Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-10-13

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

All times shown according to UTC.

Time Nick Message
01:15 britthouser joined #puppet-openstack
01:58 xingchao joined #puppet-openstack
03:07 gildub joined #puppet-openstack
03:29 snk joined #puppet-openstack
03:33 hogepodge benh57 RHEL 6 was already falling behing on a number of packages.
03:50 xarses joined #puppet-openstack
03:53 hogepodge ping crinkle
03:53 crinkle hogepodge: hi
03:54 hogepodge Lots of reverting to my module work. I was able to back out all of those nasty vagrant hacks after I figured out a clean way to make CentOS fall back to the old style device assignment.
03:54 crinkle hogepodge: awesome!
03:56 hogepodge Just pm'ed you a link to my latest box. I'd publish it wider but if too many people download I start to pay for the bandwidth. :-P
03:57 crinkle hogepodge: thanks!
04:00 crinkle hogepodge: I have a branch on puppetlabs-openstack that fixes a number of minor bugs, want to take a look at it and check my sanity? https://github.com/puppetlabs/​puppetlabs-openstack/pull/102
04:00 crinkle once that's merged I think I can do a new minor release
04:01 crinkle of stable/icehouse
04:02 hogepodge That all looks good to me. I'm not familiar with rvm, but there are a number of things there that were definitely giving me headaches.
04:12 thumpba joined #puppet-openstack
04:17 nibalizer hogepodge: isn't it free to upload to the vagrant cloud?
04:27 hogepodge nibalizer I don't know. For PL they've been linking to their own s3 images.
04:29 nibalizer https://vagrantcloud.com/
04:29 nibalizer ya looks free
04:29 nibalizer some weird preimum features for $6/mo but probably not just for box sharing
04:30 nibalizer ya if you only want public boxes then free
05:08 sputnik13 joined #puppet-openstack
05:14 artem_panchenko_ joined #puppet-openstack
05:20 sanjayu joined #puppet-openstack
05:22 f10d4 joined #puppet-openstack
06:02 thumpba joined #puppet-openstack
07:12 tlbr_ joined #puppet-openstack
07:17 michchap joined #puppet-openstack
07:23 linux-uzer joined #puppet-openstack
07:37 stamak joined #puppet-openstack
07:47 TorLaedre joined #puppet-openstack
07:52 francois1 joined #puppet-openstack
07:57 derekh joined #puppet-openstack
08:10 mmagr joined #puppet-openstack
08:12 imcsk8 joined #puppet-openstack
08:12 dgurtner joined #puppet-openstack
08:22 social joined #puppet-openstack
08:27 thumpba_ joined #puppet-openstack
08:34 gaudenz joined #puppet-openstack
08:43 thumpba joined #puppet-openstack
08:46 gildub joined #puppet-openstack
08:54 mmaglana joined #puppet-openstack
09:06 berendt joined #puppet-openstack
09:44 bogdando hello. I put a comment to pacemaker for nova::rabbitmq blueprint https://review.openstack.org/#/c/104861/. Please could you elaborate what is the current status of puppet common module for Openstack? Some links I could read?
09:45 bogdando The '::ha_openstack' wrapper is likely the same concept, so I'd like to not reinvent something is has been developed already :)
09:45 bogdando Once I have some more details about this pacemaker wrapper for Openstack services I will provide them in aforementioned blueprint first...
09:47 bogdando Speaking in another words, what is this common module for puppet Openstack modules?
09:47 bogdando and how we could contribute our pacemaker wrapper there as well
09:55 mmaglana joined #puppet-openstack
09:56 michchap bogdando: my personal opinion is that the nova::rabbitmq class shouldn't be responsible for installing rabbitmq, and that should instead be left to a separate profile/role.
09:56 michchap bogdando: maybe raise this during the meeting?
