Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-05-21

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

All times shown according to UTC.

Time Nick Message
00:11 aldavud joined #puppet-openstack
00:17 openstackgerrit Gilles Dubreuil proposed a change to stackforge/puppet-vswitch: Refacfored a more suitable vs_redhat provider  https://review.openstack.org/94504
00:19 openstackgerrit Gilles Dubreuil proposed a change to stackforge/puppet-vswitch: Refacfored a more suitable ovs_redhat provider  https://review.openstack.org/94504
00:30 openstackgerrit Sebastien Badia proposed a change to stackforge/puppet-glance: Clean tests according michael comment on review¹  https://review.openstack.org/94505
00:32 britthouser joined #puppet-openstack
00:42 kitp joined #puppet-openstack
00:46 sbadia hum?
00:47 sbadia mgagne: the original change of #94505 is already merged in master
00:47 mgagne sbadia: I amended my comment
00:47 sbadia ok, sorry :)
00:49 mgagne sbadia: git commit summary should content the essence of the change. "I did what that other guy suggested" isn't that helpful when glancing over the change list. =)
00:49 sbadia yep yep, I was too fast
00:50 mgagne sbadia: np, I have to go now, I'll review tomorrow, cya!
00:50 sbadia cya :)
00:51 uberj joined #puppet-openstack
00:53 xavpaice michchap: ping
00:53 michchap xavpaice: pong
00:54 xavpaice hey.  I was wondering how you feel about moving the hiera lookups outside of the compat.pp manifest, and having some parameter defaults instead that could be overridden when calling the class?
00:55 xavpaice although I think everyone should use hiera, moving to params would mean those without can still make use of the class
00:57 michchap xavpaice: yeah totally fine with that.
00:57 michchap xavpaice: also after talking to mgagne we're not going to move openstacklib into stackforge in its current form
00:58 xavpaice fair call - it's quite specific as it is now, for stackforge it probably wants to be more generic.
00:58 xavpaice pull request coming :)
00:58 michchap xavpaice: the 'lib' will just be for lower level things, and I'm going to work with EmilienM to get puppet-openstack-cloud doing things.
00:59 openstackgerrit Sebastien Badia proposed a change to stackforge/puppet-glance: Fix test regression: each statement isn't correctly used here.  https://review.openstack.org/94505
01:00 xavpaice great - so 'lib' will move into puppet-openstack-cloud then?
01:08 kitp joined #puppet-openstack
01:09 daneyon joined #puppet-openstack
01:12 rmoe joined #puppet-openstack
01:27 michchap xavpaice: going to figure it out as we go, but atm that's my thinking, yep
01:29 xingchao joined #puppet-openstack
01:30 xarses joined #puppet-openstack
01:56 michchap joined #puppet-openstack
01:59 shakamunyi joined #puppet-openstack
02:08 daneyon joined #puppet-openstack
02:13 aimon joined #puppet-openstack
02:15 aimon joined #puppet-openstack
02:16 alugovoi joined #puppet-openstack
02:18 aimon joined #puppet-openstack
02:20 aimon joined #puppet-openstack
03:10 badiane_ka joined #puppet-openstack
03:18 kitp joined #puppet-openstack
03:22 thumpba joined #puppet-openstack
03:30 kkirkpatrick joined #puppet-openstack
03:31 aimon joined #puppet-openstack
03:32 aimon joined #puppet-openstack
03:34 aimon joined #puppet-openstack
03:35 kitp joined #puppet-openstack
03:36 aimon joined #puppet-openstack
04:24 sanjayu joined #puppet-openstack
04:45 social joined #puppet-openstack
05:10 michchap xavpaice: the inifile provider already has puppetlabs-inifile as a dependency
05:25 social joined #puppet-openstack
06:04 social joined #puppet-openstack
06:16 dgollub joined #puppet-openstack
06:26 kitp joined #puppet-openstack
06:42 arnaud_orange joined #puppet-openstack
07:01 aldavud joined #puppet-openstack
07:03 artem_panchenko joined #puppet-openstack
07:08 arnaud_orange joined #puppet-openstack
07:09 thumpba joined #puppet-openstack
07:24 gcha joined #puppet-openstack
07:27 kitp joined #puppet-openstack
07:42 dgollub joined #puppet-openstack
07:47 openstackgerrit Sergey Vasilenko proposed a change to stackforge/puppet-keystone: Make service provider as changeble incoming parameter.  https://review.openstack.org/86551
07:47 EmilienM good morning
08:00 aldavud joined #puppet-openstack
08:11 mmagr joined #puppet-openstack
08:14 derekh_ joined #puppet-openstack
08:27 kitp joined #puppet-openstack
08:33 arnaud_orange morning
08:40 francois hi there
08:41 francois michchap, xavpaice: oh that's the first time I hear of puppet-openstack-cloud
08:42 francois is it another composition layer?
