Tuesday, 2016-03-15

EmilienMdsneddon: for https://bugzilla.redhat.com/show_bug.cgi?id=131769300:01
openstackbugzilla.redhat.com bug 1317693 in openstack-tripleo-heat-templates "RabbitMQ failures when deploying RDO with IPv6" [Unspecified,New] - Assigned to tsedovic00:01
EmilienMit sounds like you miss brackets00:01
EmilienMcurl -k --noproxy localhost --retry 30 --retry-delay 6 -f -L -o /var/lib/rabbitmq/rabbitmqadmin http://guest:guest@fd00:fd00:fd00:2000::13:15672/cli/rabbitmqadmin returned 7 instead of one of [0]00:01
*** saneax is now known as saneax_AFK00:23
*** chem has quit IRC00:24
*** Marga__ has quit IRC00:27
*** Marga_ has joined #tripleo00:28
*** mordred has quit IRC00:31
*** afazekas_ has quit IRC00:32
*** afazekas has joined #tripleo00:32
*** mordred has joined #tripleo00:33
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Normalize RabbitMQ IP for IPv6 deployments  https://review.openstack.org/29266400:58
EmilienMslagle, dsneddon: this should work ^00:58
*** stendulker has joined #tripleo01:17
*** dmacpher has joined #tripleo01:25
*** chlong has quit IRC01:26
*** shivrao has quit IRC01:35
*** tzumainn has quit IRC01:54
*** ibravo has quit IRC02:00
*** rhallisey has quit IRC02:05
*** rhallisey has joined #tripleo02:08
*** shivrao has joined #tripleo02:14
*** shivrao has quit IRC02:14
*** stendulker has quit IRC02:15
*** shivrao has joined #tripleo02:31
*** shivrao has quit IRC02:32
*** shivrao has joined #tripleo02:33
*** rhallisey has quit IRC02:35
*** shivrao has joined #tripleo02:36
*** shivrao has quit IRC02:39
*** xinwu has quit IRC02:46
openstackgerritgreghaynes proposed openstack/diskimage-builder: WIP add element to install glean.rs  https://review.openstack.org/29268902:54
*** yamahata has joined #tripleo03:02
*** yuanying has quit IRC03:22
*** lblanchard has quit IRC03:24
*** chlong has joined #tripleo03:29
*** xinwu has joined #tripleo03:31
*** stendulker has joined #tripleo03:35
*** links has joined #tripleo03:38
*** yuanying has joined #tripleo03:38
*** yuanying has quit IRC03:45
*** shivrao has joined #tripleo03:55
*** shivrao has quit IRC03:56
*** ayoung has quit IRC03:57
*** shivrao has joined #tripleo03:57
*** Marga_ has quit IRC03:59
*** Marga_ has joined #tripleo04:00
*** Marga_ has quit IRC04:00
*** stendulker_ has joined #tripleo04:04
*** stendulker has quit IRC04:04
*** yuanying has joined #tripleo04:06
*** stendulker has joined #tripleo04:07
*** stendulker_ has quit IRC04:10
*** masco has joined #tripleo04:14
*** stendulker_ has joined #tripleo04:15
*** stendulker has quit IRC04:16
*** stendulker has joined #tripleo04:16
*** stendulker_ has quit IRC04:20
*** Marga_ has joined #tripleo04:21
*** Marga_ has quit IRC04:25
*** Marga_ has joined #tripleo04:33
*** Marga_ has quit IRC04:34
*** Marga_ has joined #tripleo04:34
*** shivrao_ has joined #tripleo04:55
*** shivrao has quit IRC04:58
*** shivrao_ is now known as shivrao04:58
openstackgerritMichael Chapman proposed openstack/puppet-tripleo: Add keystone and db sync profiles  https://review.openstack.org/29271205:43
openstackgerritMichael Chapman proposed openstack/puppet-tripleo: Add keystone and db sync profiles  https://review.openstack.org/29271205:53
*** jaosorior has joined #tripleo05:57
*** rlandy has quit IRC05:58
*** saneax_AFK is now known as saneax06:00
*** akuznetsov has joined #tripleo06:07
*** shivrao has quit IRC06:07
*** shivrao has joined #tripleo06:09
*** jprovazn has joined #tripleo06:24
*** TSCHAK_ has joined #tripleo06:34
*** TSCHAK_ has quit IRC06:35
*** jcooley has quit IRC06:35
*** akuznetsov has quit IRC06:42
*** oshvartz has joined #tripleo06:53
*** mikelk has joined #tripleo07:10
*** tesseract has joined #tripleo07:13
*** tesseract is now known as Guest9751907:13
*** liverpooler has joined #tripleo07:24
*** liverpooler has quit IRC07:33
*** liverpooler has joined #tripleo07:34
*** rain has joined #tripleo07:42
*** rain is now known as leanderthal07:42
*** ccamacho has joined #tripleo07:43
*** a_petrich has quit