Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-12-04

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

All times shown according to UTC.

Time Nick Message
00:05 rcallawa joined #puppet-openstack
00:27 dgurtner joined #puppet-openstack
00:27 dgurtner joined #puppet-openstack
01:06 openstackgerrit Scott Pham proposed stackforge/puppet-openstacklib: Adding augeas insertion check. Augeas change call will fail if key does not exists in policy.json. This fix will add the key/value if it doesn't exists or will change the value to match.  https://review.openstack.org/138915
01:09 topshare joined #puppet-openstack
01:29 ToMiles joined #puppet-openstack
01:30 rmoe joined #puppet-openstack
01:36 xarses joined #puppet-openstack
01:36 michchap joined #puppet-openstack
01:39 aimon joined #puppet-openstack
01:57 ericpeterson joined #puppet-openstack
02:17 ericpeterson joined #puppet-openstack
03:03 openstackgerrit Scott Pham proposed stackforge/puppet-openstacklib: Adding augeas insertion check. Augeas change call will fail if key does not exists in policy.json. This fix will add the key/value if it doesn't exists or will change the value to match. Removed -changed from patchset 1 to allow spec test to continue Chan  https://review.openstack.org/138915
03:05 ericpeterson joined #puppet-openstack
03:13 mmaglana joined #puppet-openstack
03:17 badiane_ka joined #puppet-openstack
03:49 mmaglana joined #puppet-openstack
04:02 ToMiles_ joined #puppet-openstack
04:32 badiane_ka joined #puppet-openstack
04:33 badiane_ka joined #puppet-openstack
04:44 markvoelker joined #puppet-openstack
05:02 sanjayu joined #puppet-openstack
05:06 aimon_ joined #puppet-openstack
05:08 mmaglana_ joined #puppet-openstack
05:23 fedexo joined #puppet-openstack
05:33 rcallawa joined #puppet-openstack
05:43 markvoelker joined #puppet-openstack
06:05 ericpeterson joined #puppet-openstack
07:11 rcallawa joined #puppet-openstack
07:14 EmilienM mgagne: michchap some parts are wrong
07:14 EmilienM in https://aptira.com/blog/112-what-is-​ceilometer-s-minimum-viable-product
07:15 EmilienM the database thing is explained here: http://docs.openstack.org/developer/ceilomet​er/architecture.html#high-level-description
07:19 topshare joined #puppet-openstack
07:22 mgagne EmilienM: "finding the right queries within the Ceilometer API to extract the information you need": the query language did not include the required features we needed to query data efficiently. "it is recommended to download the data from the API in order to work on it in a separate database": why have a middle man when you can gather the events yourself and not process/store it twice. "this is fully left up to the implementor": yep and now
07:22 EmilienM mgagne: you here :O
07:22 mgagne EmilienM: not for long
07:22 mgagne EmilienM: was working on that: https://twitter.com/mgagne​/status/540266340402356224
07:22 EmilienM looking
07:23 EmilienM wow
07:23 EmilienM another mgagne side project :)
07:23 mgagne hehe
07:23 EmilienM this is nice
07:24 EmilienM mgagne: so I know it's late for us now (2.24am) but we need to plan a puppet event in Montreal
07:24 mgagne sure thing
07:24 EmilienM let's start some discussion on the ML. Do you want me to run it?
07:25 mgagne sure, I might only be available in January though
07:26 EmilienM ok, let's plan something for January then
07:26 EmilienM mgagne: I'll manage if we can get sponsor thing
07:26 EmilienM mgagne: i.e. asking Red Hat
07:26 mgagne EmilienM: I'm not looking for a big event =)
07:26 EmilienM me neither
07:27 EmilienM but if we could have a room and some lunch
07:27 EmilienM that would be a start
07:28 mgagne yes
07:28 mgagne other similar side project: http://imgur.com/hiuPDVB self-balancing robot :)
07:28 EmilienM mgagne: when do you sleep?
07:29 mgagne EmilienM: tbh, I don't know :P
07:29 mgagne EmilienM: now would be a great idea I guess
07:30 EmilienM mgagne: me too. Good night, see you tomorrow
07:30 mgagne EmilienM: good night, cya
07:52 arnaud_orange joined #puppet-openstack
07:59 dgurtner joined #puppet-openstack
07:59 dgurtner joined #puppet-openstack
08:11 stamak joined #puppet-openstack
08:17 qba73 joined #puppet-openstack
08:24 aimon joined #puppet-openstack
08:25 topshare joined #puppet-openstack
08:26 rcallawa joined #puppet-openstack
08:44 mmagr joined #puppet-openstack
09:01 igajsin joined #puppet-openstack
09:06 tlbr joined #puppet-openstack
09:22 off_rhoden joined #puppet-openstack
09:27 rcallawa joined #puppet-openstack
09:41 gchamoul joined #puppet-openstack
09:42 gchamoul joined #puppet-openstack
09:53 tfz joined #puppet-openstack
10:08 tlbr joined #puppet-openstack
10:13 topshare joined #puppet-openstack
10:24 tlbr_ joined #puppet-openstack
10:28 rcallawa joined #puppet-openstack
10:55 derekh joined #puppet-openstack
11:41 berendt joined #puppet-openstack
12:17 rcallawa joined #puppet-openstack
12:24 rcallawa joined #puppet-openstack
12:31 topshare joined #puppet-openstack
13:03 EmilienM morning!