08:45 francois as a user who started out with puppet-openstack and then migrated to puppet_openstack_builder
08:45 francois I'm getting somewhat lost with so many different composition layers ;)
08:46 arnaud_orange left #puppet-openstack
09:04 giorgiodinapoli joined #puppet-openstack
09:04 giorgiodinapoli hi
09:14 mattymo hey EmilienM francois and michchap
09:16 francois gi mattymo
09:16 francois s/g/h/
09:28 kitp joined #puppet-openstack
09:29 francois for those that were at the summit, is there any recording of the juno-summit-ops-puppet session?
09:29 francois or just the pad?
09:44 EmilienM francois: just pad
09:49 giorgiodinapoli has any one solved this as of yet?: Error: MySQL setup failed. The inferred location of the database based on the
09:49 giorgiodinapoli openstack::network::management hiera value is . The
09:58 francois ok EmilienM thanks
09:58 francois I'm interested to know of there was some agreement during the summit on a common composition layer?
10:19 aldavud giorgiodinapoli: afaik this means your host is lacking the correct network config on the interface
10:21 giorgiodinapoli aldavud but i used the default file like it is in examples. so i wonder whats worng
10:29 kitp joined #puppet-openstack
10:41 hipster joined #puppet-openstack
11:12 michchap mattymo: hey, thanks for the response
11:14 michchap francois: puppet-openstack-cloud is currently both a module for ancillary functions like loadbalancing and repo management as well as a composition layer. It's like a better designed puppet-openstack. I will be working with EmilienM over the next few weeks to see if I can convert it to use puppet_openstack_builder so it is just an ancillary module, but we'll see how we go.
11:16 michchap francois: there are a ton of changes planned for puppet_openstack_builder this cycle. We're still working out the details but we hope to have shared CI and a bunch of other things going by Juno if all goes well.
11:22 mattymo michchap, what are the main features of puppet_openstack_builder and where can I read up on it?
11:27 exed__ joined #puppet-openstack
11:28 michchap mattymo: it's a composition layer written in yaml - the main features are: roles defined as a list of classes and a list of class_groups (which themselves are a list of classes); a set of data mappings so that you can set things like 'debug' or the controller VIP in a single spot and have it propagate to the needed places; the concept of scenarios to allow encapsulation of different deployment types; and globals that allow manipul
11:28 michchap ation of the hierarchy so that we can encapsulate data for eg each glance backend or neutron plugin
11:29 mattymo globals that are modifiable as you get more granular - awesome
11:29 kitp joined #puppet-openstack
11:30 michchap mattymo: so my goal with the lib modules is to compliment this, but I don't want any composition layer things in the lib because we already have all the control we need
11:30 michchap mattymo: this cycle we're working on making it overlayable so that Cisco, Aptira and eNovance can all 'overlay' their own scenarios on top of it and override what they need to specify the configurations they actually support
11:31 mattymo right
11:31 mattymo I think this particular feature is something that duplicates what we already do "not bad" in fuel
11:31 michchap mattymo: right now it's similar to but less featured than Fuel
11:31 mattymo but it's all python
11:31 michchap mattymo: yep the concepts are all very similar :)
11:32 mattymo it's tightly bunded with our web UI and its db backend, so it's not like we can break it up
11:32 michchap mattymo: I plan on driving the yaml layer from a web UI in the medium term, but we have other things that need to be done first.
