Wednesday, 2016-05-18

*** r-mibu has quit IRC00:05
*** r-mibu has joined #tripleo00:05
*** sthillma has quit IRC00:11
*** sthillma has joined #tripleo00:11
*** rbrady has quit IRC00:15
*** dmacpher has quit IRC00:25
*** diazjf has joined #tripleo00:33
openstackgerritDmitry Ilyin proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker
*** panda has quit IRC00:48
*** panda has joined #tripleo00:48
*** akshai has joined #tripleo00:53
*** akshai_ has quit IRC00:53
*** sthillma has quit IRC01:07
*** diazjf1 has joined #tripleo01:12
*** diazjf has quit IRC01:15
*** tremble has quit IRC01:18
*** dmacpher has joined #tripleo01:20
*** tremble has joined #tripleo01:21
*** tremble has quit IRC01:21
*** tremble has joined #tripleo01:21
*** yamahata has quit IRC01:34
*** tremble has quit IRC01:42
*** tremble has joined #tripleo01:43
*** weshay has quit IRC01:47
-openstackstatus- NOTICE: Gerrit is about to be restarted to help with page timeouts01:47
*** mburned has quit IRC02:09
*** mburned has joined #tripleo02:10
*** akshai has quit IRC02:11
*** dixiaoli has joined #tripleo02:17
*** MaxPC has quit IRC02:22
*** mburned has quit IRC02:32
*** mburned has joined #tripleo02:34
*** diazjf1 has quit IRC02:40
*** mgarciam has quit IRC03:02
*** dtrainor has quit IRC03:03
*** dkelson has joined #tripleo03:07
*** tzumainn has quit IRC03:08
dkelsonI'm new to TripleO. I'm trying to customize my overcloud. I'm using OS::TripleO::NodeExtraConfigPost to run some puppet-keystone stuff. My overcloud deployment fails on ComputeNodesPostDeployment. I only need to modify the controller(s), how do I prevent from trying to do my custom config on the compute node(s)?03:12
*** coolsvap has joined #tripleo03:22
*** dixiaoli has quit IRC03:23
*** links has joined #tripleo03:26
*** mburned has quit IRC03:29
*** mburned has joined #tripleo03:31
*** dixiaoli has joined #tripleo03:37
*** akshai has joined #tripleo03:49
*** akshai has quit IRC03:50
dkelsonI think I got it03:57
*** masco has joined #tripleo04:17
*** mburned has quit IRC04:29
*** mburned has joined #tripleo04:29
*** liverpooler has quit IRC04:31
*** veteran has joined #tripleo04:31
*** veteran has quit IRC04:31
*** tremble has quit IRC04:34
*** tremble has joined #tripleo04:37
*** jcoufal has joined #tripleo04:45
openstackgerritayoung proposed openstack/tripleo-quickstart: WIP Allow for multiple undercloud nodes
*** dsariel has joined #tripleo04:55
*** SlickN1k has joined #tripleo05:05
*** hdsdman_ has joined #tripleo05:08
*** davidlenwell has quit IRC05:09
*** dmacpher has quit IRC05:10
*** sambetts has joined #tripleo05:11
*** sambetts|afk has quit IRC05:12
*** snecklifter has quit IRC05:12
*** SlickNik has quit IRC05:12
*** dsariel has quit IRC05:12
*** links has quit IRC05:12
*** dmsimard has quit IRC05:12
*** sirushti has quit IRC05:12
*** hdsdman has quit IRC05:12
*** colonwq has quit IRC05:12
*** tremble has quit IRC05:12
*** SlickN1k is now known as SlickNik05:13
*** sirushti has joined #tripleo05:13
*** dsariel has joined #tripleo05:13
*** links has joined #tripleo05:13
*** snecklifter has joined #tripleo05:13
*** tremble has joined #tripleo05:13
*** tremble has quit IRC05:13
*** tremble has joined #tripleo05:13
*** colonwq has joined #tripleo05:14
*** dmsimard has joined #tripleo05:16
*** davidlenwell has joined #tripleo05:17
*** ChanServ sets mode: +v davidlenwell05:17
marioshello world!05:18
*** dmacpher has joined #tripleo05:18
*** dsariel has quit IRC05:23
*** tremble has quit IRC05:23
*** bvandenh has quit IRC05:27
*** aufi has joined #tripleo05:37
*** bvandenh has joined #tripleo05:40
*** ccamacho has quit IRC05:46
*** rcernin has joined #tripleo05:47
*** ccamacho has joined #tripleo05:47
*** fragatin_ has joined #tripleo05:52
*** fragatin_ has quit IRC05:52
*** fragatin_ has joined #tripleo05:53
*** fragatin_ has quit IRC05:53
*** fragatin_ has joined #tripleo05:54
*** fragatina has quit IRC05:55
*** fragatin_ has quit IRC05:59
*** david-lyle has quit IRC06:05
*** jrist has quit IRC06:07
*** fragatina has joined #tripleo06:07
*** dsariel has joined #tripleo06:08
*** fragatina has quit IRC06:12
*** xinwu has quit IRC06:13
*** cmyster has quit IRC06:13
*** apetrich has joined #tripleo06:14
*** cmyster has joined #tripleo06:15
*** cmyster has quit IRC06:15
*** cmyster has joined #tripleo06:15
*** bvandenh has quit IRC06:16
*** olap has joined #tripleo06:17
*** bvandenh has joined #tripleo06:19
*** jrist has joined #tripleo06:22
*** fragatina has joined #tripleo06:24
*** bvandenh has quit IRC06:25
*** tremble has joined #tripleo06:28
*** Guest72844 has joined #tripleo06:31
*** Guest72844 has quit IRC06:31
*** anshul has joined #tripleo06:32
*** anshul is now known as Guest3201106:32
*** Guest32011 is now known as abehl06:34
*** fragatina has quit IRC06:35
*** fragatina has joined #tripleo06:36
*** bvandenh has joined #tripleo06:36
openstackgerritMarios Andreou proposed openstack/tripleo-heat-templates: Adds memcached as a composable service
openstackgerritMarios Andreou proposed openstack/puppet-tripleo: Adds the base and pacemaker profile for the memcached service
*** jtomasek has joined #tripleo06:45
openstackgerritMarios Andreou proposed openstack/puppet-tripleo: Adds the base and pacemaker profile for the memcached service
*** rcernin has quit IRC06:47
*** rcernin has joined #tripleo06:51
*** coolsvap has quit IRC06:53
*** coolsvap has joined #tripleo06:53
*** florianf has joined #tripleo06:55
*** dmacpher has quit IRC06:59
*** jprovazn has joined #tripleo06:59
ccamachoo/ morning!07:05
marioso/ ccamacho07:13
mariosccamacho: hey man thanks for the great review and fixup07:14
openstackgerritMerged openstack/tripleo-heat-templates: Move dhcp agent maps to OS::Heat::None to resource_registry
ccamachomarios, hey! np man, yesterday I got this message when deploying it without pacemaker "Error: Could not find dependency Exec[wait-for-settle] for Pacemaker::Resource::Systemd[memcached]" Im checking it up again to try to get the line on the .pp getting that error07:15
mariosccamacho: hmmm thanks. so am waiting to see what ci makes of the puppet-tripleo side and try a recheck for the tht07:19
*** xinwu has joined #tripleo07:19
*** xinwu has quit IRC07:19
mariosccamacho: cos jenkins thinks the "This change depends on a change that failed to merge."07:19
mariosmojo be all like :/07:20
*** ebarrera has joined #tripleo07:20
ccamachomarios, jaja.  Question, when making changes in puppet-tripleo.. are the changes applied to the overcloud images for a THT jenkins check? (With the depends-on tag) Or needed to have first the puppet-tripleo patch landed to be able to test it?07:22
*** mcornea has joined #tripleo07:24
mariosccamacho: i _assume_ they're included with the depends-on and it will enforce the order for landing (i.e. wont merge this until that does)07:25
ccamachomarios, xD mmm not 100% sure about if they are included in the overcloud images, that's why I'm using all the time mi CI.. anyway Ill let you know when have more details about that previous error when deploying without pacemaker07:27
hewbroccaMorning folks07:28
hewbroccahow's the CI working?07:28
mariosccamacho: ack07:30
marioshewbrocca: seen some green runs this morning fwiw07:30
marioshewbrocca: on my review rounds07:30
ccamachomorning! there are some green jobs up there
*** mburned has quit IRC07:32
*** mburned has joined #tripleo07:33
hewbroccathings running faster in general?07:34
marioshewbrocca: well here's one example
*** paramite has joined #tripleo07:34
*** dmk0202 has joined #tripleo07:35
mariosall under 2 hours, including upgrades job is a good thing07:35
mariosbut as others have said, we need to watch it over a few days07:35
hewbroccaStill, seems like it was worth the effort upgrading the machines07:36
*** devvesa has joined #tripleo07:38
*** devvesa has quit IRC07:38
*** mikelk has joined #tripleo07:39
*** devvesa has joined #tripleo07:39
*** zoli_gone-proxy is now known as zoliXXL07:41
zoliXXLgood morning07:43
*** fzdarsky has joined #tripleo07:47
*** leanderthal|afk is now known as leanderthal07:48
*** jpena|off is now known as jpena07:48
*** shardy has joined #tripleo07:53
*** jistr has joined #tripleo07:54
*** david-lyle has joined #tripleo07:57
*** zoliXXL is now known as zoli|afk07:59
EmilienMpabelanger: like I said (at least for Puppet CI), we don't use EPEL08:09
EmilienMmarios: I missed your messages last night, was it all about commit message? or something in code too?08:10
mariosEmilienM: o/08:11
mariosEmilienM: no commit message but i saw you updated it no?08:11
EmilienMmarios: I pushed a comment but did not send it08:12
mariosEmilienM: and it shows up in
EmilienMmarios: the rest is fine?08:12
EmilienMok excellent08:13
*** gfidente has joined #tripleo08:13
mariosEmilienM yeah afaics08:13
EmilienMmarios: I'm still having issue with upgrade08:13
mariosEmilienM: oh yeah... i was +2 on v16 , readding08:13
EmilienMmarios: we need to +2 it when THT is green08:13
EmilienMand upgrade still fails... I'll continue to investigate but if you can help me I'll be happy, I've been spending some time on it08:14
EmilienMmaybe fresh eyes would help08:14
*** mburned has quit IRC08:14
*** liverpooler has joined #tripleo08:14
*** liverpooler has quit IRC08:14
mariosEmilienM: sure... seems like i have a few calls today/some planning for the upgrades in mitaka but i can definitely look at reviews to satart with08:15
*** liverpooler has joined #tripleo08:15
*** ohamada has joined #tripleo08:15
*** mburned has joined #tripleo08:16
EmilienMmarios: it seems like rabbit cookie is changing when running upgrade08:20