09:57 bogdando michchap, yes, of cause. As far as I understood, the logic implemented in nova::rabbitmq could be dafely put into puppet-openstack
09:58 bogdando michchap, And ::openstack  class could create all required entities for nova in rabbitmq and install rabbit as well
09:58 bogdando like it does for keystone entities
09:59 bogdando michchap, but the idea I'd like to discuss is more generic - what would be the best way to put Openstack services under pacemaker w/o modifying their core modules
09:59 bogdando And I believe this 'ha::linux' / 'ha::openstack' definition wrapper could be the best option...
10:00 michchap bogdando: I'd say using puppet-openstacklib
10:00 michchap bogdando: unless it's going to take a lot of dependencies, in which case it might be better suited to puppet-openstack_extras
10:01 bogdando michchap, exactly. now I know at least the name of this mysterious module :)
10:01 michchap bogdando: the dependency chain is extras depends on [the modules] which depend on openstacklib
10:02 bogdando michchap, thanks. Some dev-docs about this may be?..
10:02 michchap bogdando: so in retrospect perhaps it might be best suited for extras?
10:02 bogdando michchap, perhaps. I think you got the idea. E.g. we could use some '::ha_linux { 'foo_service': provider => 'pacemaker' } ' definition in catalog in order to override given foo_service service by pacemaker control plane
10:02 michchap bogdando: only added this cycle, not sure what docs there are. We argued for some time on the ML
10:03 rcallawa joined #puppet-openstack
10:03 michchap bogdando: yep. So you'll use a collector to override provider?
10:03 bogdando michchap, so it would collect and redefine foo_service service being configured by some openstack module
10:04 bogdando michchap, yes, that is the way is under discussion in our internal dev team atm
10:04 bogdando michchap, and I'd like to make this one public :)
10:05 michchap bogdando: I find that pattern to be a bit difficult to debug, but the alternative is a lot of parameters and pushing everything into hiera/enc, which is also a bit of a pain to keep track of if it's needed everywhere.
10:05 bogdando michchap, here is related issue in puppet btw. https://tickets.puppetlabs.com/browse/PUP-3427
10:06 bogdando michchap, yes, ther are many options, we should discuss it and I will create a blueprint, starting from nova, cuz it have generic_service and will be the most complex case )
10:07 michchap bogdando: sure. IMO it's a good fit for extras and I'm keen to see what you come up with.
10:10 bogdando btw, there is agenda for meetings? https://wiki.openstack.org/wiki/​Puppet-openstack#Weekly_meetings and is this page relevant?
10:10 michchap bogdando: https://etherpad.openstack.org/p/p​uppet-openstack-agenda-10.13.2014
10:12 qba73 joined #puppet-openstack
10:12 bogdando michchap, thanks
10:13 mmaglana joined #puppet-openstack
10:14 openstackgerrit A change was merged to stackforge/puppet-horizon: should not set wsgi_socket_prefix  https://review.openstack.org/124520
10:25 bogdando michchap, updated agenda
10:33 rcallawa joined #puppet-openstack
10:38 rcallawa_ joined #puppet-openstack
10:59 linux-uzer hi, i am trying to use mysql module and to specify bind address and engine. How can i do it? It looks git mysql module had quite some changes...
11:00 linux-uzer before i could specify config_hash to mysql::server - no longer exists
11:14 mmaglana joined #puppet-openstack
11:39 openstackgerrit Tri Hoang Vo proposed a change to stackforge/puppet-keystone: Fix service keystone conflict when running in apache  https://review.openstack.org/127939
11:52 rcallawa joined #puppet-openstack
12:11 kbyrne joined #puppet-openstack
12:15 mmaglana joined #puppet-openstack
12:18 britthouser joined #puppet-openstack
12:24 ByteSore joined #puppet-openstack
12:25 pieterl joined #puppet-openstack
12:27 rcallawa joined #puppet-openstack
12:34 simonc joined #puppet-openstack
12:52 morazi joined #puppet-openstack
12:56 openstackgerrit Tri Hoang Vo proposed a change to stackforge/puppet-keystone: Fix service keystone conflict when running in apache  https://review.openstack.org/127939
13:02 michchap linux-uzer: $override_options is what you're looking for.
