Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2016-02-08

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:19 gildub joined #puppet-openstack
00:34 johnmilton joined #puppet-openstack
00:47 xingchao joined #puppet-openstack
00:50 gildub joined #puppet-openstack
01:15 gildub joined #puppet-openstack
01:39 openstackgerrit Emilien Macchi proposed openstack/puppet-mistral: Implement WSGI support for mistral-api  https://review.openstack.org/277223
01:48 xingchao joined #puppet-openstack
02:47 ilbot3 joined #puppet-openstack
02:47 Topic for #puppet-openstack is now Puppet OpenStack modules - https://wiki.openstack.org/wiki/Puppet
02:49 xingchao joined #puppet-openstack
03:58 gildub joined #puppet-openstack
04:43 ducttape_ joined #puppet-openstack
04:44 mesamoo joined #puppet-openstack
05:35 nlahouti joined #puppet-openstack
06:18 darkxploit joined #puppet-openstack
07:12 magicboiz joined #puppet-openstack
07:15 jtomasek__ joined #puppet-openstack
07:24 jtomasek_ joined #puppet-openstack
07:29 magicboiz joined #puppet-openstack
07:57 jaosorior joined #puppet-openstack
08:09 rcernin joined #puppet-openstack
08:12 dgurtner joined #puppet-openstack
08:13 rochaporto joined #puppet-openstack
08:13 samuelBartel joined #puppet-openstack
08:13 jaosorior joined #puppet-openstack
08:16 hjensas joined #puppet-openstack
08:16 hjensas joined #puppet-openstack
08:16 jpena joined #puppet-openstack
08:22 xingchao joined #puppet-openstack
08:23 openstackgerrit Xingchao Yu proposed openstack/puppet-oslo: Add define oslo::messaging::amqp  https://review.openstack.org/274764
08:33 jistr joined #puppet-openstack
08:36 dgurtner joined #puppet-openstack
08:42 darkxploit joined #puppet-openstack
08:46 openstackgerrit Martin Millnert proposed openstack/puppet-neutron: params.pp: Fix opencontrail_plugin_package name  https://review.openstack.org/277318
08:58 openstackgerrit Vasyl Saienko proposed openstack/puppet-nova: Allow to set max_concurrent_build for ironic compute  https://review.openstack.org/276422
09:02 openstackgerrit joined #puppet-openstack
09:03 jaosorior joined #puppet-openstack
09:22 openstackgerrit Vasyl Saienko proposed openstack/puppet-nova: Allow to set max_concurrent_build for ironic compute  https://review.openstack.org/276422
09:32 aarefiev joined #puppet-openstack
09:42 openstackgerrit Arnoud de Jonge proposed openstack/puppet-designate: Added Designate Pool Manager  https://review.openstack.org/275294
10:05 richm joined #puppet-openstack
10:09 mesamoo joined #puppet-openstack
10:28 gildub joined #puppet-openstack
10:47 paramite joined #puppet-openstack
11:07 tfz joined #puppet-openstack
11:20 derekh joined #puppet-openstack
11:21 rcernin joined #puppet-openstack
11:25 rcernin joined #puppet-openstack
11:31 openstackgerrit Vasyl Saienko proposed openstack/puppet-nova: Allow to set max_concurrent_build for ironic compute  https://review.openstack.org/276422
11:39 gfidente joined #puppet-openstack
11:39 gfidente joined #puppet-openstack
11:43 jaosorior joined #puppet-openstack
12:15 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: The ceph extras location is no more.  https://review.openstack.org/276991
12:43 mdavidson joined #puppet-openstack
12:43 EmilienM good morning
12:46 jaosorior joined #puppet-openstack
12:50 degorenko morning :)
13:02 mgould joined #puppet-openstack
13:06 dprince joined #puppet-openstack
13:13 ducttape_ joined #puppet-openstack
13:19 rmart04 joined #puppet-openstack
13:31 rmart04 left #puppet-openstack
13:37 dmsimard EmilienM: only for gnocchi I think
13:37 dmsimard Nova api db is backwards compatible
13:38 dmsimard The other nova neutron thing, I don't know yet
13:38 EmilienM yeah, it's in my todo for this morning
13:38 EmilienM (the nova/neutron)
13:39 dmsimard I can fetch you the log from #openstack-nova but it's not much more than (Well, it works in Devstack)
13:40 dmsimard He mentioned something about keystone v2 -> v3 but doesn't go in depth of anything
13:40 trown joined #puppet-openstack
13:40 EmilienM it always works in devstack
13:43 dmsimard I need to do an easter egg of that t-shirt at the Montreal OpenStack meetup :)
13:48 trown joined #puppet-openstack
13:56 openstackgerrit Emilien Macchi proposed openstack/puppet-mistral: Refactor puppet-mistral  https://review.openstack.org/277207
13:57 openstackgerrit Emilien Macchi proposed openstack/puppet-mistral: Bring acceptance tests  https://review.openstack.org/277209
13:58 openstackgerrit Emilien Macchi proposed openstack/puppet-mistral: Implement WSGI support for mistral-api  https://review.openstack.org/277223
14:02 zigo mwhahaha: Hey, I just got librarian-puppet-simple in Debian, it's going to reach every mirrors in a few hours now... :)
14:04 magicboiz joined #puppet-openstack
14:06 EmilienM dmsimard: ok so we have AFIK 3 things :
14:06 EmilienM 1/ https://bugs.launchpad.net/puppet-nova/+bug/1542486
14:06 openstack Launchpad bug 1542486 in puppet-nova "nova-compute stack traces with BadRequest: Specifying 'tenant_id' other than authenticated tenant in request requires admin privileges" [Undecided,New]
14:07 EmilienM 2/ nova db <-- do we have a lp # ?