11:32 mattymo michchap, this monster: https://github.com/stackforge/fuel-web/blob/m​aster/nailgun/nailgun/fixtures/openstack.yaml
11:33 michchap mattymo: wow - a giant hash!
11:34 mattymo it's modular finally now
11:37 mattymo but it doesn't do specific classes
11:37 mattymo we still hardcode all that into roles
11:38 mattymo haproxy management would be nice to see, but we're intensely customized in our vip management + haproxy that we probably won't be able to share
11:38 mattymo we're working around a few bugs by creating a network namespace just for haproxy mgmt traffic
11:41 gcha joined #puppet-openstack
11:47 michchap mattymo: I did look at Fuel hoping to 'borrow' some of your HA work but it seemed to be woven into most of the modules instead of being a piece to be added or subtracted
11:48 michchap mattymo: (to be fair, I don't really see the point of deploying non-HA)
11:48 mattymo :)
11:48 mattymo fun fact.. we realized our HA works as non-HA just fine with 0 modifications - just a single controller and storage
11:49 mattymo so we're going to just do HA as default and make a note to the user if there's only 1 controller that it can't be HA
11:50 michchap mattymo: that's pretty cool. I guess that gets harder the more stuff you add - fencing etc. Mine is in a similar situation where it can run the same role from 1 .. n controllers. It's a neat model.
11:50 mattymo our HA is all in openstack::controller_ha, with a few hacks in osnailyfacter::cluster_ha for storage and vip management
11:51 mattymo our bottleneck is UI modifications to match all the capabilities of the modules
11:51 mattymo you can't just put 50 checkboxes or text fields in front of a user when they aren't likely to change any of it
11:51 michchap mattymo: I am thinking of auto-generating the web UI from the data model. It's hard to get right though
11:52 mattymo start really really simple. go django if you can
11:53 michchap mattymo: I won't start on it until all the module doc is done - that's where I plan on putting the descriptions for all the text fields.
11:53 michchap mattymo: I nearly have the lint check done for parameter doc...
11:54 mattymo it would be nice if you could parse docstrings and generate your yaml
11:54 mattymo it's quite brutal to go from (mostly) unstructured data to rigidly structured
11:55 dgollub joined #puppet-openstack
11:55 michchap mattymo: I was going to parse the docstrings to generate the UI, and have the parameters described by those docstrings pushed into the yaml where the user chooses to make a change
11:55 mattymo are you using default passwords still? or will you generate random ones?
11:56 michchap they're user specified
11:57 michchap one for each service + one for admin user, and a couple of shared secrets
11:57 mattymo consider switching to generated - makes less work for the user
11:57 mattymo and they can retrieve the yaml and look up passwords later if they want
11:58 michchap mattymo: Generate within puppet or from the frontend?
11:58 mattymo custom admin user of course let them decide
11:58 mattymo if you do masterless, you'll get many different passwords. if you have puppetdb you can generate once and store
11:58 mattymo since we're masterless it's in frontend
11:58 michchap mattymo: I would probably generate from the frontend as well - we are also masterless
11:59 michchap mattymo: I guess the user doesn't tend to need the service user passwords...that is a good point
11:59 mattymo I guess the only other thing we do that's peculiar in fuel is parallel puppet runs
12:00 mattymo first controller runs to the end, then all other controllers in parallel, then all computes and storage in parallel
12:00 mattymo if we do mongodb, then it goes before first controller
12:00 michchap right...I've generally been doing the control nodes one at a time in the past
12:01 michchap but that makes sense
12:01 mattymo it's mostly for swift
12:03 berendt joined #puppet-openstack
12:08 aldavud giorgiodinapoli: and did you configure your network as described? i.e. having an interface with address 192.168.33.x?
