Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2015-11-17

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

All times shown according to UTC.

Time Nick Message
00:00 bklei joined #puppet-openstack
00:01 bklei_ joined #puppet-openstack
00:13 sergmelikyan joined #puppet-openstack
00:14 aimon joined #puppet-openstack
00:20 reynir joined #puppet-openstack
00:35 bogdando joined #puppet-openstack
00:36 aimon joined #puppet-openstack
00:42 jerrygb joined #puppet-openstack
00:47 bklei joined #puppet-openstack
01:10 cwolferh joined #puppet-openstack
01:30 xiaohua joined #puppet-openstack
01:37 markvoelker joined #puppet-openstack
01:38 aimon joined #puppet-openstack
01:40 aimon_ joined #puppet-openstack
01:43 bogdando joined #puppet-openstack
02:04 bogdando joined #puppet-openstack
02:29 xiaohua joined #puppet-openstack
02:35 bklei joined #puppet-openstack
02:57 fedexo joined #puppet-openstack
02:57 xarses joined #puppet-openstack
03:00 gildub_ joined #puppet-openstack
03:18 fvollero joined #puppet-openstack
03:18 mkarpin joined #puppet-openstack
03:20 xiaohua joined #puppet-openstack
03:24 timrc joined #puppet-openstack
03:27 ducttape_ joined #puppet-openstack
03:32 morazi joined #puppet-openstack
03:54 xiaohua joined #puppet-openstack
03:57 fedexo joined #puppet-openstack
04:10 ducttape_ joined #puppet-openstack
04:10 _ody joined #puppet-openstack
04:24 bklei joined #puppet-openstack
04:31 openstackgerrit Gilles Dubreuil proposed openstack/puppet-keystone: Remove occurrence of @@default_domain_id.  https://review.openstack.org/241284
04:35 zhangjn joined #puppet-openstack
04:54 rahulait joined #puppet-openstack
04:56 imcsk8_ joined #puppet-openstack
04:57 joshsamuelson joined #puppet-openstack
04:57 __nick joined #puppet-openstack
04:59 xiaohua joined #puppet-openstack
05:00 mgagne joined #puppet-openstack
05:01 jerrygb joined #puppet-openstack
05:01 kbyrne joined #puppet-openstack
05:02 mwhahaha joined #puppet-openstack
05:02 bkero joined #puppet-openstack
05:03 darkxploit joined #puppet-openstack
05:05 sbadia joined #puppet-openstack
05:19 bnemec joined #puppet-openstack
05:23 xiaohua joined #puppet-openstack
05:25 spredzy_ joined #puppet-openstack
05:28 aglarendil joined #puppet-openstack
05:38 holser joined #puppet-openstack
05:39 bpiotrowski joined #puppet-openstack
05:49 openstackgerrit Gilles Dubreuil proposed openstack/puppet-keystone: Add keystone domain specific configuration.  https://review.openstack.org/219289
05:57 sanjayu joined #puppet-openstack
06:08 xiaohua joined #puppet-openstack
06:12 bklei joined #puppet-openstack
06:17 xiaohua joined #puppet-openstack
06:22 rcernin joined #puppet-openstack
06:39 jerrygb joined #puppet-openstack
06:53 angdraug joined #puppet-openstack
07:12 openstackgerrit Nikolay Starodubtsev proposed openstack/puppet-murano: Add ability to deploy Murano CloudFoundry service broker API  https://review.openstack.org/245043
07:29 nlahouti joined #puppet-openstack
08:01 bklei joined #puppet-openstack
08:03 dgurtner joined #puppet-openstack
08:03 dgurtner joined #puppet-openstack
08:32 xiaohua joined #puppet-openstack
08:34 bkero joined #puppet-openstack
08:35 dgurtner joined #puppet-openstack
08:35 dgurtner joined #puppet-openstack
08:35 gfidente joined #puppet-openstack
08:35 gfidente joined #puppet-openstack
08:36 jpena joined #puppet-openstack
08:40 jerrygb joined #puppet-openstack
09:32 openstackgerrit joined #puppet-openstack
09:41 sleinen-AS559 joined #puppet-openstack
09:42 karimb joined #puppet-openstack
09:49 bklei joined #puppet-openstack
10:28 ddmitriev joined #puppet-openstack
10:30 xiaohua joined #puppet-openstack
10:32 openstackgerrit joined #puppet-openstack
10:32 xiaohua joined #puppet-openstack
10:43 jerrygb joined #puppet-openstack
10:46 jtomasek joined #puppet-openstack
10:54 xiaohua joined #puppet-openstack
10:54 marfx000 joined #puppet-openstack
10:56 marfx000_ joined #puppet-openstack
10:59 marfx000 joined #puppet-openstack
11:06 markvoelker joined #puppet-openstack
11:12 zhangjn joined #puppet-openstack
11:13 zhangjn joined #puppet-openstack
11:14 iurygregory good morning people =D
11:14 zhangjn joined #puppet-openstack
11:15 iurygregory Hey just FYI - the link for the agenda for today meeting is https://etherpad.openstack.org/p/pupp​et-openstack-weekly-meeting-20151117 and not https://etherpad.openstack.org/p/pup​pet-openstack-weekly-meeting-2015111
11:15 zhangjn joined #puppet-openstack
11:16 gildub_ joined #puppet-openstack
11:16 zhangjn joined #puppet-openstack
11:17 zhangjn joined #puppet-openstack
11:18 zhangjn joined #puppet-openstack
11:20 zhangjn joined #puppet-openstack
11:21 zhangjn joined #puppet-openstack
11:23 rmart04 joined #puppet-openstack
11:40 toMeloos joined #puppet-openstack
11:42 xiaohua joined #puppet-openstack
11:49 xiaohua joined #puppet-openstack
11:51 gfidente joined #puppet-openstack
12:15 xiaohua joined #puppet-openstack
12:17 EmilienM good morning folks
12:18 xiaohua good morning
12:18 mkarpin good morning!
