Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2013-10-08

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

All times shown according to UTC.

Time Nick Message
00:05 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-openstack: Preparing for 2.2.0 release.  https://review.openstack.org/50181
00:06 dmsimard joined #puppet-openstack
00:16 dafter joined #puppet-openstack
00:16 ari joined #puppet-openstack
01:09 ari joined #puppet-openstack
01:13 ryanycoleman joined #puppet-openstack
01:16 ryanycol_ joined #puppet-openstack
01:35 fandikurnia01 joined #puppet-openstack
01:46 _ilbot joined #puppet-openstack
01:46 Topic for #puppet-openstack is now Place to collaborate on Puppet/OpenStack tools: logs at http://irclog.perlgeek.de/puppet-openstack/today
01:47 fandikurnia01 joined #puppet-openstack
01:51 xingchao joined #puppet-openstack
01:58 ryanycoleman joined #puppet-openstack
02:15 ari joined #puppet-openstack
02:32 openstackgerrit A change was merged to stackforge/puppet-horizon: Preparing for 2.2.0 release.  https://review.openstack.org/50185
02:34 openstackgerrit A change was merged to stackforge/puppet-keystone: Preparing for 2.2.0 release.  https://review.openstack.org/50174
02:34 openstackgerrit A change was merged to stackforge/puppet-swift: Preparing for 2.2.0 release.  https://review.openstack.org/50183
02:35 openstackgerrit A change was merged to stackforge/puppet-nova: Preparing for 2.2.0 release.  https://review.openstack.org/50180
02:35 openstackgerrit A change was merged to stackforge/puppet-glance: Preparing for 2.2.0 release.  https://review.openstack.org/50155
02:35 openstackgerrit A change was merged to stackforge/puppet-cinder: Prepping module for 2.2.0 release.  https://review.openstack.org/50150
02:36 openstackgerrit A change was merged to stackforge/puppet-neutron: Preparing neutron (quantum) for 2.2.0 release.  https://review.openstack.org/50178
02:37 openstackgerrit A change was merged to stackforge/puppet-ceilometer: Allow to set full url for endpoints  https://review.openstack.org/47406
02:41 openstackgerrit A change was merged to stackforge/puppet-openstack: Sync storage rings from rsync server  https://review.openstack.org/47085
02:43 openstackgerrit A change was merged to stackforge/puppet-openstack: Preparing for 2.2.0 release.  https://review.openstack.org/50181
02:51 fandikurnia01 joined #puppet-openstack
03:03 ryanycoleman joined #puppet-openstack
03:14 qba73 joined #puppet-openstack
03:30 ari joined #puppet-openstack
04:18 ryanycoleman joined #puppet-openstack
04:31 dafter joined #puppet-openstack
04:31 dafter joined #puppet-openstack
05:07 fandikurnia01 joined #puppet-openstack
05:25 qba73_ joined #puppet-openstack
05:35 fandikurnia01 joined #puppet-openstack
05:40 fandikurnia01 joined #puppet-openstack
05:46 blkperl joined #puppet-openstack
05:48 fandikurnia01 joined #puppet-openstack
06:01 fandikurnia01 joined #puppet-openstack
07:29 xingchao joined #puppet-openstack
07:38 xingchao_ joined #puppet-openstack
08:03 derekh joined #puppet-openstack
08:13 nibalizer joined #puppet-openstack
08:24 Hunner joined #puppet-openstack
08:24 Hunner joined #puppet-openstack
08:25 blkperl joined #puppet-openstack
08:38 mmagr joined #puppet-openstack
08:53 blkperl joined #puppet-openstack
10:09 nightfly joined #puppet-openstack
10:47 nibalizer joined #puppet-openstack
11:24 dafter joined #puppet-openstack
11:24 dafter joined #puppet-openstack
11:45 blkperl joined #puppet-openstack
12:14 dprince joined #puppet-openstack
13:01 mjblack joined #puppet-openstack
13:02 mjblack joined #puppet-openstack
13:12 fandikurnia01 joined #puppet-openstack
13:26 openstackgerrit Matthew J Black proposed a change to stackforge/puppet-horizon: Updated user and group in debian params.  https://review.openstack.org/50173
13:34 dguitarbite joined #puppet-openstack
13:43 dafter joined #puppet-openstack
13:44 mjblack mgagne: I didnt know there was a reported bug for that change thanks for the info
13:47 mgagne mjblack: np
14:35 rcrit trying to add rspec tests to horizon::params to test values based on osfamily
14:35 rcrit I've tried contains_class('horizon::params').with('somevalue') and it always fails, comparing to nil
14:35 rcrit how does one do this?
