Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-07-14

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

All times shown according to UTC.

Time Nick Message
00:13 dgollub joined #puppet-openstack
00:37 JayJ joined #puppet-openstack
01:30 aldavud joined #puppet-openstack
01:30 xingchao joined #puppet-openstack
01:37 gildub joined #puppet-openstack
03:00 sputnik1_ joined #puppet-openstack
03:36 sputnik1_ joined #puppet-openstack
03:58 mjblack joined #puppet-openstack
04:12 sputnik1_ joined #puppet-openstack
04:17 sputnik1_ joined #puppet-openstack
04:20 sanjayu joined #puppet-openstack
04:29 sputnik1_ joined #puppet-openstack
05:18 sputnik1_ joined #puppet-openstack
05:31 JayJ joined #puppet-openstack
06:00 dgollub joined #puppet-openstack
06:10 dgollub joined #puppet-openstack
06:47 bogdando joined #puppet-openstack
06:58 dgollub joined #puppet-openstack
07:03 artem_panchenko joined #puppet-openstack
07:12 sputnik1_ joined #puppet-openstack
07:15 sputnik1_ joined #puppet-openstack
07:24 openstackgerrit Liran Cohen proposed a change to stackforge/puppet-nova: Control the number of workers for nova-conductor service  https://review.openstack.org/106656
07:40 aldavud joined #puppet-openstack
07:55 tfz joined #puppet-openstack
08:10 igajsin1 joined #puppet-openstack
08:13 derekh__ joined #puppet-openstack
08:15 ddmitriev joined #puppet-openstack
08:42 qba73 joined #puppet-openstack
08:51 social anywone for final +1 workflow for this? https://review.openstack.org/#/c/104550/\
08:51 social https://review.openstack.org/#/c/104550
08:51 qba73 joined #puppet-openstack
08:57 mmagr joined #puppet-openstack
08:58 aldavud_away joined #puppet-openstack
10:11 Zordrak mgagne: I'm not sure how best to get hold of florian - he's not on IRC. I think he may be on extended holiday
10:12 Zordrak mgagne: I just kind of assumed that this of all places would be the best place to find anyone who's been anywhere near kickstack
10:13 Zordrak I'm not explicitly tied to kickstack as a solution; I'm just looking for a way to do a reasonable proof of concept with openstack via puppet for when we want to start using it within the company.
10:14 Zordrak kickstack seemed the most appropriate way to do so; bearing in mind I'm trying to get an extremely basic default configuration and I manage puppet via TheForeman
10:21 berendt joined #puppet-openstack
10:25 JayJ_ joined #puppet-openstack
10:56 Zordrak I would really appreciate any suggestions on the correct path to follow to get from an existing foreman installation to having a development openstack all-in-one install deployed from puppet
10:58 gildub joined #puppet-openstack
11:40 saju_m joined #puppet-openstack
11:46 saju_m joined #puppet-openstack
11:59 anonitus joined #puppet-openstack
12:02 denca joined #puppet-openstack
12:06 berendt joined #puppet-openstack
12:13 gildub joined #puppet-openstack
12:20 saju_m joined #puppet-openstack
12:25 JayJ_ joined #puppet-openstack
12:47 dprince joined #puppet-openstack
12:47 morazi joined #puppet-openstack
13:04 mtl1 joined #puppet-openstack
13:14 tfz joined #puppet-openstack
13:15 csschwe joined #puppet-openstack
13:16 dgollub joined #puppet-openstack
13:25 JayJ_ joined #puppet-openstack
13:41 JayJ_ joined #puppet-openstack
13:44 JayJ_ joined #puppet-openstack
13:51 JayJ_ joined #puppet-openstack
13:57 imcsk8 joined #puppet-openstack
14:15 pradk joined #puppet-openstack
14:17 daneyon joined #puppet-openstack
14:18 daneyon joined #puppet-openstack
14:28 mjblack joined #puppet-openstack
14:29 mjblack joined #puppet-openstack
14:39 JayJ_ joined #puppet-openstack
14:52 hogepodge joined #puppet-openstack
14:52 hogepodge Good morning!