12:26 xiaohua joined #puppet-openstack
12:31 jerrygb joined #puppet-openstack
12:37 markvoelker joined #puppet-openstack
12:41 sleinen-AS5591 joined #puppet-openstack
12:42 degorenko hey :) welcome back EmilienM
12:42 EmilienM thanks!
12:46 iurygregory mornig EmilienM  o/
12:59 dprince joined #puppet-openstack
13:08 openstackgerrit Clayton O'Neill proposed openstack/puppet-openstack-specs: Enhance inifile option deprecations spec  https://review.openstack.org/240515
13:14 ducttape_ joined #puppet-openstack
13:21 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Make libvirt migration security configurable  https://review.openstack.org/246384
13:24 EmilienM spredzy: have you looked https://bugs.launchpad.net/​puppet-tuskar/+bug/1515273 ?
13:24 openstack Launchpad bug 1515273 in puppet-tuskar "Defaults for *::db classes are incorrect" [Undecided,New]
13:26 morazi joined #puppet-openstack
13:27 bklei joined #puppet-openstack
13:27 markvoelker joined #puppet-openstack
13:46 ken_barber joined #puppet-openstack
14:03 jamespage joined #puppet-openstack
14:05 richm joined #puppet-openstack
14:08 spredzy EmilienM, I did, I answered on the ML http://lists.openstack.org/pipermail/o​penstack-dev/2015-November/079387.html
14:09 spredzy Haven't done the code yet though
14:09 EmilienM spredzy: oh, I did not read all emails yet :) thanks for pointing that out
14:09 mdorman joined #puppet-openstack
14:09 openstackgerrit Jaume Devesa proposed openstack/puppet-tripleo: MidoNet services manifests  https://review.openstack.org/244184
14:11 bklei joined #puppet-openstack
14:11 xiaohua joined #puppet-openstack
14:12 bklei joined #puppet-openstack
14:16 bklei joined #puppet-openstack
14:22 karimb joined #puppet-openstack
14:23 mkarpin EmilienM  spredzy crincle sbadia guys can you please take a look (when you have time) at  https://review.openstack.org/#/c/241614/, thank you!
14:27 karimb_ joined #puppet-openstack
14:33 sergmelikyan joined #puppet-openstack
14:34 sergmelikyan joined #puppet-openstack
14:41 sergmelikyan joined #puppet-openstack
14:42 jtomasek joined #puppet-openstack
14:54 iurygregory Meeting will start in 5 minutes in #openstack-meeting-4 =)
14:58 richm joined #puppet-openstack
15:02 mlrn joined #puppet-openstack
15:06 dfisher joined #puppet-openstack
15:08 ducttape_ joined #puppet-openstack
15:09 ducttape_ joined #puppet-openstack
15:09 xarses joined #puppet-openstack
15:18 mfisch joined #puppet-openstack
15:18 samuelBartel joined #puppet-openstack
15:30 nlahouti joined #puppet-openstack
15:37 nlahouti joined #puppet-openstack
15:40 nrandall joined #puppet-openstack
15:47 jamespage joined #puppet-openstack
15:55 darkxploit joined #puppet-openstack
16:00 ntpttr joined #puppet-openstack
16:03 bnemec joined #puppet-openstack
16:05 openstackgerrit Merged openstack/puppet-keystone: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/242134
16:06 fedexo joined #puppet-openstack
16:14 sleinen-AS559 joined #puppet-openstack
16:14 openstackgerrit Alexey Deryugin proposed openstack/puppet-murano: Application resource implementation  https://review.openstack.org/220238
16:18 openstackgerrit Alexey Deryugin proposed openstack/puppet-murano: Add acceptance tests for client and application provider  https://review.openstack.org/246480
16:24 openstackgerrit Merged openstack/puppet-neutron: Add workaround for v2 tenant names  https://review.openstack.org/244900
16:28 _ducttape_ joined #puppet-openstack
16:36 cwolferh joined #puppet-openstack
16:39 EmilienM mwhahaha: https://github.com/openstack/puppet-keys​tone/blob/master/manifests/init.pp#L481
16:39 mwhahaha so i think that's always been there right?