14:36 mgagne rcrit: I don't think you test it directly as the value isn't part of the compiled catalog, isn't it?
14:36 rcrit beats me, totally new to this, got dinged in a review to add tests to the default params
14:36 mgagne rcrit: you have to think it indirectly with the horizon::init class
14:37 mgagne rcrit: think -> test
14:37 mgagne rcrit: ie. test that /var/log/horizon is owned by the correct user when on debian/ubuntu
14:38 rcrit I'm adding os-specific params and want to be sure the value is right
14:38 rcrit nothing else to test, other than string compare based on osfamily
14:38 mgagne rcrit: I know what you are talking about
14:39 rcrit mgagne, should contain_class('horizon::init').with() not working either :-(
14:39 mgagne rcrit: https://github.com/stackforge/puppet-neutron/blob/master/spec/classes/neutron_agents_linuxbridge_spec.rb
14:39 mgagne rcrit: afaik, you have to test the resources using those values
14:40 rcrit yeah, it doesn't create anything though. It is only used in a file_line, which we also can't test
14:41 mgagne rcrit: file_line is a resource
14:41 mgagne contain_file_line('horizon root').with_line('WSGIScriptAlias /horizon /os/specific/path')
14:42 rcrit mgagne, ok, I'll give it a shot, thanks
14:47 rcrit mgagne, that did the tricks, thanks a lot
14:47 mgagne np
14:55 openstackgerrit Rob Crittenden proposed a change to stackforge/puppet-horizon: When listen_ssl is true, enable the SSL engine and set a cert/key  https://review.openstack.org/49799
15:09 dafter joined #puppet-openstack
15:09 dafter joined #puppet-openstack
15:23 badiane_ka joined #puppet-openstack
15:28 ari_ joined #puppet-openstack
15:56 mjblack mgagne: did you have that issue with horizon where if debug needs to be set to true for it to work?
15:59 mgagne mjblack: yes
16:00 mjblack mgagne: do you know if there was a bug submitted for horizon on it? I ran into that and I want to grab it for tracking purposes
16:08 mmagr joined #puppet-openstack
16:11 ryanycoleman joined #puppet-openstack
16:12 ari joined #puppet-openstack
16:29 ryanycoleman joined #puppet-openstack
16:42 ari joined #puppet-openstack
16:44 ryanycoleman joined #puppet-openstack
16:44 xingchao joined #puppet-openstack
16:56 otherwiseguy joined #puppet-openstack
17:14 fvollero mgagne: a quick question, in puppet-ceilometer we gonna have 2 new services, now, the trick is to add them as alarm_evaluator.pp and alarm_notifier.pp or alarm/evaluator.pp and alarm/notifier.pp ?
17:14 mgagne fvollero: alarm/* would be fine with me
17:15 fvollero mgagne: ack
17:16 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-cinder: Prepping module for 2.2.0 release.  https://review.openstack.org/50412
17:17 hogepodge joined #puppet-openstack
17:18 hogepodge Did I get that cherry-pick review submitted correctly on puppet-cinder?
17:18 mgagne hogepodge: why was it proposed to master first?
17:19 hogepodge I wanted to get the change log in the master branch.
17:19 mgagne hogepodge: this is problematic to me
17:19 mgagne hogepodge: in its current form
17:19 hogepodge How so?
17:20 mgagne hogepodge: Modulefile is modified. What if master was released as 3.0.0 already?
17:20 hogepodge mgagne: I expected master to bump up to 3.0.0 next week, then cherry-pick back to stable/havana
17:21 mgagne hogepodge: it's pointless to bump master to 2.2.0 if it's never going to be released no?
17:22 hogepodge mgagne: the module file needs some value. I don't ever expect a release to be cut from master. I learned the hard way with 2.1 by backing everybody up
17:22 mgagne hogepodge: just wondering about the process itself
17:22 hogepodge magagne: develop on master, cherry-pick back to a stable branch, then release.