13:39 stamak joined #puppet-openstack
13:48 topshare joined #puppet-openstack
13:55 rcallawa joined #puppet-openstack
14:01 rcallawa joined #puppet-openstack
14:04 mmagr joined #puppet-openstack
14:11 ericpeterson joined #puppet-openstack
14:16 richm joined #puppet-openstack
14:37 openstackgerrit Ryan Hallisey proposed stackforge/puppet-nova: Add Ironic support into nova puppet modules  https://review.openstack.org/132704
15:05 ericpeterson joined #puppet-openstack
15:09 ericpeterson joined #puppet-openstack
15:40 prad joined #puppet-openstack
15:41 mdorman joined #puppet-openstack
15:51 markvoelker joined #puppet-openstack
15:56 berendt joined #puppet-openstack
16:00 openstackgerrit Merged stackforge/puppet-swift: Enlarging the scope of ssh module  https://review.openstack.org/113550
16:00 openstackgerrit Merged stackforge/puppet-keystone: fix apache::vhost concat test breakage  https://review.openstack.org/136887
16:01 openstackgerrit Merged stackforge/puppet-keystone: setup keystone using apache mod_wsgi  https://review.openstack.org/136885
16:07 EmilienM fc__: ping
16:09 EmilienM spredzy: FYI: https://review.openstack.org/#/c/138915/
16:11 stamak joined #puppet-openstack
16:13 fc__ EmilienM: pong
16:14 EmilienM fc__: I have three questions for you
16:14 EmilienM fc__: 1) I was wondering if you already has a look at stackforge/puppet-ceph recently
16:14 EmilienM fc__: 2) if you know the gap between enovance/puppet-ceph & stackforge/puppet-ceph
16:15 EmilienM fc__: 3) why enovance/puppet-ceph requires to run puppet 4 or 5 times to have the whole setup up and running with pools, etc
16:15 fc__ 1) I had a quick look at it, yes
16:16 fc__ 2) not so much apparently, the main thing to check is the migration path for the OSDs
16:16 fvollero_ joined #puppet-openstack
16:17 fc__ 3) it does because the security keys are autogenerated then exported then collected. stackforge/puppet-ceph has keys defined in the configuration
16:18 ericpeterson joined #puppet-openstack
16:18 aricg- joined #puppet-openstack
16:18 cwolferh_ joined #puppet-openstack
16:19 fc__ 3) same goes for the discovery/configuration of monitors which is based on exported resources in enovance/puppet-ceph and is pre-defined in stackforge/puppet-ceph
16:19 sbadia_ joined #puppet-openstack
16:20 kskurtveit joined #puppet-openstack
16:20 dachary joined #puppet-openstack
16:22 rlaurika joined #puppet-openstack
16:23 openstackgerrit Richard Megginson proposed stackforge/puppet-keystone: use openstackclient in keystone_user  https://review.openstack.org/139103
16:23 richm EmilienM: ^^^
16:23 EmilienM richm: ack
16:24 EmilienM fc__: thanks a lot; it helps.
16:24 EmilienM fc__: dachary is working on this topic FYI.
16:27 aricg joined #puppet-openstack
16:56 ericpete_ joined #puppet-openstack
16:57 dachary EmilienM: which topic ? (hi ;-)
16:57 EmilienM crinkle: so openstackclient with output in json is going to be release in a long time. So the plan B could be to keep using Aviator
16:58 EmilienM richm: ^
16:58 crinkle well we can ask them when it will be released
16:58 crinkle but then we wait for uca/rdo to keep up
16:58 crinkle so :/
16:59 EmilienM dachary: stackforge/puppet-ceph
16:59 openstackgerrit Bradley Klein proposed stackforge/puppet-monasca: Fix password for read-only user  https://review.openstack.org/139114
16:59 EmilienM crinkle: It's a huge blocker, we need to have Keystone V3 API support
16:59 crinkle okay
16:59 EmilienM I would suggest to go ahead and not waiting for a release
16:59 crinkle richm: do you feel the same way? ^
16:59 dachary EmilienM: it's a part time hobby ;-) dmsimard and off_rhoden are more involved than I am.