14:59 crinkle morning o/
15:00 marun joined #puppet-openstack
15:00 mfisch marnin
15:00 hogepodge Summertime in the north, so not sure how many people are here today.
15:00 hogepodge #meeting start
15:00 JayJ_ joined #puppet-openstack
15:01 hogepodge #topic puppet-openstacklib and puppet-openstack_extras
15:02 openstackgerrit Matt Fischer proposed a change to stackforge/puppet-keystone: Fix password update for users  https://review.openstack.org/106788
15:02 hogepodge Both libraries have been merged and set up, thanks to crinkle for that.
15:03 hogepodge Database work is under review, https://review.openstack.org/#/c/105280/
15:03 mfisch (sorry about that mp interrupting)
15:04 hogepodge Once it’s merged I’d like to see the main projects start to take advantage of that code.
15:05 mfisch I'm excited to have that lib working
15:06 hogepodge For the comon api code (starting with keystone) we should be thinking about how to authenticate against the endpoints.
15:06 hogepodge This is a pretty light meeting, and it seems like there aren’t many people here.
15:06 doo joined #puppet-openstack
15:06 hogepodge Anyone else have topics they’d like to bring up?
15:07 mfisch hogepodge: whats the plan to switch stuff over to openstacklib?
15:08 hogepodge Once code is merged into lib we start refactoring the modules. We need to be careful about deprecation, but it should be faily straightforward.
15:08 mfisch keystone and neutron would be my first two candidates
15:08 mfisch thanks
15:09 hogepodge For the types/providers I’d just like to see the back-end changed and keeping the API the same.
15:11 doo_ joined #puppet-openstack
15:14 hogepodge guess that’s it for this morning
15:15 hogepodge #meeting end
15:15 mfisch thanks hogepodge
15:16 hogepodge mfisch np
15:18 aldavud joined #puppet-openstack
15:21 openstackgerrit Matt Fischer proposed a change to stackforge/puppet-keystone: allow setting the tenant descriptions  https://review.openstack.org/106791
15:23 sputnik1_ joined #puppet-openstack
15:26 xingchao joined #puppet-openstack
15:29 mdorman joined #puppet-openstack
15:36 mdorman joined #puppet-openstack
15:40 yocum joined #puppet-openstack
15:40 yocum I probably missed the meeting, didn't I?  :(
15:40 yocum blentz: is there a log, somewhere
15:41 * yocum had to configure sasl for freenode, apparently.  :/
15:41 blentz yocum: i'll send you my channel logs
15:41 yocum blentz: thx
15:45 openstackgerrit Mathieu Gagné proposed a change to stackforge/puppet-neutron: Add ability to configure dhcp_agent_notification  https://review.openstack.org/105323
15:45 openstackgerrit Mathieu Gagné proposed a change to stackforge/puppet-neutron: Add linuxbridge mechanism support for ML2 plugin  https://review.openstack.org/106227
15:48 crow23 joined #puppet-openstack
15:53 kaiserpathos joined #puppet-openstack
15:53 rmoe joined #puppet-openstack
15:54 mdorman_ joined #puppet-openstack
16:07 hogepodge left #puppet-openstack
16:19 aimon joined #puppet-openstack
16:20 sputnik1_ joined #puppet-openstack
16:26 openstackgerrit Mathieu Gagné proposed a change to stackforge/puppet-neutron: Rename QUOTAS section for quotas  https://review.openstack.org/106812
16:30 hogepodge joined #puppet-openstack
16:32 aldavud_ joined #puppet-openstack
16:44 saju_m joined #puppet-openstack
16:49 zeus joined #puppet-openstack
17:04 csschwe joined #puppet-openstack
17:16 hogepodge_ joined #puppet-openstack
17:43 doo joined #puppet-openstack
17:47 openstackgerrit Colleen Murphy proposed a change to stackforge/puppet-openstacklib: Add db::mysql and db::mysql::host_access to openstacklib  https://review.openstack.org/105280
17:47 mmaglana joined #puppet-openstack
17:47 crinkle hogepodge: ^
17:49 ashp joined #puppet-openstack
17:54 hogepodge We should probably go through the pain of migrating the Modulefile to metadata.json during this cycle.