13:11 britthouser joined #puppet-openstack
13:16 mmaglana joined #puppet-openstack
13:18 britthouser joined #puppet-openstack
13:18 britthou_ joined #puppet-openstack
13:24 sanjayu newsnews
13:26 sputnik13 joined #puppet-openstack
13:49 JayJ joined #puppet-openstack
13:57 jproulx1 joined #puppet-openstack
14:01 openstackgerrit Tri Hoang Vo proposed a change to stackforge/puppet-keystone: Fix service keystone conflict when running in apache  https://review.openstack.org/127939
14:16 mmaglana joined #puppet-openstack
14:22 rwsu joined #puppet-openstack
14:23 emccormick joined #puppet-openstack
14:32 mdorman joined #puppet-openstack
14:37 ingard joined #puppet-openstack
14:40 ingard hi guys. anyone involved with the puppet-ceph module here?
14:40 ingard I'm wondering if there is any effort to "discover" ssd journal devices along with discovering the osd devices themselves?
14:51 EmilienM good morning o/
14:51 crinkle morning
14:57 xingchao joined #puppet-openstack
14:57 crinkle https://etherpad.openstack.org/p/p​uppet-openstack-agenda-10.13.2014
14:57 crinkle anyone want to add to the agenda? ^
14:58 bogdando left #puppet-openstack
14:58 bogdando joined #puppet-openstack
14:58 bogdando hi
14:59 EmilienM crinkle: fyi, it's thanksgiving here, we may miss some people
15:00 crinkle EmilienM: oh I didn't realize
15:00 EmilienM crinkle: isn't it in the US?
15:00 crinkle EmilienM: thanksgiving in the US isn't until november
15:01 holser left #puppet-openstack
15:01 EmilienM crinkle: oh, strange, in Canada it's today :-)
15:01 holser joined #puppet-openstack
15:01 xingchao seems time is up :)
15:02 crinkle how many are here? I'd be fine with postponing
15:03 bogdando here
15:03 EmilienM o/
15:03 mdorman +1
15:03 crinkle okay let's have a meeting
15:03 holser hi
15:03 crinkle #startmeeting
15:04 crinkle bogdando: do you want to go first?
15:04 bogdando crinkle, yes, I could do
15:04 crinkle #topic Pacemaker in openstack_extras
15:04 bogdando So, we have a nova::rabbitmq pacemaker provider blueprint as a 'prototype'
15:05 bogdando originally, it was suggested to implement it and may be do the Openstack services in [modules] as well
15:06 bogdando #link https://review.openstack.org/#/c/104861/
15:06 bogdando But we had some options to do it better :)
15:06 bogdando I put them in agenda...
15:07 bogdando idea is to not modify puppet-openstack [modules], but use a wrappers in order to put Openstack [services] under pacemaker control plane
15:08 dilyin joined #puppet-openstack
15:08 * dilyin тут
15:08 bogdando put these wrappers somewhere in puppet-extras or puppet-openstacklib
15:08 EmilienM I'm not sure, but puppet-extras is more adapted I think
15:08 bogdando and include them in catalog if we want to *redefine* pacemaker service type for Openstack services
15:08 bogdando and rabbit and whatever
15:09 bogdando So looks like I have an action item to create a basic draft for this thing
15:09 crinkle yeah I don't have a lot of familiarity with pacemaker but I would +1 putting things in openstack_extras
15:09 bogdando but I hope we could make some votes before to design... see agenda ;)
15:10 crinkle bogdando: it might be worthwhile to send something to the mailing list
15:10 crinkle since meeting attendance today is relatively low
15:10 EmilienM I like the idea
15:10 bogdando crinkle, agree
15:11 EmilienM the more we can put in common code...
