Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2015-05-27

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

All times shown according to UTC.

Time Nick Message
00:04 openstackgerrit Colleen Murphy proposed stackforge/puppet-nova: Unpin beaker-rspec (do not merge)  https://review.openstack.org/185161
00:16 markvoelker joined #puppet-openstack
00:23 markvoelker joined #puppet-openstack
00:39 aimon joined #puppet-openstack
00:43 openstackgerrit Colleen Murphy proposed stackforge/puppet-nova: Unpin beaker-rspec  https://review.openstack.org/185161
00:52 aimon_ joined #puppet-openstack
01:21 ducttape_ joined #puppet-openstack
01:34 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Refined source of authentication information  https://review.openstack.org/180407
02:03 openstackgerrit Spencer Krum proposed stackforge/puppet-nova: Dummy commit  https://review.openstack.org/185817
02:47 ducttape_ joined #puppet-openstack
02:51 fedexo joined #puppet-openstack
03:16 gothicmindfood joined #puppet-openstack
03:21 panda joined #puppet-openstack
03:23 gildub joined #puppet-openstack
03:33 stevemar joined #puppet-openstack
04:39 sks joined #puppet-openstack
05:02 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Refined authentication sources  https://review.openstack.org/180407
05:07 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-keystone: Refined authentication sources  https://review.openstack.org/181299
05:08 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Refined authentication sources  https://review.openstack.org/180407
06:21 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstack-specs: Refine authentication process for Puppet resource providers  https://review.openstack.org/181600
07:17 markvoelker joined #puppet-openstack
07:22 panda joined #puppet-openstack
07:25 tfz joined #puppet-openstack
07:30 tfz joined #puppet-openstack
07:40 paramite joined #puppet-openstack
07:47 gfidente joined #puppet-openstack
07:47 gfidente joined #puppet-openstack
07:48 dalgaaf joined #puppet-openstack
07:52 jistr joined #puppet-openstack
07:58 jpena joined #puppet-openstack
08:05 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstack-specs: Restructure authentication for resource providers  https://review.openstack.org/181600
08:09 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Restructures authentication for resource providers  https://review.openstack.org/180407
08:10 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-keystone: Restructures authentication for resource providers  https://review.openstack.org/181299
08:34 openstackgerrit Giulio Fidente proposed stackforge/puppet-nova: Add parameter to control use of rbd for the ephemeral storage  https://review.openstack.org/119093
08:34 gfidente jistr, ^^ I've updated the commit message a little
08:36 jistr cool
08:36 jistr yeah that is a good patch to have, hopefully someone from -core will find the time to look
08:38 gfidente so I think one might want to improve it further and make the epehmeral_only option possible
08:39 gfidente so ideally a new param with persistent_storage
08:39 gfidente could offer control over that?
08:40 gfidente in which case one could pass both params to true or only one of the two
08:44 edrian joined #puppet-openstack
08:44 edrian morning!
09:13 alex_bh joined #puppet-openstack
09:22 tfz joined #puppet-openstack
09:26 jtomasek joined #puppet-openstack
09:37 derekh joined #puppet-openstack
10:15 fvollero joined #puppet-openstack
10:44 gildub joined #puppet-openstack
11:00 rushiagr_away joined #puppet-openstack
11:30 slagle_ joined #puppet-openstack
11:31 abramley_ joined #puppet-openstack
11:36 slagle_ joined #puppet-openstack
11:44 tfz joined #puppet-openstack
11:52 dprince joined #puppet-openstack
12:04 markvoelker joined #puppet-openstack
12:10 ducttape_ joined #puppet-openstack
12:12 ducttape_ joined #puppet-openstack
12:16 morazi joined #puppet-openstack
12:41 rcallawa joined #puppet-openstack
12:49 prad joined #puppet-openstack
12:57 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Restructures authentication for resource providers  https://review.openstack.org/180407
12:57 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-keystone: Restructures authentication for resource providers  https://review.openstack.org/181299
13:01 rcallawa_ joined #puppet-openstack
13:08 jpena joined #puppet-openstack
13:10 sks joined #puppet-openstack
13:12 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-keystone: Restructures authentication for resource providers  https://review.openstack.org/181299
13:16 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Restructures authentication for resource providers  https://review.openstack.org/180407
13:50 jtomasek joined #puppet-openstack
13:54 gsilvis joined #puppet-openstack
13:59 chem joined #puppet-openstack
14:06 ferest joined #puppet-openstack
14:06 saneax joined #puppet-openstack
14:14 ducttape_ joined #puppet-openstack
14:15 ericpeterson joined #puppet-openstack
14:24 ducttape_ joined #puppet-openstack
14:32 kkirkpatrick joined #puppet-openstack
15:05 rcallawa joined #puppet-openstack
15:13 mdorman joined #puppet-openstack
15:22 panda joined #puppet-openstack
15:25 rbrooker_ joined #puppet-openstack
15:25 dprince joined #puppet-openstack
16:04 dfisher joined #puppet-openstack
16:09 Fin1te joined #puppet-openstack
16:11 stevemar joined #puppet-openstack
16:21 derekh_ joined #puppet-openstack
16:29 cdelatte joined #puppet-openstack
16:46 dfisher fast question about https://bugs.launchpad.net/bugs/1458915 - are each 'also affects' the same bug or a different bug?  If they're the same do I tag the commit messages with Partial-Bug: #1458915 instead of Closes-Bug: ?