IRC07:52
*** yamahata has quit IRC07:52
*** apetrich has joined #tripleo07:52
*** athomas has joined #tripleo07:54
*** Guest97519 has quit IRC08:00
*** tesseract has joined #tripleo08:05
*** tesseract is now known as Guest9179408:06
*** bvandenh has joined #tripleo08:08
*** apetrich has quit IRC08:09
*** aufi has joined #tripleo08:13
*** paramite has joined #tripleo08:17
*** ohamada has joined #tripleo08:19
*** shivrao has quit IRC08:25
*** apetrich has joined #tripleo08:30
*** chlong has quit IRC08:32
*** mannidi has joined #tripleo08:37
*** rasca_ has joined #tripleo08:40
*** rasca has quit IRC08:42
*** Guest91794 has quit IRC08:45
*** mkovacik has joined #tripleo08:49
*** xinwu has quit IRC08:50
*** mbound has joined #tripleo08:56
*** Guest91794 has joined #tripleo08:59
*** dmacpher has quit IRC09:13
*** mkovacik has quit IRC09:18
*** jistr has joined #tripleo09:20
*** snecklifter has quit IRC09:21
*** rcernin has joined #tripleo09:31
*** lucas-dinner is now known as lucasagomes09:33
*** akrivoka has joined #tripleo09:36
*** panda has quit IRC09:40
*** paramite is now known as paramite|afk09:40
*** akuznetsov has joined #tripleo09:40
*** panda has joined #tripleo09:41
*** paramite|afk is now known as paramite09:42
jaosoriorjistr: Hey dude, this commit adds ssl to the overcloud deployment in the non-ha gate, got time to check it out? https://review.openstack.org/#/c/281988/13 would be nice to finally have that tested in CI09:45
*** derekh has joined #tripleo09:47
*** stendulker_ has joined #tripleo09:48
jistrjaosorior: looking09:49
*** mgould has joined #tripleo09:50
*** stendulker has quit IRC09:50
jistrjaosorior: how does it work with the CN of the certificate in that patch? I unencoded it locally and i see it's hardwired to
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891509:53
jistri mean, do we count on the fact that we're just deploying overcloud once in the CI, and the public VIP will always be does it happen always like that?09:53
jaosoriorjistr: when bnemec first commited that, it used to have network isolation, so then the IP would be predictable. But AFAIK it wasn't working on the non-ha gate, so he ended up doing it without netiso09:53
jaosoriorjistr: So yeah, it's pretty much counting on that the first IP deployed every time has been
jaosoriorI haven't seen it as much of a problem as it's something we can get rid of later09:54
jaosoriorbut netiso needs to work in the non-ha case first09:55
openstackgerritJuan Antonio Osorio Robles proposed openstack-infra/tripleo-ci: TEST: overcloud SSL with netiso  https://review.openstack.org/29278710:01
*** olap has joined #tripleo10:02
jistrhm i wonder if we could use ControlFixedIPs instead of predicting auto-assignment https://github.com/openstack/tripleo-heat-templates/blob/0b5459c178b8fdbda36b2fd2c5362b06303b2b6d/overcloud.yaml#L128410:02
jaosoriorjistr: Well, lets see if that does the trick ^^ if netiso is working there it should fail cause of the mismatch in the CN of the cert10:02
jistri +1'd the change and mentioned the above on the review10:04
jaosoriorjistr: By the way, do you know why the clouddomain is not included in these values of /etc/hosts? https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/controller.yaml#L1701 and why is it the format it's in?10:05
*** dshulyak has joined #tripleo10:07
jaosoriorI guess that mapping is not really used, so it's still somewhat trivial to change it10:09
jistrhmm not sure, but it looks like this should control the IP for FQDN: PRIMARYIP PRIMARYHOST.DOMAIN PRIMARYHOST  (and CloudDomain is included there)10:09
jaosoriorjistr: It's included for the primary ip, but not for any other10:09
*** jcoufal has joined #tripleo10:10
jistrand i'd expect that these "EXTERNALIP EXTERNALHOST" etc. are just "utility" hostnames when you need to talk to a server on a specific network using a hostname instead of IP10:10
jistrjaosorior: would it help something if the other ones had CloudDomain included?10:11