16:39 openstackgerrit Mykyta Karpin proposed openstack/puppet-nova: Use OpenstackClient for nova providers auth  https://review.openstack.org/226862
16:40 mwhahaha I thought there was also an email about this where it only affected the people deploying via source and not packages as the packages provide a sane default
16:40 mwhahaha also that param is not actually used
16:41 ducttape_ joined #puppet-openstack
16:41 mwhahaha https://github.com/openstack/puppet-keyst​one/blob/master/manifests/logging.pp#L126
16:41 mwhahaha so we want the init.pp log_dir to be undef and the logging.pp log_dir to be specified
16:41 mwhahaha to allow for backwards compatibility
16:45 EmilienM mwhahaha: sounds like something is not consistent here
16:45 mwhahaha that's part of the original logging change
16:45 mwhahaha which is different then the os_service_default
16:48 openstackgerrit Denis Egorenko proposed openstack/puppet-sahara: Switch Sahara to $::os_service_default  https://review.openstack.org/245178
16:48 openstackgerrit Merged openstack/puppet-nova: Use os_package_type for libvirt service name  https://review.openstack.org/243912
16:49 openstackgerrit Athlan-Guyot sofer proposed openstack/puppet-keystone: Keystone_endpoint match service by name/type.  https://review.openstack.org/239434
16:50 sergmelikyan joined #puppet-openstack
16:55 marfx000 joined #puppet-openstack
16:56 ntpttr EmilienM: Hi, when you have a moment would you mind checking out my initial commit of puppet-magnum :)? https://review.openstack.org/#/c/242164/
16:57 openstackgerrit Athlan-Guyot sofer proposed openstack/puppet-keystone: Resource keystone_identity_provider for Keystone.  https://review.openstack.org/202689
16:59 ntpttr left #puppet-openstack
17:03 nlahouti joined #puppet-openstack
17:04 nlahouti joined #puppet-openstack
17:13 EmilienM clayton: can I assign https://bugs.launchpad.net/​puppet-tuskar/+bug/1515273 + others to you ?
17:13 openstack Launchpad bug 1515273 in puppet-tuskar "Defaults for *::db classes are incorrect" [Undecided,New]
17:14 chem mwhahaha: the patch has been merged, so I won't look into it.  But I think I know why you had this behaviour.  If the default domain is not defined in the catalog then it's not removed from the name.  As domain Default is by default, not defined in any puppet catalog, then "Default" is appended.
17:14 nlahouti joined #puppet-openstack
17:15 ntpttr joined #puppet-openstack
17:15 mwhahaha yes
17:15 EmilienM chem: is it a problem to have this merged?
17:16 chem EmilienM: no
17:16 mwhahaha the problem (and their may be others) is that this code if executed on a different node than where the tenant is created it runs into that case. The patch was to at least temporarily continue to support the v2 keystone_tenant on neutron resources until we completely ditch the domain-less tenant names
17:17 mwhahaha so yes, you're right but it was an issue created by our naming convention change and since it was in the neutron providers it was not found during the keystone tests
17:17 chem mwhahaha: I agree :)
17:18 chem mwhahaha: but the /problem/ is not that it's not executed on a different server, it's that there is not default domain definition in the catalog
17:18 mwhahaha both are true
17:18 chem mwhahaha: to sumarize, adding "keystone_domain {'Default', is_default => true}"
17:19 mwhahaha because if i had keystone_tenant in my catalog it would have never done the lookup
17:19 chem to the catalog would suffice
17:19 mwhahaha how does that impact the keystone_tenant provider?
17:19 chem mwhahaha: no, because the naming would be adjusted as Default would be recognized as default domain and then be removed from the naming
17:20 mwhahaha where?