EmilienMif you open the journal on the controller, you'll see:08:21
EmilienM/Stage[main]/Rabbitmq::Config/Rabbitmq_erlang_cookie[/var/lib/rabbitmq/.erlang.cookie]/content: The rabbitmq erlang cookie was changed08:21
EmilienM2 times08:21
EmilienMI need to check if we have that on other jobs08:21
*** dsariel has quit IRC08:21
EmilienMok I checked another patch and the cookie does not change08:23
EmilienMmarios: can you review again please?08:26
*** mburned has quit IRC08:27
EmilienMcan it be the salt: {get_param: RabbitCookieSalt} ?08:27
*** mburned has joined #tripleo08:27
openstackgerritCarlos Camacho proposed openstack/tripleo-heat-templates: Composable roles within services - NTP
*** dtantsur|afk is now known as dtantsur08:29
*** lucas-dinner is now known as lucasagomes08:32
*** zoli|afk is now known as zoli08:34
*** zoli is now known as zoliXXL08:35
*** pblaho has joined #tripleo08:36
*** mbound has joined #tripleo08:36
mariosEmilienM: sry was on a call08:38
*** _milan_ has joined #tripleo08:40
mariosEmilienM: looking08:41
EmilienMgfidente: can you look also?08:41
*** derekh has joined #tripleo08:42
*** wendar_ has joined #tripleo08:43
*** spredzy_ has joined #tripleo08:44
*** honza_ has joined #tripleo08:44
*** shardy_ has joined #tripleo08:44
*** lucasagomes_ has joined #tripleo08:45
*** masco_ has joined #tripleo08:45
mariosEmilienM: do you mean ? the failed upgrades job... i see it goes create complete for stack create but the update fails
*** shardy has quit IRC08:45
*** milan has quit IRC08:45
*** telnoratti has quit IRC08:45
*** ramishra has quit IRC08:45
*** honza has quit IRC08:45
*** onovy has quit IRC08:45
*** saneax_AFK has quit IRC08:45
*** mgagne has quit IRC08:45
*** gfidente has quit IRC08:45
*** tremble has quit IRC08:45
*** masco has quit IRC08:45
*** lucasagomes has quit IRC08:45
*** spredzy has quit IRC08:45
*** Nakato has quit IRC08:45
*** wendar has quit IRC08:45
*** vkmc has quit IRC08:45
*** Nakato has joined #tripleo08:45
*** mgagne has joined #tripleo08:45
*** vkmc has joined #tripleo08:45
*** ramishra has joined #tripleo08:45
*** ramishra has quit IRC08:45
EmilienMmarios: like I said, the upgrade fails, because it regenerate a RabbitCookie08:45
*** vkmc has quit IRC08:45
*** vkmc has joined #tripleo08:45
*** ramishra has joined #tripleo08:45
marioswhat on earth08:46
EmilienMirc netsplit08:46
*** mgagne is now known as Guest9229208:46
*** telnoratti has joined #tripleo08:46
*** onovy has joined #tripleo08:46
EmilienMmarios: something is regenerating a RabbitCookie08:47
*** tremble has joined #tripleo08:47
mariosEmilienM: yeah makes sense ... so it could be the salt as you said - perhaps it is changed by the client or sthing for update?08:47
*** ramishra has quit IRC08:47
*** ramishra has joined #tripleo08:48
*** panda has quit IRC08:48
*** hrybacki has quit IRC08:48
*** marcusvrn_ has quit IRC08:48
EmilienMmarios: well, I have no idea what is going on, that's why i'm asking here :)08:48
*** lucasagomes_ is now known as lucasagomes08:49
*** panda has joined #tripleo08:49
*** gfidente has joined #tripleo08:50
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Update 60-setup-testenvs to match what we use
*** jaosorior has joined #tripleo08:51
openstackgerritMichele Baldessari proposed openstack/tripleo-heat-templates: Tighten the access rules for galera
*** saneax_AFK has joined #tripleo08:53
*** saneax_AFK is now known as saneax08:53
bandinigfidente: I tested your patch and rebased mine on top of it and it all worked correctly08:53
* hewbrocca is still not sure about the Rabbit Cookies08:53
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles
gfidentebandini, yeah thanks for testing but the 'default' in the templates might avoid us the tripleoclient backport08:54
mariosEmilienM: i can't see the client doing anything with that fwiw08:55
gfidenteit'd set a default password unless the user customizes it08:55
bandinigfidente: well let's pick the approach that is most used currently to set passwords, whichever that is?08:55
gfidentebandini, you think that is acceptable?08:55
gfidentethe most used is to generate it from the client yes08:55
EmilienMmarios, gfidente: what would generate a new RabbitCookie in my patch?08:56
mariosEmilienM: one thign i did notice is where the param is removed from controller @ I see a comment about the default value for RabbitCookie but I am not sure of the relevance to what we're seeing on update yet08:56
EmilienMbecause this is happenning08:56
gfidentein which case I'll do the tripleoclient ports08:56
bandinigfidente: I think the tripleoclient backport should not be that much of an effort, so I'd go with the most common approach08:56
bandinibut I am totally open here08:56
gfidenteEmilienM, yeah so I am looking at it now08:57
EmilienMmarios: why would we keep it in controller.yaml?08:57
*** marcusvrn_ has joined #tripleo08:57
gfidentethis is the same problem I pointed out in bandini's change08:57
*** ccamacho has quit IRC08:58
mariosEmilienM: i did not say that,i am wondering about the relevance of that comment to what we see is happening. the     default: ''  # Has to be here because of the ignored empty value bug08:58
gfidenteEmilienM, if we relocate the randomstring into a new resource heat doesn't know it has to preserve it08:58
gfidentebandini, ^^08:58
mariosgfidente: so then we keep it there ... as is this the same as the rabitmq_network staying there08:58
gfidenteand shardy suggested this has to be worked around in tripleo-common08:59
gfidentewhich I think is fine08:59
gfidenteso we'll have to pull those out and re-paste as parameters08:59
gfidenteuntil then what marios said +08:59
EmilienMhow can it be workarounded?08:59
gfidenteleave it there08:59
bandinigfidente: so are you preferring the tripleoclient change or something else?08:59
* bandini confused now :)09:00
gfidentebut we'll need to figure how to pick it up from the role09:00
*** hrybacki has joined #tripleo09:00
EmilienMI'm confused too09:00
gfidentebandini, so EmilienM is actually facing with rabbitmq cookie the same issue which the current submission for galera password would introduce for the clustercheck user09:00
gfidentewhich is the reason why we should do it differently for clustercheck user, while we can09:01
bandinigfidente: "do it differently" == "set the password via tripleoclient"?09:01
bandiniack, fine by me09:01
bandiniI +1 your change and submitted mine depending on your tripleoclient change09:02
gfidente but for rabbitmq we can't because it's already in, so we need to pull it out and re-submit it09:02
gfidenteEmilienM, let's see together how we can pass that to the role though09:02
shardy_marios: AFAIK that comment is old and bogus09:02
bandiniI see09:02
*** ccamacho has joined #tripleo09:02
shardy_I'm not aware of any reasone we need to do it, clearly a bug reference in the comment would have helped tho :)09:02
EmilienMgfidente: I don't understand why Heat re-generate it, honestly09:03
gfidenteEmilienM, because we're deleting a resource and creating another09:03
EmilienMwe're moving it, not deleting, no?09:04
openstackgerritMartin André proposed openstack/tripleo-common: Validations actions and workbook
gfidenteEmilienM, right this is a point for shardy09:04
gfidentenot sure how bad is to consider 'the same' a resource with same name and implementation from a nested stack09:04
shardy_EmilienM: sorry, joing another meeting, bbi ~30mins to discuss09:04
*** mgould has joined #tripleo09:05
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci: WIP: DONT MERGE Running tests
*** nijaba has quit IRC09:05
*** akrivoka has joined #tripleo09:11
*** jtomasek_ has joined #tripleo09:12
*** nijaba has joined #tripleo09:12
*** nijaba has quit IRC09:12
*** nijaba has joined #tripleo09:12
gfidented0ugal, do you see issues backporting down to liberty or even kilo?09:16
mariosshardy_: heh which one09:20
d0ugalgfidente: no, it looks simple enough09:21
gfidented0ugal, ack I remember the tests were a bit different before but the functionality should be the same right?09:21
d0ugalgfidente: Yeah, sounds correct. It's been a while tho'09:22
gfidented0ugal, as far as I remember we need to be careful on landing the template change first otherwise heat will fail validation on a non-existing parameter09:23
gfidentebecause initially we weren't using parameter_defaults09:24
gfidentebut that's all09:24
d0ugalgfidente: yup, well remembered09:25
gfidenteack tks09:25
*** olap has quit IRC09:27
*** olap has joined #tripleo09:29
*** jtomasek_ has quit IRC09:32
openstackgerritEmilien Macchi proposed openstack/instack-undercloud: Deploy and Upgrade Ironic to run in mod_wsgi
openstackgerritMartin André proposed openstack/tripleo-common: Validations actions and workbook
*** zoliXXL is now known as zoli|afk09:43
*** dsariel has joined #tripleo09:43
openstackgerritMerged openstack/tripleo-heat-templates: Ensure manage_service and enabled are false for Pacemaker
*** stendulker has joined #tripleo09:46
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci: Add timeout to get_host_info task
*** electrofelix has joined #tripleo09:49
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: Remove commented out os-cloud-config
openstackgerritMerged openstack/puppet-tripleo: Remove manage_service and enabled from TripleO manifests
openstackgerritMartin André proposed openstack/tripleo-common: Validations actions and workbook
*** zoli|afk is now known as zoli09:56
openstackgerritDougal Matthews proposed openstack/tripleo-common: Deployment workflows for remote execution, etc.
