Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2015-11-05

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

All times shown according to UTC.

Time Nick Message
00:03 zhangjn joined #puppet-openstack
00:14 openstackgerrit Alan Pevec proposed openstack/puppet-openstack-integration: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/241825
00:31 jerrygb joined #puppet-openstack
00:40 karmix EmilienM: Did puppet-neutron have some issues with CI related to ovs on red hat a few weeks ago?  I can't figure out why Change 237530 by Lukas Bezdicka triggered a bunch of ovs errors in CI.
00:47 markvoelker joined #puppet-openstack
01:00 zhangjn joined #puppet-openstack
01:13 jmccrory joined #puppet-openstack
01:40 karmix Yep.  CI is broken on stable/kilo, for at least the puppet-neutron repo.  It looks like it has been broken for a few weeks, and likely has something to do with changes in puppet-openstack-integration.
01:50 dgurtner joined #puppet-openstack
01:50 dgurtner joined #puppet-openstack
02:08 jmccrory joined #puppet-openstack
02:33 karmix I was mistaken on the puppet-openstack-integration part.
02:40 EmilienM karmix: it should not
02:45 EmilienM this patch merged yesterday: https://review.openstack.org/#/c/238615/
02:45 EmilienM in stable/kilo
02:45 karmix CI is broken on stable/kilo for puppet-neutron because puppet_spec_helper is cloning the master branch into spec/fixtures/modules/keystone.  Master on keystone includes change 228744 (commit: 8461e9ac), which alters how the tenant/project is set in a way that is no longer compatible with the neutron-puppet unit tests in stable/kilo.
02:45 karmix EmilienM: I'll look at that patch real quick.  I think the solution here is to pin the repos to stable/kilo in .fixtures.yml.
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 karmix EmilienM: Thanks, that explains why I was seeing puppet-openstack-integration clone the modules until I switched to the review branch.
02:50 karmix EmilienM: I am still seeing one unit test fail on the tip of stable/kilo.  Have you seen any issues with the "neutron::agents::ml2::ovs on Debian platforms configures neutron plugin ovs agent with ml2 plugin should contain Class[neutron::params]" test, lately?
02:55 openstackgerrit Doug Knight proposed openstack/puppet-neutron: Drop nova_admin_tenant_id_setter  https://review.openstack.org/237530
03:10 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: run_tests: install modules after installing puppet  https://review.openstack.org/241853
03:14 openstackgerrit Emilien Macchi proposed openstack/puppet-neutron: DNM - CI test  https://review.openstack.org/241774
03:15 zhangjn_ joined #puppet-openstack
03:15 guessi joined #puppet-openstack
03:18 guessi left #puppet-openstack
03:22 EmilienM I broke integration CI, but I have a patch: https://review.openstack.org/#/c/241853 - please review it
03:44 openstackgerrit Adam Vinsh proposed openstack/puppet-swift: WIP multi port swift config  https://review.openstack.org/241726
03:49 nlahouti joined #puppet-openstack
04:10 nlahouti1 joined #puppet-openstack
04:17 nlahouti joined #puppet-openstack
04:20 fedexo joined #puppet-openstack
04:55 markvoelker joined #puppet-openstack
04:56 degorenko joined #puppet-openstack
05:06 jamespage joined #puppet-openstack
05:28 openstackgerrit shravya Gaddam proposed openstack/puppet-trove: log file for guestagent is modified  https://review.openstack.org/241884
05:39 nihilifer joined #puppet-openstack
05:39 sanjayu joined #puppet-openstack
05:41 zhangjn joined #puppet-openstack
06:15 rahulait joined #puppet-openstack
06:33 nlahouti joined #puppet-openstack
06:37 nlahouti joined #puppet-openstack
06:50 jtomasek joined #puppet-openstack
06:56 markvoelker joined #puppet-openstack
06:57 nlahouti joined #puppet-openstack
07:05 openstackgerrit Denis Egorenko proposed openstack/puppet-horizon: Try to use zuul-cloner to prepare fixtures  https://review.openstack.org/241908
07:05 dgurtner joined #puppet-openstack
07:05 dgurtner joined #puppet-openstack
07:07 nlahouti joined #puppet-openstack
07:56 zigo joined #puppet-openstack
08:00 zhangjn joined #puppet-openstack
08:01 dgurtner joined #puppet-openstack
08:01 dgurtner joined #puppet-openstack
08:11 openstackgerrit Mykyta Karpin proposed openstack/puppet-nova: Update parameters for nova  https://review.openstack.org/241614
08:17 openstackgerrit joined #puppet-openstack
08:29 openstackgerrit Mykyta Karpin proposed openstack/puppet-keystone: add policy driver option for keystone  https://review.openstack.org/240946
08:30 karimb joined #puppet-openstack
08:32 karimb_ joined #puppet-openstack
