Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-12-03

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

All times shown according to UTC.

Time Nick Message
00:13 xavpaice joined #puppet-openstack
00:19 gildub joined #puppet-openstack
00:21 vinsh joined #puppet-openstack
01:08 topshare joined #puppet-openstack
01:13 saneax_-_ joined #puppet-openstack
01:27 xarses joined #puppet-openstack
01:30 rcallawa joined #puppet-openstack
01:31 rcallawa joined #puppet-openstack
01:36 rmoe joined #puppet-openstack
02:04 xingchao joined #puppet-openstack
02:59 cwolferh joined #puppet-openstack
03:59 openstackgerrit Mike Dorman proposed stackforge/puppet-neutron: Add service_name parameter to neuron::server class  https://review.openstack.org/136423
04:01 sanjayu joined #puppet-openstack
04:49 xavpaice neuron server? I want one of those.
05:24 markvoelker joined #puppet-openstack
05:35 fedexo joined #puppet-openstack
05:44 fedexo joined #puppet-openstack
05:46 ToMiles_ joined #puppet-openstack
05:52 markvoelker joined #puppet-openstack
05:57 sanjayu joined #puppet-openstack
06:14 britthouser joined #puppet-openstack
06:34 markvoelker joined #puppet-openstack
06:41 aimon joined #puppet-openstack
06:43 ikkeT joined #puppet-openstack
06:45 ToMiles joined #puppet-openstack
06:45 ikke-t joined #puppet-openstack
06:47 simonmcc_ joined #puppet-openstack
06:47 simonmcc_ joined #puppet-openstack
06:50 tfz joined #puppet-openstack
07:25 aimon_ joined #puppet-openstack
07:29 tfz joined #puppet-openstack
07:31 omolchanov joined #puppet-openstack
07:31 omartsyniuk joined #puppet-openstack
07:50 stamak joined #puppet-openstack
07:50 openstackgerrit joined #puppet-openstack
08:14 omolchanov joined #puppet-openstack
08:33 dgurtner joined #puppet-openstack
08:33 dgurtner joined #puppet-openstack
08:37 rcallawa joined #puppet-openstack
08:39 openstackgerrit Bogdan Dobrelya proposed stackforge/puppet-openstack_extras: Add pacemaker provider for HA services  https://review.openstack.org/136708
08:42 topshare joined #puppet-openstack
08:48 rcallawa_ joined #puppet-openstack
08:53 rcallawa joined #puppet-openstack
09:06 sanjayu joined #puppet-openstack
09:10 markvoelker joined #puppet-openstack
09:23 berendt joined #puppet-openstack
09:33 rcallawa joined #puppet-openstack
09:41 qba73 joined #puppet-openstack
09:59 igajsin joined #puppet-openstack
10:13 openstackgerrit Michal Skalski proposed stackforge/puppet-swift: Enlarging the scope of ssh module  https://review.openstack.org/113550
10:19 TorLaedre joined #puppet-openstack
10:39 sanjayu joined #puppet-openstack
10:50 derekh joined #puppet-openstack
10:50 * social looks around for anyone to review backport https://review.openstack.org/#/c/138681/
11:05 igajsin joined #puppet-openstack
11:06 social joined #puppet-openstack
11:16 omolchanov joined #puppet-openstack
11:18 stamak joined #puppet-openstack
11:39 michchap_ social: done.
11:42 social michchap_: thanks
11:47 omartsyniuk joined #puppet-openstack
12:00 topshare joined #puppet-openstack
12:51 sanjayu joined #puppet-openstack
12:57 openstackgerrit Bogdan Dobrelya proposed stackforge/puppet-openstack_extras: Add pacemaker provider for HA services  https://review.openstack.org/136708
12:58 omartsyniuk joined #puppet-openstack
13:15 morazi joined #puppet-openstack
13:16 * social annoys with https://review.openstack.org/#/c/106144/9
13:16 social EmilienM: ^^
13:23 openstackgerrit Bogdan Dobrelya proposed stackforge/puppet-openstack_extras: Add pacemaker provider for HA services  https://review.openstack.org/136708
13:37 hogepodge joined #puppet-openstack
13:41 bogdando hello. Any help with tests failure cause http://logs.openstack.org/08/136708/10/check/gate-puppet-openstack_extras-puppet-unit-3.3/d15bffa ?
13:41 bogdando I cannot reproduce the failure at my PC
13:42 bogdando I have all tests passed
13:42 bogdando https://review.openstack.org/136708 ^^
13:45 topshare joined #puppet-openstack
13:45 xingchao_ joined #puppet-openstack
13:58 richm joined #puppet-openstack
14:02 dalgaaf joined #puppet-openstack
14:19 michchap_ beddari: I changed my mind again :) https://github.com/michaeltchapman/puppet_openstack_builder/blob/r7/hiera/hiera.yaml#L11
14:24 rcallawa joined #puppet-openstack
14:34 xingchao joined #puppet-openstack
14:35 beddari michchap_: better :P
15:03 patcable hmm. thoughts on adding HA-ability to the openstack composition module?