sshnaidmgfidente, hi10:08
gfidentesshnaidm, pon10:08
*** zoli is now known as zoli|lunch10:09
sshnaidmgfidente, looking on this patch  - is 13386 the right port? I see tempest tried to connect to 13786 :
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci: WIP: DONT MERGE Running tests
shardy_jistr, marios: Hey I'm trying to make a start breaking down the upgrade related tasks - are you OK if I raise a placeholder blueprint for composable upgrades?10:15
gfidentesshnaidm, 13386 is where haproxy binds yes, where is tempest getting 13786? was the job started before the fix?10:15
shardy_e.g adding interfaces to composable services to enable per-service definition of what happens during upgrade, as we discussed at summit10:15
jistrshardy_: we just had a call about this earlier with marios and social, we'd probably do it ourselves if that's ok?10:15
jistrshardy_: well we then also discussed on #tripleo that it might not be the best approach to take though, as we're unable to control ordering within a single "hook"10:16
shardy_jistr: Sure - if possible can you raise the blueprints today, even if you set them all to drafting state and we then work out the details over the next few days?10:16
mariosshardy_: hey... so the proposal is to rework the text a bit in to act as an umbrella blueprint for newton10:16
mariosshardy_: yeah exactly10:16
openstackgerritCarlos Camacho proposed openstack/tripleo-docs: Composable roles within services - NTP
mariosshardy_: we (me jistr and social if that is lucas?) have taken on to file a blueprint each with specific things like10:17
shardy_marios: yup - basically I'm trying to get a prioritized list of BPs for N-2, so we can ensure we have some features we can track and focus on for the next milestone10:17
socialmarios: yep10:17
sshnaidmgfidente, haproxy in nonha jobs also?10:17
shardy_we didn't deliver any BPs for N-1, which probably is an indication that our BPs are all to large in scope I think10:18
gfidentesshnaidm, haproxy in nonha as well yes, nonha is the only job using tls10:18
mariosshardy_: like fix/evolve the existing workflow to upgrade mitaka to newton upstream , fix/evolve the minor version package updates workflow for newton and rework the upgrades procedure entirely to enable independent upgrades of services rather than nodes10:18
gfidentesshnaidm, so it's the only one using 13386 instead of 838610:18
mariosshardy_: we will then add those three blueprints as dependents to
shardy_marios, jistr: ack, that's perfect, thanks10:18
mariosshardy_: overall we can target the umbrella to n3 i guess (since it covers the upgrades work over newton)10:18
shardy_sorry to chase, I'm being chased myself for status here, so just trying to get things organized :)10:18
mariosshardy_: and each of the individual ones to whatever10:19
marioswell n2 probably10:19
shardy_marios: yup, that's fine, but it'd be good to identify the first steps as separate BPs and target them to N210:19
gfidentesshnaidm, 13786 is manila fwiw10:19
mariosshardy_: yeah we have an ai to at least file the blueprints toady. I'll ping you later before reworking the existing (to serve as umbrella) blueprint10:19
shardy_marios: ack, thanks that would be great10:20
shardy_you can just add all the new BPs as depenencies from the current umbrella one as you said10:20
mariosshardy_: great as jistr said we agreed on these various points earlier in a call - i was going to update you/catchup about that reworking later10:20
jistrshardy_: yeah i think the high level plan is to amend the existing monolithic upgrades first, so that we have a ready and working "plan B", as we don't have a concept how to refactor the upgrades just yet. What we've discussed at summit might probably corner us into situations where some upgrades wouldn't be possible to do.10:20
mariosshardy_: am glad it sounds like a good way to go10:21
shardy_jistr: +1, getting the existing upgrade script working, then decomposing sounds like a good plan10:21
mariosshardy_: yeah - we will have to deal with composable services as well as the evolved pacemaker arch there so even that may not be as simple/easy as it may appear10:23
shardy_what would be good is improving the CI upgrade job so we can test/prove the monolithic implementation10:24
shardy_then, we can make incremental improvements with assurance things still work10:24
shardy_evidently there are still walltime challenges there, but perhaps we can think about if/how that might be possible e.g with cached images etc10:24
shardy_probably we need an actually-test-upgrades bug for CI10:25
mariosshardy_: yeah the initial deploy in full (image build etc) doesn't have to (also) be exercised for the upgrade - so using images would proably save quite a bit of time. The greater challenge however is automating the current upgrades procedure (jistr has wip docs at btw) - in the worst case with a script :/ possibly with mistral but I am not sure yet about that10:27
openstackgerritMerged openstack/python-tripleoclient: Generate a password for the Mysql 'clustercheck' user
matbuhey guys, i have a ansible role for upgrade10:28
shardy_marios: as a first step can we just run the controller upgrade?10:28
mariosshardy_: the fact of the matter is it is an operator driven procedure involving 3 heat stack updates and the 'upgrade_non_controller'sh' for each non controller (except cinder)10:28
matbuit could be use in CI10:28
jistrmarios: i think script should be fine10:29
mariosshardy_: jistr i guess for the purposes of ci pinning nova may not be a big deal (thinking of just upgrading the controller)10:29
jistrmarios: i mean, i'm not sure if it makes sense to automate the current procedure in mistral if the procedure might change a lot10:29
jistre.g. will we even be using heat for the upgrades going forward...10:29
openstackgerritCarlos Camacho proposed openstack/tripleo-docs: Composable roles within services - NTP
mariosshardy_: jistr i.e. without running sthing to pin nova-compute rpc version on the compute10:30
jistrmarios, shardy_: re just controller upgrade -- not sure if we could run converge afterwards. Maybe it would work in many cases, but it's not the general case. At the very least we'd feed into the converge step that it needs to keep nova pinned.10:30
jistrmarios: without pinning nova might stop working (i.e. pingtest after upgrade might fail)10:31
mariosjistr: i think we'd need to run converge ... for delivering sthing like rabbit password change to all nodes10:31
jistrso i'm wondering if we can skip the compute upgrade10:31
mariosjistr: e.g. if there is sthing which lands in /etc/foo/foo.conf on controller which also affects a service running on _other_node_10:31
jistrbut compute upgrade should be reasonably fast actually10:31
shardy_Ok, few things to work out here - would it be a good idea to raise a bug, outline the issues, and discuss the options?10:32
shardy_we can target that at n2 and ensure someone has bandwidth to work on it10:32
mariosshardy_: yeah... so jcoufal mentioned someone was/is already looking at the ci for upgrades10:32
mariosjistr: do you remember?10:33
jistrnot sure if it was "someone is" or "someone will be"10:33
jistralso not sure who "someone" was :))10:33
shardy_This is why I want a bug - it makes it very easy to see who's assigned to the work :)10:33
mariosshardy_: OK I can file that10:33
jistrmarios: thanks. I can also go through the bug once it's filed and post more info if i can think of something relevant on it.10:34
mariosshardy_: jistr thanks for chat. matbu sorry missed your ping ... link to upgrades role? (I think i saw it /you mentioned it few days ago) interesting to keep in mind as a candidate10:35
* jistr lunch now10:35
* marios irc quiet time slower response so i can actually do some of the stuff I said i would :)10:35
sshnaidmgfidente, there is a problem, sahara listens to 13386, but configured as it was on 13786:
matbumarios: no worries10:37
matbumarios: jistr shardy_
sshnaidmgfidente, you can see in logs:
matbumarios: jistr shardy_ it could do minor or major upgrade10:37
shardy_ needs a +2/A - easy review (passing CI)10:37
sshnaidmgfidente, I think that's why tempest search for it in 13786 port..10:37
matbumarios: jistr shardy_ on top of tripleo-quickstart deployment for example10:38
sshnaidmadarazs, hi10:38
gfidentesshnaidm, I think it's an older version of tht which doesn't include the change you pointed to10:38