12:12 morazi joined #puppet-openstack
12:22 giorgiodinapoli aldavud, if i use the default values, i think this error can not be related telling me : MySQL setup failed. The inferred location of the database based on the
12:22 giorgiodinapoli openstack::network::management hiera value is . The
12:22 giorgiodinapoli explicit address from openstack::controller::address::management
12:22 giorgiodinapoli is 172.16.33.4  <-- but management value is not .
12:22 francois michchap: thanks for the details, sounds quite interesting
12:22 uberj joined #puppet-openstack
12:27 giorgiodinapoli actually i tried this:  $management_network = hiera('openstack::network::management') notify {$management_network:} but its not notifiing anything
12:30 kitp joined #puppet-openstack
12:37 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-cinder: Enable cinder server to be run in SSL mode  https://review.openstack.org/94395
12:41 morazi joined #puppet-openstack
12:44 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Data mappings for Galera in full_ha deployments  https://review.openstack.org/92949
12:44 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Drop obsolete Galera references  https://review.openstack.org/92902
12:45 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-nova: Enable nova server to be run in SSL mode  https://review.openstack.org/94399
12:46 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-cinder: Enable cinder server to be run in SSL mode  https://review.openstack.org/94395
12:46 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-nova: Enable nova server to be run in SSL mode  https://review.openstack.org/94399
12:52 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Minor fixups for the stacktira bootstrap script  https://review.openstack.org/92561
12:53 daneyon joined #puppet-openstack
12:54 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Conversion script skipping first in list  https://review.openstack.org/92750
12:55 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Add options for instance resizing  https://review.openstack.org/92301
12:56 dprince joined #puppet-openstack
12:57 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-heat: Enable heat server to be run in SSL mode  https://review.openstack.org/94412
13:03 sanjayu joined #puppet-openstack
13:03 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-neutron: Enable neutron server to be run in SSL mode  https://review.openstack.org/94432
13:13 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Added 'tee' to the logging command in install.sh  https://review.openstack.org/92451
13:16 Mr_D joined #puppet-openstack
13:22 kitp joined #puppet-openstack
13:35 openstackgerrit A change was merged to stackforge/puppet-swift: Add swift-ring-builder multi region support  https://review.openstack.org/93115
13:40 openstackgerrit A change was merged to stackforge/puppet-keystone: Make service provider as changeble incoming parameter.  https://review.openstack.org/86551
13:43 [1]Mr_D joined #puppet-openstack
13:47 [2]Mr_D joined #puppet-openstack
13:47 dmsimard Merge storm !
13:49 hipster joined #puppet-openstack
13:59 prad_ joined #puppet-openstack
14:07 openstackgerrit Yanis Guenane proposed a change to stackforge/puppet-nova: Enable nova server to be run in SSL mode  https://review.openstack.org/94399
14:15 rwsu joined #puppet-openstack
14:30 shakamunyi joined #puppet-openstack
14:42 kkirkpatrick joined #puppet-openstack
15:06 thumpba joined #puppet-openstack
15:08 mdorman joined #puppet-openstack
15:11 alugovoi joined #puppet-openstack
15:16 openstackgerrit Francois Deppierraz proposed a change to stackforge/puppet_openstack_builder: Pin modules to the stable/havana branch  https://review.openstack.org/94645
15:19 kkirkpatrick joined #puppet-openstack
15:25 kkirkpatrick joined #puppet-openstack
15:51 dachary joined #puppet-openstack
16:06 ryanycoleman joined #puppet-openstack
16:09 alugovoi joined #puppet-openstack
16:15 cbaesema joined #puppet-openstack
16:16 ryanycoleman joined #puppet-openstack
16:34 kkirkpatrick joined #puppet-openstack
16:34 hogepodge joined #puppet-openstack
16:37 rmoe joined #puppet-openstack
16:38 aimon joined #puppet-openstack
16:40 openstackgerrit Mark T. Voelker proposed a change to stackforge/puppet_openstack_builder: Fix Cinder deprecation warning  https://review.openstack.org/94671
16:49 kkirkpatrick joined #puppet-openstack
17:01 ryanycoleman joined #puppet-openstack
17:02 xarses joined #puppet-openstack
17:20 xarses joined #puppet-openstack
17:24 kaleng joined #puppet-openstack
17:30 openstackgerrit Emilien Macchi proposed a change to stackforge/puppet-keystone: enabling tokens flushing using cron  https://review.openstack.org/93917
17:30 EmilienM sbadia: mgagne : I made a new patchset with your inputs
17:30 mgagne EmilienM: and saw you fixed the doc =)
17:30 mgagne EmilienM: openstack doc
17:31 EmilienM mgagne: you see everything
17:31 EmilienM my god
17:31 badiane_ka joined #puppet-openstack
17:35 aldavud joined #puppet-openstack
17:40 dmsimard EmilienM: He does.