15:04 patcable I think there would need to be a hiera option for using_ha and then ability to specify the correct interfaces for a particular host, and the cluster addresses
15:06 EmilienM social: will look
15:09 rickerc joined #puppet-openstack
15:17 mschmitt joined #puppet-openstack
15:19 ericpeterson joined #puppet-openstack
15:37 openstackgerrit Dmitry Ilyin proposed stackforge/puppet-openstack_extras: Add pacemaker provider for HA services  https://review.openstack.org/136708
15:38 berendt joined #puppet-openstack
15:41 omartsyniuk_ joined #puppet-openstack
15:42 EmilienM social: let me read https://bugs.launchpad.net/openstack-manuals/+bug/1375746 and go back to the review. It's on my todo for today.
15:47 social EmilienM: np
15:50 rcallawa joined #puppet-openstack
15:56 mdorman joined #puppet-openstack
16:00 rickerc joined #puppet-openstack
16:05 EmilienM social: have you seen https://bugs.launchpad.net/openstack-manuals/+bug/1375746/comments/6 ?
16:06 EmilienM social: OVS plugin is dead
16:06 EmilienM everything is ML2 right now.
16:07 rickerc joined #puppet-openstack
16:12 social EmilienM: so I should poke packagers to fix it
16:13 EmilienM social: yes.
16:16 markvoelker joined #puppet-openstack
16:41 alokj joined #puppet-openstack
16:51 prad joined #puppet-openstack
17:00 rcallawa joined #puppet-openstack
17:08 imcsk8 joined #puppet-openstack
17:16 ToMiles joined #puppet-openstack
17:17 rmoe joined #puppet-openstack
17:19 cwolferh joined #puppet-openstack
17:32 xarses joined #puppet-openstack
17:47 ddmitriev joined #puppet-openstack
17:49 hogepodge joined #puppet-openstack
18:01 rcallawa joined #puppet-openstack
18:13 RichardRaseley joined #puppet-openstack
18:27 ericpeterson joined #puppet-openstack
18:36 rcallawa joined #puppet-openstack
18:45 openstackgerrit Rushi Agrawal proposed stackforge/puppet-horizon: Update comment documentation  https://review.openstack.org/138818
19:01 rcallawa joined #puppet-openstack
19:03 stamak joined #puppet-openstack
19:04 aimon joined #puppet-openstack
19:14 openstackgerrit Ryan Bak proposed stackforge/puppet-monasca: Fixed checks_d directory permissions  https://review.openstack.org/138830
19:20 openstackgerrit Merged stackforge/puppet-horizon: Update comment documentation  https://review.openstack.org/138818
19:37 EmilienM nibalizer: hey, could you re-trigger https://review.openstack.org/#/c/102020/ ? we have the job now
19:42 nibalizer EmilienM: no prob
19:42 nibalizer but you ccan recheck same as me
19:42 nibalizer it doesn't care who said what or who owns the patch
19:42 EmilienM nibalizer: oh, by running "recheck" command?
19:42 EmilienM ah ok, thx !
19:43 nibalizer but i think i have work to do there to adress concerns from crinkle and sbadia
19:43 nibalizer so its not actually ready yyet
19:44 mmaglana joined #puppet-openstack
19:55 openstackgerrit Emilien Macchi proposed stackforge/puppet-tempest: Allow to not manage Tempest code in Puppet  https://review.openstack.org/138838
19:57 EmilienM sbadia: if you can have a look ^
19:57 EmilienM this module is a bit crappy
20:13 openstackgerrit Merged stackforge/puppet-monasca: Fixed checks_d directory permissions  https://review.openstack.org/138830
21:01 rcallawa joined #puppet-openstack
22:05 tfz joined #puppet-openstack
22:06 mschmitt joined #puppet-openstack
22:08 EmilienM mgagne: crinkle: I would be happy if some core could review these old patches: https://review.openstack.org/#/q/status:open+topic:bp/common-openstack-identity-resource,n,z
22:08 mgagne EmilienM: doing it now
22:08 EmilienM thanks
22:09 mgagne EmilienM: awesome one: "Ironic Networking Service"
22:09 EmilienM mgagne: yes.