17:20 mwhahaha (in code)
17:20 chem I show you 2s
17:20 mwhahaha sure
17:23 chem mwhahaha: it's a bit involving :)
17:24 chem mwhahaha: 1/ https://github.com/openstack/puppet-k​eystone/blob/master/lib/puppet/provid​er/keystone_tenant/openstack.rb#L73
17:24 chem mwhahaha: 2/ https://github.com/openstack/puppet-keystone/blo​b/master/lib/puppet/provider/keystone.rb#L50-L59
17:24 chem mwhahaha:  3/ https://github.com/openstack/puppet-k​eystone/blob/master/lib/puppet_x/keys​tone/type/default_domain.rb#L52-L59
17:24 mwhahaha ha yea that is totally not ovious
17:24 mwhahaha obvious
17:25 mwhahaha we might want to put a note in the keystone readme about the keystone_domain work around
17:25 chem the naming is checking default_domain which is set by the keystone_tenant/keystone_user/keystone_user_role *type* when they need it (at catalog compilation time)
17:25 chem in the @@default_domain
17:25 chem then in 3/ you have the catalog searching for default_domain
17:26 chem mwhahaha: I didn't though of this problem :)
17:26 chem before
17:27 EmilienM degorenko, clayton: one of you can take the bugs https://bugs.launchpad.net/​puppet-tuskar/+bug/1515273 ? so we can track the work in progress
17:27 openstack Launchpad bug 1515273 in puppet-tuskar "Defaults for *::db classes are incorrect" [Undecided,New]
17:27 chem mwhahaha: anyway that's why just to being coherent with people using the default_domain parameter in keystone::init manifest
17:28 mwhahaha but that doesn't solve where i'm not pulling in keystone for my other resources on other catalogs
17:28 mwhahaha right?
17:28 mwhahaha so i hit this in neutron and if that's not on the same host as my keystone::init, i need to specify keystone_domain in the neutron catalog
17:28 chem mwhahaha: if you're using different catalag, that doesn't work, you're right
17:29 chem mwhahaha: ^^ that's correct
17:29 mwhahaha yea so we need to document that :D
17:29 degorenko EmilienM, as i understand - just need to move under service_default, right?
17:29 dfisher got a dependency question for you all.  why is the puppetlabs-rabbitmq dependency pinned at such an old version?
17:30 chem mwhahaha: there is this https://etherpad.openstack​.org/p/keystone_no_domain
17:30 dfisher or, it was for juno .. nevermind
17:30 EmilienM degorenko: yes, I think. Could be done in a separated patch
17:31 chem and this  https://review.openstack.org/#/c/219127/ that need some love
17:31 degorenko EmilienM, ok, then i will fix it
17:31 degorenko for sahara too, of course :D
17:31 degorenko btw, for sahara it is on reivew
17:31 EmilienM dfisher: we're using 5.2.3 and the latest is 5.3.1
17:31 mwhahaha for me etherpad != documentation (nor specs for that matter). something like this needs to be visible in the puppet module read me or something of the source
17:31 chem I think putting this information in the blueprint would make sense
17:32 chem mwhahaha: I agree
17:32 EmilienM dfisher: please submit a patch against puppet-openstack-integration
17:32 dfisher right.  Juno capped it at < 4.0.0
17:32 mwhahaha because if i'm an end user i just see the readme and thats it
17:32 dfisher nono.  kilo is capped at < 6.0.0
17:32 EmilienM ah, kilo
17:32 dfisher yeah, we're fine.
17:32 chem mwhahaha: ha, so the spec is no good, neither
17:32 chem ?
17:32 mwhahaha no
17:32 mwhahaha no operator is going to go looking in the specs for a given release
17:33 mwhahaha if they are they have more time on their hands then they should
17:33 mwhahaha something like that is working with the module and needs to be packaged with the module
17:33 mwhahaha i would defer to others if they disagree, but when I'm consuming a puppet module i'm looking at the read me for important usage information
17:35 chem mwhahaha: oki
17:36 rcernin joined #puppet-openstack
17:39 sergmelikyan joined #puppet-openstack
17:43 degorenko EmilienM, for glance 237970 - i guess i can use https://review.openstack.org/237973 as dependency and just update my patch with os_service_default
17:46 chem mwhahaha: reading the code again, the problem shouldn't have appeared as 'Default' is the fallback for default domain
17:46 openstackgerrit Ivan Berezovskiy proposed openstack/puppet-cinder: Support of PyMySQL driver for MySQL backend  https://review.openstack.org/241214
17:46 chem mwhahaha: and, as such, should have been removed from the name
17:46 mwhahaha hmm ok
17:47 chem mwhahaha: if you got to see 'blah::Default' in the listing of the resource, then, it's a bug in the keystone module
17:47 mwhahaha yea that's what i got
17:47 IvanBerezovskiy left #puppet-openstack
17:48 EmilienM degorenko: ok
17:48 chem mwhahaha: Oki. I check.