17:45 kaleng joined #puppet-openstack
17:46 hogepodge Looking over this today. https://github.com/openstack/nova-spec
17:46 hogepodge Seems fairly straight forward.
17:50 mgagne +s
17:51 openstackgerrit Emilien Macchi proposed a change to stackforge/puppet-keystone: enabling tokens flushing using cron  https://review.openstack.org/93917
17:56 xarses joined #puppet-openstack
18:01 mgagne aimon: in which timezone are you?
18:02 aimon PST
18:02 aimon mgagne ˆ
18:04 openstackgerrit A change was merged to stackforge/puppet-keystone: enabling tokens flushing using cron  https://review.openstack.org/93917
18:04 berendt joined #puppet-openstack
18:09 EmilienM make sense to backport this one IMO ^
18:12 thumpba joined #puppet-openstack
18:19 ryanycoleman joined #puppet-openstack
18:23 gcha joined #puppet-openstack
18:24 mgagne hogepodge: stackforge/puppet-openstack-specs plz =)
18:24 hogepodge I’ll get the project in for review today. Definitely need it. :-P
18:25 mgagne hogepodge: just to make sure the name is right =)
18:25 hogepodge I caught a cold in Atlanta. Yesterday was dead for me. Monday was slightly better.
18:25 mgagne hogepodge: I saw inconsistent in previous email, _ vs - =)
18:25 hogepodge Yeah, I keep forgetting the last s
18:25 hogepodge For the rest of the projects we want to have the _
18:26 hogepodge Puppet modules don’t do well with _ in the name. So puppet-blah_blah
18:26 hogepodge first _ s -
18:26 mgagne hogepodge: are you referring to openstacklib?
18:26 hogepodge “Puppet modules don’t do well with - in the name.”
18:26 hogepodge Yeah.
18:27 mgagne hogepodge: puppet-openstacklib ?
18:27 hogepodge Oh, there was discussion about that a couple of days ago. People really favored puppet-openstack(-|_)lib
18:28 hogepodge Whatever the names shake out to be I’ll announce the correct ones. :-P
18:28 mgagne hogepodge: yes please =)
18:29 mgagne hogepodge: I don't like much the _, openstacklib looks more pleasant to my eyes =)
18:29 mgagne hogepodge: like puppetlabs-stdlib =)
18:30 aimon openstacklib is much nicer IMHO
18:30 hogepodge what about openstackcommon vs openstack_common?
18:30 aimon and consistent with stdlib as mgagne mentions
18:31 mgagne hogepodge: openstack_common is a bit long but I can do with the underscore
18:31 aimon I prefer the _ in that case.. I think lib works due to it being truncated form of full word
18:35 mgagne aimon: +1
18:51 aldavud_ joined #puppet-openstack
18:54 hogepodge https://github.com/hogepod​ge/puppet-openstack-specs
18:54 hogepodge Please read the README.md and template.rst files and comment, send pull requests. This will be the seed for the stackforge project.