15:11 bogdando so I could take care a ML as well
15:12 bogdando so, should we make a vote maybe?..
15:12 bogdando there is some more options, e.g. to put ha_mode option for all [modules]...
15:12 EmilienM not sure we have enough core contributors here.
15:12 bogdando okay (
15:13 holser also we have 4th solution - introduce $high_availability option in every openstack class and configure services according to it
15:13 bogdando ofc that would require quite a lot of changes
15:13 EmilienM yes... and maintain this code...
15:13 bogdando and some basic OCF scripts to be shipped with [modules]
15:13 bogdando HA out of box :)
15:14 bogdando wrappers just would put all this stuff behind the scenes and would make it completely up to devs
15:15 dilyin making all OS modules shipped with HA mode out of the box would be the best solution obviously but do we have resources enough to implement it
15:15 dilyin wrappers could be a good compromise that would allow to use unmodified upstream modules but still get HA support without much effort
15:15 xingchao bogdando: I see the discussion on gerrit, maybe you could provide a prototype of :ha_openstack to make it more clear
15:16 bogdando xingchao, yes, dilyin is working on a PoC
15:16 crinkle I think HA stuff is best placed in extras
15:16 xingchao bogdando: dilyin:   great
15:17 crinkle each service's module is great at doing things on one node, HA is a different matter
15:17 bogdando xingchao, we will provide a link to BP (and ML thread) to it once polished a bit...
15:17 mmaglana joined #puppet-openstack
15:17 dilyin yes we can place such overrides into a separate module. if you download and include it it would turn unmodified upstream modules into HA enabled ones
15:18 bogdando crinkle, indeed. And pacemaker always suggests there would no be interference from OS side, from systemd or upstart or system V etc.
15:18 xingchao bogdando: I'm waiting for :)
15:18 rharrison joined #puppet-openstack
15:18 dilyin or we can start to build HA support into upstream modules
15:18 bogdando dilyin, yes, exactly, that would be an option 4 holser mentioned above.
15:19 dilyin and i'm to exaclty sure waht would be harder to support
15:19 EmilienM it could be a lot of dupp code I think
15:19 crinkle ^
15:19 dilyin yes there definitly will be some along with a lot of klugdes
15:19 bogdando EmilienM, but we could put it into puppet-openstacklibs?
15:20 dilyin building HA support into upstream modules can make a lot clearer code of cause
15:20 crinkle openstacklib is more for stuff that the modules are going to consume
15:20 bogdando /offtop btw guys, how to announce an action item for meeting bot?
15:20 crinkle this sounds more like it should be wrapping around/consuming the modules
15:20 EmilienM ( crinkle tell me if I'm wrong ) openstacklib is more for common providers. I would rather see that in extras.
15:20 crinkle EmilienM: yes that's how I see it
15:20 crinkle bogdando: I don't think we have a real meeting bot but I would say #action <action>
15:21 holser crinkle, I am with you. since every module has own OCF file and own dependencies, HA should be per module
15:21 bogdando EmilienM, I mean, if we start supporting HA in puppet modules for Openstack, we could deduplicate it to libs
15:21 xingchao yes, I agree with emilienM, these codes should put into extras
15:21 bogdando EmilienM, though, it would be hard
15:22 crinkle I really don't think HA belongs inside the modules at all, either in the actual module or in openstacklib
15:22 dilyin well actually yes we can make pacemaker::service definition and just call it from every OS module and not only OS
15:22 EmilienM another thing also. afik, we have all different ways to implement HA
15:22 dilyin that would be easier to support then other solutions
15:23 bogdando #action bogdando send a ML about options for Pacemaker Openstack services provider
15:23 crinkle EmilienM: agreed
15:23 EmilienM HA does not mean "enable something in one option"
15:24 EmilienM as you probably know, it's very complicated and can be reached by 1 000 ways. So we need a flexible code to fit a maximum of use cases.