22:09 mgagne EmilienM: now stuck with it
22:09 EmilienM mgagne: it's my fault, and I'll fix it in another patch
22:09 EmilienM mgagne: I'm the author of this module
22:09 EmilienM mgagne: to be honest, I'm not sure something is using this module
22:09 mgagne EmilienM: oh, I'm not looking for someone to blame ;)
22:09 EmilienM someone*
22:10 mgagne EmilienM: if you don't mind, I would like to restore my old patches to deprecate the plethora of parameters we have to se to configure endpoint (address, port, protocol, etc.) so someone only has to provide url instead
22:11 EmilienM mgagne: please do. Does that affect my patches? let me know
22:11 mgagne EmilienM: no and it shouldn't be done in the same patch anyway
22:13 openstackgerrit Merged stackforge/puppet-ironic: Refactorise Keystone resources management  https://review.openstack.org/136388
22:14 openstackgerrit Merged stackforge/puppet-heat: Refactorise Keystone resources management  https://review.openstack.org/136384
22:16 openstackgerrit Merged stackforge/puppet-sahara: Refactorise Keystone resources management  https://review.openstack.org/136394
22:18 openstackgerrit Merged stackforge/puppet-manila: Re-factorise Keystone resources management  https://review.openstack.org/137494
22:18 openstackgerrit Merged stackforge/puppet-trove: Refactorise Keystone resources management  https://review.openstack.org/136403
22:21 openstackgerrit Merged stackforge/puppet-cinder: Refactorise Keystone resources management  https://review.openstack.org/136372
22:22 openstackgerrit Merged stackforge/puppet-neutron: Refactorise Keystone resources management  https://review.openstack.org/135842
22:22 openstackgerrit Merged stackforge/puppet-ceilometer: Refactorise Keystone resources management  https://review.openstack.org/136381
22:23 mgagne EmilienM: what's special about puppet-swift that require backward compat code?
22:23 EmilienM mgagne: I just respect the code which was present
22:23 openstackgerrit Merged stackforge/puppet-glance: Refactorise Keystone resources management  https://review.openstack.org/135841
22:23 EmilienM I just respect backward compatibility here
22:24 mgagne ok, we should revisit that part later
22:25 openstackgerrit Merged stackforge/puppet-nova: Refactorise Keystone resources management  https://review.openstack.org/134493
22:27 openstackgerrit Merged stackforge/puppet-swift: Refactorise Keystone resources management  https://review.openstack.org/136402
22:28 mgagne EmilienM: do you manage mongodb users with puppetlabs-mongodb?
22:28 EmilienM mgagne: no
22:28 EmilienM mgagne: we don't do auth, it's in our backlog somewhere
22:28 EmilienM mgagne: i've seen your issues
22:29 mgagne EmilienM: hehe
22:29 EmilienM mgagne: using mongodb for ceilometer?
22:29 mgagne EmilienM: was about to say: since you can't answer my question, I'll warn you about the pitfalls to come =)
22:29 mgagne EmilienM: yes
22:30 mgagne EmilienM: but we might replace ceilometer soon as we found it to not be appropriate for what we are doing with it
22:31 mgagne EmilienM: this guy summarizes my opinion on Ceilometer: https://twitter.com/aptira/status/532115110366814208
22:34 EmilienM mgagne: reading now
22:35 EmilienM mgagne: ok cool
22:35 EmilienM mgagne: I don't have much knowledge to take position here.
22:35 EmilienM mgagne: hey I wanted to show you something
22:36 mgagne EmilienM: what is it?
22:36 EmilienM mgagne: something I'm doing in our composition layer: https://review.openstack.org/#/c/138846/
22:36 EmilienM it's really WIP and PoC
22:36 EmilienM but the idea is here
22:37 mgagne right
22:37 mgagne EmilienM: I'm still thinking about the service validation thing
22:38 EmilienM mgagne: is it bad?
22:38 mgagne EmilienM: with $::cloud::validate_services, you can't tell to not check for specific services, can't you?
22:39 EmilienM mgagne: right.
22:39 EmilienM for now
22:39 EmilienM but again, if we realize it's a blocker, let's bring it flexible, I just did not want to add too much parameters
22:39 mgagne EmilienM: I didn't have time to reply to the data bindings thread but I feel that it might be one of the key to come up with a super clean design
22:40 EmilienM mgagne: I see your point. That mean you would control the service validation with data binding in cloud::compute::api and not in cloud::init, right?
22:41 mgagne yes
22:41 mgagne EmilienM: so someone could enable the built-in service validation
22:41 mgagne EmilienM: use data bindings to override the parameters of the class
22:42 mgagne EmilienM: parameters which could be whatever the module feels is required to configure the service validation
22:42 EmilienM mgagne: I have the feeling that a composition layer is now useless if you do 100% data binding. Am I wrong?
22:42 EmilienM I mean:
22:42 EmilienM having manifests like this
22:42 mgagne EmilienM: and someone could disable or override the built-in one with his own *resource*
22:42 mgagne EmilienM: you still need a composition layer
22:42 EmilienM instead, we should have a role based composition layer using "include", riht?
22:42 EmilienM include nova::api
22:42 mgagne EmilienM: neutron-agent won't install by itself by including nova::compute, won't it?