16:53 mdorman it’s the same bug id
16:53 dfisher ok.
16:53 mdorman and you do closes-bug.  garret knows to correlate the project hwere the commit containing closes-bug is to know which ‘also affects’ project to mark as resolved in the bug report
16:53 dfisher i assume gerrit/launchpad is smart enough to link things correctly?
16:53 mdorman yup, exactly
16:53 mdorman thanks for working on that!
16:53 dfisher cool.
16:54 dfisher hey, i gotta start somewhere :)
16:54 mdorman yup yup
16:54 dfisher was it your slide in specific that said "no posix users/groups/modes!" from last week?
16:56 aimon joined #puppet-openstack
16:57 jpena joined #puppet-openstack
16:58 mdorman i don’t think that was one of mine, maybe it was something that mfisch called out?   i don’t recall that
16:58 marvin999 joined #puppet-openstack
16:58 dfisher it was one of the two of you :)
16:59 dfisher anyway, i figured that was low enough hanging fruit for a newbie to handle
16:59 mdorman cool
16:59 dfisher might take me a bit to get through all of them.
16:59 dfisher as I have a million other things to actually get working on Solaris :/
17:00 mdorman yup, well certainly any help is appreciated :)
17:00 dfisher and, our legal / FOSS process is horrid.
17:01 rcallaw__ joined #puppet-openstack
17:02 dfisher ok, got some failures in puppet-ceilometer:  http://paste.openstack.org/show/240317/  - Solaris doesn't currently have ceilometer so are these false failures?
17:09 mdorman no i don’t think so, the rspec tests should be able to run sucessfully regardless of if the thing is actaully really intsalled.
17:11 dfisher hrm.
17:11 dfisher ok.
17:11 dfisher back to the hacking...
17:12 mdorman do those tests fail out of the box, or is it after your patch?
17:14 dfisher checking....
17:16 dfisher no, doesn't look like it.
17:16 dfisher bah!
17:17 dfisher oh ffs.
17:17 dfisher ignore me.
17:18 RichardRaseley joined #puppet-openstack
17:18 mdorman haha :)
17:19 dfisher is there a way to re-run just the failed tests?
17:20 mfisch dfisher mdorman no POSIX was Emilian
17:20 dfisher ah, thanks for the clarification :)
17:22 dfisher mdorman:   Finished in 2 minutes 7.4 seconds
17:22 dfisher 660 examples, 0 failures
17:22 mdorman dfisher:   no i think you have to run all of them
17:23 dfisher yeah, no worries.   2 minutes isn't too long :)
17:23 mdorman tbh, rspec remains kind of a mystery to me so i am defintely not a real authoritative source for info on it
17:23 dfisher at least you maybe understand ruby beyond 'hello world'
17:23 dfisher my ruby code is shameful.