adarazssshnaidm: hello10:38
gfidentethe purpose of that fix was precisely to point the public_url to 1338610:38
gfidentewhich was in fact incorrectly set to 13786 as per logs10:39
gfidentesshnaidm, how old is the tempest run?10:40
*** ccamacho is now known as ccamacho|lunch10:41
sshnaidmgfidente, about a week.. what it tht?10:41
*** thrash|g0ne is now known as thrash10:41
sshnaidmadarazs, I think you told that you had a problem with tempest configuration because there is no cinder, right? was it about quickstart?10:41
gfidentesshnaidm, looks like 4 weeks10:41
sshnaidmgfidente, 15-May-201610:42
adarazssshnaidm: I don't recall saying that. not recently anyway. maybe panda said it at some point?10:42
gfidentesshnaidm, ok we'd need to check what version of tht is there then10:42
sshnaidmadarazs, oh, maybe, I was sure somebody told this, sorry :)10:42
pandanot me10:42
gfidentesshnaidm, is it deploying mitaka/tht?10:43
sshnaidmgfidente, sorry, what is tht?10:43
*** dixiaoli has quit IRC10:43
sshnaidmgfidente, it's master10:43
sshnaidmgfidente, it's a regular nonha job, master branch10:43
sshnaidmgfidente, I'll try to see tht version ther10:44
openstackgerritMerged openstack/python-tripleoclient: Remove commented out os-cloud-config
gfidentesshnaidm, back in a few, going to grab something for lunch10:44
sshnaidmgfidente, I think this one: openstack-tripleo-heat-templates-2.0.1-0.20160511130113.0d3dace.el7.centos.noarch10:45
sshnaidmgfidente, sure, no critical10:45
*** mburned is now known as mburned_out10:48
*** mburned_out is now known as mburned10:50
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role
*** zoli|lunch is now known as zoli10:53
openstackgerrityolanda.robla proposed openstack/diskimage-builder: Add dib element to generate logical volumes
*** dtantsur is now known as dtantsur|brb10:59
*** florianf has quit IRC10:59
*** florianf_ has joined #tripleo10:59
*** stendulker has quit IRC11:08
*** morazi has quit IRC11:09
*** weshay has joined #tripleo11:09
mariosshardy_: jistr social
openstackLaunchpad bug 1583125 in tripleo "There is no 'major version' upgrades job for ci " [High,Triaged]11:16
mariosshardy_: i assigned n211:16
*** fzdarsky is now known as fzdarsky|lunch11:17
*** Goneri has joined #tripleo11:20
jaosoriorHey guys, do you know what the plan for HAProxy will be in the context of composable roles?11:21
jaosoriorwill there be a HAProxy role?11:21
mariosjaosorior: looks like11:22
marios and jaosorior11:22
jaosorior`hadn't seen that11:23
mariosjaosorior: np11:23
jaosoriorIt seems that we'll need some refactoring for the loadbalancer.pp in puppet-tripleo though11:24
jaosoriorright now we're passing that manifest this -> hiera('controller_node_ips')11:25
*** spredzy_ has quit IRC11:25
jaosoriorwhich assumes that the controllers contain all the load-balanced serviecs11:25
jaosoriorand that won't necessarily be the case anymore11:25
hdsdman_can someone provide me with a rhel register example11:25
hdsdman_i always get WARNING: Support for registering with a username and password is deprecated11:25
hdsdman_during overcloud deploy11:26
*** olap has quit IRC11:26
*** ccamacho|lunch is now known as ccamacho11:27
*** florianf_ is now known as florianf11:27
*** lucasagomes is now known as lucas-hungry11:38
*** julim has joined #tripleo11:40
jistrmarios: re LP bug -- perfect, thanks :)11:40
mariosjistr: ack thanks11:41
*** MaxPC has joined #tripleo11:43
jistrmatbu: good stuff! I linked your on the bug page ^^. This is pretty much what could be in the upgrade CI job for overcloud
jistrcc marios social ^11:44
jistrs/your/your repo/11:44
matbujistr: cool11:44
mariosjistr: ack thanks (yeah made sure to include the link to matbu role in the irc chat on bug )11:45
*** jpena is now known as jpena|lunch11:46
jistrwhich reminds me... i wonder if we could do the "new" upgrades with Mistral + Ansible rather than Heat + Bash. Seems like it might be a bit better fit.11:46
jistrbut it still doesn't address the ordering of the composable parts at all IIUC11:47
jistrfor that we'd need to perhaps do some Puppet tricks, or reimplement the world in Prolog ideally11:47
*** spredzy has joined #tripleo11:47
gfidentespredzy !11:48
jistro/ welcome :)11:48
spredzygfidente !11:49
spredzyjistr: o/11:49
jistri played with Prolog a bit over the weekend, and i got it to output ordering of actions pretty nice based on *totally arbitrary ordering rules* on input11:51
*** cmyster has quit IRC11:51
jistrbut i have no idea how to integrate it with anything in the real world just yet11:52
*** dtantsur|brb is now known as dtantsur11:52
jistrso meh.. :)11:52
*** rhallisey has joined #tripleo11:53
hewbroccaProlog yay!11:54
*** cmyster has joined #tripleo11:55
*** cmyster has quit IRC11:55
*** cmyster has joined #tripleo11:55
jistr...just in case we wanted to raise the TripleO entry barrier even higher :D11:55
hewbroccajistr: let's rewrite everything in elisp11:56
hewbroccaWhen emacs can deploy openstack, I will truly be able to retire11:56
rhalliseyhewbrocca, haha11:57
hewbroccaITYM (((((((((((:)))))))))))11:57
*** rodrigods has quit IRC11:58
*** rodrigods has joined #tripleo11:58
*** mcornea has quit IRC11:58
*** mcornea has joined #tripleo11:59
mariosbandini: i heard you like adjectives11:59
*** rbrady has joined #tripleo12:03
*** dprince has joined #tripleo12:05
mariosbandini: any better/newer version for
*** rbrady has quit IRC12:07
*** honza_ is now known as honza12:07
bandinimarios: you don't like the colors? (it should be fairly uptodate, although I had zero time to hack on lightweight lately)12:08
jaosoriorjistr haha Prolog is fun dude. Never got to do anything useful but the concepts of it are interesting12:10
*** rbrady has joined #tripleo12:11
*** trown|outtypewww is now known as trown12:12
*** jayg|g0n3 is now known as jayg12:12
openstackgerritDougal Matthews proposed openstack/tripleo-heat-templates: Use the correct capitalization of OpenStack
bandinijistr: I think it is one of the most downloaded files on my website (I guess students taking their assignments off the the internet ;)12:13
jistrjaosorior: yea i kinda liked playing with it. Expressing a problem like -- "I have these actions, and these orderings between them. This few have already been executed, the others are still pending, tell me which one to execute now." -- has been pretty concise actually. It was about 1 screen of code. No way one could do this in an imperative language as concise.12:13
jistrI have it on my home computer, might fetch it and paste it later just to show what it looks like. It's pretty interesting imo.12:14
openstackgerritDougal Matthews proposed openstack/tripleo-image-elements: Use the correct capitalization of OpenStack
socialjistr: get prolog to generate the ansible to run for upgrade12:14
mariosjistr: fwiw, the prof that taught us prolog was a genius. he was also a raging alcoholic. just saying12:15
jistrmarios: haha hoping i won't get that far12:15
bandiniI like the second part already12:15
mariosbandini: thanks just checking cos I'll be referring to it12:15
hewbroccamarios: sounds like my kinda guy, he looking for a job?12:15
mariosbandini: can you check my comment on the spec light review?12:16
marioshewbrocca: haha12:16
mariosbandini: (i used that word again... )12:16
bandinimarios: ahah right, I need to review the terminology all over12:17
bandinimarios: thanks for raising the point, it is important12:17
mariosbandini: :D12:17
jistrsocial: been thinking about this a bit... the exact integration is still a bit hazy for me though. E.g. we'd probably want to orchestrate particular tasks rather than roles, which isn't very inviting.12:17
*** zoli is now known as zoli|afk12:19
socialjistr: well I think the upgrades and updates should be handled by same code as the deployment, only thing different should be that the packages get updated instead of installed >.>12:20
socialjistr: I'd even trigger db-sync on update the same way installation would as db-syncs are idempotent12:21
*** fultonj has joined #tripleo12:21
gfidentejaosorior, hey I just noticed the version of enable-tls in tripleo-ci is broken as it misses fixes we put only in tht12:21
gfidentejaosorior, do we need to pass EndpointMap from tripleo-ci enable-tls.yaml or can we get that from the version of enable-tls.yaml in tht?12:22