14:07 EmilienM 3/ gnocchi change that require new packaging but also new manifest change <-- do we have lp # ?
14:07 EmilienM dmsimard: right ?^
14:09 EmilienM I'm assigning 1/ for now, I'm working on it
14:11 jaosorior joined #puppet-openstack
14:11 dmsimard EmilienM: 2 has a related bug
14:11 dmsimard 3 as well
14:12 EmilienM dmsimard: how can we merge https://review.openstack.org/#/c/276932/ without a promotion? it's also failing on ubuntu since their packages are not updated quite often
14:13 johnmilton joined #puppet-openstack
14:13 dmsimard I don't know
14:13 dmsimard :(
14:13 johnmilton joined #puppet-openstack
14:14 dmsimard I haven't been able to identify the commit that broke things
14:15 EmilienM for centos jobs, I suggest you do a promotion without puppet jobs (if we know there is only this blocker)
14:15 EmilienM for ubuntu, we'll have to trick our nova.pp in puppet-openstack-integration to configure things differently on ubuntu & centos7
14:15 EmilienM that's the only way to make it pass I think
14:15 yankcrime anyone here coming to the eu-ops meetup in manchester, uk next week?
14:16 EmilienM dmsimard: wdyt?
14:16 yankcrime i ask as i'm moderating the puppet session - it'd be good to get some points for discussion up on the etherpad here: https://etherpad.openstack.org/p/MAN-ops-puppet
14:16 EmilienM dmsimard: I can write the p-o-i patch
14:17 dmsimard Afk getting daughter to school
14:17 EmilienM yankcrime: looking
14:17 EmilienM yankcrime: your etherpad is empty
14:17 EmilienM yankcrime: have you attended our puppet sessions before? I don't recall your nickname
14:17 yankcrime EmilienM: that's the session-specific one, let me dig out the main one
14:18 yankcrime https://etherpad.openstack.org/p/MAN-ops-meetup
14:19 yankcrime EmilienM: and nope, i haven't been to any of the previous ops meetups
14:19 v1k0d3n joined #puppet-openstack
14:19 EmilienM yankcrime: I'll ask you to reach openstack-dev with [puppet] tag to make sure you reach all of us
14:20 EmilienM I'm sure we have some folks from our group going there
14:20 yankcrime EmilienM: cool, will do - thanks
14:20 EmilienM and I would be happy if they could lead the session
14:20 yankcrime EmilienM: sure, there's been a fair amount of discussion on the openstack-operators list about moderators, so far noone else has stepped forward however
14:21 jaosorior joined #puppet-openstack
14:22 bdossant joined #puppet-openstack
14:23 EmilienM ok 3/ is https://bugs.launchpad.net/puppet-gnocchi/+bug/1542069
14:23 openstack Launchpad bug 1542069 in puppet-gnocchi "puppet-gnocchi fails Exec[manage_gnocchi_high_policy] on latest RDO" [Undecided,In progress] - Assigned to David Moreau Simard (dmsimard)
14:26 yankcrime EmilienM: i should probably point out that it's not specifically development-focussed, given it's an operator's meetup with the title of the session being 'experience with puppet deployments'
14:26 EmilienM yankcrime: most of our groups are operators
14:26 EmilienM so I want to know if mfisch or clayton or mdorman and others are going there
14:27 EmilienM they're part of our group and could also lead the session
14:27 EmilienM yankcrime: that's why asking on openstack-dev [puppet] would help
14:29 dmsimard EmilienM: back
14:30 EmilienM dmsimard: I rechecked https://review.openstack.org/#/c/276489/ so integ jobs will run
14:30 EmilienM dmsimard: (I enabled it last night)
14:30 EmilienM if it pass, we can merge it, right?
14:30 EmilienM or you think it will fail
14:30 dmsimard EmilienM: I tested that patch with current-passed-ci and it failed
14:30 EmilienM damn
14:31 EmilienM dmsimard: so we need a ninja promotion
14:31 dmsimard One of my comments has a link to the failing logs
14:31 dmsimard Doh, logs.openstack deleted the logs o_O
14:32 yankcrime EmilienM: i'll suggest to the organisers that they should widen the email audience when calling for moderators to this session - i think your point could be applied not just to puppet
14:32 yankcrime although i'd assume that most of the operators that care would be subscribed to openstack-operators as well anyway ;)
14:34 EmilienM dmsimard: we have 2 options, bump our CI to trunk or do a RDO promotion without puppet jobs
14:35 dmsimard EmilienM: Packstack and RDO manager still aren't passing, we're not close to a promotion. Packstack needs the api_db patch still and it looks more complicated than expected.