18:57 ianw- joined #puppet-openstack
18:59 ianw-- joined #puppet-openstack
19:00 mgagne hogepodge: can we reword "Puppet Openstack projects" for "Puppet for Openstack projects". The former could be misinterpreted as an official project/endorsement from OpenStack.
19:00 hogepodge Yup. Just pull request it! :-)
19:00 hogepodge Old school for now :-P
19:04 rickerc_ joined #puppet-openstack
19:05 rickerc joined #puppet-openstack
19:13 aldavud_ joined #puppet-openstack
19:16 kkirkpatrick joined #puppet-openstack
19:30 schofield joined #puppet-openstack
19:59 ryanycoleman joined #puppet-openstack
20:02 aldavud joined #puppet-openstack
20:04 schofield left #puppet-openstack
20:08 EmilienM spredzy: I let you a question/comment on https://review.openstack.org/#/c/94399/
20:21 xarses joined #puppet-openstack
20:25 daneyon joined #puppet-openstack
20:58 ryanycoleman joined #puppet-openstack
21:09 ryanycoleman joined #puppet-openstack
21:27 xarses joined #puppet-openstack
21:31 michchap joined #puppet-openstack
21:31 dgollub joined #puppet-openstack
22:03 ryanycoleman joined #puppet-openstack
22:05 michchap joined #puppet-openstack
22:09 xarses dmsimard: poke https://review.openstack.org/#/c/90556/ https://review.openstack.org/#/c/90558/
22:09 dmsimard xarses: pong
22:10 dmsimard xarses: I've been extremely busy recently, i'll try and check tonight or tomorrow morning. Fair ?
22:10 xarses Summit was fun, whish you could have made it
22:11 xarses dmsimard: Just trying to keep up myself, they are dachary's anyway =)
22:11 xarses get too them when you can
22:12 dmsimard xarses: only thing that sort of bothers me
22:12 dmsimard is the usage of sgdisk
22:12 dmsimard I probably commented about that on other reviews
22:12 dmsimard sgdisk doesn't come bundled with Ubuntu afaik, sfdisk is
22:13 dmsimard But it's not enough to prevent stuff from going through
22:13 xarses sgdisk is a requirement of ceph-disk afaict
22:14 dmsimard oh, I hadn't thought that it could come bundled as a ceph dependency, you're right
22:15 * xarses opens up an upstart file
22:15 xarses erm pre-seed even
22:15 dmsimard xarses: Do we really want to support cuttefish ? https://review.openstack.org/#/c/90556/
22:16 dmsimard We should at least cut to dumpling
22:16 xarses getting quite close on that point
22:16 dmsimard Or cut the repo in branches for releases like the puppet-openstack
22:17 dmsimard Otherwise we're never going to end if we try to support everything for every release :p
22:17 xarses I don't think we are complete enough to do that yet
22:17 dmsimard right.
22:19 openstackgerrit Chris Ricker proposed a change to stackforge/puppet_openstack_builder: Add variables for active-passive Galera HA proxy  https://review.openstack.org/94738
22:20 xarses dmsimard: it looks like sgdisk comes with gdisk package
22:20 dmsimard I have a feeling all those regexes are going to come back and bite us in the rear. We need to find a more reliable way.