17:58 mgagne hogepodge: and force us to define a support matrix =)
17:59 xarses joined #puppet-openstack
17:59 hogepodge mgagne does it force us to do that?
17:59 hogepodge or are you saying we should opt in to that?
17:59 mgagne hogepodge: do you know what kind of info metadata.json has?
18:00 xarses joined #puppet-openstack
18:00 hogepodge http://docs.puppetlabs.com/puppet/latest/referenc​e/modules_publishing.html#fields-in-metadatajson
18:01 mgagne hogepodge: I see operatingsystem_support
18:01 hogepodge operatingsystem_support is optional
18:01 hogepodge Not that we wouldn’t want to use it.
18:01 mgagne hogepodge: shouldn't we provide it anyway%
18:13 mgagne hogepodge: this screenshot shows puppet version support: http://docs.puppetlabs.com/puppet/latest​/reference/images/forge_new_release.png where is it filled?
18:20 hogepodge mgagne it’s a non-documented field
18:20 shakamunyi joined #puppet-openstack
18:21 hogepodge It was a feature that was added, then caused problems. It probably won’t work for us right out of the gate anyway, since we have a hole in our puppet support matrix (3.0)
18:27 saju_m joined #puppet-openstack
18:30 hogepodge_ joined #puppet-openstack
18:49 openstackgerrit Emilien Macchi proposed a change to stackforge/puppet-neutron: agent/lbaas: fix HAproxy papckage relationship  https://review.openstack.org/106839
18:50 EmilienM hogepodge: sounds like I did not miss any meeting, it was off today here
18:51 rwsu joined #puppet-openstack
18:52 openstackgerrit Emilien Macchi proposed a change to stackforge/puppet-neutron: agent/lbaas: fix HAproxy package relationship  https://review.openstack.org/106839
19:50 mgagne EmilienM: still there?
19:51 EmilienM mgagne: was leaving, please let me a message in query if needed
19:51 mgagne EmilienM: could you reapprove? I rebased https://review.openstack.org/#/c/106227/
19:51 EmilienM mgagne: ok
19:51 EmilienM done
19:51 mgagne ty
19:55 openstackgerrit A change was merged to stackforge/puppet-neutron: Add linuxbridge mechanism support for ML2 plugin  https://review.openstack.org/106227
20:18 hogepodge EmilienM we had a short one. Can you look at the db request? maybe mgagne too. I had crinkle take out the dependency on the db service (and the include) because we can’t be certain what installed the database (or if it’s even colocated on the node).
20:18 hogepodge EmilienM have a great day and holiday, there’s not a huge rush.
20:18 mgagne hogepodge: which patch set got it removed?
20:19 hogepodge mgagne 14
20:19 hogepodge https://review.openstack.org/#/c/105280/
20:19 mgagne ok cool
20:19 mgagne but
20:19 mgagne resources still depend on Class['mysql::server']
20:20 mgagne what's the use case we wish to support?
20:22 mgagne hogepodge: ^
20:22 crinkle it is essentially copying mysql::db from the mysql module, it needs the server to exist before it can create a database and a grant
20:23 hogepodge Server installation feels like it should be an extra, and depending explicitly on server I think is a bug. But I could be wrong.
20:23 hogepodge Plus, it’s possible that you want to set up the database on a different node than where it’s located.
20:23 mgagne hogepodge: this isn't supported by puppetlabs-mysql
20:23 mgagne hogepodge: provisioning has to happen on the same node
20:24 bodepd I'm about to be running some integration tests. I'm verifying that I can silently receive and use build events to trigger test openstack deploys
20:24 bodepd if I wind up spewing events, it's a bug and I will be fixing it.
20:24 mgagne hogepodge: so we basically did a copy/paste of mysql::db?
20:24 bodepd or spewing gerrit comments
20:25 crinkle mgagne: yes, but using a password hash instead of plaintext password
20:25 hogepodge mgagne it was a refactoring (so a kind of copy paste)
20:26 mgagne crinkle: it isn't using ensure_resource anymore, any particular reason?
