind testing too it would be much appreciated.21:04
cloudnullbut it looks like its going21:04
sigmavirus24cool21:05
* sigmavirus24 goes to test21:05
cloudnullhttp://104.130.7.7:8181/reports/release-master-report.json http://104.130.7.7:8181/pools/cryptography-vectors/ http://104.130.7.7:8181/pools/cryptography/21:06
* sigmavirus24 is running21:10
sigmavirus24cloudnull: looks good from that21:10
*** tlian2 has joined #openstack-ansible21:11
*** Mudpuppy has joined #openstack-ansible21:14
*** tlian has quit IRC21:14
*** yaya has joined #openstack-ansible21:17
cloudnullsigmavirus24: https://jenkins05.openstack.org/job/os-ansible-deployment-dsvm-check-commit/489/console is running now21:18
* cloudnull fingers crossed :)21:19
sigmavirus24yeah mine's building everything and hasn't croaked yet21:19
*** persia has quit IRC21:23
*** persia_ has joined #openstack-ansible21:24
*** persia_ has quit IRC21:24
*** javeriak has quit IRC21:24
*** persia has joined #openstack-ansible21:25
*** Mudpuppy has quit IRC21:27
*** Mudpuppy has joined #openstack-ansible21:28
*** javeriak_ has joined #openstack-ansible21:28
*** javeriak_ has quit IRC21:39
*** javeriak has joined #openstack-ansible21:41
*** yaya has quit IRC21:47
*** fawadkhaliq has joined #openstack-ansible21:48
cloudnullsigmavirus24:  is everything right with the world  ?21:50
cloudnullgate wise its past keystone so LGTM https://jenkins05.openstack.org/job/os-ansible-deployment-dsvm-check-commit/489/console21:51
*** fawadkhaliq has quit IRC21:53
sigmavirus24cloudnull: so it fails on galera_containers21:57
sigmavirus24when it tries to install python-memcached it can't find a version of six that satisfies memcached's requirements (which has no freaking bounds)21:58
cloudnullit seems to be working in gate commit21:58
sigmavirus24hm21:59
sigmavirus24this seems to be a name normalization problem21:59
sigmavirus24maybe time to scrap this aio22:01
*** JRobinson__ has joined #openstack-ansible22:01
*** sdake_ is now known as sdake22:02
* sigmavirus24 shrugs22:02
cloudnullme too .22:02
cloudnulli just work here...22:02
cloudnull:)22:02
sigmavirus24well22:07
sigmavirus24time to switch over to working on cryptography so they can ship 1.022:07
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: [WIP] Backported tempest roles from Master  https://review.openstack.org/19244522:10
cloudnull+1 for upstream work22:11
*** persia has quit IRC22:23
*** Mudpuppy has quit IRC22:24
*** 21WACGOA9 has joined #openstack-ansible22:25
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Allow Horizon to access multiple regions  https://review.openstack.org/19030222:25
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Generate consistent length passwords in AIO  https://review.openstack.org/19222022:25
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Implement service_region user_group_var  https://review.openstack.org/18966422:26
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Add read/write_affinity settings for Swift  https://review.openstack.org/19102322:27
javeriakhello ansible people22:27
javeriakwas wondering if the this BP needs to get approved, as the code and spec are both merged now.. https://blueprints.launchpad.net/openstack-ansible/+spec/add-plumgrid-neutron-plugin22:28
*** 21WACGOA9 is now known as persia22:30
*** persia has quit IRC22:31
*** persia has joined #openstack-ansible22:31
*** galstrom is now known as galstrom_zzz22:34
*** openstackgerrit has quit IRC22:38
*** openstackgerrit has joined #openstack-ansible22:39
cloudnulljaveriak: done.22:40
cloudnulllooks like a clerical miss, from the spec approval.22:40
javeriakcloudnull, thanks, also question, if  wanted to add a new host afterwards during a play, how would i go about that22:40
cloudnullnew host?22:41
javeriaksaw the add_host module, it can add hosts to a new group22:41
cloudnullis it a host that will always exist? if so you should add it in the *_user_config.yml file22:41
cloudnullif its something that you will add in dynamically then the add_host module should be used.22:42
javeriaksay ive got another type of host_group that doesnt lie in rpc_user_config22:42
cloudnulli'd say add_host is what you'd want to go with22:42
javeriakit will always exist, so i should then add anew host_group to *_user_config.yml with the new hostnames/ips, the dynamic inventroy will pick them up right22:43
cloudnullyes it will22:43
javeriakoh okay, thanks22:43