14:35 dmsimard The Triple-O patch is here, maybe you can review it: https://review.openstack.org/#/c/276790/
14:36 dmsimard Once the api_db fix merges in both, we'll know if they also hit that nova-neutron problem
14:36 EmilienM yankcrime: I'll check with our group who is going
14:36 dmsimard my head still hurts from all the searching I did friday :/
14:36 yankcrime EmilienM: perfect, thanks
14:36 EmilienM dmsimard: waiting for CI on https://review.openstack.org/#/c/276790/
14:36 EmilienM and I'll ask gfidente and dprince to land it
14:36 yankcrime EmilienM: and if they could respond to tom fitfield / matt jarvis's email regarding a call for moderators, that'd be perfect
14:38 EmilienM dmsimard: ok, lot of deps here
14:39 dmsimard EmilienM: I think you might want to find a hash that works for you in current and we can set you back to passed-ci when we figure out all that pita
14:39 EmilienM that's sounds like a good idea
14:40 EmilienM dmsimard: I'm testing the latest hash
14:40 EmilienM make it land
14:40 EmilienM and get back to you
14:40 EmilienM degorenko: I'll need your help for reviews
14:40 degorenko EmilienM, whatever for you :)
14:41 degorenko EmilienM, btw, take a look on my comments here: https://review.openstack.org/276961 when you will have time
14:42 dmsimard EmilienM: so were you taking over https://review.openstack.org/#/c/276932/ or do you expect me to submit a new PS ?
14:42 EmilienM dmsimard: I'll take over all
14:42 dmsimard EmilienM: okay, sure
14:43 EmilienM I'll even do different patchsets so we can revert some parts later
14:43 EmilienM degorenko: ok will look after the storm
14:44 degorenko EmilienM, so, what's kind of help you need from me? :) bunch of patches or what?
14:44 dprince dmsimard: +2 on the nova_api patch
14:44 EmilienM degorenko: I'll let you know in a few minutes
14:44 degorenko ack
14:45 dmsimard dprince: cool, that was my first OoO patch :P
14:45 dmsimard thanks
14:45 dmsimard dprince: I know you were planning to do it since you added support in puppet-nova for it but you weren't around so I figured I'd take a stab at it
14:46 dprince dmsimard: yep, no problem. I was on travel last week so I was slow :)
14:47 hjensas joined #puppet-openstack
14:51 ducttape_ joined #puppet-openstack
14:53 chem joined #puppet-openstack
14:53 EmilienM dmsimard: also taking over https://review.openstack.org/#/c/276932/
14:54 dmsimard EmilienM: okay
14:54 EmilienM dmsimard: did you find which commit for nova?
14:54 dmsimard EmilienM: no. I asked the developers if they could point me at a recent change that could explain why it broke but was skillfully ignored
14:55 EmilienM so nice
14:55 dmsimard There's a bazillion commits in nova every day and I don't even have a timeframe to narrow it down as to when it broke since the issue was hidden by the nova-api database thing
14:55 EmilienM sure, we don't have time for that
15:13 magicboiz joined #puppet-openstack
15:13 Anticimex this channel "owns" puppet-glance, right?
15:13 Anticimex (s/channel/project ?)
15:14 Anticimex assuming so, I'm getting bug with stable/kilo: Error 400 on SERVER: Could not autoload puppet/type/glance_image: Could not autoload puppet/provider/glance_image/glance: superclass mismatch for class Glance on node openstack-dev-glance2.srv.lab.cloud.ipnett.se
15:16 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Align nova-neutron configuration for keystone v3  https://review.openstack.org/276932
15:16 EmilienM dmsimard: ^
15:17 dmsimard EmilienM: doesn't look like that actually changes anything (as in: the resulting manifest)
15:17 dmsimard and my patch didn't pass tests
15:18 EmilienM https://review.openstack.org/#/c/276932/1..2/manifests/network/neutron.pp
15:18 EmilienM it does
15:18 EmilienM yeah it won't pass tests now
15:18 EmilienM I'm working on it
15:18 rochaporto joined #puppet-openstack
15:19 dmsimard EmilienM: the errors for my patch are available in the last run here https://review.openstack.org/#/c/276492/
15:19 EmilienM we'll have to use depends-on
15:19 dmsimard It's already there
15:19 EmilienM dmsimard: yeah but also in puppet-nova
15:19 EmilienM I guess?
15:19 dmsimard there's a depends on on the gnocchi patch and the nova-neutron patch
15:20 EmilienM dmsimard: you mean tempest still does not pass? damn
15:20 dmsimard EmilienM: with my patch http://logs.openstack.org/92/276492/7/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/7a3c0c4/console.html.gz#_2016-02-05_22_43_19_258
15:21 EmilienM yeah
15:21 dmsimard but the error is different http://logs.openstack.org/92/276492/7/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/7a3c0c4/logs/nova/nova-compute.txt.gz#_2016-02-05_22_37_35_061
15:21 EmilienM http://logs.openstack.org/92/276492/7/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/7a3c0c4/logs/nova/nova-compute.txt.gz#_2016-02-05_22_37_45_578
15:21 dmsimard looks related to keystone config
15:21 EmilienM right
15:21 dmsimard I'm not super familiar with v3 and domains
15:22 dmsimard but I saw puppet-openstack-integration seemed to set it up with sensible defaults
15:22 EmilienM is richm around?
15:22 baker joined #puppet-openstack
15:23 clayton EmilienM: mfisch and I both aren't going to be at the ops mid-cycle.  it's just too much travel time for the 2 days
15:23 EmilienM clayton: ok, thx
15:23 EmilienM maybe mdorman?