08:33 IvanBerezovskiy joined #puppet-openstack
08:36 openstack joined #puppet-openstack
08:36 myatsenko Guys please review this patch https://review.openstack.org/#/c/203675/
08:38 jpena joined #puppet-openstack
08:46 rahulait joined #puppet-openstack
08:51 paramite joined #puppet-openstack
08:55 dgurtner joined #puppet-openstack
08:55 dgurtner joined #puppet-openstack
08:56 ddmitriev joined #puppet-openstack
08:56 markvoelker joined #puppet-openstack
09:00 openstackgerrit Maksym Yatsenko proposed openstack/puppet-nova: [WIP]  https://review.openstack.org/233011
09:06 gildub joined #puppet-openstack
09:09 derekh joined #puppet-openstack
09:24 gfidente joined #puppet-openstack
09:24 jistr joined #puppet-openstack
09:32 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-cinder: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/241214
09:57 markvoelker joined #puppet-openstack
10:00 bogdando joined #puppet-openstack
10:20 bogdando joined #puppet-openstack
10:22 zerda joined #puppet-openstack
10:23 zhangjn joined #puppet-openstack
10:25 openstackgerrit Merged openstack/puppet-openstacklib: Fallback to default rcfile  https://review.openstack.org/227762
10:35 zhangjn_ joined #puppet-openstack
10:50 chem joined #puppet-openstack
10:58 jerrygb joined #puppet-openstack
11:00 sergmelikyan joined #puppet-openstack
11:13 markvoelker joined #puppet-openstack
11:49 khushbu joined #puppet-openstack
11:54 openstackgerrit Iury Gregory Melo Ferreira proposed openstack/puppet-keystone: Support for Keystone as Service Provider  https://review.openstack.org/216821
11:59 iurygregory myatsenko, take a look at my comment in your patch =)
12:08 khushbu_ joined #puppet-openstack
12:08 zhangjn joined #puppet-openstack
12:17 openstackgerrit Alexey Deryugin proposed openstack/puppet-murano: Add basic acceptance test  https://review.openstack.org/233591
12:30 openstackgerrit Iury Gregory Melo Ferreira proposed openstack/puppet-keystone: Support for Keystone as Service Provider  https://review.openstack.org/216821
12:35 EmilienM hello
12:36 chem EmilienM: hé
12:36 EmilienM degorenko: can you  look https://review.openstack.org/#/c/241853 please ?
12:36 iurygregory morning =D
12:37 degorenko hey o/
12:37 degorenko EmilienM, ack, looking
12:37 EmilienM degorenko: the current integration testing jobs are broken withot it
12:37 EmilienM degorenko: tested with https://review.openstack.org/#/c/241774/
12:39 zhangjn joined #puppet-openstack
12:40 zhangjn joined #puppet-openstack
12:41 degorenko EmilienM, now we are trying to run puppet module list when puppet doesn't installed yet?
12:41 EmilienM yes
12:41 degorenko lol :)
12:41 EmilienM it's because of me
12:41 EmilienM let me tell you why
12:42 EmilienM yesterday we discovered our integration jobs were using a different version of puppet than beaker jobs so I added the macro that purge puppet in the job workflow and some bash code that install puppet
12:42 EmilienM #2 in the wrong place :)
12:44 iurygregory wow
12:44 degorenko EmilienM, wow :) PTO time is dark time - you don't know what's happens during this period :D
12:44 EmilienM the good thing is now our stable/kilo is green again and our integration jobs too
12:44 degorenko yep :) i saw
12:44 EmilienM degorenko: don't worry - next week I'm in your situation, I'm PTO all week
12:44 degorenko bth, i've reupload fix for horizon
12:44 degorenko EmilienM, nice :)
12:45 degorenko btw*
12:45 degorenko thanks for notification
12:45 degorenko who will be also on PTO?
12:45 EmilienM degorenko: I don't know
12:45 degorenko ah, ok :)
12:46 degorenko also, i've read weekly meeting on this week
12:46 degorenko very short :D
12:48 sergmelikyan joined #puppet-openstack
12:48 degorenko EmilienM, do we have some progress with os_service_default?
12:49 EmilienM degorenko: puppet-cinder should be covered, look at the trello card
12:49 degorenko EmilienM, yep, i read comments - may be something out of this comments was here in irc?
12:49 openstackgerrit Athlan-Guyot sofer proposed openstack/puppet-keystone: Support for services different only by type.  https://review.openstack.org/238954
12:50 EmilienM degorenko: nope
12:50 degorenko EmilienM, ok, thanks
12:50 EmilienM degorenko: still in japan?
12:51 degorenko EmilienM, no, already in Moscow :) last day in PTO. Because of jetlag :)
12:54 degorenko EmilienM, also, i saw your patches for switching to Liberty GA repos
12:54 iurygregory jetlag sucks =/ I've slept for 14hours =P
12:54 degorenko it seems that we don't have trove
12:54 EmilienM degorenko: it's blocked
12:54 EmilienM degorenko: right, RDO team is finishing the GA repo
12:55 degorenko it is only one problem, i guess?