15:24 dilyin well... having different ways to implement HA is an argument for wrappers... of for a single definition because you would be able to rerite it as you with
15:24 crinkle yep
15:24 EmilienM bogdando: to be honest, I did not spend too much time on your BP, but I'll do it ASAP and let a review.
15:24 dilyin *rewrite it as you wish
15:24 bogdando #action dilyin provide a PoC manifests for basic options for implementing Pacemaker Openstack services provider
15:25 EmilienM bogdando: and please publish the link on the review, then we can catch up
15:25 EmilienM (the specs review)
15:25 bogdando EmilienM, well, I guess it is missing this wrapper related stuff yet and more close to option 4
15:25 dilyin here is a demo how can you override a simple module to add ha support https://github.com/dmitryilyin/puppet_ha_overrides
15:25 bogdando #link https://github.com/dmitryilyin/puppet_ha_overrides
15:26 aimon joined #puppet-openstack
15:26 bogdando okay, looks clear more or less... what about Nova::rabbitmq's destiny? :)
15:27 bogdando I believe, this wrapper class could be moved to puppet-openstack - install, configure, create entities...
15:27 bogdando but anyway it would be nice to have it as is for some deprecation period
15:28 bogdando and I believe we should introduce support for rabbitmq>=4.0
15:28 dilyin it does not matter where these wrappers are placed as long as you have them included
15:28 bogdando there were many nice changes since v1.0 currently supported in nova::rabbitmq
15:28 EmilienM bogdando: are you mentioning github/stackforge/puppet-openstack ?
15:29 bogdando EmilienM, yes. We could make all rabbit's stuff there as well? could we?
15:29 EmilienM bogdando: I don't think puppet-openstack module has a future.
15:29 dilyin HA stuff for rabbit? yes, we can place in there too
15:29 rharrison joined #puppet-openstack
15:29 bogdando dilyin, yes, but currently nova installs rabbit via nova::rabbitmq
15:29 dilyin without ha?
15:29 EmilienM michchap, I and some folks were thinking about depreciation. There is no more contributions on it.
15:30 bogdando dilyin, it contains its own wrapper, yes, no Ha yet
15:30 bogdando EmilienM, but there then?..
15:30 bogdando s/where
15:30 dilyin it's ok for nova to use rabbit module. we just need to override rabbit service provided by rabbit modules
15:31 bogdando dilyin, I like this option also. Use the same wrapper for rabbitmq service, ignoring the way it was installed and configured - just override it to pacemaker.
15:32 EmilienM crinkle: did we have some thoughts about moving rabbitmq common things in openstack-extras?
15:32 bogdando but we are also speaking about moving nova::rabbitmq , right?..
15:32 dilyin an wrappers give you a freedom NOT to use them or write your own wrappers instead
15:32 crinkle EmilienM: iirc we were thinking of moving rabbitmq stuff to openstacklib
15:32 crinkle but maybe it's more appropriate to move it to extras
15:33 crinkle I don't really know how rabbit works
15:33 bogdando well, nova's one cannot be skipped, it creates some entites now, w/o them nova won't be configured and running
15:33 EmilienM crinkle: it's not ruby providers, it's just rabbitmq manifests code
15:33 dilyin there just a  set of native types to create users and other stuff
15:33 bogdando yup
15:34 EmilienM anyway, moving rabbitmq is another topic.
15:34 stamak joined #puppet-openstack
15:34 bogdando EmilienM, yes, but it was at agenda...
15:34 bogdando michchap asked me to discuss it as well
15:35 bogdando so, do we want to add rabbitmq>=4.0 support for nova::rabbitmq? I have a WIP branch for it :)
15:35 bogdando there is a similar for mysql in modules, so why not?
15:35 EmilienM bogdando: can you share the link? I may have miss it
15:36 blentz joined #puppet-openstack
15:36 bogdando EmilienM, e.g.  https://github.com/stackforge/puppet-nov​a/blob/master/manifests/db/mysql.pp#L38
15:36 bogdando #link https://github.com/stackforge/puppet-nov​a/blob/master/manifests/db/mysql.pp#L38
15:38 xingchao bogdando: introduce rabbitmq>=4.0 will bring some compatbliity problem ?