15:25 EmilienM dmsimard: I'm diff'ing the nova.conf from devstack and ours
15:26 dfisher joined #puppet-openstack
15:28 EmilienM dmsimard: devstack nova.conf has http://127.0.0.1:35357/v3
15:28 EmilienM while we don't have the version
15:30 EmilienM yeah that's it
15:30 EmilienM http://logs.openstack.org/92/276492/7/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/7a3c0c4/logs/nova/nova-compute.txt.gz#_2016-02-05_22_37_45_246
15:30 EmilienM dmsimard: ^
15:30 EmilienM http://127.0.0.1:35357/auth/tokens does not work I think
15:30 dmsimard makes sense
15:31 EmilienM we need to set the version (again)
15:31 * EmilienM taking care of it
15:31 dmsimard so does this make mandatory for operators to run v2 and v3 side by side ?
15:31 EmilienM it sounds like, yes
15:31 EmilienM though i'm not the best person to confirm it
15:32 Anticimex ok, re-ran stable/liberty puppet-glance now with stable/liberty openstacklib, on clean puppet server and clean puppet agent host. still get this bug
15:32 EmilienM Anticimex: which bug?
15:34 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Align nova-neutron configuration for keystone v3  https://review.openstack.org/276932
15:36 EmilienM dmsimard: not sure how we'll merge the puppet-nova change actually
15:36 dmsimard EmilienM: it really sucks.
15:36 EmilienM because it will never pass CI on ubuntu
15:36 dmsimard EmilienM: is v3 mandatory now or something ? We need answers..
15:36 EmilienM it seems like yes
15:37 dmsimard so it wasn't before
15:37 dmsimard and I checked the release notes and saw no mention of that
15:37 dmsimard I RTFM
15:37 EmilienM dmsimard: I asked on #o-nova
15:37 Anticimex EmilienM: https://gist.github.com/Millnert/d1c2a9938156ca6bdbcf
15:37 Anticimex i'll check local puppet apply next
15:37 EmilienM Anticimex: it would help if we could read your manifest
15:38 xarses_ joined #puppet-openstack
15:38 Anticimex i'll add it
15:38 EmilienM stable/liberty runs fine in our CI
15:38 Anticimex good to know, thanks; let me put that manifest up
15:38 Anticimex ... i was stuck thinking it's passenger module caching bug
15:39 Anticimex EmilienM: gist updated.  but i haven't even looked there yet myself
15:39 Anticimex ie didn't look into manifest yet. :s  doing now
15:39 EmilienM dmsimard: or, I disable voting for our integ jobs
15:40 dmsimard EmilienM: why do you expect ubuntu to fail ?
15:40 EmilienM dmsimard: because it does not do the v3 thing by default
15:41 EmilienM so let's say we want to configure nova::network::neutron in nova.pp in puppet-openstack-integration
15:41 dmsimard puppet-keystone/puppet-openstack-integration sets up things differently by distro ?
15:41 EmilienM we'll have to patch first puppet-openstack-integration and then puppet-nova
15:41 EmilienM not puppet-keystone
15:41 EmilienM the issue is in nova.conf now
15:42 EmilienM puppet-nova needs a patch in puppet-openstack-integration to land
15:42 EmilienM but vice-versa
15:42 EmilienM so we're doomed I guess
15:42 dmsimard oh
15:42 dmsimard I see
15:42 dmsimard man, we got like 3 chicken and egg going, this is awesome
15:42 EmilienM only for ubuntu
15:42 richm EmilienM: hello
15:42 EmilienM yeah I'm going to disable voting on the integ jobs
15:43 dmsimard EmilienM: because Ubuntu is not up to date ?
15:43 EmilienM yes
15:43 EmilienM richm: hey, we had a question on keystone v3. Do you know if nova made it mandatory to talk with neutron notifs?
15:43 dmsimard Can James help us with that ?
15:43 EmilienM who is james
15:43 dmsimard James Page ?
15:43 EmilienM ah james
15:43 dmsimard lol :P
15:43 EmilienM they update packaging every milestone
15:44 EmilienM so not very helpful in our case
15:44 dmsimard ouch
15:44 EmilienM m1, m2, m3, etc
15:44 hjensas joined #puppet-openstack
15:44 hjensas joined #puppet-openstack
15:44 EmilienM I don't even know where we are now
15:44 dmsimard We're past M2, M3 in a while
15:45 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Bump RDO packaging and update nova.conf on RedHat systems  https://review.openstack.org/276492
15:45 dmsimard M3 at the end of february
15:45 EmilienM I'm about to disable voting for our integ jobs (/me sad panda)
15:45 richm EmilienM: I do not know
15:45 richm EmilienM: maybe ayoung would know
15:47 darkxploit joined #puppet-openstack
15:55 farid_ joined #puppet-openstack
15:56 rcernin joined #puppet-openstack
15:59 rahulait joined #puppet-openstack
16:04 trown|brb joined #puppet-openstack
16:05 mdorman joined #puppet-openstack
16:11 EmilienM mdorman: you going in manchester?
16:13 mdorman EmilienM:   i am not
16:13 trown joined #puppet-openstack
16:14 EmilienM ok, will check with degorenko mwhahaha nibalizer or crinkle if they go, otherwise I guess nobody in our group is going
16:14 mdorman alright.  yeah, kind of a bummer
16:18 Anticimex EmilienM: seems my bug is indeed a puppet Passenger bug.