cloudnullif its something that you want osad to build containers for you can create an environment entry for the same thing22:44
cloudnullbut if a host group is all you need and its percistant then adding to user config is the simplest route.22:45
*** tlian2 has quit IRC22:45
javeriakno containers, but i want to my external playbooks to be able to run on that host with host: new_host22:45
cloudnullsn entry in the user_config should do it22:45
*** Mudpuppy has joined #openstack-ansible22:45
javeriakyes user config sounds right22:45
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: [WIP] Backported tempest roles from Master  https://review.openstack.org/19244522:53
*** rrrobbb has quit IRC23:21
stevelleJust to be sure I understand the thinking, why do we not install our ansible plugins globally?23:35
cloudnullstevelle: the libraries ?23:35
stevelleyes23:35
cloudnullhttps://github.com/stackforge/os-ansible-deployment/tree/master/playbooks/plugins/lookups and https://github.com/stackforge/os-ansible-deployment/tree/master/playbooks/library ?23:36
stevelleparticularly playbooks/plugins23:36
cloudnulli guess you could. but im not sure why you would want to as it would dirty the global ansible namespace.23:37
cloudnullit would be better to pr the plugin upstream if it needed to be global23:38
cloudnullbut in our case that lookup plugin is so specific to us that i dont think anisble would accept it23:38
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: [WIP] Backported tempest roles from Master  https://review.openstack.org/19244523:39
stevelleI wrote a plugin filter last night that I need for haproxy.  no this isn't a great use case anyway23:40
stevelletrying to decide what to do with it23:40
cloudnullis it general enough to move into osad ?23:41
*** sigmavirus24 is now known as sigmavirus24_awa23:42
stevelleallows you to filter a host group based on an individual var (from hostvars) matching a fixed value23:42
cloudnulli mean does it improve haproxy and how we do haproxy deployments? it sounds like it might? if so i'd say spec it and pr it into master.23:43
stevellecould be tweaked to do regex instead of isequal or w/e23:43
*** stevemar has quit IRC23:44
*** stevemar has joined #openstack-ansible23:44
stevellefor a cross-regional identity cloud I have a template to config an haproxy-per-region and use this filter to only get the members of groups['keystone_all'] which belong to a keystone_service_region -- which I'm setting with container_vars.  Is it generally useful?  only for a multi-node dev env.23:45
stevelle(and even then one that build a multi-regional keystone)23:45
stevelleunless we think of another reason to filter hosts by vars23:45
cloudnullhum...23:47
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: [WIP] Backported tempest roles from Master  https://review.openstack.org/19244523:48
stevellebut here, don't use this in production ;)23:48
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Updated Juno for new dev work  https://review.openstack.org/18897523:50
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Backported tempest roles from Master  https://review.openstack.org/19244523:51
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Updated Juno for new dev work  https://review.openstack.org/18897523:51
cloudnullstevelle: been pondering, but idk . . . i mean it couldn't be used with anything but haproxy right ?23:52
stevellewell, it could be used anywhere you want a subset of groups['something'] because that is what it gives you23:52
stevellespecifically a subset based on variation in vars23:53
stevellemaybe it makes container_vars more useful when the values vary by host but otherwise you shouldn't have much variation there23:54
stevelleand hostvars don't include facts do they?23:54
stevelledoesn't seem to at least23:54
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Backported tempest roles from Master  https://review.openstack.org/19244523:55
openstackgerritKevin Carter proposed stackforge/os-ansible-deployment: Updated Juno for new dev work  https://review.openstack.org/18897523:55
cloudnullhostvars are set from inventory.23:56
cloudnullfacts are generated.23:56
cloudnullbut you access facts using the hostvars entry when accessing them from another host.23:57
cloudnullbut id say through it up as a wip and let some folks review it23:58
stevelleok, accessing facts using the hostvars entry might be a useful tool23:59
« Monday, 2015-06-15 Index Wednesday, 2015-06-17 »

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