jaosoriorjistr: It would, and I will probably submit a patch adding the clouddomain there10:12
*** liverpooler has quit IRC10:12
jaosoriorjistr: In this case, I have all the nodes registered as FreeIPA hosts, and to register them, they should be in the same domain, as it's used as the kerberos domain10:12
jaosorior* kerberos realm10:12
*** liverpooler has joined #tripleo10:12
*** mkovacik has joined #tripleo10:14
jaosoriorso, what I'm trying is to enable SSL for some internal interfaces; And that /etc/hosts file seems to have a separate entry for the internal IP... so what I'm doing is adding it as a separate FreeIPA host. to try to make things simple and not depend on having DNS fully working10:14
jistrah i see... interesting. I was wondering why would you want to register the same host under 2 different FQDNs10:15
*** rasca_ has quit IRC10:16
jaosoriorjistr: Yeah, it's not the cleanest solution. And I would prefer there would be only one host registered. But then again there are different IPs working in different networks (potentially... in the netiso case), so this just works without having tripleo require a full fledged DNS working10:17
*** mannidi_ has joined #tripleo10:19
*** mannidi_ has quit IRC10:20
*** mannidi_ has joined #tripleo10:21
*** electrofelix has joined #tripleo10:21
*** mannidi_ has quit IRC10:21
*** mannidi has quit IRC10:23
*** athomas has quit IRC10:25
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Change /etc/hosts format and add domain  https://review.openstack.org/29279910:29
*** rasca_ has joined #tripleo10:30
*** athomas has joined #tripleo10:31
*** rodrigods has quit IRC10:35
*** rodrigods has joined #tripleo10:36
jaosorioraww damn... openstack CI is broken :(10:38
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Basic beaker one node test.  https://review.openstack.org/28137610:38
*** shardy has joined #tripleo10:39
*** paramite is now known as paramite|afk10:42
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Add a service provider.  https://review.openstack.org/28612410:47
*** derekh changes topic to "TripleO | CI jobs currently not running Zuul appears to be stuck | CI status: http://tripleo.org/cistatus.html | Docs: http://tripleo.org/"10:49
*** stendulker has joined #tripleo11:01
*** stendulker_ has quit IRC11:04
*** stendulker has quit IRC11:06
*** thrash|g0ne is now known as thrash11:11
*** rhallisey has joined #tripleo11:12
*** masco has quit IRC11:12
-openstackstatus- NOTICE: Gerrit is going to be restarted11:15
rhalliseyderekh, ya /tmp is read only on atomic11:15
rhalliseyonly /var and /etc can be changed11:15
rhalliseyderekh, you addition should work11:15
derekhrhallisey: If switched it to /var/tmp but now gerrit is having trouble so it hasn't run ion CI yet11:15
derekhrhallisey: cool11:16
*** tosky has joined #tripleo11:24
*** akuznetsov has quit IRC11:28
rhalliseyok gerrit looks to be back11:31
-openstackstatus- NOTICE: Gerrit had to be restarted because was not responsive. As a consequence, some of the test results have been lost, from 08:30 UTC to 10:30 UTC approximately. Please recheck any affected jobs by this problem.11:34
-openstackstatus- NOTICE: Gerrit had to be restarted because was not responsive. As a consequence, some of the test results have been lost, from 09:30 UTC to 11:30 UTC approximately. Please recheck any affected jobs by this problem.11:37
*** weshay has joined #tripleo11:39
jaosoriorshardy: Hey dude, if I want to pass an array via Heat to a softwaredeployment, if that is passed as an input to a softwareconfig, will it be in the form of an array in bash?11:39
*** derekh changes topic to "TripleO | CI status: http://tripleo.org/cistatus.html | Docs: http://tripleo.org/"11:44
EmilienMderekh: ping about https://review.openstack.org/#/c/292359/11:45
*** paramite|afk has quit IRC11:45
derekhEmilienM: yup11:45
EmilienMderekh: my -1 was more about why we don't test ceph again?11:45
EmilienMin the same patch11:45
EmilienMI see toci_gate_test.sh  is not updated yet11:46