16:18 Anticimex with conflicting modules in r10k multi-environments
16:19 _ducttape_ joined #puppet-openstack
16:19 Anticimex so working around it in site conf instead
16:23 degorenko EmilienM, where we should go? :)
16:26 EmilienM degorenko: in manchester, ops meetup
16:26 degorenko EmilienM, oh, no, i'm sorry, but i'm not in England
16:26 EmilienM cool no worries
16:41 ducttape_ joined #puppet-openstack
16:42 bogdando joined #puppet-openstack
16:46 nlahouti joined #puppet-openstack
16:55 yankcrime EmilienM: at the risk of stating the obvious, if there's anything you would like those attending to discuss / consider - even in your group's absence - then please add it to that etherpad all the same
16:56 EmilienM yankcrime: well, if you lead the sessions, you can gather some feedback
16:56 yankcrime EmilienM: of course
17:08 IvanBerezovskiy joined #puppet-openstack
17:10 EmilienM nibalizer: any chance you can look at https://review.openstack.org/#/c/277457/ when you have 1 min?
17:18 EmilienM ok integration jobs should be non-voting in a few time
17:19 EmilienM so when https://review.openstack.org/#/c/276492/ is green, we'll merge puppet-nova & puppet-gnocchi changes
17:19 EmilienM dmsimard: ^
17:19 EmilienM and then if all is good, you run your promotion jobs
17:19 EmilienM an when you do so, we'll go back to tested repo
17:19 EmilienM and re-enabling voting for integ jobs
17:19 dmsimard We're going to need some damn good release notes for Mitaka, lots of breaking changes
17:19 EmilienM dmsimard: is it a good plan?
17:20 EmilienM yes, I'll work on this
17:21 dmsimard Promotion jobs will require triple-o and packstack to be fixed as well but with puppet-openstack fixed, it shouldn't be too hard to figure out what to do for them
17:21 EmilienM build RPM for packstack
17:21 EmilienM and poke dprince for ooo patches
17:21 EmilienM dmsimard: did you patch both undercloud & overcloud?
17:21 dmsimard EmilienM: I only submitted a tht patch for the nova api database
17:22 dmsimard I'm not familiar with the rest of the changes required
17:22 EmilienM we need undercloud too
17:22 dmsimard dprince: ^
17:22 EmilienM instack-undercloud
17:36 baker joined #puppet-openstack
17:37 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Align nova-neutron configuration for keystone v3  https://review.openstack.org/276932
17:38 xarses dmsimard: https://review.openstack.org/#/c/274951/
17:39 dmsimard yeah I had looked at PS1, let me diff between that and PS4
17:39 xarses thanks
17:40 dmsimard xarses: lgtm
17:41 dmsimard -5.5k lines :)
17:41 dmsimard Love those patches
17:41 colonwq dmsimard, my patches (finally) passed this weekend. thanks for helping me get them into shape.
17:42 dmsimard colonwq: cool I'll try and look
17:44 colonwq dmsimard, thanks.
17:45 xarses dmsimard: removing lines is the best =)
17:48 openstackgerrit Merged openstack/puppet-ceph: Drop deprecated rspec-system tests and releases  https://review.openstack.org/274951
17:58 openstackgerrit Denis Egorenko proposed openstack/puppet-openstack-integration: Remove managing of cinder types for switching on a new scheme  https://review.openstack.org/277514
17:59 openstackgerrit Denis Egorenko proposed openstack/puppet-cinder: Replace defines for managing cinder types with providers  https://review.openstack.org/273513
18:19 openstackgerrit Merged openstack/puppet-ceph: Split the Rados GW tests into parts  https://review.openstack.org/252664
18:51 brandon joined #puppet-openstack
18:57 openstackgerrit Brandon Weeks proposed openstack/puppet-ceph: Add support for puppetlabs-concat 2.x  https://review.openstack.org/276969
19:04 openstackgerrit Brandon Weeks proposed openstack/puppet-ceph: Add support for puppetlabs-apt 2.x  https://review.openstack.org/276970
19:08 _ducttape_ joined #puppet-openstack
19:11 dmsimard EmilienM: ooohhh, we got a success on the keystone v3 nova patch gj
19:11 openstackgerrit Muhammad Salman Khan proposed openstack/puppet-neutron: Change of PLUMgrid plugin package and addition of new parameters.  https://review.openstack.org/276515
19:11 EmilienM where do you see that man
19:11 dmsimard EmilienM: https://review.openstack.org/#/c/276932/
19:11 dmsimard Lots of failed but the two important ones passed :P
19:12 EmilienM oh that
19:13 EmilienM why does it work on ubuntu?
19:13 EmilienM oh yeah i know
19:13 EmilienM let me fix unit tests
19:13 EmilienM and dig why one trusty job is failing
19:13 EmilienM oh no http://logs.openstack.org/32/276932/4/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/ff90cca/console.html#_2016-02-08_18_55_00_494
19:14 dmsimard looks like ceph timeout
19:14 EmilienM the bug we had in tripleo also
19:14 EmilienM not sure it's consistent
19:16 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: The ceph extras location is no more.  https://review.openstack.org/276991
19:16 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Align nova-neutron configuration for keystone v3  https://review.openstack.org/276932
19:18 EmilienM dmsimard: it's a good news for puppet-nova, it means current packaging does not need promotion to test v3
19:18 EmilienM so you were right, only gnocchi is chicken and eg
19:22 ducttape_ joined #puppet-openstack
19:23 openstackgerrit Brandon Weeks proposed openstack/puppet-ceph: Bump allowed version of puppetlabs-concat  https://review.openstack.org/276969
19:23 brandon dmsimard: ^ better?
19:24 dmsimard brandon: nitpick would be to keep commit message within 80 columns but otherwise yes :)
19:24 dmsimard EmilienM: do we actually gate against commit messages now?