22:20 xarses so it should be around for everyone
22:21 uberj joined #puppet-openstack
22:23 xarses dmsimard: yep, they are hard to read. At some point they might stop working, but the integration tests should prevent it from being merged
22:23 xarses well they should anway
22:24 dmsimard For https://review.openstack.org/#/c/94287/
22:24 dmsimard I wish we could've got https://review.openstack.org/#/c/64388/ merged
22:25 dmsimard I need to compare the two before I weigh in on the newer commit
22:26 xarses We might as well pull it back in, I wanted it too
22:26 dmsimard I'll add a comment in the meantime
22:29 tdb_ joined #puppet-openstack
22:30 fvollero_ joined #puppet-openstack
22:30 francois1 joined #puppet-openstack
22:31 exed_- joined #puppet-openstack
22:33 tfz1 joined #puppet-openstack
22:33 michchap_ joined #puppet-openstack
22:34 aldavud joined #puppet-openstack
22:34 social__ joined #puppet-openstack
22:34 ashp joined #puppet-openstack
22:35 spredzy_ joined #puppet-openstack
22:36 adreyer joined #puppet-openstack
22:37 kaleng joined #puppet-openstack
22:42 uberj joined #puppet-openstack
22:43 aldavud_ joined #puppet-openstack
22:44 openstackgerrit A change was merged to stackforge/puppet-ceph: ceph::osd : add support for mon data directories  https://review.openstack.org/90556
22:45 openstackgerrit A change was merged to stackforge/puppet-ceph: ceph::osd : indent tests  https://review.openstack.org/90557
22:45 openstackgerrit A change was merged to stackforge/puppet-ceph: ceph::osd : add tests for osd data on an existing partition  https://review.openstack.org/90558
22:49 openstackgerrit A change was merged to stackforge/puppet_openstack_builder: Fix Cinder deprecation warning  https://review.openstack.org/94671
22:49 ianw joined #puppet-openstack
22:55 rharrison joined #puppet-openstack
23:12 rwsu joined #puppet-openstack
23:13 spredzy_ joined #puppet-openstack
23:13 michchap_ joined #puppet-openstack
23:17 aimon joined #puppet-openstack
23:17 kitp joined #puppet-openstack
23:18 hogepodge https://review.openstack.org/#/c/94743/
23:21 mgagne hogepodge: I suggest the gerribot change to be split
23:21 mgagne hogepodge: omnibus commits aren't recommended
23:21 hogepodge Figured. :-P
23:22 hogepodge It’s just one line, necessary anyway. I shouldn’t have said anything about it.
23:24 michchap_ hogepodge: +1 to the convo way up about not having _ in module names.
23:25 michchap_ hogepodge: if we aren't going to nuke puppet-openstack I may as well just start adding things to that tbh.
23:30 aimon joined #puppet-openstack
23:30 hogepodge michchap_ the reason for the _ weirdness is because of the Puppet module tool.
23:31 hogepodge - in names is going to be less supported as time goes on.
23:31 hogepodge from what I understand.
23:31 hogepodge I’m great with openstacklib
23:31 hogepodge but openstackcommon?
23:32 hogepodge mgagne that review above is going to take a long while.
23:32 michchap_ hogepodge: well...we can just put it in openstack - we're not deleting it are we?
23:33 hogepodge Not deleting it, the source needs to be there. People depend on it.
23:33 michchap_ if we're not deleting it I may as well put the common stuff there.
23:33 hogepodge mgagne infra backlog is over 100 deep post summit
23:33 hogepodge Yes, but not everybody will use it.
23:35 michchap_ why is that a problem? Not everybody uses everything in puppet-openstack today. I just used it for repo management from havana onwards and ignore the rest of it.
23:39 hogepodge I’m not using puppet-openstack at all any more.
23:39 hogepodge mgagne https://review.openstack.org/#/c/94748/
23:39 hogepodge common/lib is meant to serve that role
23:40 michchap_ then what role does puppet-openstack serve?
23:43 michchap_ at the moment it has auth_file, test_file, provision, repo and swift being used by either Cisco or myself.
23:45 hogepodge we should maintain those then, but have that functionality replaced with the other modules.
23:45 hogepodge The role should be deployment and testing example.
23:45 hogepodge I’d like to replace it with the puppetlabs module I’m maintaining, or some other suitable module.
23:47 michchap_ given the argument for not deleting it was that people are depending on it I'm not sure how we can replace it.
23:48 hogepodge It’s marked as deprecated. Those deprecated functions will go elsewhere. It can’t happen overnight.
23:48 michchap_ it's been marked that way for an entire cycle now
23:48 michchap_ I feel that's enough time
23:50 openstackgerrit joined #puppet-openstack

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