13:02 dprince joined #puppet-openstack
13:03 EmilienM degorenko: yes
13:03 EmilienM we're working on it
13:03 EmilienM I think today or tomorrow we can have the packages
13:06 degorenko cool :)
13:07 openstackgerrit Merged openstack/puppet-openstack-integration: run_tests: install modules after installing puppet  https://review.openstack.org/241853
13:08 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/241825
13:11 jistr joined #puppet-openstack
13:11 jerrygb joined #puppet-openstack
13:12 iurygregory we can try recheck now EmilienM ?
13:14 markvoelker joined #puppet-openstack
13:14 degorenko iurygregory, i can TRY :)
13:18 markvoelker_ joined #puppet-openstack
13:22 xarses joined #puppet-openstack
13:27 morazi joined #puppet-openstack
13:27 openstackgerrit Emilien Macchi proposed openstack/puppet-manila: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242045
13:28 openstackgerrit Emilien Macchi proposed openstack/puppet-aodh: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242046
13:29 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242048
13:29 sergmelikyan joined #puppet-openstack
13:34 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
13:35 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
13:36 social grr git
13:40 social oh it's not git it's gerrit
13:41 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
13:49 openstackgerrit Maksym Yatsenko proposed openstack/puppet-nova: Floating IP range support in Nova network  https://review.openstack.org/233011
13:55 richm joined #puppet-openstack
14:04 jerrygb joined #puppet-openstack
14:09 bklei joined #puppet-openstack
14:10 openstackgerrit Merged openstack/puppet-neutron: Drop nova_admin_tenant_id_setter  https://review.openstack.org/237530
14:20 EmilienM paramite: I think you need to update https://review.openstack.org/#/c/239680/
14:20 EmilienM paramite: yesterday, we merged a change in puppet-keystone (cc chem) and it probably broke puppet-heat
14:23 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-cinder: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/241214
14:24 paramite EmilienM, perfect news :) ... ok, will check that out after z3 stuff
14:25 rgowrishankar joined #puppet-openstack
14:26 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242048
14:27 EmilienM paramite: thx. Also you can talk with chem to understand what we changed
14:29 paramite EmilienM, definitely
14:33 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
14:36 dgurtner joined #puppet-openstack
14:37 zigo EmilienM: Can you review that one? https://review.openstack.org/#/c/241615/
14:37 zigo Sorry to insist, but it's very important for me.
14:41 iurygregory Hey EmilienM, puppet-keystone CI is broken? https://jenkins05.openstack.org/job​/gate-puppet-keystone-puppet-beaker​-rspec-dsvm-trusty/231/consoleFull
14:41 iurygregory https://jenkins06.openstack.org/job/gate-puppet-ke​ystone-puppet-unit-3.3-dsvm-centos7/96/consoleFull
14:41 openstackgerrit Maksym Yatsenko proposed openstack/puppet-nova: Floating IP range support in Nova network  https://review.openstack.org/233011
14:42 EmilienM chem: have you seen that error before?
14:42 EmilienM [1;31mError: /Stage[main]/Main/Keystone_us​er_role[adminv3::admin_domain@openstackv3::admin_domain]: Could not evaluate: undefined method `empty?' for nil:NilClassm
14:42 EmilienM iurygregory: have you rebased your patch?
14:45 dfisher joined #puppet-openstack
14:46 omartsyniuk joined #puppet-openstack
14:46 _matthias_ joined #puppet-openstack
14:47 karmix joined #puppet-openstack
14:47 partner joined #puppet-openstack
14:47 jprs joined #puppet-openstack
14:48 trash joined #puppet-openstack
14:49 iurygregory joined #puppet-openstack
15:01 ducttape_ joined #puppet-openstack
15:08 openstackgerrit Gael Chamoulaud proposed openstack/puppet-modulesync-configs: Remove class_parameter_defaults puppet-lint check  https://review.openstack.org/242105
15:10 dhollinger joined #puppet-openstack
15:12 jesusaurus joined #puppet-openstack
15:14 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-cinder: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/241214
15:14 ntpttr joined #puppet-openstack
15:14 paramite chem, hey, you're there?
15:15 EmilienM mgagne, mfisch: can we merge https://review.openstack.org/#/c/240998/ and https://review.openstack.org/#/c/238631 please ?