17:24 openstackgerrit Drew Fisher proposed stackforge/puppet-ceilometer: remove POSIX users, groups and file modes Closes-Bug: #1458915  https://review.openstack.org/186092
17:25 openstackgerrit Colleen Murphy proposed stackforge/puppet-openstack-specs: Restructure authentication for resource providers  https://review.openstack.org/181600
17:42 openstackgerrit Yanis Guenane proposed stackforge/puppet-openstacklib: Create a common openstacklib::db::sync resource  https://review.openstack.org/185601
17:43 spredzy mdorman, mfisch crinkle can I get your opinion on ^ the global objective is https://review.openstack.org/#/q/status:open+branch:master+topic:db_sync,n,z
17:47 spredzy clayton, ^
17:50 mdorman spredzy:  makes sense to me
17:50 spredzy mdorman, cool :)
17:51 mfisch I'll look in a bit
18:05 rcallawa joined #puppet-openstack
18:08 aimon joined #puppet-openstack
18:09 openstackgerrit Colleen Murphy proposed stackforge/puppet-openstack-specs: Fix specs sphinx build  https://review.openstack.org/186107
18:09 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
18:39 jtomasek joined #puppet-openstack
18:45 tfz joined #puppet-openstack
18:47 Fin1te joined #puppet-openstack
19:05 clayton spredzy: do we have a policy on specifying path in exec resources?  I know the puppet-ceph guys at one point argued that it shouldn't be set
19:06 clayton spredzy I like the idea, but I'm not convinced it actually saves much (if any) code
19:06 spredzy clayton, good question I am not sure. The issue IMO of not specifying it is people trying it quickly and not working out of the box due to lack of path
19:07 spredzy clayton, its not much for saving code but uniformizing the way things are done across all modules
19:07 clayton the argument I got from the ceph guys was that people should set path as a global default on exec resources if the default isn't appropriate
19:07 clayton and I tend to agree that that is cleaner than having people override it all over the place
19:07 spredzy also it would allow to include the class only on the sync db node in the case of an ha infra
19:07 spredzy so the ENC can return it and glbaly set sync_db to false
19:08 clayton yeah, I'm in favor of the cinder::db::sync type of classes
19:08 clayton I'm just not sure the openstacklib part really adds much on top of that.
19:09 spredzy clayton, since it is something that is used in almost every module it just made sense to me to add it there
19:09 spredzy like postgresql https://github.com/stackforge/puppet-openstacklib/blob/master/manifests/db/postgresql.pp or mysql openstacklib defines
19:09 clayton I've only looked at the cinder implementation
19:09 _ryanv_ joined #puppet-openstack
19:10 clayton do you anticipate a need to have openstacklib::db::sync be more than a thin exec wrapper?
19:11 spredzy at the moment no, but if anychange needs to be apply this will be the single point of change
19:11 spredzy no need to change it across all modules as they grow
19:11 clayton why is everything -1 on jenkins right now, because it is lacking the openstacklib dep?
19:12 spredzy For example it currently impact 14 modules
19:13 clayton nod, I think I'm mostly on board with the changes, other than the path bit
19:13 spredzy clayton, it claims it can't find openstacklib::db::sync, not sure the Depends-On worked, will check that tomm
19:13 gfidente joined #puppet-openstack
19:13 gfidente joined #puppet-openstack
19:14 spredzy clayton, could you comment on the change about the exec path so we keep an history of that convo if you dont mind
19:17 aimon joined #puppet-openstack
19:20 jtomasek joined #puppet-openstack
19:35 openstackgerrit Colleen Murphy proposed stackforge/puppet-openstack-specs: Fix RST tests  https://review.openstack.org/186107
19:39 mgagne clayton: I see little value in openstacklib::db::sync tbh
19:39 mgagne clayton: it's only a wrapper around an exec resource
19:40 mgagne clayton: which offers little added value except to satisfy a "DRY" principle
19:50 clayton mgagne: nod, but I agree it's nice to standardize it, and I'm not willing to redo the work that spredzy already did, so I'd be ok with it if the path stuff is reworked
19:50 clayton it's probably a bit overengineered, but we might want to shove something else in there one day
19:51 clayton the other thing I was thinking is that it's possible that having a shared underlying resource might be useful in a collector one day
19:51 mgagne clayton: overengineered just like java with 294 abstraction layers with 1 line code method implementation =)
19:52 mgagne clayton: it's already complicated the way it is, I suggest we do not complicate it unless we have real actual use cases and values to add
19:52 mgagne clayton: otherwise I have people coming to my office asking: - can you explain me how puppet works? - lets see, take this module for example... ho wait, too complicated. lets take that one instead, oops, too many indirections, etc.
19:54 mgagne clayton: if we don't have any real plan yet for added value, I suggest we don't do it yet. We might just end up with a wrong design anymore based on theoretical needs we don't have today.
19:54 mgagne that's my take on it
19:54 dfisher haha.  "how puppet works?  Let's just go get a beer and talk about it, shall we?"
19:54 mgagne yep
19:55 clayton well, I'm -1 on the base review because of the path, but I'm flexible on it the other issues.  I take your point, but I see the value in standarizing it across the different modules in general
19:55 mgagne and now you switch to the wrong tab in your editor and it's ansible code. and the guy: wow, this looks simple, what is it? :D /troll
19:56 clayton I think ansible makes the simple things a lot easier and the hard things a lot harder :)
19:56 dfisher ^ tag line
19:56 dfisher Ansible:  A Win More utility
19:56 clayton We use it a good bit for deployments and such, but I've never really tried using it for config management.  Perhaps my opinion would change if I did
19:57 dfisher if only ansible wasn't GPL3.