17:57 markvoelker_ joined #puppet-openstack
17:57 serg_melikyan joined #puppet-openstack
18:00 bnemec joined #puppet-openstack
18:02 openstackgerrit Alexey Deryugin proposed openstack/puppet-murano: Add acceptance tests for client and application provider  https://review.openstack.org/246480
18:05 timrc joined #puppet-openstack
18:06 sergmelikyan joined #puppet-openstack
18:06 sbadia joined #puppet-openstack
18:09 richm joined #puppet-openstack
18:10 sanjayu joined #puppet-openstack
18:11 openstackgerrit Mykyta Karpin proposed openstack/puppet-glance:  Add support of os_service_default fact in glance spec  https://review.openstack.org/237973
18:20 chem mwhahaha: another question about your running environment.  Where you use the neutron module, you don't create any tenant in the catalog ?
18:20 chem mwhahaha: is that correct ?
18:20 mwhahaha correct
18:21 mwhahaha fuel uses task based deployments so we only pull in the code required to do what we are working on
18:21 mwhahaha so the tenants are setup elsewhere
18:21 mwhahaha and neutron comes along to configure it's vips w/o the tenant/user definitions
18:21 mwhahaha the equivalent would be setting up keystone on server A and neutron on server B
18:22 mwhahaha or at least a similar condition
18:22 mwhahaha specifically the item i was running into was for the neutron_subnets resource
18:22 chem mwhahaha: so no tenant/user/user_role creation in the catalog.  Oki, I pinpoint the bug, then.
18:22 mwhahaha correct
18:23 openstackgerrit Denis Egorenko proposed openstack/puppet-glance: Introduce glance::db class  https://review.openstack.org/237970
18:23 chem mwhahaha: it's only when one of those resource is present that the default domain is set
18:23 mwhahaha yea that's what i figured which is where documenting the keystone_domain { 'Default': is_default => true } would be imporant
18:23 mwhahaha important
18:24 chem mwhahaha: humm ... I think that the keystone module should be fixed for this case as well.
18:25 mwhahaha probably
18:25 chem mwhahaha: once again the keystone_domain {is_default} is just a hack, you shouldn't need it to make it work
18:26 openstackgerrit Denis Egorenko proposed openstack/puppet-tuskar: Switch Tuskar to $::os_service_default  https://review.openstack.org/246541
18:33 nlahouti joined #puppet-openstack
18:41 aimon joined #puppet-openstack
18:48 aglarendil joined #puppet-openstack
19:02 ntpttr joined #puppet-openstack
19:05 morazi joined #puppet-openstack
19:10 holser joined #puppet-openstack
19:17 bpiotrowski joined #puppet-openstack
19:20 gildub_ joined #puppet-openstack
19:21 _ody I noticed things referencing $::os_service_default when I read today's meetings notes; are we now wanting to block commits using the old "<SERVICE DEFAULT>" pattern now?
19:36 chem richm: EmilienM mwhahaha gildub_ : I have created this bug report for capturing the problem they had with the neutron module: https://bugs.launchpad.net/p​uppet-keystone/+bug/1517187
19:36 openstack Launchpad bug 1517187 in puppet-keystone "Default domain not calculated when no keystone_user/keystone_tenant/keystone_user_role is defined." [Undecided,New]
19:37 EmilienM chem: interesting
19:37 EmilienM I'm happy we see that bug before the release
19:40 EmilienM chem: have you seen https://bugs.launchpad.net​/puppet-heat/+bug/1516299 ?
19:40 openstack Launchpad bug 1516299 in puppet-heat "heat::keystone::auth configuration failing" [Undecided,New]
19:41 EmilienM _ody, Hunner: big highlight on https://review.openstack.org/#/c/245874/
19:43 chem EmilienM: richm gildub_ do we agree on this  "When not set, Keystone default domain id is 'default'"
19:43 chem EmilienM: richm gildub_ from https://review.openstack.org/#/c/2​19127/6/specs/kilo/api-v3-support.rst
19:43 EmilienM chem: yes or 'Default', no?
19:44 chem EmilienM: ?
19:44 EmilienM default = Default ?
19:44 EmilienM or we don't care?
19:44 iurygregory no!
19:44 iurygregory it's different
19:44 _ody sbadia has a review open about that issue he was looking for feeback on.