15:38 EmilienM bogdando: I was asking for your WIP branch
15:38 bogdando EmilienM, it is deprecated now, but was not at v4.0 of puppet-nova, So we cannot just move to new rabbit as well - if new one introduced, we had to maintain it backward compatible for some time
15:38 bogdando xingchao, it should not, I'm trying to do it in a backwards compatible way
15:39 EmilienM what I would like to see, is https://github.com/stackforge/puppet-n​ova/blob/master/manifests/rabbitmq.pp in openstack-specs. Because the code is always the same.
15:39 EmilienM but again, I may be wrong, it's just how I think :)
15:39 crinkle EmilienM: openstack-specs? do you mean openstacklib ?
15:39 EmilienM crinkle: not specs... extra sorry
15:39 crinkle okay
15:40 bogdando EmilienM, yes, I have a WIP but it is a bit obsolete from upstream pov and should be rebased onto current puppet-nova
15:40 EmilienM crinkle: same for nova::keystone bits, I would like to use your provider later
15:40 crinkle if the code is always the same then I think it should be in openstacklib. If it's something that can be added to provide extra functionality then I would want it in openstack-extras
15:41 EmilienM +1
15:41 crinkle (I'm not sure which category this falls under)
15:41 bogdando EmilienM, actually it is now based on Fuel fork, but I work on contributing it... https://github.com/bogdando/fuel-library​-1/blob/sync_rabbit_adapt/deployment/pup​pet/nova/manifests/rabbitmq.pp#L108-150
15:41 EmilienM +1 for a new folder 'messaging' in https://github.com/stackforge/puppet​-openstacklib/tree/master/manifests
15:42 EmilienM crinkle: ^
15:42 crinkle okay
15:42 bogdando link #https://github.com/bogdando/fuel-library​-1/blob/sync_rabbit_adapt/deployment/pup​pet/nova/manifests/rabbitmq.pp#L108-150
15:42 EmilienM and inside, messaging/rabbitmq.pp
15:42 bogdando just an example, please ignore all ugly code around
15:42 EmilienM with qpid support, etc
15:43 EmilienM crinkle: I have that feeling it's still unclear what lands in lib & extras... :(
15:43 crinkle EmilienM: yeah :(
15:43 EmilienM maybe we should start by here.
15:43 EmilienM and define some "rules"
15:44 bogdando So, if you aprove I could submit this new rabbit module support, completely backwards compatible, to puppet-nova
15:44 bogdando even if we decide to move it somethere else, we have to deprecate its location first, I believe :)
15:44 crinkle bogdando: I think changes to rabbit should be made across all the modules, it will get messy if we let one module march ahead of the others
15:44 EmilienM bodepd_: we will do like with mysql things
15:44 xingchao bogdando: the codes seems good, it looks like what we have done in mysql module before
15:45 crinkle so the question is whether to do it in all the modules, move it to openstacklib, or move it to openstack_extras
15:45 bogdando crinkle, other modules do not install rabbit in puppet openstack ones, only nova does
15:45 EmilienM crinkle: actually, you're right. Messaging bits need to land in lib. It's the same thing as DB
15:46 EmilienM crinkle: I just added the point to the agenda
15:47 crinkle bogdando: oh I see
15:47 bogdando EmilienM, crinkle so, what I'm suggesting: 1) add support for new rabbit in nova::rabbitmq, 2) deprecate nova::rabbitmq with warnings, 3) move nova::rabbitmq to extras (or lib?) later, lets say in K or H, 4) deprecate rabbit version and use the latest one
15:47 EmilienM bogdando: the code looks good. Indeed. We could start a new thread about this bits
15:47 crinkle EmilienM: bogdando points out that they're actually not all the same...
15:47 bogdando makes sense?