gfidenteI know we need the certs, but can we drop the EndpointMap param?12:22
gfidentejaosorior, looks like in CI we need to make it point to IP_ADDRESS instead of CLOUDNAME to make it work?12:24
*** morazi has joined #tripleo12:24
jistrsocial: we actually do trigger db-sync every stack update and count on it to not do anything IIRC. It's just that e.g. during minor update we want no downtime, so things need to be applied node-by-node rather than all at once. During major upgrade it's the opposite. We must stop all services at the same time because DB syncs that actually do something can only be safely run when services are switched off.12:25
*** Goneri has quit IRC12:25
jistrsocial: perhaps that idea could be achievable though and worth investigating further. Just mentioning that we have some constraints to think of, sooner rather than later :)12:26
bandinimarios: how about "next generation" ?12:27
bandinipino|work: you should be proud that I manage to quote star trek12:27
mariosbandini: like it12:28
mariosbandini: but more like "next generation! (tm)"12:28
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron server
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron metadata service
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron l3 service
*** links has quit IRC12:31
*** lucas-hungry is now known as lucasagomes12:32
*** links has joined #tripleo12:35
*** Goneri has joined #tripleo12:37
openstackgerritMarios Andreou proposed openstack/tripleo-heat-templates: Composable manila tripleo-heat-templates side
sshnaidmtrown, hola12:43
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role
trownsshnaidm: hi12:44
sshnaidmtrown, the quickstart doesn't install cinder, right? do you know maybe reasons for it?12:45
*** rlandy has quit IRC12:45
*** akshai has joined #tripleo12:45
sshnaidmtrown, overcloud, of course12:45
trownsshnaidm: umm... it definitely install cinder12:45
*** rlandy has joined #tripleo12:45
sshnaidmtrown, ook, let me check again..12:45
trownsshnaidm: the overcloud is just a normal deploy using `openstack overcloud deploy`12:46
trownthere isnt (or at least shouldnt be) anything quickstart specific about deploy12:46
jaosoriorgfidente: That's the case, we need to change the default there12:47
jaosoriorwhich will be a bit weird12:47
*** akshai_ has joined #tripleo12:47
gfidentejaosorior, acl12:47
jaosoriorgfidente: Well... that was part of the reason why I was trying to simplify things with this CR
jaosoriorthere are too many places that need maintenance for the TLS-related things12:48
*** panda has quit IRC12:48
*** itamarl has joined #tripleo12:48
itamarlova list12:48
sshnaidmtrown, yeah, mea culpa.. tempest configuration just failed because didn't find something with cinder, did you see something like this?12:49
*** panda has joined #tripleo12:49
jaosoriorbut I guess I'll be abandoning that CR soon12:49
jaosoriorunless there would be a compromise to get that accepted12:50
*** akshai has quit IRC12:50
pabelangerderekh: replied to ML about AFS.  If you'd like to see the comments12:50
trownsshnaidm: ya, need the cinder_tests package added to undercloud image12:51
trownor cinder-tests actually I think12:51
sshnaidmtrown, yeah, thanks, found it.. python-cinder-tests12:52
trownsshnaidm: ya thats it12:52
sshnaidmtrown, I think I can improve a bit the tempest support in quickstart.12:52
trownsshnaidm: surely, go for it!12:53
sshnaidmtrown, cool, will do then12:53
trownsshnaidm: I would like to have the nonha jobs in the promote running tempest12:53
trownsshnaidm: but it is always breaking because of silly stuff like that12:53
sshnaidmtrown, you mean in centos ci?12:53
trownsshnaidm: ya
sshnaidmyeah, sure12:54
socialjistr: marios: to minor version updates, what about system library/kernel update?12:58
*** liverpooler has quit IRC13:01
*** athomas has joined #tripleo13:02
*** jpena|lunch is now known as jpena13:02
openstackgerritGiulio Fidente proposed openstack-infra/tripleo-ci: Sync up EndpointMap in enable-tls with default enable-tls from THT
EmilienMgfidente: %{hiera('')}13:04
jistrsocial: kernel updates are handled by a subsequent reboot, we don't automate that part currently13:06
gfidenteEmilienM, ack thanks, I'll try that out :)13:07
EmilienMgfidente: well, thanks to you for your help this morning and also to marios13:07
socialjistr: so in minor update we want to run yum update; but that updates everything even system libraries qemu and so on, which means for example that libvirt can get updated and libvirtd will be reloaded in process13:07
jistrshardy_, marios, social, jcoufal: submitted the per-service upgrade blueprint and filled the whiteboard with some questions/investigation points. Might be worth a read to get a clearer picture of the scope of what we're talking about, and also please if you can think of other questions that need answering but i might have missed them, feel free to add them.
shardy_jistr: will do, thanks!13:12
openstackgerritGiulio Fidente proposed openstack-infra/tripleo-ci: Sync up EndpointMap in enable-tls with default enable-tls from THT
mariosjistr: thanks jistr13:16
*** links has quit IRC13:16
*** mbound has quit IRC13:17
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map
mariosjistr: shardy_ social my one at
*** zoli|afk is now known as zoli13:27
*** akshai_ has quit IRC13:27
hdsdman_no possibility to add 14:52 < trown> sshnaidm: ya thats it13:28
*** pino|work has quit IRC13:28
hdsdman_sorry, no posibility to rhel_reg_auto_attach as command argument for overcloud deploy?13:28
sshnaidmhdsdman_, why?13:28
hdsdman_sshnaidm: sorry, still something in my paste buffer :/13:29
openstackgerritMatt Young proposed openstack/tripleo-common: WIP: add support to image build yaml input to handle env vars
*** jprovazn has quit IRC13:30
*** akshai has joined #tripleo13:31
*** zeroshft has joined #tripleo13:32
*** zeroshft has quit IRC13:32
*** zeroshft has joined #tripleo13:33
mandredprince: hi, regarding your comment in, do you have a preference where the new keypair for validations should go?13:33
mandredprince: can we rely on instack-undercloud's script to create the keypair?13:33
openstackgerritMichele Baldessari proposed openstack/tripleo-heat-templates: Tighten the access rules for galera
*** athomas has quit IRC13:38
hdsdman_mandre: i had issues with rabbitmq credentials while rerunning the undercloud script, i had to refeed the old credentials13:38
*** athomas has joined #tripleo13:39
mandrehdsdman_: I'm a bit out of context, what's your issue?13:40
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.
jcoufaljistr: thanks!13:42
hdsdman_mandre: just fyi, the issues i had with the script13:43
mariosjcoufal: sorry missed to include you earlier, the 'adapt existing workflow' one is at
mariosjcoufal: (I also updated the google doc)13:44
*** egafford has joined #tripleo13:44
jcoufalmarios: awesome!13:44
mariosshardy_: around? ok for me to go ahead and update ? proposal - we already started tracking thigns there13:49
mariosshardy_: hmm probably raw is more readable
shardy_marios, sure - go for it13:50
shardy_feel free to juggle milestones to make a plan you feel is achieveable also13:51
*** tzumainn has joined #tripleo13:51
*** fzdarsky|lunch is now known as fzdarsky13:51
mariosshardy_: done thx will revisit tomorrow morning13:51
*** zoli is now known as zoli|brb13:51
mariosjcoufal: FYI tracker at
mariosjcoufal: oh yeah is the same one already in the googledoc, just updated to serve as tracker13:52
mandrehdsdman_: I don't recall talking about an issue with you, sorry :-/13:52
mariosshardy_: my thought was to target the umbrella to N3?13:52
mariosshardy_: as in the goal is to get all or as much of these sub things in for newton !#13:53
*** ccamacho has quit IRC13:53
mariosjistr: social fyi updated ^^^13:53
mariossocial: can you add your one in the UPDATEME?13:53
mariossocial: and add it as a dependency13:54
*** ccamacho has joined #tripleo13:54
socialmarios: ack13:55
jistrmarios: nice, thanks13:56
*** chem`` has joined #tripleo13:57
*** absubram has quit IRC13:59
openstackgerritMichele Baldessari proposed openstack/tripleo-specs: Add initial next generation architecture spec
bandinimarios: ^ ;)14:00
*** pradk has joined #tripleo14:00
openstackgerritAthlan-Guyot sofer proposed openstack/tripleo-heat-templates: WIP: integration of the new puppet pacemaker.