19:44 iurygregory Default is name
19:44 iurygregory default is id
19:44 iurygregory for the default domain =)
19:45 chem EmilienM: iurygregory that's correct, and that's what is said in the review
19:45 EmilienM great
19:45 chem domain_id = default means domain_name 'Default'
19:45 iurygregory yep ^^
19:46 chem EmilienM: richm gildub_ because the corner raised by mwhahaha cannot be fixed by any other way
19:46 chem EmilienM: richm gildub_ the catalog cannot be searched at self.instances time
19:47 iurygregory i think is valid consider when not set domain_id is default
19:48 angdraug joined #puppet-openstack
19:48 _ody EmilienM: Looking into a good recommendation on the version string matching.  The referenced change might only be valid for that 3.8.x release of PE.  PE versioning changes after that and heavily relies on the version numbers assocaited with the all-in-one agent.
19:48 chem iurygregory: EmilienM richm gildub_ when I say "not set", I mean that the keystone_user/keystone_tenant/keystone_user_role has no ::domain part
19:49 chem iurygregory: EmilienM richm gildub_ in this case, then we consider that it's 'Default'
19:50 chem iurygregory: EmilienM richm gildub_ so, is that ok to implement the behavior in this review ?
19:51 chem iurygregory: EmilienM richm gildub_ relative to what we have now that means: no catalog searching at all
19:51 chem iurygregory: EmilienM richm gildub_ relative to what we have now that means: and simply using "Default"
19:51 iurygregory can you point the review?
19:52 iurygregory i think this approach is valid
19:52 chem iurygregory: https://review.openstack.org/#/c/2​19127/6/specs/kilo/api-v3-support.rst
19:53 chem EmilienM: richm gildub_ iurygregory I have leave now.  Please comment in the bugzilla or somewhere I can find it tomorrow.
19:53 iurygregory tonigh i'll take a look
19:53 EmilienM chem: will look
19:54 chem EmilienM: I think the puppet-heat bug is related, have to look.  Tomorrow..
19:54 EmilienM chem: thanks and good night!
19:54 chem EmilienM: the good news is that "Default" will certainly fix all those bug
19:55 iurygregory nice to know chem, good night o/
19:55 chem thanks, bye
19:56 EmilienM chem: thanks a lot
20:03 cgfbee joined #puppet-openstack
20:11 chem EmilienM: iurygregory richm gildub_ : I have sum up my though on the matter there https://review.openstack.org/#/c/219127/ (the new comments, not the old one)
20:12 karimb joined #puppet-openstack
20:13 iurygregory ok chem =)
20:13 cwolferh joined #puppet-openstack
20:50 rmart04 joined #puppet-openstack
20:52 rmart04 left #puppet-openstack
20:53 rahulait joined #puppet-openstack
20:58 mfisch anyone had issues with the nova providers not handling regions?
20:58 mfisch once I put host aggregates into multinode I see an issue, its only hitting our east endpoint
21:06 zigo EmilienM: Hi man!
21:06 zigo How was your holidays?
21:07 zigo Guys, I really need this one to be merged: https://review.openstack.org/#/c/241218/
21:08 zigo It has a huge impact on so many things for me.
21:09 zigo mwhahaha: ^
21:10 mwhahaha oh the pesky collectstatic
21:11 mwhahaha you know i'm not sure that works correctly for ubuntu either
21:12 zigo mwhahaha: This one is a distraction, what I need is all the rest.
21:12 zigo Ie, correct .conf name, etc.
21:12 mwhahaha yea
21:13 mwhahaha so did you ever see my response about the openstack-dashboard pulling in openstack-dashboard-apache?
21:13 zigo mwhahaha: It's IMO a package's decision to run collectstatic, and puppet should never deal with it.
21:13 mwhahaha because it's a recommended package or whatever
21:13 zigo I'm not sure I did.
21:13 mwhahaha so when i apt-get install openstack-dashboard it automatically is install openstack-dashboard-apache
21:14 zigo Well, you don't have to run with recommended turned on...
21:14 dfisher zigo: i agree with you (about collectstatic).  for solaris, we collectstatic/compress before packaging.
21:14 zigo apt-get install --no-install-recommends also would work.
21:14 mwhahaha true, it's just the default
21:14 zigo I run *everything* with recommended turned off.
21:15 mwhahaha we should probably update fuel to set that
21:15 zigo If this is a blocker, I can turn the recommended into a suggest, I don't mind much.
21:16 zigo Though really, if this patch is merged, everything is solved either ways, because we get the correct configuration file modified by puppet.