19:24 dmsimard I think I saw you commit something for that
19:25 EmilienM dmsimard: no we won't, it has been rejected by infra, the job is actually not doing what we wanted
19:25 dmsimard EmilienM: oh ?
19:25 EmilienM it's for DocImpact in fact
19:25 EmilienM and we don't have that
19:26 openstackgerrit Brandon Weeks proposed openstack/puppet-ceph: Bump allowed version of puppetlabs-concat  https://review.openstack.org/276969
19:29 farid_ joined #puppet-openstack
19:33 xarses_ joined #puppet-openstack
19:42 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Bump RDO packaging and update nova.conf on RedHat systems  https://review.openstack.org/276492
19:51 hjensas joined #puppet-openstack
19:51 hjensas joined #puppet-openstack
19:54 dfisher EmilienM: do you have a pointer to your full functional test that installs a single-node via puppet?
19:55 EmilienM s/your/our/
19:55 EmilienM everything is here: https://github.com/openstack/puppet-openstack-integration
19:55 * dfisher nods
19:55 dfisher thanks!
19:57 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Initial work for multinode support  https://review.openstack.org/220297
20:03 baker joined #puppet-openstack
20:05 dprince joined #puppet-openstack
20:07 openstackgerrit Brandon Weeks proposed openstack/puppet-ceph: Bump required version of puppetlabs-apt  https://review.openstack.org/276970
20:07 farid_ joined #puppet-openstack
20:11 _ody While I had my head in the sand doing production cloud launch stuff...did CI get smarter?  I just noticed that simple docs/README changes aren't trickerring the full testing suite.
20:13 EmilienM _ody: yeah, I did a patch for that
20:13 EmilienM is it working well?
20:14 EmilienM _ody: https://review.openstack.org/274137
20:14 gildub joined #puppet-openstack
20:14 EmilienM trying to save CI resources
20:17 baker joined #puppet-openstack
20:19 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: [WIP] Run puppet  https://review.openstack.org/220635
20:20 EmilienM dmsimard: can you stop workflow https://review.openstack.org/#/c/276489/ please ?
20:20 EmilienM we need this patch
20:23 EmilienM dmsimard: gordon +1'ed, in french! lol
20:27 dgurtner joined #puppet-openstack
20:34 baker joined #puppet-openstack
20:38 EmilienM dmsimard: https://review.openstack.org/#/c/276932/ is green (almost except ceph)
20:39 EmilienM mwhahaha: can you look at it please?
20:39 mwhahaha k
20:39 EmilienM and also https://review.openstack.org/#/c/276489/
20:40 EmilienM it's weird ceph is not passing though, ping xarses_ & dmsimard about this one
20:40 EmilienM dgurtner: also maybe
20:41 EmilienM it sounds like ceph is failing on ubuntu http://logs.openstack.org/32/276932/5/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/86d1c42/console.html#_2016-02-08_20_10_02_425
20:45 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: [WIP] Run puppet  https://review.openstack.org/220635
20:53 _ody EmilienM: Yeah.  Very cool thing to add.
20:53 baker joined #puppet-openstack
20:56 ducttape_ joined #puppet-openstack
20:56 baker joined #puppet-openstack
20:56 openstackgerrit Merged openstack/puppet-neutron: Deprecate outdated auth_ options  https://review.openstack.org/274368
20:58 EmilienM _ody: can you look https://review.openstack.org/#/c/276489/ please ,
20:59 _ody looking
20:59 brandon joined #puppet-openstack
21:00 EmilienM mwhahaha: I commented
21:00 mwhahaha the problem is for us who take master and integrate it
21:01 EmilienM mwhahaha: well, we already discussed about that I think
21:01 openstackgerrit Merged openstack/puppet-barbican: Use uppercase 'S' in word "OpenStack"  https://review.openstack.org/276600
21:01 EmilienM we consider master something we could break
21:01 mwhahaha while technically yea it wasn't released, it may be in use so if someone git pulls, it breaks everything
21:01 mwhahaha k let me just double check something and i'll update
21:01 EmilienM http://specs.openstack.org/openstack/puppet-openstack-specs/specs/kilo/master-policy.html
21:05 openstackgerrit David Gurtner proposed openstack/puppet-ceph: Update supported version information  https://review.openstack.org/277562
21:06 EmilienM is deprecation respected between liberty & mitaka for this param?
21:07 johnmilton joined #puppet-openstack
21:07 baker joined #puppet-openstack
21:10 EmilienM _ody: and https://review.openstack.org/#/c/276932/ when you got time
21:13 Fin1te joined #puppet-openstack
21:14 _ody ok. np.  I'll take a look now.
21:14 EmilienM thanks
21:15 xarses I need some help with the auth in openstacklib, I need to use it against an env with out v3.0 API, I've set OS_IDENTITY_API_VERSION=2.0 but it's still attempting to do domain commands
21:21 baker joined #puppet-openstack
21:21 _ody EmilienM: I approved https://review.openstack.org/#/c/276932/ but quickly noticed that gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty is failed.  Looks like it has been succesful otherwise.  I'll keep an eye on it to see if it fails on the gate too.
21:22 _ody nvm.  It is failing on other nova stuff
21:22 _ody lgtm
21:23 EmilienM _ody: right, it's failing on ceph
21:23 EmilienM we need to investigate, but in the meantime we need to bump our CI
21:23 EmilienM _ody: this is the last one: https://review.openstack.org/#/c/276492/
21:23 EmilienM that will fix what we wanted at the end
21:27 brandon_ joined #puppet-openstack
21:29 _ody EmilienM: That last one implments the changes from https://review.openstack.org/#/c/276932/, yeah?