22:43 EmilienM no it isn't
22:43 mgagne EmilienM: there is also the question of "reusable" config values
22:43 EmilienM mgagne: do you have a composition layer like puppet-openstack-cloud? what do you think about it honestly?
22:43 mgagne EmilienM: like the database host for example, a lot of services requires it. it doesn't feel right to have to copy the value for all modules
22:44 mgagne EmilienM: bodepd proposed a solution with his custom data binding module
22:44 mgagne EmilienM: I have one and I can't see a world without it
22:44 EmilienM hehe
22:44 EmilienM brb
22:44 mgagne EmilienM: https://github.com/bodepd/scenario_node_terminus#data-mappings
22:45 mgagne EmilienM: his solution
22:46 vinsh joined #puppet-openstack
22:46 mgagne EmilienM: I however like the concept of attributes used by Chef where you have a place where anyone can read from
22:47 mgagne EmilienM: puppetlabs introduced a similar approach when refactoring mongodb and postgresql module: https://github.com/puppetlabs/puppetlabs-mongodb/blob/master/manifests/globals.pp https://github.com/puppetlabs/puppetlabs-postgresql/blob/master/manifests/globals.pp
22:55 * EmilienM back
22:55 EmilienM mgagne: I already read about bodepd thing, pretty interesting
23:01 gildub joined #puppet-openstack
23:07 mgagne EmilienM: I'm not a fan of his approach due to the fact that you need to load a new node_terminus
23:08 mgagne EmilienM: but I like the idea to have a way to auto propagate what is +/- a global parameter
23:08 mgagne EmilienM: while still having a way to override it in each service
23:16 michchap_ EmilienM: scenario node terminus has limitations. You cannot apply any logic in hiera, this is one reason why I stopped using it.
23:17 michchap_ EmilienM: it also is very different to how data is laid out in normal hiera instances, so there is a learning step for most people. This makes it hard to get adoption.
23:23 vinsh joined #puppet-openstack
23:27 mgagne michchap_: ohai! have you tested rabbitmq_policy?
23:37 michchap_ mgagne: Good evening! I don't know that I have, why?
23:37 mgagne michchap_: well, kind of forgot that I should contribute it upstream, was looking for feedback
23:38 michchap_ mgagne: there's a patch here that's been going on forever...https://github.com/puppetlabs/puppetlabs-rabbitmq/pull/192
23:39 mgagne michchap_: that's why I love Gerrit
23:39 mgagne michchap_: you don't have to go through 32 commits to get the proposed change
23:39 michchap_ mgagne: yeah github is not good for review :(
23:40 michchap_ mgagne: also I don't know there's a way I can rebase it for him since it's his PR from his repo. So it's actually crap for collaboration.
23:40 mgagne michchap_: "but but you preserve history and you can see the changes over time that the patch had!" What if I tell you this belong in a code review tool? :O
23:40 michchap_ mgagne: preaching to the choir :D
23:41 * mgagne circlejerks
23:41 michchap_ patcable: there are many composition modules. Some of them are HA already.
23:41 mgagne this guy tries to do too much in the same pull request
23:42 michchap_ yes. makes it very hard to merge.
23:42 patcable michchap_: I was thinking of the puppetlabs one
23:43 crinkle You can check out a pr, rebase it, and submit a new one
23:43 crinkle Hunner made a comment on that one that was never addressed, is why we haven't merged it yet
23:46 mgagne crinkle: I'm just gonna fix it (policy support) by submitting a new PR with code I have been using in production
23:48 mgagne crinkle: now situation is worst, there is now 5 PR for rabbitmq_policy support
23:48 crinkle :(
23:48 mgagne 7 in fact
23:49 crinkle gah poor rabbitmq module
23:50 mgagne crinkle: I think I was the first to propose a ghetto solution to manage policy which in the end managed to break stuff. Someone retook what I posed in a issue and made a pull request with it which got reverted. =)
23:51 michchap_ mgagne: that was me. there was policy support via an exec for a week :D
23:51 mgagne michchap_: haha, I like you
23:51 ericpeterson joined #puppet-openstack
23:51 crinkle I think they reverted it because of #192
23:52 crinkle but then didn't merge #192
23:52 michchap_ mgagne: it worked fine for me. I was happy!
23:52 mgagne awesome
23:52 * crinkle looks
23:52 mgagne michchap_: I was too until I upgraded to 3.2.0+ :O
23:53 michchap_ mgagne: yeah I figured that would mess everything up. I was on centos6 at the time.
23:54 mgagne michchap_: btw, I don't know who wrote this but I like it: https://aptira.com/blog/112-what-is-ceilometer-s-minimum-viable-product
23:55 mgagne michchap_: hit the nail right on the head
23:55 michchap_ mgagne: he is @manpageman on twitter

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