15:25 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
15:25 sergmelikyan joined #puppet-openstack
15:28 sergmelikyan joined #puppet-openstack
15:30 markvoelker joined #puppet-openstack
15:33 ntpttr joined #puppet-openstack
15:35 markvoelker_ joined #puppet-openstack
15:40 dgurtner joined #puppet-openstack
15:40 dgurtner joined #puppet-openstack
15:44 mdorman joined #puppet-openstack
15:50 paramite joined #puppet-openstack
15:52 openstackgerrit Sergey Kolekonov proposed openstack/puppet-openstack-integration: Switch from nova_admin_* options for Neutron to actual  https://review.openstack.org/242126
15:57 openstackgerrit Lukas Bezdicka proposed openstack/puppet-tempest: Switch id setters to openstack client  https://review.openstack.org/242052
15:57 angdraug joined #puppet-openstack
16:01 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-keystone: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/242134
16:04 sergmelikyan joined #puppet-openstack
16:10 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242048
16:10 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242048
16:11 jpena does anyone know why puppet-neutron is duplicating auth configuration in neutron.conf and /etc/neutron/api-paste.ini? I've seen https://github.com/openstack/puppet-ne​utron/tree/master/manifests#L327-L337 and several other instances, and I'm pretty sure that api-paste.ini file is not even parsed in RDO
16:15 jerrygb joined #puppet-openstack
16:15 EmilienM jpena: mgagne filed a bug similar for cinder iirc
16:17 jpena EmilienM: ok, I'll file another bug and prepare a patch in a few days
16:17 IvanBerezovskiy EmilienM: hi, I've found that this commit https://review.openstack.org/#/c/227762/ broke keystone unit tests - e.g. http://logs.openstack.org/21/216821/29/​check/gate-puppet-keystone-puppet-unit-​3.3-dsvm-centos7/52e266c/console.html
16:17 IvanBerezovskiy what should we do? Update keystone tests or revert change?
16:19 IvanBerezovskiy I think I can research and fix the problem
16:19 openstackgerrit Mykyta Karpin proposed openstack/puppet-keystone: add policy driver option for keystone  https://review.openstack.org/240946
16:19 EmilienM IvanBerezovskiy: you can file a bug, and send a patch if you see the solution.
16:20 IvanBerezovskiy EmilienM: ok, sure
16:22 EmilienM chem: see https://review.openstack.org/#/c/227762/
16:24 EmilienM our whole might be broken because of ^
16:24 IvanBerezovskiy https://bugs.launchpad.net/p​uppet-keystone/+bug/1513533
16:24 openstack Launchpad bug 1513533 in puppet-keystone "unit tests are broken" [Undecided,New] - Assigned to Ivan Berezovskiy (iberezovskiy)
16:24 openstackgerrit Mykyta Karpin proposed openstack/puppet-keystone: add policy driver option for keystone  https://review.openstack.org/240946
16:27 sanjayu joined #puppet-openstack
16:30 fedexo joined #puppet-openstack
16:30 EmilienM IvanBerezovskiy: I think we need to patch puppet-keystone
16:30 IvanBerezovskiy I'm working on it
16:31 EmilienM cool
16:31 jerrygb joined #puppet-openstack
16:36 chem IvanBerezovskiy: I have a patch at the ready for the problem, is it ok or do you have it as well ?
16:37 IvanBerezovskiy I didn't publish it yet
16:38 chem IvanBerezovskiy: your choice, if you have it go ahead :)
16:45 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-keystone: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/242134
16:45 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-keystone: Fix unit tests according to fallback for openrc file  https://review.openstack.org/242153
16:45 nlahouti joined #puppet-openstack
16:47 openstackgerrit Mykyta Karpin proposed openstack/puppet-keystone: add policy driver option for keystone  https://review.openstack.org/240946
16:50 severion joined #puppet-openstack
16:52 mjblack- joined #puppet-openstack
16:52 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-keystone: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/242134
16:52 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-keystone: Fix unit tests according to fallback for openrc file  https://review.openstack.org/242153
16:54 Aegil_ joined #puppet-openstack
16:55 chem EmilienM: could you +2 the Ivan's patch, degorenko is not there anymore, another core would be nice to get this stuff in master quickly.
16:55 EmilienM sure
16:55 chem EmilienM: thanks
16:56 EmilienM if richm is around ^
16:58 timrc_ joined #puppet-openstack
16:59 m3ntat joined #puppet-openstack
17:01 nlahouti joined #puppet-openstack
17:02 ntpttr1 joined #puppet-openstack
17:03 simonmcc_ joined #puppet-openstack
17:03 dprince joined #puppet-openstack
17:03 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:03 EmilienM jayg: ^
17:03 EmilienM that's wip
17:06 rahulait joined #puppet-openstack
17:06 openstackgerrit Nate Potter proposed openstack/puppet-magnum: puppet-magnum: Initial commit  https://review.openstack.org/242164
17:07 bastelfreak joined #puppet-openstack
17:07 chem EmilienM: richm I have a question about the keystone_endpoint patch: the bug is that the type is not given so it does all sort of weird things.  Can I make this a requirement (breaking existing code) without a deprecation period, or should I add a deprecation warning (while it's really a bug, not a change of feature) ?