21:16 mwhahaha but anyway so openstack-dashboard-apache conflicts with the horizon::wsgi::apache class
21:16 mwhahaha i think suggests also get automagically installed as well
21:16 zigo Because of the double-defined WSGI definition?
21:16 mwhahaha yea
21:16 zigo Well, that patch fixes that !!!
21:16 zigo That's the whole point of it.
21:16 mwhahaha nope
21:16 mwhahaha it doesn't it makes it worse :D
21:16 zigo -            $httpd_config_file     = '/etc/apache2/sites-availabl​e/openstack-dashboard.conf'
21:16 zigo +            $httpd_config_file     = '/etc/apache2/sites-available/ope​nstack-dashboard-alias-only.conf'
21:17 zigo So, where's that thing, so that I get it to learn what file to use? :)
21:17 zigo (and add it to this patch...)
21:17 mwhahaha so down the puppet rabbit hole we go
21:17 mwhahaha https://github.com/openstack/puppet-hori​zon/blob/master/manifests/wsgi/apache.pp
21:17 zigo Thanks.
21:17 mwhahaha specifically, https://github.com/openstack/puppet-horizon​/blob/master/manifests/wsgi/apache.pp#L106
21:18 mwhahaha er wait
21:18 mwhahaha you might be right
21:18 zigo LOL ! :)
21:18 mwhahaha it's totally different in fuel so maybe we're doing it elsewhere
21:18 mwhahaha need to try out a clean version of the module and package at some point
21:19 mwhahaha nope here it is
21:19 mwhahaha https://github.com/openstack/puppet-horizo​n/blob/master/manifests/wsgi/apache.pp#L85
21:19 zigo clean package != the one from Ubuntu ... :P
21:19 zigo ?!?
21:19 zigo There's no vhost involved here ...
21:20 zigo It's only an Alias.
21:20 mwhahaha right but that's where the wsgi stuff gets duped
21:20 sergmelikyan joined #puppet-openstack
21:20 mwhahaha so i guess you're right, by fixing the http_config_file out puppet should clear that out
21:20 mwhahaha https://github.com/openstack/puppet-horizon/b​lob/master/manifests/wsgi/apache.pp#L106-L116
21:20 mwhahaha ok let me help you out with tests on that review
21:20 mwhahaha and then i think it should be good
21:20 zigo :)
21:21 zigo Cool, you're my hero, really.
21:21 zigo It was so far a nightmare to deal with, you're the only light I see at the end of the tunnel.
21:22 mfisch clayton: near as I can tell none of the nova providers can handle regions
21:51 mwhahaha zigo: how does the debian version of the package handle COMPRESS_OFFLINE in the local_settings.py?
21:52 zigo mwhahaha: It's added by the package.
21:52 mwhahaha so we set it in puppet
21:52 zigo echo "COMPRESS_OFFLINE=True" >>$(CURDIR)/debian/openstack-dashboard/usr/​share/openstack-dashboard/local_settings.py
21:52 zigo That's what I do in the debian/rules.
21:53 zigo I guess it's not so much of a trouble to have this defined twice though ...
21:53 zigo mwhahaha: Or is it?
21:53 mwhahaha well no so i was looking at the change and techincally the collectstatic is only run if compress_offline is true
21:54 zigo You mean false?
21:54 mwhahaha so normally it's true (and would be set) and your appending may be overridden
21:54 mwhahaha no it appears that if it's false it doesn't get added
21:54 zigo I'm not sure I get it...
21:54 zigo collectstatic is in fact not needed when you have COMPRESS_OFFLINE=True.
21:54 zigo But it is if it is False ...
21:55 zigo Hum...
21:55 zigo I'm mistaking with no compression at all! :)
21:55 mwhahaha https://github.com/openstack/puppet-hor​izon/blob/master/manifests/init.pp#L390
21:55 zigo Now I'm confused.
21:55 zigo What's the logic behind this?
21:55 mwhahaha so that's the line that actually triggers the collectstatic
21:55 mwhahaha no idea
21:55 mwhahaha was like this when i got here :D
21:56 mwhahaha refreshonly => true on the exec means it only runs when it's notified, and we only notify it when the config file is changed and if compress_offline is true
21:57 mwhahaha https://github.com/openstack/puppet-horizon/co​mmit/720ce5cf845d57ce04d9d602cbd58b45f3144121
21:57 mwhahaha was the original change that put this thing in there
21:59 mwhahaha my assumption as that whole collectstatic thing needs to get pulled
21:59 EmilienM zigo: hey
21:59 zigo EmilienM: \o
22:01 zigo mwhahaha: Agreed.
22:01 EmilienM zigo: the beach was awesome, I already miss it
22:01 zigo EmilienM: Where did you go?