15:47 crinkle bogdando: I think that sounds good but I would take that to the mailing list as well
15:48 bogdando crinkle, ok please, do or should I ?..
15:48 EmilienM bogdando: it looks good but indeed, a discussion in the ML would be perfect to be sure we all agree
15:48 EmilienM bogdando: go ahead!
15:48 bogdando perhaps, some one else wants some action items :)
15:48 xingchao bogdando: +1 for it
15:49 crinkle okay let's move on
15:49 bogdando #action bogdando suggest to the ML: 1) add support for new rabbit in nova::rabbitmq, 2) deprecate nova::rabbitmq with warnings, 3) move nova::rabbitmq to extras (or lib?) later, lets say in K or H, 4) deprecate rabbit version and use the latest one
15:49 bogdando okay guys, thanks for your time
15:49 EmilienM bogdando: thx for your insights
15:49 EmilienM crinkle: out of topic, but I'll add a bot here for having real meetings.
15:49 EmilienM I'm frustrated to not having meeting logs.
15:50 crinkle EmilienM: haha
15:50 bogdando oh, I expected to see it rendered in html :(
15:50 crinkle #topic Paris summit agenda
15:50 EmilienM crinkle: are we done?
15:50 EmilienM ah nop
15:50 EmilienM bogdando: next meeting will be. No worries
15:51 crinkle I think I will work with sbadia to create a formal agenda and then submit it to the mailing list
15:51 crinkle rather than discuss it now
15:51 crinkle it sounds like HA will be a major topic
15:51 crinkle #action crinkle and sbadia to work on it
15:52 crinkle anything else?
15:52 ericpeterson joined #puppet-openstack
15:52 EmilienM crinkle: do we have a room?
15:52 crinkle EmilienM: hogepodge is working on it I think
15:52 EmilienM that is awesome
15:52 xingchao :)
15:52 EmilienM because we need a room
15:53 crinkle :)
15:53 crinkle okay
15:53 crinkle #endmeeting
15:54 EmilienM crinkle: thanks. Have a nice day everyone ! (off for me :-P )
15:54 crinkle EmilienM: happy thanksgiving!
15:54 EmilienM crinkle: thanks!
15:54 hogepodge http://kilodesignsummit.sched.org
15:54 xingchao EmilienM: I see you propose a cool project tripleo-puppet-elements recently :)
15:54 hogepodge http://kilodesignsummit.sched.org/ev​ent/f9ed6b21149364e38fa4ab03b80a7d74
15:55 crinkle hogepodge: \o/
15:55 xingchao hogepodge: great LOL
15:57 rmoe joined #puppet-openstack
16:02 EmilienM crinkle: fyi: https://review.openstack.org/128004
16:02 EmilienM xingchao: yes, we are going to discuss about this at the next summit with dprince & other tripleo folks. Feel free to join
16:03 crinkle EmilienM: another possibility is moving meetings to #openstack-meeting
16:03 crinkle I hear there is a slot open, just not at our current meeting time ...
16:04 EmilienM crinkle: staying here would let us the time flexibility.