20:26 mgagne crinkle: I would like those explanation to be included in the commit message if you don't mind
20:27 crinkle mgagne: sure, I might actually change that back to use ensure_resource
20:28 crinkle that would be safer
20:28 mgagne crinkle: whatever you choose, I would like a note about mysql::db being dropped (so we can use a password_hash instead) and why ensure_resource was dropped (if we still intent to drop it)
20:30 crinkle mgagne: what do you think about ensure_resource - I don't see why we would need it, since that user and db shouldn't be created outside of the db::mysql resource
20:30 crinkle but it doesn't hurt
20:30 crinkle I will change the commit message
20:31 mgagne crinkle: if we drop mysql::db for various reasons, I was more or less expecting an explanation and notes about why it got refactored to drop ensure_resource
20:32 mgagne crinkle: as I was +/- expecting a copy/paste from nova::db::mysql
20:33 crinkle mgagne: I will add explanations, but I'm rethinking my decision to drop ensure_resource
20:34 morazi joined #puppet-openstack
20:34 mgagne crinkle: whatever you come up with, just add a note and people will be able to weight in their opinion. Your explanation could be as simple as: ensure_resource isn't that great to start with for those reasons. =)
20:34 crinkle mgagne: okay
20:46 openstackgerrit Colleen Murphy proposed a change to stackforge/puppet-openstacklib: Add db::mysql and db::mysql::host_access to openstacklib  https://review.openstack.org/105280
20:52 mgagne crinkle: superawesome
20:57 aldavud joined #puppet-openstack
20:58 crinkle mgagne: :)
21:04 mjblack joined #puppet-openstack
21:08 EmilienM hogepodge: it was bastille day :-)
21:15 rstarmer joined #puppet-openstack
21:29 hogepodge EmilienM Vive la France!
21:29 EmilienM ahah
21:32 openstackgerrit joined #puppet-openstack
21:47 rstarmer joined #puppet-openstack
21:55 hogepodge mgagne for now moving to metadata.json is a bad idea
21:56 mgagne hogepodge: should we implement this bad idea then? =)
21:56 mmaglana joined #puppet-openstack
21:56 rstarmer joined #puppet-openstack
21:56 hogepodge mgagne right now pmt has a bug that only impacts you if you’re using a preexisting metadata.json file.
21:57 hogepodge If you’re still using Modulefile the generated metadata.json and checksums work just fine.
21:57 hogepodge So we shouldn’t move forward until the bug is fixed upstream.
21:59 mgagne hogepodge: what about people using older versions?
22:00 mgagne hogepodge: puppetlabs is known to only fix in the latest version and forget about the older ones or only include the fix in the next x.y.0 version
22:05 hogepodge magane the impact is on how the checksum is calculated for the metadata.json file itself, and is a big problem during packaging. I think it was introduced in 3.6, so older versions shouldn’t be a problem.
22:15 hogepodge mgagne it’s a bundling bug from 3.6. If you have a metadata.json file it computes a checksum of it, then normalizes the metadata.json file. That ordering is wrong. If metadata.json doesn’t exist to begin with the bogus checksum is never computed.
22:16 mgagne :-/
22:16 hogepodge using the ‘puppet module build’ command. So the version impact is strongly localized, but anything that’s been published with bad data needs to be republished. None of the puppet-openstack modules are impacted.
22:26 aldavud joined #puppet-openstack
22:38 aldavud joined #puppet-openstack
22:48 marun joined #puppet-openstack
22:59 rstarmer joined #puppet-openstack
23:09 mgagne if we aren't using puppet 3.6 to build our module packages, why should we care about this particular bug?
23:09 mgagne hogepodge: ^
23:11 hogepodge mgagne I’ve been using 3.6 for packaging, I just haven’t adoped the metadata.json file. Since it doesn’t exist, it never gets rewritten by the tool. So I’m not going to use metadata.json until the bug is fixed, because I don’t want the packaging to break or have to add extra steps to get it to work.
23:11 mgagne hogepodge: alright, makes sense
23:24 gildub joined #puppet-openstack
23:29 rstarmer joined #puppet-openstack
23:29 doo joined #puppet-openstack
23:53 rstarmer joined #puppet-openstack

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