mariosbandini: SHIPIPT14:02
marioswe couldn't even shipt it14:02
mariosit was so advanced14:02
*** yamahata has joined #tripleo14:03
mariosbandini: you must make sure to include 'new frontiers' in the description of the spec/opening paragraph please on next pass14:04
bandiniwe need a star trek consultant14:05
thrashTo boldly go...14:05
bandiniwhere is pino when I need him14:05
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Dump IPs configuration as hieradata
*** pino|work has joined #tripleo14:06
*** yolanda has joined #tripleo14:08
yolandahi, it's first time i run tripleo, and i'm hitting a problem. One of the puppet bits tries to run semodule, but my stack user doesn't seem to be allowed to do it14:09
yolandaanyone had this problem?14:09
*** links has joined #tripleo14:10
bandiniyolanda: usually the stack user in the undercloud has full sudo powers, so it should be able to run this stuff as root without asking for a password14:11
shardy_gfidente: did you make any progress wrt sharing data between nodes, e.g the ML thread EmilienM started?14:12
shardy_I'm trying to move SwiftDevicesAndProxyConfig out of overcloud.yaml, and it's the same problem I think14:12
gfidenteshardy_, so we discussed that a little on a call14:13
*** zoli|brb is now known as zoli14:13
gfidenteand allNodesConfig seemed the quickest win14:14
*** shardy_ is now known as shardy14:14
*** lblanchard has joined #tripleo14:14
shardygfidente: Hmm, possibly, it doesn't provide a drop-in replacement for stuff like SwiftDevicesAndProxy config tho14:15
shardywe really need a way to distribute that data to only the nodes running the proxy and swift services14:15
shardyand the current composable services interface doesn't provide a place where we can wire data between ResourceGroups like that14:17
shardymaybe dprince has some ideas, I may try a few things also14:17
*** mbound has joined #tripleo14:17
ccamachohey marios, quick question, in overcloud-resource-registry-puppet.yaml (From your THT memcached patch) why are you calling a pacemaker yaml file? You are doing that in puppet-pacemaker.yaml. Should you call there puppet/services/memcached.yaml instead??? (Without pacemaker)14:18
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-tripleo: WIP: integration of the new puppet pacemaker.
yolandabandini, i think i got a false positive, i had my install failing on 'dib-run-parts', '/usr/libexec/os-refresh-config/configure.d' , but i guess it's run by root?14:18
*** yamahata has quit IRC14:18
*** tserong has quit IRC14:19
mariosccamacho: thanks man can you add it to the review - have a call starting in a sec14:19
mariosccamacho: i will have  a look14:19
gfidenteshardy, so for both the devices list and the memcached list we need to collect the list of nodes running the service14:19
bandiniyolanda: don't recall the exact detail tbh14:19
mariosccamacho: may well be a nit you have found14:19
*** mbound has quit IRC14:23
*** dkelson has quit IRC14:24
*** tserong has joined #tripleo14:25
*** Guest92292 is now known as mgagne14:25
*** mgagne has quit IRC14:25
*** mgagne has joined #tripleo14:25
gfidenteshardy and yes ideally this is the same problem we have to gather the list of nodes hosting redis, rabbit, galera ...14:25
gfidenteshardy, or the list of nodes hosting an api to populate the haproxy backends list14:25
*** dsariel has quit IRC14:30
hdsdman_can somebody have a quick look? i got issues with the network configuration on one controller during deploy14:33
gfidenteshardy, I was thinking to collect in allnodesconfig the list of nodes->services deployed14:33
hdsdman_according to os-net-config all is fine but still my is fucked14:33
shardygfidente: ah, yeah that may work14:34
socialmarios: I think I can't change UPDATEME on
socialmarios: jistr: shardy:
mariosack thanks social14:35
*** marios is now known as marios|call14:35
*** tserong has quit IRC14:37
*** tserong has joined #tripleo14:38
*** links has quit IRC14:39
jistrsocial: cool, thanks14:46
*** coolsvap has quit IRC14:53
*** coolsvap has joined #tripleo14:54
*** dsariel has joined #tripleo14:56
hdsdman_how is this possible?14:57
*** jaosorior has quit IRC14:59
*** jaosorior has joined #tripleo14:59
*** jistr is now known as jistr|mtg15:02
openstackgerritMatt Young proposed openstack/tripleo-common: WIP: add support to image build yaml input to handle env vars
chem``hi, I've noticed that neutron code is kinda duplicated in puppet-tripleo and THT, like this for insntance and
chem``the whole ovs file is duplicated.15:06
chem``I understand that it's the preparation of the composition layer, but I'll ask anyway if is this expected ?15:06
shardywe're in the process of switching over to using the new puppet-tripleo profiles15:07
shardyafter which the duplication in t-h-t will be gone15:07
chem``oki, so meanwhile, both code exists at different place, unlike the keystone and glance migration that has been done in one go15:08
shardychem``: the support has to land in puppet-tripleo first, then t-h-t switches over to use it15:08
chem``shardy: oki then.  Thanks.15:08
shardyglance and keystone were the same, only they have landed and the neutron patches are still under review15:08
*** masco_ has quit IRC15:09
*** diazjf has joined #tripleo15:10
chem``shardy:, ack15:10
*** dmacpher has joined #tripleo15:10
hdsdman_guys, can someone help me a bit to debug this?15:11
hdsdman_os-net-config states that no changes are needed -> network configuration OK15:12
hdsdman_however after/during software deployment my network gets screwed (virtual nics and ports dissepear)15:12
shardyhdsdman_: that just means it applied the config you asked for, it doesn't necessarily follow the configuration is OK15:12
hdsdman_while os-net-config is still running happely and states that no changes are needed15:13
hdsdman_shardy: thx for feedback. my network is correct during first boot/deploy15:13
shardyhdsdman_: and it breaks on a subsequent update?15:14
hdsdman_manually running os_net-config gives me exirt code 0 o_O15:14
hdsdman_shardy: yes and it is only happeing on one controller node, the other ones are correct15:14
marios|callsocial: i'll update it now15:15
marios|callsocial: thx15:15
*** athomas has quit IRC15:16
hdsdman_than the network tests fail towards the ip of the deleted virtual interface15:16
hdsdman_and deploy fails15:16
marios|callsocial: i think the dependency should be the other way around15:16
marios|callsocial: i.e. add to
marios|callsocial: done15:17
*** athomas has joined #tripleo15:18
EmilienMdprince, gfidente: can we land rabbit? it works now15:18
gfidenteEmilienM ;)15:18
EmilienM and
hdsdman_they use the same template15:19
hdsdman_end result ->
*** marios|call is now known as marios15:19
*** jistr|mtg is now known as jistr15:19
openstackgerritDan Prince proposed openstack/tripleo-common: WIP: Add baremetal node introspection workflows.
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Dump IPs configuration as hieradata
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-tripleo: WIP: integration of the new puppet pacemaker.
*** mcornea has quit IRC15:23
dprinceEmilienM: +2 on both from me15:23
dprinceEmilienM: thanks15:23
EmilienMdprince: thx15:23
EmilienMnext: haproxy15:23
*** athomas has quit IRC15:23
*** ebarrera has quit IRC15:23
*** ramishra has quit IRC15:24
*** dprince has quit IRC15:24
*** rcernin has quit IRC15:24
dtantsurtrown, I know you also didn't like the bulk introspection principle, could you please weigh in on ? I'm trying to prevent this terrible bulk introspection to carry over to new shiny API..15:26
*** rhallisey has quit IRC15:27
trowndtantsur: sure I have not played with mistral yet, that will be a good opportunity to try it out15:27
*** ccamacho has quit IRC15:28
*** rhallisey has joined #tripleo15:28
openstackgerritMerged openstack/puppet-tripleo: Composable role for RabbitMQ
EmilienMgfidente: ok to +A
gfidenteEmilienM, ack15:30
socialmarios: yeah I couldn't change that15:31
mariossocial: ack, i added to the description and changed the dependency order too thanks man15:31
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.
*** dtrainor has joined #tripleo15:32
*** mbound has joined #tripleo15:34
openstackgerrityolanda.robla proposed openstack/diskimage-builder: Add dib element to generate logical volumes
*** paramite is now known as paramite|afk15:36
*** dsariel has quit IRC15:37
*** dtrainor has quit IRC15:37
*** dmk0202 has quit IRC15:37
*** dtrainor has joined #tripleo15:37
*** dmk0202 has joined #tripleo15:37
*** leanderthal is now known as leanderthal|afk15:37
*** tremble has quit IRC15:38
*** mikelk has quit IRC15:40
*** jprovazn has joined #tripleo15:41
*** aufi has quit IRC15:42
*** thrash is now known as thrash|f00dz15:43
openstackgerritMatt Young proposed openstack/tripleo-common: WIP: add support to image build yaml input to handle env vars
*** dmk0202 has quit IRC15:51
openstackgerritMartin André proposed openstack/tripleo-common: Validations actions and workbook
openstackgerritMartin André proposed openstack/tripleo-common: Allow running validation against different plans
mariosmmmm brno thursday morning breakfasts15:53
* marios hungry almost time for dinner here15:53
mariosjistr: reserve some donuts for me pls15:53
*** itamarl has quit IRC15:54
jistrmarios: sure, i'll put some aside and give them to you when we see each other (next summit?)15:54
mariosjistr: yeah together with that tshirt if it arrives from the docs guys and also some mystery meat wrap from across the road please15:55
*** penick has joined #tripleo15:55
mariosjistr: that'd be grand!15:55
d0ugalhmm, os-refresh-config just failed for me during an openstack undercloud install.15:57
d0ugalAre others having issues with this?15:57
d0ugaloh, actually, I think it is a puppet error15:57
*** ccamacho has joined #tripleo15:57
d0ugalso I guess I should use the source puppet. /me looks for the correct env var15:57
jistrmarios: :D15:58
*** dkelson has joined #tripleo15:58
*** xinwu has joined #tripleo15:58
*** mbound has quit IRC16:00
*** paramite|afk is now known as paramite16:00
*** jistr has quit IRC16:00
* marios EOD bai16:00
*** florianf has quit IRC16:01
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.