17:07 bastelfreak joined #puppet-openstack
17:08 bkero joined #puppet-openstack
17:08 chem EmilienM: richm in both case all the keystone manifests have to be reworked
17:08 EmilienM are we breaking the interface?
17:08 chem EmilienM: yes definitively
17:09 chem (https://review.openstack.org/#/c/23943​4/8/lib/puppet/type/keystone_endpoint.rb)
17:09 chem EmilienM: ^
17:09 richm Do we have to do this now or can we postpone?
17:09 chem hum ... it's a quite severe bug.  and is required to have keystone triple-o integration
17:09 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:09 EmilienM argh
17:10 EmilienM we're breaking some stuffs in liberty
17:10 chem EmilienM: richm: the bug is here https://review.openstack.org/#/c/234704
17:10 bnemec joined #puppet-openstack
17:10 EmilienM we need good communication: ML, doc, etc
17:10 chem (triple o stuff)
17:10 karmix joined #puppet-openstack
17:11 chem EmilienM: the thing is that it was broken, but working in simple case: only one type per service name
17:11 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:11 chem richm: ^
17:11 richm chem: so there is no way we can just wait an use openstackclient with the fix?  We have to fix in puppet-keystone?
17:11 chem EmilienM: richm in triple-o they need to have nova/compute and nova/computev3
17:12 richm they can't have novav3/computev3?
17:12 chem richm: dunno, I'm not sure I understand how it would work ^
17:12 chem richm: ^ I don't know :)
17:13 richm This is different than the problem where you could not view endpoints with v2 that were created with v3?
17:14 chem I think it's related as puppet completly messed up the enpoints/service database
17:14 bkero joined #puppet-openstack
17:14 chem not certain though
17:16 chem richm: EmilienM I can quickly try to change the stuff to have a deprecation warning, instead of a failure.  I don't know if it's doable (it's more complicated) and will let the bug open in some strange way I think, but I can investigate (one houre or so)
17:17 richm If we are proposing this as a fix for the triple-o keystone problem, we should investigate that problem more to find the root cause of the problem
17:17 EmilienM if the bug occurs somewhere else, we need to fix it asap
17:17 EmilienM even if that's something not compatible
17:17 EmilienM folks will understand that and make the change
17:18 EmilienM for that we need good communication
17:19 chem EmilienM: richm the bug is there for everyone :) But not a lot of people seems to do different than in the manifests, so fixing the manifest should render the stuff transparent for most people.  I can start a discussion on the ML for the other people to let them know about the problem and the necessary fix.
17:20 EmilienM chem++
17:21 chem EmilienM: ack, I'll do that.
17:22 dgurtner joined #puppet-openstack
17:22 dgurtner joined #puppet-openstack
17:24 richm ack
17:25 chem richm: the root cause of the problem is that they uncovered this puppet bug by having a naming convention that few people seems to have.  In fact it is impossible with puppet to create such service (same name different type), while it's completly legit (fixed there https://review.openstack.org/#/c/238954/), that's why for people only using puppet they cannot detect the bug.  The problem appeared here because the installation was
17:25 chem bootstrapped by some other tool and puppet apply after.
17:25 chem anyway, I'll write the mail.
17:28 openstackgerrit Athlan-Guyot sofer proposed openstack/puppet-keystone: Keystone_endpoint match service by name/type.  https://review.openstack.org/239434
17:31 sergmelikyan joined #puppet-openstack
17:32 richm chem: It sounds to me like they are trying to use a "new" feature that until now, no one has ever used and no one has ever tried to do it with puppet-keystone keystone_endpoint
17:33 richm and in the process found a bug in openstackclient
17:33 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:33 chem richm: humm, but having a service named nova/compute and nova/computev3 is not a new feature ?
17:33 chem richm: and the bug in not in openstackclient
17:34 chem richm: openstackclient in this case does exactly what puppet ask it to do
17:34 EmilienM IvanBerezovskiy: thx a lot for the bugfix, nice work
17:34 chem richm: or maybe I'm missing something ?
17:37 richm chem: no, not a new feature exactly, but they are doing something different than what was done in the past with respect to service names types
17:37 richm chem: the usual way was to have nova/compute and novav3/computev3
17:38 richm chem: but they are trying to do nova/compute and nova/computev3
17:38 chem richm: ah, oki.
17:38 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:38 chem richm: but it's not forbiden by the API, the usual way was just a convention
17:39 chem richm: which make puppet _appears_ to work
17:40 chem richm: can we explain that you should use the convention and not use anything else when you use puppet while it's completly possible to do it when you use the cli directly ?
17:40 chem richm: (it's really a question, no irony here)
17:44 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:45 dfisher ^ no solaris.  /sobs
17:45 richm chem: If tripleo has a legitimate use case, that they can do in no other way, then we have to fix it.