19:57 clayton I like it for orchestration stuff
19:57 mgagne clayton: tbh, I don't know how different it would really be
19:58 clayton me either, I know we have some pretty horribly gross ansible code too though
19:58 mgagne clayton: you paste your config values in placeholder or a global attribute bag and something configures it like ini_file.
20:00 mgagne it's really not that complicated when you think about it
20:00 clayton well, I'm sure it's workable, seems like a number of people do it
20:00 mgagne and I feel like the way puppet makes you pass values all around the place doesn't help
20:01 clayton we don't actually do a lot of that, a lot of what we end up just putting in hiera.
20:01 clayton which ends up being gross in other ways sometimes
20:01 mgagne sure but there is all that "namespacing" stuff making it not that intuitive
20:01 clayton I suspect a lot of your ansible config would end up looking a lot like hiera too
20:01 mgagne I would probably organize it anyway
20:02 mgagne same I currently do with a super openstack module where I just have a super cinder class (for example) accepting a union of all configs you find across all our classes
20:02 mgagne and the module just selects and passes them as needed to puppet-cinder classes
20:03 mgagne it becomes "fun" when you have drivers and plugins but tbh, it isn't that bad
20:08 jtomasek joined #puppet-openstack
20:17 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
20:34 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
20:35 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
20:39 RichardRaseley What is the standard procedure around patches that have been sitting out for a while, waiting for a +2 (e.g. https://review.openstack.org/#/c/177975/) ?
20:40 clayton funny, that's the third fix I've seen for that issue
20:40 clayton I did one and so did sebastian
20:40 RichardRaseley Ah, has one been merged already?
20:40 clayton nope, the code is pretty much broken as written, and I don't know that we have consensus on how it should be fixed
20:40 clayton reading that one now
20:41 RichardRaseley Awesome, thank you clayton. I don't have a vested interest in it, but just had reviewed it and noticed it was still hanging out.
20:41 openstackgerrit Clayton O'Neill proposed stackforge/puppet-swift: Handle both string and array for memcache param.  https://review.openstack.org/177975
20:42 clayton I rebased the changeset, I'd like to see it pass against the 4.0 unit tests, since this is really a future parser fix
20:42 clayton it's failing right now on another issue that has already been fixed.
20:42 RichardRaseley Ah, excellent - thank you.
20:43 clayton sbadia had a problem with the code in general, in that there is nothing that says a memcached instance running locally has to be configured to connect on 127.0.0.1
20:44 clayton which is true, but I think a bigger issue.  I'd like to address that separately from the future parser issue
20:47 RichardRaseley So, if I am to understand, the issue you outlined above is more about the configuration assumptions in the proposed change (re: memcached listen address) vs. the actual proposed mechanism?
20:48 clayton yeah
20:49 clayton the class should probably take a boolean parameter that indicates whether or not memcached is local or not, instead of trying to figure it out
20:50 RichardRaseley That makes sense.
20:53 RichardRaseley Hunner: I'm idling on IRC now. Happy? =P
20:55 Hunner \o/
21:04 gildub joined #puppet-openstack
21:05 rcallawa_ joined #puppet-openstack
21:05 openstackgerrit Merged stackforge/puppet-neutron: Beaker: install APT repo with openstack_extras  https://review.openstack.org/183609
21:11 rcallawa joined #puppet-openstack
21:14 fvollero joined #puppet-openstack
21:41 fvollero_ joined #puppet-openstack
21:58 fvollero joined #puppet-openstack
22:04 fvollero joined #puppet-openstack
22:06 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
22:07 gfidente joined #puppet-openstack
22:07 gfidente joined #puppet-openstack
22:42 openstackgerrit Richard Raseley proposed stackforge/puppet-swift: Improve Endpoint Configurability  https://review.openstack.org/186205
22:57 openstackgerrit Merged stackforge/puppet-cinder: Beaker: install APT repo with openstack_extras  https://review.openstack.org/182046
22:57 openstackgerrit Merged stackforge/puppet-cinder: Bring Redhat support to acceptance tests  https://review.openstack.org/182668
23:00 openstackgerrit Richard Raseley proposed stackforge/puppet-nova: Improve Endpoint Configurability  https://review.openstack.org/186109
23:00 dfisher hunner:  ping?
23:03 chem joined #puppet-openstack
23:05 openstackgerrit Richard Raseley proposed stackforge/puppet-swift: Improve Endpoint Configurability  https://review.openstack.org/186205
23:28 openstackgerrit Richard Raseley proposed stackforge/puppet-swift: Improve Endpoint Configurability  https://review.openstack.org/186205
23:45 openstackgerrit Gilles Dubreuil proposed stackforge/puppet-openstacklib: Restructures authentication for resource providers  https://review.openstack.org/180407

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