*** sthillma has joined #tripleo16:03
*** ifarkas has quit IRC16:05
openstackgerritMartin André proposed openstack/instack-undercloud: Validation to detect if undercloud disks use 512e
openstackgerritMartin André proposed openstack/instack-undercloud: TripleO Ansible validations
*** sthillma has quit IRC16:08
*** mbound has joined #tripleo16:11
*** zoli is now known as zoli_gone-proxy16:13
*** mbound has quit IRC16:13
*** mbound has joined #tripleo16:13
openstackgerritMartin André proposed openstack/tripleo-common: Reference upstream repo for ansible validations
*** abehl has quit IRC16:20
*** devvesa has quit IRC16:23
*** xinwu has quit IRC16:28
*** xinwu has joined #tripleo16:29
*** thrash|f00dz is now known as thrash16:30
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Switch Cinder Api/Scheduler/Volume to composable roles
*** xinwu has quit IRC16:33
openstackgerritSteven Hardy proposed openstack/puppet-tripleo: Add puppet profile for swift proxy/storage
*** leanderthal|afk has quit IRC16:43
dkelsonwhat's the easiest way to test a OS::TripleO::NodeExtraConfigPost that uses puppet without having to re-deploy for each iteration?16:46
shardydkelson: Just run openstack overcloud deploy without deleting the stack between attempts16:47
shardypuppet will be re-applied on all nodes, so it will take a while, but you won't have to wait for all the nodes to be rebuilt16:48
dkelsonshardy, thanks16:49
openstackgerritSteven Hardy proposed openstack/tripleo-heat-templates: Add step to ObjectStorage RingBuilder deployment
*** dprince has joined #tripleo16:50
gfidenteshardy, I think for the rings we need to change strategy a little16:52
gfidentegenerate it on one node and distribute to the others16:52
*** anshul has joined #tripleo16:52
gfidentebecause the resulting rings depend on historical data so they differ when generated on new nodes after scale up16:53
*** anshul is now known as Guest5522516:53
shardydkelson: there is one other approach which is to log into a node, find the failing manifest in /var/liber/heat-config/heat-config-puppet and apply it manually, then modify it until it works and copy the result back into the template16:53
*** mgarciam has joined #tripleo16:54
*** dtantsur is now known as dtantsur|afk16:54
shardygfidente: Hmm, so the ringbuilder stuff is more broken than I thought then16:55
*** mgould has quit IRC16:55
gfidenteyeah we got a BZ for this16:55
gfidenteit works as long as you don't scale up16:55
*** akshai has quit IRC16:56
shardygfidente: Ok, what I just posted was simply a fix for initial deployment failing16:56
gfidenteshardy, the puppet module has an rsync class I was looking into16:56
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy Loadbalancer as a composable role
gfidenteas that could do the sync across nodes16:56
shardydprince added a step conditional to the ringbuilder manifest recently which we don't pass via the ObjectStorage post config16:56
*** akrivoka has quit IRC16:56
shardygfidente: Ok, thanks for the info, sounds like we can fix the deployment then look into that16:57
EmilienMgfidente: the puppet class in puppet-swift that does cross nodes stuff requires puppetdb16:57
gfidenteEmilienM, yeah I remember we discussed that16:57
dkelsonshardy, I'm a bit new to puppet and tripleo so pardon the novice questions. I have some files I want to install onto the overcloud controller. Can I use a puppet File resource? What would I use for the source => ?16:57
*** diazjf has quit IRC16:58
gfidenteshardy, on the resource naming/step naming match16:59
gfidentedoes it matter if we skip _Step2 and call it _Step3 ?16:59
shardygfidente: probably not, but that seemed worse to me as it's more user-visible17:00
gfidenteah right the resource list17:00
*** derekh has quit IRC17:02
dprinceshardy: which patch are we talking about here?17:02
shardydkelson: np, yes you could use puppet to do that, or just a shell script, depends on your preference17:03
openstackgerritMerged openstack/tripleo-heat-templates: composable neutron l3 service
dprinceshardy: thanks, looking17:03
shardyI just tried to deploy with an ObjectStorage node via --swift-storage-scale 1 and hit a problem17:03
shardyI think we just need to feed the step into swift-storage-post17:04
openstackgerritMerged openstack/tripleo-heat-templates: composable neutron metadata service
dprinceshardy: lgtm, feel free to send once CI passes I think17:04
shardymaybe we can add coverage to CI now we've got a bit more capacity17:04
shardydprince: thanks!17:04
dprinceshardy: actually, on the CI front I'd rather not cover this17:04
dprinceshardy: ideally, eventually when we are more composable all the roles would be the same...17:05
dprinceshardy: so I'd rather push for that I think instead17:05
shardyYeah, I guess so17:05
trownslagle: the removal of the undercloud-package-install element changed the default for puppet to source:
openstackLaunchpad bug 1583263 in tripleo "The default for installing puppet modules in instack-undercloud was changed to source" [High,Triaged]17:05
shardyFWIW I just pushed a first cut of breaking the proxy and storage manifests into a puppet-tripleo profile17:05
shardyI'll try wiring those in via composable services tomorrow17:05
dprinceshardy: nice, could you add them to the etherpad?17:06
shardydprince: done so already, will add t-h-t patches tomorrow17:07
dprinceshardy: great. I see it17:07
*** akshai has joined #tripleo17:08
dkelsonshardy, on the undercloud, where do I place a file so that I can reference it with source => 'puppet:///modules/myfile' inside of a Puppet File resource?17:08
*** mgould has joined #tripleo17:08
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Pass enabled and manage_service to Neutron/L3 when Pacemaker
*** absubram has joined #tripleo17:11
slagletrown: ok, i think we ought to fix that in the puppet-modules element, to respect the default install type17:12
*** mgould has quit IRC17:12
slaglei can work on it in a few if you aren't already17:12
dprinceshardy: hey, so at the end of the Neutron patches, when we finish the vendors decomposition too I'll push one big "remove all the unused neutron parameters patch"17:12
shardydprince: Ok, that wfm, I'm just pushing to remove them because it makes the custom roles thing easier17:12
shardyI'll remove my objection17:13
dprinceyep, I get you17:13
trownslagle: cool, just noticed it while working on ironic packaging17:13
dprinceshardy: I am fine to fix it in my patch now17:13
trownslagle: I had not started on a fix yet17:13
dprinceshardy: just wanted you to know there are several more unused parameters coming too17:13
shardydkelson: Hmm, that probably won't work unless you can inline the file and reference it via a heat get_file17:15
shardye.g look at puppet/controller.yaml where we reference the static hieradata files:17:15
shardyget_file is resolved on the undercloud while processing the templates, and then formatted into the resulting deployment to the nodes17:16
*** fzdarsky is now known as fzdarsky|afk17:16
shardya potentially easier approch is to use deploy artefacts:17:16
shardythat will upload a tarball to swift, then simply unpack it on each node on deployment17:17
shardyno template changes needed at all17:17
dkelsonok, that might be useful17:17
trowndkelson: makes that even easier17:17
trowncheck my last comment though... using it is a bit fickle wrt trailing/leading slashes17:18
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Dump IPs configuration as hieradata
* shardy has to go now, see you tomorrow all17:19
*** shardy has quit IRC17:19
*** jpena is now known as jpena|off17:22
*** Guest55225 has quit IRC17:25
openstackgerritMerged openstack/instack-undercloud: Don't use ironic-conductor element
*** fragatina has quit IRC17:31
*** ohamada has quit IRC17:32
*** xinwu has joined #tripleo17:34
*** athomas has joined #tripleo17:36
*** trown is now known as trown|lunch17:36
*** rhallisey has quit IRC17:36
*** rhallisey has joined #tripleo17:37
*** electrofelix has quit IRC17:37
*** lucasagomes is now known as lucas-dinner17:44
*** akshai has quit IRC17:53
*** akshai has joined #tripleo17:53
*** pkunicki has joined #tripleo17:54
*** gfidente is now known as gfidente|afk17:54
*** yamahata has joined #tripleo18:00
*** wendar_ is now known as wendar18:04
*** fragatina has joined #tripleo18:05
*** jcoufal has quit IRC18:15
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.
*** coolsvap has quit IRC18:20
*** fragatin_ has joined #tripleo18:26
*** sthillma has joined #tripleo18:26
*** isq has joined #tripleo18:27
*** michchap has joined #tripleo18:28
*** diazjf has joined #tripleo18:28
*** michchap has quit IRC18:28
*** michchap has joined #tripleo18:29
*** fragatina has quit IRC18:29
*** fragatina has joined #tripleo18:29
*** fragatin_ has quit IRC18:29
*** michchap_ has quit IRC18:30
*** michcha__ has joined #tripleo18:30
*** akshai has quit IRC18:30
*** akshai has joined #tripleo18:33
*** michchap has quit IRC18:34
*** trown|lunch is now known as trown18:44
openstackgerritMerged openstack-infra/tripleo-ci: Sync up EndpointMap in enable-tls with default enable-tls from THT
*** sambetts is now known as sambetts|afk18:47
openstackgerritJames Slagle proposed openstack/tripleo-puppet-elements: Honor $DIB_DEFAULT_INSTALLTYPE in puppet-modules
openstackgerritMatt Young proposed openstack/tripleo-common: Add support to image build yaml input to handle env vars
openstackgerritJames Slagle proposed openstack/tripleo-puppet-elements: Honor $DIB_DEFAULT_INSTALLTYPE in puppet-modules
slagletrown: ^19:11
*** kbyrne has quit IRC19:16
*** kbyrne has joined #tripleo19:20
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron server
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent
openstackgerritMerged openstack/tripleo-heat-templates: Add step to ObjectStorage RingBuilder deployment
pradkslagle, hey19:28
pradkslagle, so i'm looking to see how we can get the aodh upgrade idea prototyped to see if its viable19:29
openstackgerritJeff Peeler proposed openstack/tripleo-common: Remove unnecessary package dhcp-client
slaglepradk: ok, the first think i would probably do is try it entirely manually19:30
slaglepradk: deploy liberty, and see if you can migrate it to use aodh from mitaka19:30
openstackgerritMatt Young proposed openstack/tripleo-common: Add support to image build yaml input to handle env vars
slagleget all the steps worked out, etc19:30
pradkslagle, sure, thought you already tried that? no?19:30
slaglepradk: no. i showed what the yum upgrade would do19:31
slaglethat's all19:31
slaglewhat we'll probably need is a script that:19:31
slagledisables the ceilometer pcmk resource , drops constraints, etc19:31
pradkslagle, understood, i'll do it manually, but what i want to clarify is how the integration with upgrade process happens19:31
slaglestops the alarm service19:32
slagleenables mitaka repos19:32
slaglethen a puppet manifest that configures aodh19:32
slagleand installs19:32
pradkso the puppet manifest is wired into heat template somewhere?19:33
pradkwhat invokes it?19:33
slaglepradk: eventually, yes19:33
pradkand we probably want to reuse the hiera data for uris etc19:33
slaglefor just a manual test, write the manifest, then puppet apply19:33
slagleyea so for the actual integration into the templates, we could add the softwaredeployment resoruce(s) before the full package upgrade to mitaka19:34
slaglethe new hiera data should already be laid down by then19:34
slaglethen we run the deployment that executes the single manifest to migrate to aodh19:34
slaglethen after that we can continue on with the rest of the full upgrade to mitaka19:35
pradkslagle, "the new hiera data should already be laid down by then" -> is this the data with aodh params as well from mitaka? would that already be there when deploy runs?19:36
slaglepradk: yea, i'm thinking it would have to be the hieradata with mitaka params19:36
slaglesince we'd be configuring aodh from mitaka19:36
*** akshai has quit IRC19:36
pradkslagle, what i'm confused about is, is the new mitaka code getting exceuted to install aodh or a stand alone manifests that duplicates the mitaka code in overcloud_controller.pp19:37
*** fragatina has quit IRC19:38
slagleit'd be duplicated from overcloud_controller_pacemaker.pp, mostly this:
*** fragatina has joined #tripleo19:38
slaglethat is not really considered "duplication" in the puppet world19:38
slagleas it's just includes and classes19:39
slagleor so i've been told :)19:39
openstackgerritMerged openstack/diskimage-builder: Add for syncing units
slaglepradk: where we might run into trouble is if any of those aodh classes start fiddling with other openstack services19:39
pradkyep thats what i was referring to, so that block of code would be in a standalone file and gets wired in through softwaredeploy ment19:40
pradkwith new mitaka overcloud.yaml and controller.yaml19:40
pradkgot it19:40
pradkslagle, so the extra code we're writing is really the softwaredeploy heat block to invoke the aodh.pp19:41
pradk+ a script to drop existing pacemaker constraints19:41
slaglepradk: yes, and the bash script to do any pre/post steps19:41
slaglewe might want a script to run after the puppet to validate that aodh is up and fucntioning19:41
slaglewhatever that might look like19:41
EmilienMdprince: your neutron patches made my rabbitmq merge conflict :-) I'm rebasing and maybe we can re-approve it19:42
pradkslagle, cool, makes sense .. ok i'll do a liberty install and verify this actually does whats expected .. i might need some help with wiring softwaredeployment heat stuff19:43
pradkslagle, i think i already added the script to drop pacemaker constraints19:43
pradklemme find it19:43
slaglek, i can help on the template side if you need it19:44
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role
slaglei was trying to resurrect my liberty environment to try some of it out19:44
*** dsariel has joined #tripleo19:44
*** dprince has quit IRC19:44
openstackgerritPaul Belanger proposed openstack-infra/tripleo-ci: Stop rebuilding locels on centos-7
pradkslagle, i pushed that with my original aodh patch already.. might need some amending19:45
pradkso thats already in mitaka19:46
*** diazjf has quit IRC19:46
*** diazjf has joined #tripleo19:47
*** akshai has joined #tripleo19:47
*** mgarciam has quit IRC19:52
*** gfidente|afk is now known as gfidente19:56
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Make sure delorean.repo takes precedence over delorean-deps.repo
openstackgerritPaul Belanger proposed openstack-infra/tripleo-ci: Stop rebuilding locales on centos-7
*** jprovazn has quit IRC20:11
*** diazjf has quit IRC20:13
*** thrash is now known as thrash|biab20:16
*** pkunicki has quit IRC20:17
*** akshai has quit IRC20:20
*** diazjf has joined #tripleo20:20
*** jayg is now known as jayg|g0n320:25
openstackgerritJames Slagle proposed openstack/instack-undercloud: Consolidate undercloud* elements
openstackgerritJames Slagle proposed openstack/instack-undercloud: Add timestamps to puppet apply output
openstackgerritJames Slagle proposed openstack/instack-undercloud: Minimize element usage
*** akshai has joined #tripleo20:27
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move 02-undercloud-stack-heat-metadata to python module
openstackgerritJames Slagle proposed openstack/instack-undercloud: Remove truncate nova config workaround
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-net-config element
openstackgerritMatt Young proposed openstack/tripleo-common: Add support to image build yaml input to handle env vars
*** absubram has quit IRC20:46
*** akshai has quit IRC20:47
*** lblanchard has quit IRC20:56
*** fzdarsky|afk has quit IRC20:57
*** akshai has joined #tripleo21:00
*** dsariel has quit IRC21:02
*** fultonj has quit IRC21:04
*** trown is now known as trown|outtypewww21:08
*** akshai_ has joined #tripleo21:08
*** akshai has quit IRC21:12
*** thrash|biab is now known as thrash21:14
*** ccamacho has quit IRC21:18
*** ayoung has quit IRC21:18
*** ayoung has joined #tripleo21:18
bandinislagle: ever seen this when running --undercloud ?
slaglebandini: similar things, yes21:23
slaglebandini: try rm -rf /opt/stack, then rerun21:24
bandinislagle: perfect thanks!21:25
*** morazi has quit IRC21:36
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: [WIP] Set glance-api bind host within role template
*** xinwu has quit IRC21:45
*** diazjf has quit IRC21:50
*** MaxPC has quit IRC21:51
*** weshay has quit IRC21:52
*** myoung|remote has quit IRC21:52
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: [WIP] Set glance-api bind host within role template
EmilienMgfidente: is now failing non-ha, I did recheck even if it was +2 +A before21:59
EmilienMI want to make sure CI pass before we try to merge it again21:59
gfidenteunrelated to the cookie though I think22:00
EmilienMgfidente: we +2 +A ?22:00
gfidenteI think it's just undercloud failing to spawn the controller node22:02
gfidenteso yes I'd +A22:03
*** mburned is now known as mburned_out22:03
*** mburned_out is now known as mburned22:03
*** julim has quit IRC22:04
openstackgerritMerged openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Pass MysqlVirtualIP via EndpointMap
*** paramite has quit IRC22:15
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display
*** athomas has quit IRC22:22
*** mburned is now known as mburned_out22:24
openstackgerritQasim Sarfraz proposed openstack/tripleo-heat-templates: use puppet-tripleo plumgrid profile
*** rhallisey has quit IRC22:29
*** pradk has quit IRC22:41
openstackgerritQasim Sarfraz proposed openstack/puppet-tripleo: add plumgrid neutron profile
*** gfidente has quit IRC22:43
*** mbound has quit IRC22:54
*** yamahata has quit IRC22:58
*** akshai_ has quit IRC22:58
*** jaosorior has quit IRC22:58
*** jaosorior has joined #tripleo22:59
*** dkelson has quit IRC23:00
*** zeroshft has quit IRC23:01
openstackgerritQasim Sarfraz proposed openstack/puppet-tripleo: add plumgrid neutron profile
*** apetrich has quit IRC23:34
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron server
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent
*** mbound has joined #tripleo23:55
*** diazjf has joined #tripleo23:58

Generated by 2.14.0 by Marius Gedminas - find it at!