17:46 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
17:46 richm chem: Especially since it is not explicitly forbidden to do so by the keystone_endpoint api
17:46 EmilienM the fact is they did not use puppet to manage keystone resources before
17:46 EmilienM that's why we found the bug
17:47 aimon joined #puppet-openstack
17:53 chem Oki, I'll try and see if triple-o people could accept to change their naming convention to novav3/computev3 to check the time frame we have for fixing this bug.
17:53 chem richm: EmilienM ^
17:53 richm chem: ack
17:54 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: tempest: use $::id fact for permissions  https://review.openstack.org/242179
18:02 cfloare joined #puppet-openstack
18:05 sergmelikyan joined #puppet-openstack
18:10 jfluhmann joined #puppet-openstack
18:19 chem richm: I've found that => https://bugzilla.redhat.co​m/show_bug.cgi?id=1236989
18:19 openstack bugzilla.redhat.com bug 1236989 in rhel-osp-director "RFE: change computev3 service name to novav3" [Unspecified,New] - Assigned to calfonso
18:20 chem richm: I know how to fix this :)
18:20 chem richm: so it seems they won't need to be convinced after all :)
18:20 iurygregory left #puppet-openstack
18:20 iurygregory joined #puppet-openstack
18:22 richm chem: excellent
18:24 chem richm: do you a link that shows the convention used in the official doc or something (if you don't have something don't bother)
18:25 richm chem: no - it was just a convention, so nothing official afaik
18:35 openstackgerrit Merged openstack/puppet-keystone: Fix unit tests according to fallback for openrc file  https://review.openstack.org/242153
18:36 openstackgerrit Iury Gregory Melo Ferreira proposed openstack/puppet-keystone: Support for Keystone as Service Provider  https://review.openstack.org/216821
18:37 mdorman joined #puppet-openstack
18:41 _ody mwhahaha: Having to statically set '<SERVICE DEFAULT>' in rspec-puppet tests?  Seems counter to the spirit of the $::os_service_default variable in the first place, yes?
18:41 _ody I am not sure where the best place to put a "constant" is in rspec.
18:41 mwhahaha not necessarily
18:42 mwhahaha depends on where you are setting <SERVICE DEFAULT>, like are you trying to check if it's actually <SERVICE DEFAULT> and the fact is being picked up correctly (that's valid)
18:43 mwhahaha or are we defining the default expected value is <SERVICE DEFAULT> because our tests are poorly structured (needs work)
18:45 _ody Agreed.  I was just thinking we'd use a variable set locally to rspec so that if the value of $::os_service_default was changed in the future an msync run would update all the module according.
18:46 _ody It would still mean tests would have to be updated but the testing of the actual value would be seperate from the actual fact.
18:46 _ody ...just not have to be updating in as many places
18:47 mwhahaha it would be nice if we could use a constant for that in rspec, i mentioned yesterday that we should probably look at rspec-puppet-facts to managing our facts in the tests
18:47 mwhahaha that way we could stop doing the debian/redhat facts all over the place
18:48 mwhahaha which is really annoying to update
18:48 _ody Oh yes.  Indeed.
18:48 mwhahaha because then we could just reference facts[:os_service_default] or something
18:53 _ody rspec-puppet-facts is probably worth meeting agenda/mailing list discussion.  I'd be happy to convert a couple modules.
18:53 * _ody dislikes duplicating facts in every test
18:57 mdorman joined #puppet-openstack
18:58 jerrygb joined #puppet-openstack
18:59 nlahouti joined #puppet-openstack
19:01 degorenko joined #puppet-openstack
19:01 clayton mwhahaha nod, I was thinking tyhe same thing, I don't like putting the string in the tests either
19:04 openstackgerrit Nate Potter proposed openstack/puppet-magnum: puppet-magnum: Initial commit  https://review.openstack.org/242164
19:07 ddmitriev joined #puppet-openstack
19:27 iurygregory CI for puppet-keystone still broken? http://logs.openstack.org/21/216821/30/check/gat​e-puppet-keystone-puppet-beaker-rspec-dsvm-trust​y/0c26bfe/console.html#_2015-11-05_19_15_52_102
19:29 chem iurygregory: I look into it
19:29 iurygregory thanks chem =)
19:29 chem iurygregory: can you give me the link to your patch
19:29 chem ?
19:29 iurygregory yes, https://review.openstack.org/#/c/216821/
19:30 chem iurygregory: thanks
19:30 iurygregory np ^^
19:45 dhollinger joined #puppet-openstack
19:50 mwhahaha hey i'm trying to use the new keystone module with fuel and i'm getting errors about openstack COuld not determine suitable URL for the plugin from the keystone providers, anyone have any thoughts about what I might be missing/need to fix?