21:30 EmilienM _ody: yeah
21:30 EmilienM that's why we have depends-on
21:30 EmilienM so we can test it
21:30 _ody Cool.  Looks good.
21:30 EmilienM the failure on ubuntu jobs is unrelated but in my radar
21:32 crinkle xarses: I think openstacklib always uses v3 now, why do you need it to do v2?
21:32 xarses crinkle: because I'm installing kilo
21:32 EmilienM with openstacklib master?
21:33 xarses no, liberty
21:33 EmilienM we don't support that
21:33 crinkle xarses: kilo should do v3
21:33 EmilienM http://specs.openstack.org/openstack/puppet-openstack-specs/specs/kilo/master-policy.html#proposed-change
21:34 EmilienM and yeah, crinkle is right, v3 is supported since kilo
21:34 nlahouti joined #puppet-openstack
21:37 xarses well the problem comes from that I can't use openstacklib from kilo because at least one module requires openstack_config and our patch for dealing with our butt old version of mysql isn't in either, and some other things
21:38 xarses well v3 may be supported, but openstack client has no clue what to do with the domain calls
21:38 xarses I'm not sure why still, as you said v3 is supposed to be supported
21:39 mwhahaha it was probably supported but not in openstack client
21:41 openstackgerrit Merged openstack/puppet-nova: Align nova-neutron configuration for keystone v3  https://review.openstack.org/276932
21:47 dgurtner joined #puppet-openstack
21:47 dgurtner joined #puppet-openstack
21:49 baker joined #puppet-openstack
21:49 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Initial work for multinode support  https://review.openstack.org/220297
21:53 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: Add support for civetweb  https://review.openstack.org/254548
21:58 xarses I still think the question stands, how do I get auth to stop trying openstack client domain calls
21:58 xarses Honestly I don't see how we even default to v3 when it was v2 in kilo and just that should be enough to get me going for now
22:01 dgurtner colonwq: hope I didn't confuse you on https://review.openstack.org/#/c/276991/6
22:01 EmilienM dgurtner: hey, have you seen the errors we have in our CI? on http://logs.openstack.org/92/276492/9/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/e1c20cc/
22:02 dgurtner EmilienM: just looked at it, but don't know what's happening
22:02 EmilienM xarses: using OS_IDENTITY_API_VERSION=2, keystone will use 'default' domain
22:03 EmilienM dgurtner: we're starting to gate the module in our integration jobs
22:03 EmilienM dgurtner: http://logs.openstack.org/92/276492/9/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/e1c20cc/console.html#_2016-02-08_20_35_42_669
22:03 EmilienM dgurtner: we just run that: https://github.com/openstack/puppet-openstack-integration/blob/master/manifests/ceph.pp
22:04 EmilienM xarses: we documented everything here: http://specs.openstack.org/openstack/puppet-openstack-specs/specs/kilo/api-v3-support.html
22:04 dgurtner EmilienM: it timeouts during "ceph-disk activate /srv/data"
22:04 EmilienM right
22:06 xarses EmilienM: thanks, I'll give that a try. I was going off of it needing to be set to 2.0 as https://github.com/openstack/puppet-openstacklib/blob/master/spec/unit/provider/openstack/auth_spec.rb#L50
22:07 xarses dgurtner: how long of a timeout?
22:07 EmilienM default exec timeout I guess
22:07 EmilienM it's not failing on centos7, waird
22:07 EmilienM weird*
22:08 dgurtner EmilienM: is there a simple way to run puppet with --trace ?
22:08 EmilienM 9.2.0 on ubuntu 0.94.5 on centos7
22:08 EmilienM dgurtner: https://github.com/openstack/puppet-openstack-integration/blob/master/run_tests.sh#L47
22:10 xarses EmilienM: how big is /srv/one?
22:11 xarses actually, activate should be almost instant
22:11 xarses is the monitor actually up and not expecting another member?
22:11 EmilienM xarses: http://logs.openstack.org/92/276492/9/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/e1c20cc/logs/df.txt.gz
22:11 EmilienM you can read all logs
22:13 jaosorior joined #puppet-openstack
22:14 xarses http://logs.openstack.org/92/276492/9/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-trusty/e1c20cc/logs/syslog.txt.gz
22:14 xarses ceph-mon in there
22:14 xarses looks like it didn't stay running
22:15 EmilienM well, I'm not sure what puppet-ceph is testing, but it's not working on ubuntu trusty
22:15 xarses trusty =(
22:15 EmilienM xarses: that's Ubuntu LTS
22:16 EmilienM the distro used in production
22:16 EmilienM and we use ceph.com packages
22:16 xarses did this stop working this morning?
22:16 EmilienM very recently yes
22:16 EmilienM let me tell you
22:17 * EmilienM goes on http://status.openstack.org/openstack-health/#/
22:17 xarses we merged https://review.openstack.org/#/c/274951/ today
22:19 EmilienM it used to work on Friday
22:19 dgurtner xarses: but the only change outside of spec tests is this https://review.openstack.org/#/c/274951/4/manifests/profile/params.pp
22:19 xarses ya I know
22:19 EmilienM yeah, and we set the uuid
22:19 EmilienM or the repo url?