17:23 hogepodge mgagne: If that process doesn't work, I'm happy to adapt to something else.
17:23 mgagne hogepodge: I just don't know why master should know about the stable/grizzly release (mainly the version dump which IMO has nothing to do in master)
17:24 hogepodge mgagne: what should the version be in master, then? The modulefile needs some value. Also, should master know anything about the release notes?
17:25 mgagne hogepodge: changelog management is an other problem by itself which I don't know how to address yet.
17:25 hogepodge mgagne: for right now, the changes have been applied, so this is more of a discussion for future releases. But I understand your concerns.
17:26 mgagne hogepodge: I don't know if we should have bumped master to 3.0.0 as soon as the stable branch was cut
17:27 hogepodge mgagne: rc, then?
17:27 hogepodge mgagne: milestones?
17:27 mgagne hogepodge: I mean when we should bump it, 1 day before the release? 1 day after the release?
17:28 hogepodge mgagne: I'd like to see the modules release as soon as we can verify a stable havana deployment. openstack stable has bug fixes built into the release cycle
17:29 hogepodge mgagne: unless there are critical features that we're missing.
17:29 hogepodge mgagne: What I don't want is to wait three months past the release cycle like happened with grizzly.
17:30 mgagne hogepodge: sure, I do understand that and I wish to not repeat the same mistake
17:34 hogepodge mgagne: I'm looking at how nova manages versions.
17:36 mgagne hogepodge: they use tags for releases and have a version to bump in setup.cfg.
17:36 mgagne hogepodge: version is bumped as soon as an rc1 is cut
17:37 hogepodge mgagne: I see. year.number
17:37 mgagne hogepodge: the version in setup.cfg does not contain the patch version.
17:38 mgagne hogepodge: (yet) as the version in setup.cfg is new since havana
17:38 hogepodge mgagne: Is this something you'd like to take to the mailing list before we do a Havana release? I feel like the horse is out of the gate with 2.2.0, but I want to make sure that we manage this by consensus.
17:39 mgagne hogepodge: better doing it as originally planned.
17:39 mgagne hogepodge: it is something we can improve in the next version.
17:40 mgagne hogepodge: however I feel the release process has too much overhead
17:40 mgagne hogepodge: changelog, bump version, cherry-pick, tag, manual tarball, manual upload, etc.
17:40 hogepodge mgagne: One way that it will improve is with automatic release to forge based on tagging.
17:41 hogepodge mgagne: We have a bot that will catch tags on github and push a release automatically.
17:41 hogepodge magane: I have to plug it in, which takes some setup on this side. That should help.
17:41 mgagne hogepodge: this bot should listen to gerrit instead =)
17:42 mgagne hogepodge: as it is where things are happening
17:42 hogepodge mgagne: :-)
17:42 mgagne hogepodge: github just happens to be ONE of the git mirrors.
17:42 hogepodge mgagne: no technical reason it can't
17:44 hogepodge mgagne: one of the weird things about the process is that I have to write the release notes based on the stable branch, but commit them to master. That's strange to me. but not all patches got back-ported for compatibility reasons.
17:44 mgagne hogepodge: this is what bothered me
17:45 hogepodge magagne: It means that I need to review logs past the previous release to make sure I get all of the changes in.
17:47 hogepodge mgagne: I'll write up an overview of the current release process, immediate plans on improving it (automatic packaging, for example), and I'll ask for feedback on how to improve it. It's important, and I don't want anyone to feel like PL is acting in a vacuum without any feedback.
17:47 mgagne hogepodge: =)
17:48 hogepodge magane: to get back to my first question, did I do the review to stable/grizzly correctly? When I did the git review, it popped up '* [new branch]      HEAD -> refs/publish/stable/grizzly'
17:48 hogepodge the [new branch] worried me a tad, but it looks ok in the review page
17:49 otherwiseguy joined #puppet-openstack
17:49 mgagne hogepodge: https://review.openstack.org/#/c/50412/ targeted to stable/grizzly, it's ok
17:50 mgagne hogepodge: click on the change-id to see that it has already been proposed (and merged) to master
17:50 hogepodge Jenins barfed on it
17:50 hogepodge Jenkins.