19:50 mwhahaha the errors are https://ci.fuel-infra.org/job/master.fuel-librar​y.pkgs.ubuntu.neutron_vlan_ha/2551/artifact/logs​/2551/fail_error_deploy_neutron_vlan_ha-fuel-sna​pshot-2015-11-05_18-42-10.tar.xz.filtered.log
19:50 iurygregory mwhahaha, i think chem is looking into it =)
19:50 jfluhmann joined #puppet-openstack
19:51 mwhahaha i think mine might be a v2 vs v3 thing
19:51 iurygregory looks like the problem i'm having in https://review.openstack.org/#/c/216821/
19:51 chem iurygregory: this is not the same error...
19:51 iurygregory oh ok =)
19:51 dgurtner joined #puppet-openstack
19:51 dgurtner joined #puppet-openstack
19:51 chem iurygregory: i think :)
19:51 mwhahaha i think we had previously patched our implementation of keystone to do something slightly different so it might be a url parameter thing
19:52 mwhahaha just was wondering if anyone had any thoughts, the googles returns something about needing to provide a v2 url instead of a v3 but i'm not sure which one is assumed by the keystone module
19:53 jerrygb joined #puppet-openstack
19:54 chem iurygregory: for your error I think that the ci is not broken (I have no confirmation of it yet ... waiting for a recheck).  but the error message is very unhelpful :) My guess is that the apache server is not properly restarted.  I have to go now for ~1h, when I'm back I'll have a vm with your code in in running the beaker, so I will be able to fully debug it.  See you in 1h30 :)
19:55 iurygregory chem, ok ;) i'm leaving the office in about 1h but in about 3h:30min i'll be at home =)
19:56 iurygregory you can ping me on irc ^^
19:57 chem richm: EmilienM : I have to update a lot of thing but basically I got this for review https://review.openstack.org/#/c/242198/ this should buy us some time for the fix in puppet-keystone to be implemented
19:57 EmilienM looking
19:57 openstackgerrit Merged openstack/puppet-cinder: Fix cinder default log_dir  https://review.openstack.org/241691
19:57 EmilienM chem: the problem is upgrade I think
19:58 EmilienM chem: you should ask slague to look
19:58 chem EmilienM I had the discussion with "bnemec"
19:58 EmilienM ok
20:01 EmilienM crinkle: what do you think about https://review.openstack.org/#/c/242162/ ?
20:03 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
20:08 crinkle EmilienM: cool! i made a comment
20:09 openstackgerrit Merged openstack/puppet-openstack_extras: spec: drop ::foo fake module  https://review.openstack.org/240998
20:10 EmilienM crinkle: I don't think we need a Vagrantfile. People just need this kind of script, imho
20:11 crinkle EmilienM: fine with me
20:11 openstackgerrit Merged openstack/puppet-openstack_extras: Try to use zuul-cloner to prepare fixtures  https://review.openstack.org/238631
20:12 dgurtner xarses: dachary dmsimard what's your thought on https://review.openstack.org/#/c/236539/
20:14 dgurtner xarses: dachary dmsimard personally I'm not entirely happy with disabling tests, but the RGW/CentOS issues are blocking everything else - and from my understanding it's not the module's fault.
20:14 dgurtner mwhahaha: feel free to chime in ^
20:15 ntpttr1 joined #puppet-openstack
20:16 dmsimard dgurtner: I was conflicted as well for that particular patchset. I think the cleanest action would be to split the test disable into another commit, file a bug and revert it when we figure the fix
20:17 dmsimard And then rebase the other patches onto master once the test disable is merged
20:20 dgurtner joined #puppet-openstack
20:20 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
20:21 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Implement all-in-one.sh script  https://review.openstack.org/242162
20:26 jfluhmann joined #puppet-openstack
20:30 dmsimard dgurtner: thoughts ?
20:32 mwhahaha probably would be a good idea to do the test one separately so it could be reverted at a later date. That being said I'd assume the changes required to make it work would probably require updating the tests anyway so i don't think relying on a revert is a must
20:35 dmsimard I'm not exactly sure about that since the third party CI isn't broken
20:43 ntpttr1 joined #puppet-openstack
20:49 dgurtner mwhahaha: it's impossible to do the test separately: to merge it needs to pass the gate - to be able to submit a new patchset the .gitreview change is required
20:49 mwhahaha no it's not
20:49 mwhahaha you just need to have your gerrit setup correclty
20:49 mwhahaha i've done it before would you like me to do it for you?
20:49 dgurtner dmsimard: from my understanding we are now having one more issue on the RGW, before it just failed the swift check, but now it actually fails to start...
20:50 dgurtner mwhahaha: I'd be happy to get pointers - my experience is you need .gitreview pointing to openstack/puppet-ceph, but if I change that it's an uncommited change in my repo and then gerrit refuses to submit the patch...
20:51 mwhahaha git stash it before git review
20:51 mwhahaha if you've already got a gerrit remote in your working dir it'll submit just fine
20:52 dgurtner it didn't... maybe it's a difference if you commit on top of an existing patchset or create a new one?