17:00 EmilienM dachary: I do running for hobby
17:00 aimon joined #puppet-openstack
17:00 dachary running what ? CI ?
17:00 dachary :-P
17:01 dmsimard hmm ?
17:02 rcallawa joined #puppet-openstack
17:03 rcallawa joined #puppet-openstack
17:04 badiane_ka joined #puppet-openstack
17:06 dachary dmsimard: hi ! How is the puppet-ceph CI doing ?
17:07 dmsimard dgurtner fixed the CI at 100% with the latest commits AFAIK. I was trying to find ways to speed up the CI runs but got sidetracked by work.
17:07 dgurtner joined #puppet-openstack
17:08 dmsimard gerritexec's not running but I could fire it up. I'm just not satisfied with the time it takes to do the CI runs, given the hardware it's running on.
17:09 richm EmilienM: crinkle: Do we require json output?  Is that the only reason we're waiting for python-openstackclient 1.0?
17:10 crinkle or 0.4.2
17:10 crinkle yes right now it's parsing JSON
17:12 richm right - but - can we use whatever output it is producing now?  Which admittedly is harder to parse than json
17:12 crinkle we could
17:12 crinkle but then we're kind of losing one of the reasons we wanted to move to it
17:12 richm ok
17:13 richm then - I'm fine with continuing to use aviator for now
17:13 crinkle okay
17:13 crinkle like I said in my email, once the initial work is done then changing our minds isn't that hard
17:13 richm the implementation (aviator, openstackclient) seems to be well hidden inside of puppet-openstacklib
17:13 crinkle yeah
17:13 richm except for a few files/classes named "openstack" which shouldn't cause any problems
17:14 richm I think we should leave them named "openstack"
17:14 richm unless that casues problems
17:14 richm hmm - well things are already named aviator, so that's ok
17:15 richm so never mind
17:15 crinkle i think it's conventional to name them after the thing that the provider is wrapping, but it doesn't cause any real problems
17:15 richm agreed
17:15 richm ok - so what's the next step
17:15 richm crinkle: are you going to revise your puppet-keystone keystone_tenant commit to use aviator?
17:15 mdorman joined #puppet-openstack
17:15 crinkle richm: I had an original one that i'll update
17:15 crinkle sec
17:16 crinkle https://review.openstack.org/#/c/116754/
17:16 crinkle i made improvements in the openstackclient version that i'll add there
17:16 richm crinkle: ok
17:16 crinkle will also update openstacklib with those improvements
17:16 richm ok
17:16 crinkle and then do the rest of the providers
17:17 crinkle I don't think I will get to it today, I'm trying to focus on the rabbitmq module
17:17 crinkle but now that we've figured this out I'll get to it in the next couple of days
17:20 richm ok
17:33 richm crinkle: can I rebase 116754 on top of the latest master branch?  the spec tests are now failing due to the apache::vhost thing
17:34 crinkle sure
17:34 arnaud_orange left #puppet-openstack
17:36 crinkle mgagne: your rabbitmq_policy is the best. It just needs docs and I had a minor question.
17:36 mgagne crinkle: did you add comment to the PR? I'll address them today
17:36 crinkle mgagne: yep
17:36 crinkle thank you
17:38 openstackgerrit Richard Megginson proposed stackforge/puppet-keystone: Use aviator as a REST interface for keystone_tenant provider  https://review.openstack.org/116754
17:39 mgagne crinkle: I often saw people saying you can combine multiple parameters to create the namevar. How can I do it?
17:40 crinkle mgagne: I don't know, I've been told it is hard and gross
17:40 crinkle I think the name@vhost name is fine
17:40 mgagne crinkle: I will implement the first proposed solution which consists of splitting and taking the last item for the vhost
17:41 mgagne crinkle: I was thinking about adding the ability to override the policy name with name and the vhost with vhost (yha) instead of relying on parsing the resource title/name
17:41 crinkle that might work too
17:42 mgagne crinkle: might do it in v2 then =)
17:42 crinkle :)
17:43 mgagne crinkle: my use case is about the same that we have with mysql_access, I programatically generate policies based on a list of strings. Only way I found to do it with old parser is with prefix()
17:47 JayJ joined #puppet-openstack
17:49 todin joined #puppet-openstack
17:51 mgagne crinkle: you can have @ in both policy names and virtual hosts
17:52 crinkle mgagne: oh :(
17:52 crinkle hrm
17:52 mgagne crinkle: will address policy names with @ first
17:54 rcallawa joined #puppet-openstack
17:56 thumpba joined #puppet-openstack
18:01 rcallawa joined #puppet-openstack
18:03 mschmitt joined #puppet-openstack
18:07 openstackgerrit Merged stackforge/puppet-monasca: Fix password for read-only user  https://review.openstack.org/139114
18:20 britthou_ joined #puppet-openstack
18:21 crinkle btw if you're working on puppet stuff today, we're having a hack day in #puppethack
18:26 ericpeterson joined #puppet-openstack
18:27 ericpete_ joined #puppet-openstack
18:28 mgagne crinkle: ok, might not have time for it. but I addressed all comments in PR
18:29 EmilienM mgagne: you using mongodb+sharding?