22:01 EmilienM zigo: Saint Martin
22:01 bkero Ooo, nice part of france
22:02 zigo Nice.
22:02 EmilienM an Island in the middle of nowhere, without anything but beach and sun. Exactly what I needed before canadian winter ;-)
22:02 zigo Well, assuming he was on the French half ...
22:02 EmilienM I was
22:02 EmilienM zigo: what do you need from now, how can I help?
22:03 EmilienM I'm about boarding to my flight to Paris
22:03 zigo I need this fix to be merged: https://review.openstack.org/#/c/241218/
22:03 EmilienM looking
22:03 bkero So much for the canadian winter
22:03 mwhahaha i'm writing tests for -^
22:03 zigo Otherwise, I'm in a big shit, with the sky falling down on me, customers unhappy, and colleagues pointing fingers at me ... :D
22:03 mwhahaha also trying to fix this logic for this collectstatic crap
22:03 zigo I'm only half joking here...
22:04 mwhahaha do we know if redhat needs the collecstatic?
22:04 mwhahaha it was specifically put in to work around debian last year
22:05 EmilienM mwhahaha: I don't think so but I'm not sure
22:05 mwhahaha so i'm going to wrap the whole lot up in a only for redhat for now
22:05 mwhahaha and if we determine it's not need for redhat we can kill it later
22:05 zigo :)
22:05 zigo Sounds like a good plan.
22:06 zigo mwhahaha: BTW, in what timezone are you? Sunnyvale?
22:06 mwhahaha central until next week then mountain
22:06 dfisher the best time zone.
22:06 dfisher (mountain)
22:06 mwhahaha moving back to a proper state
22:07 mwhahaha one with seasons other than hot and hotter
22:07 dfisher where you moving to?
22:07 mwhahaha aurora, co
22:08 dfisher nice.  I'm in Broomfield :)
22:08 zigo That's Denver, right?
22:08 mwhahaha yea
22:08 * mwhahaha stalks dfisher
22:08 * dfisher makes it easier
22:08 mwhahaha i moved to austin ~2 years ago from denver
22:08 dfisher welcome home.
22:09 zigo You should have wait until next summit, so you could have been "local" ! :)
22:09 zigo (rare enough...)
22:09 mwhahaha ah i'll come back for the summit
22:09 zigo I hope so.
22:09 mwhahaha my wife is dying from the allergens here
22:09 mwhahaha can't stay
22:09 dfisher snowed here last night.   everything died.
22:10 zigo They announced snow here for next weekend...
22:10 zigo About time !!!
22:10 zigo Skiiiiiiiiiiiiiiiiiii ! :)
22:10 dfisher usually the snow isn't quite good enough for skiing until after the new year.
22:10 dfisher but, we're due to get a TON this season
22:11 zigo Over here, it's enough if you go high enough. But this year, even very high, they had none.
22:11 ntpttr skiing in Portland on mt hood has been suffering the last few years as well
22:12 bkero This year's not going to be any better either.
22:12 zigo http://www.tignes.net/en/skii​ng-in-tignes/webcams-290.html <--- Nothing, even at 2100m ...
22:12 zigo (and higher)
22:12 ntpttr :(
22:19 openstackgerrit Alex Schultz proposed openstack/puppet-horizon: Do not run collectstatic or compress in Debian  https://review.openstack.org/241218
22:19 mwhahaha zigo: -^ now with tests
22:21 zigo mwhahaha: Thanks.
22:21 zigo Let's wait for the gate ...
22:25 * mwhahaha wanders off
22:25 mwhahaha i'll check later to make sure everything went ok, but the tests passed locally
22:27 rahulait joined #puppet-openstack
22:31 openstackgerrit Matt Fischer proposed openstack/puppet-nova: Add region support to nova providers  https://review.openstack.org/246632
22:32 mfisch fix for nova
22:32 mfisch lack of region name comprehension
22:33 openstackgerrit Matt Fischer proposed openstack/puppet-nova: Add region support to nova providers  https://review.openstack.org/246632
22:35 ntpttr joined #puppet-openstack
22:37 dmsimard mfisch: probably want to replace conf['neutron'] by conf['nova']
23:05 zigo mwhahaha: I don't think we care much about it, but the current stable version for Debian is 8.2, not 6.
23:08 xavpaice joined #puppet-openstack
23:12 sergmelikyan joined #puppet-openstack
23:20 * _ody found the OpenStack packaging for squeeze unusable
23:21 nlahouti joined #puppet-openstack
23:43 dfisher1 joined #puppet-openstack
23:43 dfisher1 left #puppet-openstack

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