22:20 xarses just checked, there is a fsid in the ceph.conf
22:20 EmilienM it could be https://github.com/openstack/puppet-ceph/commit/cba5340f8abb5d3abe594470c26e171df4ef4a2d
22:20 dgurtner xarses: and the fsid is different from the dropped old one, so should be fine...
22:20 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: The ceph extras location is no more.  https://review.openstack.org/276991
22:24 EmilienM weird, the repo change is not the root cause I think
22:24 EmilienM the packages are mirrored
22:24 EmilienM oh no I found out
22:26 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: repos: update ceph repo apt pin url  https://review.openstack.org/277593
22:26 EmilienM dgurtner, xarses ^
22:26 EmilienM can I add integ jobs to puppet-ceph gate?
22:27 EmilienM so it won't happen again
22:29 rahulait joined #puppet-openstack
22:33 xarses EmilienM: ya, the integration jobs can always use some love
22:33 dgurtner EmilienM: fine with me
22:34 xarses this also leads me to believe that the uca packages don't work with puppet-ceph
22:34 EmilienM no
22:34 EmilienM they are really broken
22:34 dgurtner EmilienM: but tbh. there's already quite some complaints about how difficult it is to contribute to the module, and not everyone who uses ceph is interested in openstack
22:34 EmilienM that's true
22:34 EmilienM but having CI jobs thay say "this use-case is broken" brings you free feedback
22:37 xarses dgurtner: it goes back to me wanting to (but lacking the time) to separate the integration tests between ceph rgw and external projects so that we can have them non-voting/gating as necessary
22:38 dgurtner xarses: but then I wonder if we actually ever would merge something that fails tests?
22:38 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Initial work for multinode support  https://review.openstack.org/220297
22:39 EmilienM dgurtner: that's a very pragmatic question :)
22:39 EmilienM if you feel confortable to merge patches that are failing CI, I guess you know what you do
22:40 xarses dgurtner: hmm, deep...
22:41 EmilienM my point is, I really want to test glance/nova/cinder with rbd backends, so I need to deploy ceph using puppet-ceph in our integration jobs. If I do that, I also want to gate puppet-ceph on those jobs so we don't have failures every morning
22:41 EmilienM that said, I can disable voting for those jobs, on puppet-ceph
22:41 EmilienM so it'll be free feedback for you, without the constrainst to block patches that fail
22:42 dgurtner works for me
22:42 xarses well separating them would still allow for better inspection of where is broken, if random contributor cant deal with the openstack side, we can still coach / fix it as needed
22:42 EmilienM but eventually one day, you'll see someone coming up and say "hey why puppet-ceph is broken in puppet openstack modules?" :-)
22:42 EmilienM xarses: separated how?
22:42 EmilienM ceph is currently running on scenario001 https://github.com/openstack/puppet-openstack-integration#description
22:43 EmilienM and i'm working on getting glance/cinder/nova/gnocchi using rbd backend
22:44 xarses I wanted to move the openstack integration tests out of ceph_rgw_spec.rb into a separate job so that its very clear to see which is failing
22:44 EmilienM xarses: we don't run rgw at all
22:44 EmilienM we run mon+osd for now, and that's all we need
22:44 EmilienM I'm not sure we talk about the same thing
22:45 EmilienM I'm talking about puppet openstack integration jobs, not puppet-ceph integration jobs https://github.com/openstack/puppet-openstack-integration
22:45 xarses we also only test cephx as part of keystone/swift so we should have another test on the internal side before kicking that over to keystone/swift side
22:45 xarses hehe
22:45 xarses ya I'm talking about https://github.com/openstack/puppet-ceph/blob/master/spec/acceptance/ceph_rgw_spec.rb
22:45 EmilienM not me :)
22:46 xarses I noticed
22:46 EmilienM but yeah, converging CI would be great
22:46 EmilienM not sure if you guys feel confortable with that
22:46 EmilienM dmsimard, xarses, dgurtner ^
22:46 xarses I'm up to seeing how it would work
22:47 dgurtner to me more tests are better ;)
22:47 xarses especially if we have them voting on puppet-ceph
22:47 xarses the keystone/swift/rgw test would probably make more sense there
22:48 EmilienM dgurtner, xarses, dmsimard: please vote https://review.openstack.org/277598
22:54 openstackgerrit Merged openstack/puppet-neutron: Support the LibreSwan driver in Red Hat platforms  https://review.openstack.org/273471
22:54 EmilienM dgurtner, xarses: I'm taking over pabelanger's work in our CI to have multi node jobs, hopefully it will also help you to test it
22:54 dgurtner EmilienM: that would make them voting, correct?
22:55 EmilienM nope
22:55 EmilienM they are currently non voting
22:55 baker joined #puppet-openstack
22:55 dgurtner EmilienM: yes. that would help a lot - single node ceph is a bit weird
22:55 EmilienM but if they become again we can disable voting for puppet-ceph if that's what you want
22:57 rahulait joined #puppet-openstack
22:59 openstackgerrit David Gurtner proposed openstack/puppet-ceph: Update supported version information  https://review.openstack.org/277562
23:02 openstackgerrit David Gurtner proposed openstack/puppet-ceph: Update supported version information  https://review.openstack.org/277562
23:07 openstackgerrit Vasyl Saienko proposed openstack/puppet-nova: Allow to set max_concurrent_build for ironic compute  https://review.openstack.org/276422
23:14 xarses dgurtner: should we get into package sources in this CR?
23:27 dgurtner xarses: ? not sure I understand your question
23:53 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: Add support for civetweb  https://review.openstack.org/254548

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary