Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2013-04-11

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

All times shown according to UTC.

Time Nick Message
00:02 openstackgerrit Dan Bode proposed a change to stackforge/puppet-openstack_dev_env: This code refactors the openstack test script  https://review.openstack.org/26689
00:04 openstackgerrit A change was merged to stackforge/puppet-openstack_dev_env: This code refactors the openstack test script  https://review.openstack.org/26689
00:04 JoeJulian Holy cow that's a lot of hoops to fix the spelling of "enalbles".
00:10 bodepd_ JoeJulian: yeah, it's not the lightest process
00:11 bodepd_ you can clone from stackforge and just run git review once everything is installed
00:18 openstackgerrit Joe Julian proposed a change to stackforge/puppet-keystone: Fixed typo  https://review.openstack.org/26692
00:24 openstackgerrit A change was merged to stackforge/puppet-keystone: Fixed typo  https://review.openstack.org/26692
00:28 mgagne that was fast =)
00:28 mgagne joejulian: and you forgot "Optioal" on the same line :P
00:29 bodepd_ yes, its those easy ones.
00:30 mgagne can't wait to use a spell checker against those puppet modules :P
00:31 JoeJulian hah! Oh well. It'
00:31 JoeJulian 's slightly more important that the actual option name is spelled right.
00:31 mgagne it sure is It'
00:32 mgagne What's the plan for puppetdoc? I don't see docs for classes on puppetforge, only defined types
00:37 openstackgerrit Dan Bode proposed a change to stackforge/puppet-openstack_dev_env: fix quantum issues  https://review.openstack.org/26693
00:39 mgagne bodepd_: Can someone mention me? I want to test my irc client =)
00:41 bodepd_ mgagne: mention
00:41 mgagne ty sir, it's working great
00:42 bodepd_ is anyone using dev_env besides me?
00:42 bodepd_ I'm debating if it even belongs on stackforge
00:42 mgagne I'm just reading about it for review purposes =)
00:42 bodepd_ and also if I should just be approving my own code
00:43 bodepd_ it's mainly what I use for CI, and development, its also intended to be an easy way to get started
00:43 bodepd_ I'll keep merging my own code (only there)
00:43 bodepd_ hopefully at some point someone will yell at me
00:44 mgagne ztop! =)
00:44 mgagne where the hell is openstack::test_file coming from?
00:44 mgagne ok, forget it, sleepy :P
00:45 bodepd_ that is next, I rewrote all fo the the quantum stuff for puppet-openstack
00:45 mgagne bodepd_: I can review but I don't feel comfortable approving stuff
00:45 bodepd_ no worries. I will have to figure out if that means it should not be here
00:45 mgagne as I'm not using anything from the project now
00:47 mgagne bodepd_: parameter names aren't clear (to me) about their purposes so I got confused
00:47 bodepd_ like what?
00:48 bodepd_ mgagne: ^^^
00:49 mgagne floating_ip for example, wasn't sure why it's a boolean, I had to check the puppetdoc
00:49 mgagne why isn't it an IP or subnet? :P
00:49 bodepd_ ah, just if its there, then it means assign one?
00:49 bodepd_ b/c its hard-coded by the vagrant environment
00:49 bodepd_ I think I see your point
00:49 bodepd_ I am sure there are lots of UI issues with the modules in general
00:50 mgagne =)
00:50 bodepd_ those are the hardest thigns to fix b/c it generally breaks backwards compat
00:50 mgagne yes, unfortunately
00:51 mgagne We have to be severe about new parameters, they have to be named correctly from the start so we don't have to rename them later. :P
00:51 aimon bodepd this tested successful: https://review.openstack.org/#/c/26664/
00:52 openstackgerrit A change was merged to stackforge/puppet-openstack_dev_env: fix quantum issues  https://review.openstack.org/26693
01:05 mgagne bodepd_: who else at puppetlabs is managing puppet modules on github?
01:09 openstackgerrit Dan Prince proposed a change to stackforge/puppet-swift: Fix swift_ring_builder output parsing.  https://review.openstack.org/26213
01:12 openstackgerrit Dan Prince proposed a change to stackforge/puppet-swift: Drop Swift requirement on keystone:python.  https://review.openstack.org/26215
01:17 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Only set special_service_provider on RHEL.  https://review.openstack.org/26227
01:20 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Only start messagebus on RedHat.  https://review.openstack.org/26228
02:23 openstackgerrit Dan Bode proposed a change to stackforge/puppet-openstack: quantum for folsom  https://review.openstack.org/26703
02:32 openstackgerrit Dan Bode proposed a change to stackforge/puppet-openstack: update cinder::base to cinder  https://review.openstack.org/26704
02:33 bodepd_ mgagne: its kind of a best effort
02:33 bodepd_ mgagne: but not enough folks :)
02:35 bodepd_ aimon: thanks for looking at that. and thanks for pointing out that critical issue :)
02:35 aimon sure np :)
02:42 openstackgerrit A change was merged to stackforge/puppet-nova: Allow to use of rabbit_hosts  https://review.openstack.org/26508
02:51 mgagne bodepd_: Thanks for testing on Folsom
02:52 mgagne bodepd_: about puppetlabs modules, thanks =)
02:58 openstackgerrit A change was merged to stackforge/puppet-nova: Set network defaults.  https://review.openstack.org/26251
04:54 openstackgerrit Aimon Bustardo proposed a change to stackforge/puppet-openstack: remove extraneous whitespace  https://review.openstack.org/26715
04:54 openstackgerrit Aimon Bustardo proposed a change to stackforge/puppet-openstack: rework openstack_compute_spec.rb  https://review.openstack.org/26716
04:55 aimon eww thats not what I meant to do
05:17 openstackgerrit A change was merged to stackforge/puppet-swift: Ensure python-swiftclient is installed to use swift-dispersion  https://review.openstack.org/26583
05:43 openstackgerrit Aimon Bustardo proposed a change to stackforge/puppet-openstack: add rabbit_virtual_host passthrough to cinder::base  https://review.openstack.org/26718
05:54 dachary joined #puppet-openstack
06:25 openstackgerrit A change was merged to stackforge/puppet-swift: Allow customise log_facility in storage server configuration  https://review.openstack.org/26581
06:32 openstackgerrit Yu Xingchao proposed a change to stackforge/puppet-keystone: Change keystone::python to keystone::client  https://review.openstack.org/26719
06:33 d3u joined #puppet-openstack
06:51 openstackgerrit Mehdi Abaakouk proposed a change to stackforge/puppet-swift: Fix swift storage service start on debian  https://review.openstack.org/26582
07:54 openstackgerrit Yu Xingchao proposed a change to stackforge/puppet-keystone: Change keystone::python and keystone::client  https://review.openstack.org/26719
08:17 openstackgerrit Yu Xingchao proposed a change to stackforge/puppet-swift: Add swift::storage::disk to support disk device.  https://review.openstack.org/26723
09:03 openstackgerrit Yu Xingchao proposed a change to stackforge/puppet-swift: Add swift::storage::disk to support disk device.  https://review.openstack.org/26723
09:04 dachary joined #puppet-openstack
09:22 openstackgerrit Yu Xingchao proposed a change to stackforge/puppet-swift: Add swift::storage::disk to support disk device.  https://review.openstack.org/26723
09:26 dachary joined #puppet-openstack
09:40 openstackgerrit Mehdi Abaakouk proposed a change to stackforge/puppet-horizon: Fix debian compatibility  https://review.openstack.org/26579
10:17 d3u joined #puppet-openstack
11:02 dprince joined #puppet-openstack
11:33 d3u joined #puppet-openstack
13:44 dprince joined #puppet-openstack
14:04 openstackgerrit Dan Prince proposed a change to stackforge/puppet-swift: Update keystone fragment to use egg.  https://review.openstack.org/26211
14:11 openstackgerrit François Charlier proposed a change to stackforge/puppet-swift: Remove traling / from keystone admin_url endpoint  https://review.openstack.org/26748
14:17 dachary joined #puppet-openstack
14:19 e1mer joined #puppet-openstack
14:43 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Only set special_service_provider on RHEL.  https://review.openstack.org/26227
14:52 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Add support for configuring qpid.  https://review.openstack.org/26230
15:04 openstackgerrit Dan Prince proposed a change to stackforge/puppet-glance: Add swift_store_auth_version to swift backend.  https://review.openstack.org/26158
15:09 openstackgerrit Dan Prince proposed a change to stackforge/puppet-glance: Add swift_store_auth_version to swift backend.  https://review.openstack.org/26158
15:30 openstackgerrit Dan Prince proposed a change to stackforge/puppet-glance: Add swift_store_auth_version to swift backend.  https://review.openstack.org/26158
15:38 openstackgerrit Dan Prince proposed a change to stackforge/puppet-swift: Fix swift_ring_builder output parsing.  https://review.openstack.org/26213
16:24 dachary joined #puppet-openstack
16:48 openstackgerrit A change was merged to stackforge/puppet-swift: Update keystone fragment to use egg.  https://review.openstack.org/26211
16:50 openstackgerrit A change was merged to stackforge/puppet-nova: Only set special_service_provider on RHEL.  https://review.openstack.org/26227
16:54 openstackgerrit A change was merged to stackforge/puppet-swift: RHEL/Fedora replicator updates.  https://review.openstack.org/26249
17:17 openstackgerrit A change was merged to stackforge/puppet-swift: Remove traling / from keystone admin_url endpoint  https://review.openstack.org/26748
17:22 bodepd_ can I get some love here: https://review.openstack.org/#/c/26664/
17:22 bodepd_ aimon: I am really not sure about your last note
17:22 bodepd_ I cannot imagine how that code woudl create the same issue
17:23 bodepd_ as I mentioned inline, my gut is that you have the wrong version where you are seeing that issue
17:24 mgagne bodepd_: Sorry, I don't have time now for tests. I sure would like to test it on my setup.
17:55 d3u joined #puppet-openstack
17:56 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Only start messagebus on RedHat.  https://review.openstack.org/26228
18:23 d3u joined #puppet-openstack
18:25 openstackgerrit A change was merged to stackforge/puppet-swift: Fix swift storage service start on debian  https://review.openstack.org/26582
18:26 openstackgerrit A change was merged to stackforge/puppet-nova: Only start messagebus on RedHat.  https://review.openstack.org/26228
18:27 dachary joined #puppet-openstack
18:32 bodepd_ aimon: ping
18:38 dprince bodepd_: So this ringbuilder branch... does it break grizzly?
18:38 dprince bodepd_: if not... then why not now?
18:38 mgagne anyone knows how to purge a ini config? The value undef and '' aren't working
18:39 dprince mgagne: purge => true?
18:39 mgagne on the value?
18:39 mgagne if it works, I'm stupid
18:40 dprince mgagne: I think you would use purge => true first and then just remove any key/values you don't want.
18:40 mgagne dprince: might not be the answer I expected =)
18:41 mgagne This: ceilometer_config { 'DEFAULT/rabbit_host': purge => true }
18:41 mgagne Results in: Invalid parameter purge
18:46 dprince mgagne: Perhaps it is broken too. See this? https://review.openstack.org/#/c/26664/
18:46 bodepd_ now,
18:47 bodepd_ resources { 'ceilometer_config': purge => true }
18:47 mgagne dprince: I don't want to purge the whole file and rebuild it only with values from the catalogue. I would like to be able to purge or "ensure => absent" a single value. I'll check the patch and see if something answers my need
18:47 bodepd_ also, the provider needs to implement: def self.file_path
18:47 mgagne ensure => absent did the trick
18:48 bodepd_ mgagne: you should be able to ensure  -=> abent
18:48 mgagne but it doesn't work with "empty" value
18:48 dprince mgagne: bodepd_ is your man. Listen to him :)
18:48 bodepd_ I don't understand 'empty' value?
18:48 mgagne "rabbit_host=" will result in the config being untouched.
18:49 bodepd_ mgagne: that is a bug. can you open an issue: https://github.com/cprice-pup​pet/puppetlabs-inifile/issues
18:50 mgagne I'll have to check if the problem is with ceilometer_config or the inifile module so I can provide a way to reproduce the problem.
18:51 bodepd_ it sounds like an issue with inifile
18:55 mgagne bodepd_: Could we get a new release of puppetlabs/rabbitmq? I'm still fighting with librarian-puppet, half of the time I'm getting a version from github and the other from puppetforge. I would need the cluster support.
18:55 bodepd_ mgagne: It would take more time than I have until after the summit.
18:56 mgagne bodepd_: alright, I can wait
18:56 bodepd_ mgagne: maybe try opening an issue, or a pull requests for the release (including changelog and a proposed nex version)?
18:56 bodepd_ I was crazy enough to sign up for a hands on lab next week
18:56 mgagne issues are disabled, only pull requests are accepted =)
18:57 bodepd_ rabbitmq?
18:57 bodepd_ then submit a PR, update CHANGELOG, and Modulefile with a propesed version
18:57 mgagne https://github.com/puppetlabs/puppetlabs-rabbitmq
18:57 mgagne I'll check that out
18:57 bodepd_ or just use a source from github for now
18:58 bodepd_ or cut a branch in your own fork and point Puppetfile at that
18:58 mgagne that's the point, librarian-puppet isn't idempotent
18:58 bodepd_ (that is the real beauty of puppetfile)
18:58 mgagne mod "rabbitmq",
18:58 mgagne :git => "https://github.com/puppetla​bs/puppetlabs-rabbitmq.git"
18:58 mgagne this doesn't work
18:58 bodepd_ why not?
18:59 mgagne Nova also requires puppetlabs/rabbitmq and librarian-puppet seems to resolve the dependency to puppetforge instead of github
18:59 bodepd_ maybe change the order?
18:59 mgagne I tried, believe me =)
18:59 bodepd_ (by the way, debugging librarian-puppet is the worst)
19:00 mgagne it's hell
19:00 mgagne I'll try to disable the forge completely and use exclusively github. not sure about how it will go
19:05 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Add support for configuring qpid.  https://review.openstack.org/26230
19:19 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Add support for configuring qpid.  https://review.openstack.org/26230
19:34 openstackgerrit A change was merged to stackforge/puppet-swift: Fix swift_ring_builder output parsing.  https://review.openstack.org/26213
19:34 mgagne bodepd_: puppetlabs-rabbitmq 2.1.0, is this the correct version? this version adds a new functionality (cluster support)
19:39 dprince mgagne: Might you relax your -2 here https://review.openstack.org/#/c/26230/
19:44 mgagne coming
19:46 bodepd_ 1.2.3
19:46 bodepd_ 1 = breaks backwards compat
19:46 bodepd_ 2 = new features, no backwards compat breakage
19:46 bodepd_ 3 = bug fix
19:47 mgagne I read about it, just looking for approval before the pull request =)
19:48 bodepd_ dprince: testing your qpid changes against grizzly Ubuntu now
19:49 mgagne commenting now
19:49 bodepd_ dprince: I would usually object for the problems you mentioned before, but given the urgency and the fact those changes have to be added, I am happy to concede as long as everyone else is om with it
19:49 bodepd_ s/you/I/
19:49 dprince bodepd_, mgange: much obliged
19:50 dprince bodepd_: problems I mentioned?
19:50 bodepd_ bodepd_: that I mentioned.
19:50 bodepd_ I have a lot of concerns around:
19:51 mgagne bodepd: I have nothing else to propose regarding the unification of queue parameters.
19:51 bodepd_ 1. proliferation of parameters to the lower level modules
19:51 mgagne even openstack isn't handling this case, they have different names
19:51 bodepd_ 2. adding parameters to a single interface that map to multipe backeends
19:51 bodepd_ (I wish Puppet inheritance better supported #2)
19:52 bodepd_ like, I said, I conceded
19:52 bodepd_ testing now
19:52 dprince bodepd_: My preference would be to avoid overloading params as much as possible.
19:52 dprince bodepd_: *and* if possible name things like the openstack- project does.
19:52 dprince bodepd_: *and* consistency across our puppet-* modules.
19:53 dprince bodepd_: In general I do also by limiting our use of params in favor of just using nova_config.
19:54 dprince bodepd_: But this is an exception to that since... well it is the message queue (an integral and required piece). To not allow qpid... is to say lets remove all of our rabbit stuff to.
19:54 dprince too
19:55 mgagne I would have a suggestion but it would break backward compatibility
19:56 mgagne Force nova::init to require Nova_config[rpc_backend]. Split rabbit and qpid in 2 new classes each with their own parameters, both would define Nova_config[rpc_backend].
19:57 mgagne You would be able to use the helper classes (nova::queue::rabbitmq / nova::queue::qpid) or provide the configuration yourself with Nova_config[rpc_backend]
19:58 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Add support for configuring qpid.  https://review.openstack.org/26230
19:59 dprince bodepd_: wierd. joe-topjian-v 'approved' my storeconfigs stuff but they aren't landing.
20:00 mgagne he probably doesn't have the right to do so.
20:00 mgagne he only has +1 and no veto.
20:00 dprince mgagne: gerrit let him mark it approved. That is the weird part.
20:01 mgagne will it automatically approve if there's 2 x +1 from non-core reviewer?
20:02 mgagne he is part of puppet-manager-core.
20:08 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Remove storeconfigs from nova::init.  https://review.openstack.org/26660
20:08 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Remove storeconfigs from nova::db::mysql.  https://review.openstack.org/26661
20:08 openstackgerrit Dan Prince proposed a change to stackforge/puppet-nova: Remove storeconfigs from nova::compute.  https://review.openstack.org/26662
20:10 mgagne bodepd_: here we go for rabbitmq: https://github.com/puppetlabs​/puppetlabs-rabbitmq/pull/44
20:11 bodepd_ that look reasonable to merge and release.
20:11 bodepd_ hold on.
20:12 dprince bodepd_: I'm taking off tomorrow BTW. Will try to review things more this evening though.
20:12 dprince bodepd_: Also, I need a puppetlabs shirt :)
20:12 dprince bodepd_: or puppet openstack
20:14 bodepd_ dprince: I'll also be out tomorrow. I have to go to San Jose to give a talk
20:15 bodepd_ we'll have puppet openstack stickers and shirts at the conference
20:15 dprince bodepd_: sweet
20:22 mgagne bodepd_: How is the Puppet CLA related to puppet modules? How does it stack up with the puppet module license? Is the CLA required when contributing to modules under the puppetlabs github account or is it only required for Puppet core?
20:25 bodepd_ idk
20:25 bodepd_ may I wasnt suposed to merge that if you havent signed the CLA
20:26 mgagne it was only a typo in the Modulefile :P
20:26 openstackgerrit A change was merged to stackforge/puppet-nova: Remove storeconfigs from nova::init.  https://review.openstack.org/26660
20:29 openstackgerrit A change was merged to stackforge/puppet-nova: Remove storeconfigs from nova::db::mysql.  https://review.openstack.org/26661
20:30 bodepd_ yeah, I'm just not sure what the policy is
20:31 openstackgerrit A change was merged to stackforge/puppet-nova: Remove storeconfigs from nova::compute.  https://review.openstack.org/26662
20:41 mgagne bodepd_: can't figure out why the build is failing https://review.openstack.org/#/c/26704/
21:03 bodepd_ ah, yeah.
21:04 bodepd_ I was having issues with that
21:04 bodepd_ it looked like a puppet bug
21:04 bodepd_ I understand why it is failing, but I thought the :: woudl fix it
21:07 mgagne I don't see any :: in the latest patch set?
21:07 bodepd_ ah, well there should be, but I tried that on my locla sytem adn it did not work as expected.
21:08 bodepd_ I mucked with it for 10 mins or so, asked in #puppet about it and moved ont
21:10 mgagne well, I do have a similar setup on my side and ::cinder works fine
21:10 mgagne this problem is only caused by openstack module having an openstack::cinder class.
21:10 bodepd_ can you run: rake spec?
21:11 bodepd_ yes. that is the problem I am referring to
21:11 bodepd_ :: should fix it, but I could not get the spec tests to pass
21:11 mgagne I'll check
21:21 mgagne bodepd_: you will have to prefix all cinder classes by :: or puppet gets confused: "cinder::volume::iscsi" would resolve to "openstack::cinder::volume::iscsi".
21:26 bodepd_ only if those classes exist
21:32 d3u joined #puppet-openstack
21:48 mgagne bodepd_: openstack::cinder exists. we have to prefix everything referring to cinder. puppet won't fallback to the "global" scope.
22:18 aimon bodepd_ Which Quantum module you guys looking at?
23:33 bodepd_ I am currently using the stackforge branch from my repo (bodepd)
23:34 bodepd_ it is based on cisco's folsom branch
23:34 bodepd_ ciscosystems
23:34 bodepd_ aimon: can I ask you about that nova patch?
23:34 bodepd_ aimon:  I feel that is a critical issue and really want to get that patch merged.

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