17:51 hogepodge mgagne: postgres failure
17:51 mgagne hogepodge: no change has been proposed to puppet-cinder since puppetlabs-postgresl has been refactored.
17:52 hogepodge I see. I'll figure out which patches need to be back-ported.
17:52 mgagne hogepodge: someone has to pin the fixture to a known working/compatible version before you change can go in
17:56 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-cinder: Pin to posgresql 2.5 module  https://review.openstack.org/50428
18:00 dafter joined #puppet-openstack
18:07 ryanycoleman joined #puppet-openstack
18:10 ryanycoleman joined #puppet-openstack
18:12 qba73 joined #puppet-openstack
18:20 dafter joined #puppet-openstack
18:29 openstackgerrit A change was merged to stackforge/puppet-cinder: Pin to posgresql 2.5 module  https://review.openstack.org/50428
18:33 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-cinder: Prepping module for 2.2.0 release.  https://review.openstack.org/50412
18:39 dafter joined #puppet-openstack
18:40 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-glance: Preparing for 2.2.0 release.  https://review.openstack.org/50436
18:42 tvb|afk joined #puppet-openstack
18:42 tvb|afk joined #puppet-openstack
18:46 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-horizon: Preparing for 2.2.0 release.  https://review.openstack.org/50440
18:53 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-neutron: Preparing quantum for 2.2.0 release  https://review.openstack.org/50442
18:55 ari_ joined #puppet-openstack
18:56 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-nova: Preparing for 2.2.0 release.  https://review.openstack.org/50443
18:58 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-openstack: Preparing for 2.2.0 release.  https://review.openstack.org/50444
18:59 derekh joined #puppet-openstack
18:59 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-swift: Preparing for 2.2.0 release.  https://review.openstack.org/50445
19:11 ryanycoleman joined #puppet-openstack
19:17 ryanycoleman joined #puppet-openstack
19:29 digicloseup joined #puppet-openstack
19:31 dafter joined #puppet-openstack
19:40 mjblack_ joined #puppet-openstack
19:48 ari joined #puppet-openstack
19:52 fandikurnia01 joined #puppet-openstack
19:57 dmsimard joined #puppet-openstack
19:59 tvb|afk joined #puppet-openstack
19:59 tvb|afk joined #puppet-openstack
20:00 ryanycoleman joined #puppet-openstack
20:00 ryanycoleman joined #puppet-openstack
20:00 badiane_ka joined #puppet-openstack
20:03 otherwiseguy joined #puppet-openstack
20:08 dmsimard1 joined #puppet-openstack
20:49 bodepd mgagne: I think I agree with you that I should not have merged those Modulefiles into master
20:49 bodepd mgagne: but I don'
20:49 bodepd t see how it breaks anything
20:49 mgagne bodepd: it's not about breaking stuff, it's about having the right procedure
20:53 dafter joined #puppet-openstack
20:53 dafter joined #puppet-openstack
21:00 hogepodge joined #puppet-openstack
21:03 ianw joined #puppet-openstack
21:03 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-keystone: Preparing for 2.2.0 release.  https://review.openstack.org/50480
21:04 ianw joined #puppet-openstack
21:07 ari_ joined #puppet-openstack
21:22 ryanycoleman joined #puppet-openstack
21:30 ryanycoleman joined #puppet-openstack
21:30 fandikurnia01 joined #puppet-openstack
21:35 ryanycoleman joined #puppet-openstack
21:36 ryanycoleman joined #puppet-openstack
21:41 ryanycoleman joined #puppet-openstack
21:50 ryanycoleman joined #puppet-openstack
21:54 hogepodge bodepd: it's worth thinking about what the Modulefiles should look like on master. what information about the branch should it capture?
21:55 hogepodge bodepd: I went to master just to maintain development consistency, but I also expected some feedback. On the whole, I think it's moving along more smoothly now than last time.
21:55 ari joined #puppet-openstack
21:56 hogepodge boded, mgagne: I'm glad we're thinking about this prior to the havana release.