20:52 dgurtner and again, if I stash it, it refuses on the basis that the stackforge/puppet-ceph namespace is no longe known
20:52 mwhahaha weird i managed to do it on another repo
20:52 mwhahaha let me see
20:53 dgurtner but please feel free to split the two patches and resubmit if you know how! having it stuffed together like that is just plain ugly
20:54 dmsimard dgurtner: I'll give it a stab
20:54 dgurtner dmsimard: given we don't even have civetweb support in the RGW I'm happy to remove the RGW test and concentrate on getting the core part of the module running
20:55 openstackgerrit Alex Schultz proposed openstack/puppet-ceph: Disable the RGW acceptance test  https://review.openstack.org/242217
20:55 mwhahaha -^
20:56 karimb joined #puppet-openstack
20:57 jmccrory left #puppet-openstack
20:57 dmsimard he beat me to it :p
20:59 openstackgerrit Alex Schultz proposed openstack/puppet-horizon: Switch debian to openstack-dashboard  https://review.openstack.org/242218
21:00 dgurtner mwhahaha++
21:00 dgurtner I still don't understand gerrit :/
21:00 mwhahaha http://goo.gl/YeZlt8
21:02 openstackgerrit David Moreau Simard proposed openstack/puppet-ceph: Disable the RGW acceptance test  https://review.openstack.org/242217
21:02 dmsimard I added a bug reference ^
21:11 dgurtner joined #puppet-openstack
21:11 dgurtner joined #puppet-openstack
21:11 EmilienM mwhahaha: lol
21:12 EmilienM I love dogs & computes
21:12 mwhahaha :)
21:12 EmilienM mwhahaha: http://goo.gl/DK6fWP
21:16 dgurtner ¯\_(ツ)_/¯
21:16 EmilienM crinkle, mfisch: if you wanna have a look, https://review.openstack.org/#/c/242162/ is ready for review. I'll update the doc so people can start using it
21:21 rahulait joined #puppet-openstack
21:22 aimon joined #puppet-openstack
21:25 mdorman joined #puppet-openstack
21:26 aimon joined #puppet-openstack
21:39 chem iurygregory: so it's definitively not the gate :)
21:40 chem iurygregory: looking what is happening right now
21:42 nlahouti joined #puppet-openstack
21:42 chem iurygregory: http://fpaste.org/287445/67597751
21:43 chem iurygregory: that's what I though, apache is not restarting properly, but we don't see the full puppet log, which make this obvious:
21:44 chem http://fpaste.org/287446/46759842
21:44 chem iurygregory: the full puppet log ^
21:45 chem iurygregory: but the  undefined method `collect' for nil:NilClass is not helpful, I'll see if the error message can be improved here
22:08 nlahouti joined #puppet-openstack
22:09 chem iurygregory: I've found the solution.  Put it in the review :)
22:10 dhollinger joined #puppet-openstack
22:13 EmilienM chem: is https://review.openstack.org/#/c/238954/ ready?
22:13 dgurtner joined #puppet-openstack
22:13 dgurtner joined #puppet-openstack
22:14 chem EmilienM: yeap
22:14 EmilienM chem: lgtm
22:15 _morgan joined #puppet-openstack
22:15 _morgan left #puppet-openstack
22:16 chem EmilienM: cool
22:17 openstackgerrit David Moreau Simard proposed openstack/puppet-ceph: Update .gitreview for new namespace  https://review.openstack.org/236539
22:18 dmsimard mwhahaha: that was quick
22:18 mwhahaha :D
22:21 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: Implement Gnocchi Statsd support  https://review.openstack.org/240663
22:22 openstackgerrit Emilien Macchi proposed openstack/puppet-gnocchi: [DO NOT MERGE] test with pure RDO Liberty testing repo (stage CBS repos)  https://review.openstack.org/242048
22:22 bastelfreak joined #puppet-openstack
22:35 msimonin joined #puppet-openstack
22:36 msimonin left #puppet-openstack
22:36 mwhahaha i don't know what's up with the rubygems & openstack infra today
22:36 mwhahaha all the puppet jobs have been super flakey :/
22:40 dmsimard rubygems is unstable
22:40 dmsimard :)
22:52 _ducttape_ joined #puppet-openstack
23:06 openstackgerrit Merged openstack/puppet-ceph: Disable the RGW acceptance test  https://review.openstack.org/242217
23:16 gildub joined #puppet-openstack
23:17 karimb joined #puppet-openstack
23:22 aimon joined #puppet-openstack
23:33 jerrygb joined #puppet-openstack
23:49 jerrygb_ joined #puppet-openstack
23:53 openstackgerrit Emilien Macchi proposed openstack/puppet-heat: DNM - CI test  https://review.openstack.org/242271

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