18:29 EmilienM puppetlabs modules does not support it, I'm about doing it
18:29 mgagne EmilienM: no sharding yet, just replicaset
18:29 crinkle mgagne: tyty
18:29 mgagne crinkle: yw
18:31 xarses joined #puppet-openstack
18:32 ericpeterson joined #puppet-openstack
18:50 openstackgerrit joined #puppet-openstack
19:15 ericpeterson joined #puppet-openstack
19:20 prad joined #puppet-openstack
19:41 openstackgerrit Ryan Hallisey proposed stackforge/puppet-ironic: Fix Ironic modules so services properly run in Juno  https://review.openstack.org/139166
19:44 ericpeterson joined #puppet-openstack
19:46 rcallawa joined #puppet-openstack
19:51 openstackgerrit Ryan Hallisey proposed stackforge/puppet-ironic: Fix Ironic modules so services properly run in Juno  https://review.openstack.org/134048
19:59 tfz joined #puppet-openstack
19:59 openstackgerrit Scott Pham proposed stackforge/puppet-openstacklib: Adding augeas insertion check. Augeas change call will fail if key does not exists in policy.json. This fix will add the key/value if it doesn't exists or will change the value to match. Added unit test. Change-Id: Ibd97228c431fa6b23e8a15ce548f82f649131cc  https://review.openstack.org/138915
20:00 badiane_ka joined #puppet-openstack
20:04 openstackgerrit joined #puppet-openstack
20:05 csharpsteen left #puppet-openstack
20:18 rcallawa_ joined #puppet-openstack
20:21 openstackgerrit Ryan Bak proposed stackforge/puppet-monasca: Removed dashboard class  https://review.openstack.org/139176
20:26 rcallawa joined #puppet-openstack
21:11 tfz joined #puppet-openstack
21:33 dgurtner joined #puppet-openstack
21:37 bitblt joined #puppet-openstack
21:46 prad joined #puppet-openstack
21:51 ddmitriev1 joined #puppet-openstack
22:13 openstackgerrit Merged stackforge/puppet-monasca: Removed dashboard class  https://review.openstack.org/139176
22:29 richm I've got a poser involving hiera
22:29 richm https://gist.github.com/richm/21474ce1671336eec931
22:29 richm Can someone explain why hiera('does not exist', undef) does _not_ return undef?????
22:34 crinkle does it return '' ?
22:35 xavpaice richm - hiera('does not exist') will usually return ''
22:35 xavpaice hiera_undef is what you want
22:35 xavpaice https://forge.puppetlabs.c​om/camptocamp/hiera_undef
22:36 crinkle there's a bug about this https://projects.puppetlabs.com/issues/20923
22:37 richm It returns ''
22:37 crinkle yeah that's a won't-fix bug
22:37 crinkle at least not till puppet 4 i think
22:50 richm xavpaice: ok - so what dependency do I add to my Gemfile? to use hiera_undef?
22:58 xavpaice not sure about that - hiera_undef is just a forge module
22:58 richm ok - so not available except via git
23:00 xavpaice well, it's not a gem at least
23:00 * xavpaice admits to not reading the gist before jumping in on this convo
23:01 richm xavpaice: well you did answer my question and provide a solution
23:01 xavpaice I guess you could add the hiera_undef as a module dep
23:03 openstackgerrit Ryan Hefner proposed stackforge/puppet-cinder: Automates generation of NFS config file  https://review.openstack.org/139254
23:09 openstackgerrit Ryan Hefner proposed stackforge/puppet-manila: Currently, the manifest uses an array to store share helpers but doesn't use the value right.  Transitioning this value to be a comma separated string so that it works correctly.  https://review.openstack.org/137536
23:09 mschmitt joined #puppet-openstack
23:14 openstackgerrit Ryan Hefner proposed stackforge/puppet-cinder: Automates generation of NFS config file  https://review.openstack.org/139254
23:35 gildub joined #puppet-openstack
23:50 rcallawa joined #puppet-openstack
23:52 dgurtner joined #puppet-openstack

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