21:58 tvb|afk joined #puppet-openstack
22:04 ryanycoleman joined #puppet-openstack
22:04 hogepodge bodepd, mgagne: My initial thought are to have the Modulefile track the current features targeted for openstack devel, perhaps with milestone tags. Use current release tags in the branches (updated when a release is ready).
22:05 hogepodge bodepd, mgagne: That way master captures the current state of the development.
22:05 mgagne hogepodge: +1
22:08 otherwiseguy joined #puppet-openstack
22:22 nightfly joined #puppet-openstack
22:25 ryanycoleman joined #puppet-openstack
22:41 harsha joined #puppet-openstack
22:42 harsha Hey Guys.. i have a bit of situation using puppet-openstack module on a HA setup
22:43 harsha i have 4 controller nodes, behind a load balancer, so i need to bring up the fist node by applying openstack::controller class,
22:44 harsha then install keystone, nova, glance packages to other nodes and point conf to use a VIP node ( initial controller node)
22:45 harsha in reality controller class is being applied on all controller nodes hence each node is trying to initialize database, trying to modify keystone-endpoint etc which is messing my setup :(
22:45 mjblack_ what are you doing for mysql replication?]
22:46 harsha mysql and rabbitmq is installed on a separate node
22:46 harsha if i have one controller node and one db/mq node and number of compute nodes all works fine
22:47 harsha but i need to add multiple controller node and that will be behind a load balancer
22:48 harsha here's what i am doing.. bringing up the db/mq node first then one controller node
22:48 harsha and no of compute nodes.. all works get till this point
22:48 mjblack_ right, but it shouldnt try to create the databases again if they exist
22:48 harsha it's not creating databases, it is trying to initialize it
22:49 mjblack_ so if you bring up the first controller and everything looks fine, the next 3 should see the databases there and start working
22:50 harsha next nodes are trying to add endpoint etc.. so my output with keystone endpoint-list
22:50 harsha is messing up as soon as i am adding a new node
22:51 mjblack_ do you mean duplicate endpoints are being created?
22:51 harsha mjblack_ so you are saying i don't need to modify any class after adding a new node?
22:52 harsha keystone endpoint-list is having a new id for each ccn node and it is adding up a new entry
22:52 mjblack_ you shouldnt, I run with a 3 node cluster
22:52 mjblack_ I have seen that too but I deleted the duplicate endpoints
22:53 harsha here's what i notice in puppet run
22:53 harsha Notice: /Stage[main]/Keystone::Endpoint/Keystone_endpoint[region-one/keystone]/public_url: public_url changed 'http://cccc.aaa.xx.121:5000/v2.0' to 'http://ccc.aaa.xx.122:5000/v2.0'
22:54 harsha deleted manually??
22:54 mjblack_ so if I understand what is happening is that you are using $fqdn instead of the vip address
22:54 harsha that's correct
22:55 mjblack_ you should be using the vip address
22:57 mjblack_ the issue you can run into by using fqdn can have bigger issues than the endpoint addresses changing
22:57 badiane_ka joined #puppet-openstack
22:58 harsha ok let me try.. i need to rush for a meeting, will be back in 1 hr and talk again
22:58 harsha thanks a lot mjblack_
22:59 mjblack_ harsha: we run a 3 node cluster with haproxy in front of it, the only issue we ran into is duplicate endpoints being created
23:00 harsha and how do you remove it manually? or do you have a puppet way to remove it?
23:00 harsha will you be available after 1 hr?? please let me know??
23:00 harsha i need to dig deeper in this.. i'll really appreciate your help
23:01 mjblack_ I might be available, depends on what comes up, its 7pm here
23:04 ryanycoleman joined #puppet-openstack
23:17 ryanycoleman joined #puppet-openstack
23:32 odyi joined #puppet-openstack
23:41 bodepd The enabled param was intended to solve this
23:42 bodepd but it may go too far
23:44 bodepd I would look at having an active/passive category in your hiera hierarchy
23:44 bodepd and just let that config files tune the system to how it needs to be set for active/passicve
23:44 bodepd it may be a little more complicated than that :)
23:51 fandikurnia01 joined #puppet-openstack
23:53 dmsimard joined #puppet-openstack
23:56 ryanycoleman joined #puppet-openstack
23:58 dmsimard joined #puppet-openstack

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