16:04 crinkle yeah
16:04 EmilienM because michchap and xingchao are very far
16:05 EmilienM crinkle: but you may have a definitive solution. For now, let's use our channel I think
16:05 crinkle sgtm
16:06 xingchao EmilienM: I would like to attend it :)
16:07 EmilienM xingchao: everything happens on openstack-dev ML
16:07 EmilienM I'm off, see you guys
16:07 xingchao EmilienM: get it, see you
16:08 xingchao joined #puppet-openstack
16:18 mmaglana joined #puppet-openstack
16:20 sputnik13 joined #puppet-openstack
16:21 xingchao joined #puppet-openstack
16:42 rwsu joined #puppet-openstack
17:02 bitblt joined #puppet-openstack
17:03 Guest13820 left #puppet-openstack
17:17 nextrevision joined #puppet-openstack
17:18 nextrevision left #puppet-openstack
17:20 mschmitt joined #puppet-openstack
17:30 EmilienM crinkle: seems like we will have to choose a meeting slot on openstack meeting channels: https://review.openstack.org/#/c/128004/
17:31 crinkle EmilienM: :/
17:34 EmilienM crinkle: there is one slot at 2pm UTC
17:35 crinkle that is 7am where I am :(
17:37 xarses joined #puppet-openstack
17:41 EmilienM crinkle: I sent a mail to the ML
17:45 stamak joined #puppet-openstack
17:58 nibalizer so i can't make it to the 8am meeting, 7am only makes it less likely that i'll come
18:42 hogepodge sbadia ping
19:12 shakamunyi joined #puppet-openstack
19:15 daneyon joined #puppet-openstack
19:17 shakayumi joined #puppet-openstack
19:21 berendt joined #puppet-openstack
19:29 openstackgerrit Matt Fischer proposed a change to stackforge/puppet-keystone: Fix deprecated LDAP config options  https://review.openstack.org/128066
19:42 openstackgerrit Matt Fischer proposed a change to stackforge/puppet-keystone: Fix deprecated LDAP config options  https://review.openstack.org/128066
19:43 daneyon_ joined #puppet-openstack
19:55 prad joined #puppet-openstack
20:07 shakamunyi joined #puppet-openstack
20:32 openstackgerrit Matt Fischer proposed a change to stackforge/puppet-keystone: Switch to new Juno defaults  https://review.openstack.org/128091
20:37 daneyon_ left #puppet-openstack
20:57 todin left #puppet-openstack
21:49 shakayumi joined #puppet-openstack
21:58 clarkb joined #puppet-openstack
21:59 clarkb hello https://review.openstack.org/#/c/126424 seems like something that should get some consensus from this group. Would be great if we could get people to weigh in
22:02 crinkle clarkb: I think the general consensus on our mailing list was that it is safe to remove 2.7
22:03 clarkb awesome (I think it is the right thing to do :) ) crinkle want to +1 that change with an all clear to remove 2.7?
22:03 crinkle clarkb: I +1'd
22:04 clarkb perfect
22:24 hogepodge +1 from me too (community member emeritus)
22:31 rcallawa joined #puppet-openstack
22:46 shakayumi joined #puppet-openstack
22:57 rcallawa joined #puppet-openstack
23:03 gildub joined #puppet-openstack
23:05 openstackgerrit joined #puppet-openstack
23:16 LeslieCarr joined #puppet-openstack
23:17 shakayumi joined #puppet-openstack
23:19 LeslieCarr hey, so i am wondering if the impossible is possible - we want to use nova-network on icehouse, is that possible? and even better, is it possible to do with modifying the awesome puppetlabs openstack module in a not insane way ?
23:28 emccormickva joined #puppet-openstack
23:35 openstackgerrit joined #puppet-openstack
23:52 aimon joined #puppet-openstack
23:55 xarses LeslieCarr: yes, nova-network still works in icehouse and will still for a while longer
23:55 xarses I however cant speak for puppetlabs-openstack
23:57 LeslieCarr thanks xarses - it’s a pretty awesome module and was hoping to be able to reuse it, but it uses neutron
23:58 LeslieCarr i was reallly hoping someone has done it before … but it looks like i’ll probably have a lot of work cut out for me
23:58 hogepodge LeslieCarr I can offer a bit more help. You're going to need to remove this line https://github.com/puppetlabs/puppetlabs-opens​tack/blob/master/manifests/common/nova.pp#L59
23:58 hogepodge Not use any of the neutron profiles (that's easy)
23:59 hogepodge And add classes from this directory to the nova api and compute roles (or common nova) https://github.com/stackforge/puppet​-nova/tree/master/manifests/network
23:59 LeslieCarr cool, and i’m guessing also 61-68 of that file ?
23:59 hogepodge Yup, that too. -:-)

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