*** paramite has joined #tripleo11:47
derekhEmilienM: ahh, this patch was supposed to switch it, https://review.openstack.org/#/c/260466/11:47
EmilienMderekh: ok it makes sense now, dropping -1 :)11:47
derekhEmilienM: thanks, your right though, we should have merged that first, gfidente isn't around at the moment, I'll check with him once I see him on11:48
*** paramite is now known as paramite|afk11:48
*** dmacpher has joined #tripleo11:50
derekhok, shardy's move of tripleo.sh has passed CI, lets merge it now and remove the one in tripleo-common befor it goes into conflict11:51
derekhbnemec: slagle ^11:53
slaglenow let's kill the other one11:53
slaglederekh: oh, guess we need to merge https://review.openstack.org/#/c/290002/ too?11:55
slagledang, giulio -2'd it11:56
slaglewe are stuck11:56
derekhslagle: ok, lets resubmit the tripleo-ci patch without the depends on11:57
derekhslagle: merg it and then submit and other patch to remove tripleo.sh11:57
derekhfrom tripleo-common11:57
derekhWe we'd be bypassing the -2 from gfedente but he only did that to prevent any accidents11:59
slaglederekh: yea, let's do that12:00
*** trown has joined #tripleo12:00
shardyslagle, derekh Hey, yeah I just commented, probably need to remove the DependsOn12:01
shardythanks for picking this up, been travelling so not had time to look at it12:01
derekhshardy: yup, I was about to do it, are you doing it ?12:01
shardyderekh: If you don't mind doing it then please do, I'm about to head out12:01
derekhshardy: on it, safe trip12:02
openstackgerritJames Slagle proposed openstack/tripleo-common: Remove tripleo.sh (now in tripleo-ci repo)  https://review.openstack.org/29284212:02
slaglederekh: ^12:02
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Move tripleo.sh into tripleo-ci repo  https://review.openstack.org/27221012:03
derekhslagle: ^ ok, lets merge both of these12:03
slaglek, should be done12:04
*** mburned_out is now known as mburned12:05
derekhcool beans12:05
*** bvandenh has quit IRC12:06
openstackgerritDougal Matthews proposed openstack/tripleo-docs: Update the location of tripleo.sh  https://review.openstack.org/29284512:07
shardyWhile we're merging things, https://review.openstack.org/#/c/255228/ needs a final +2/A - allows passing the nodes json in yaml, and not having to specify the type explicitly12:08
*** jaosorior has quit IRC12:08
*** jaosorior has joined #tripleo12:09
d0ugalslagle, shardy - a quick doc fix for the new location of tripleo.sh https://review.openstack.org/#/c/292845/12:10
shardyGuys, I may not make the meeting today due to $travel, apologies12:10
shardyThe one thing I wanted to discuss (again) was the timeline re mitaka branching12:10
shardyif anyone feels like bringing that up, that would be good :)12:11
*** paramite|afk is now known as paramite12:11
*** MaxPC has joined #tripleo12:11
openstackgerritDougal Matthews proposed openstack/tripleo-common: Remove the TODO from the README.rst  https://review.openstack.org/29284612:12
*** jayg|g0n3 is now known as jayg12:13
*** jaosorior has quit IRC12:14
*** akuznetsov has joined #tripleo12:15
*** shardy has quit IRC12:16
*** saneax is now known as saneax_AFK12:17
*** paramite is now known as paramite|afk12:17
*** liverpooler has quit IRC12:18
*** liverpooler has joined #tripleo12:19
*** paramite|afk is now known as paramite12:21
*** ayoung has joined #tripleo12:23
*** rdopiera has quit IRC12:28
openstackgerritMiles Gould proposed openstack/tripleo-common: Return list of new nodes from register_all_nodes  https://review.openstack.org/29286012:34
*** jaosorior has joined #tripleo12:36
jaosoriorthe tripleo meeting is in one hour and a half, right?12:39
*** rlandy has joined #tripleo12:39
derekhjaosorior: I believe so, 1400 UTC http://git.openstack.org/cgit/openstack-infra/irc-meetings/tree/meetings/tripleO-team-meeting.yaml12:41
EmilienMyeah, all meetings are UTC12:43
*** trown has quit IRC12:43
*** shardy has joined #tripleo12:44
jaosorioralright, thanks12:44
*** lucasagomes is now known as lucas-hungry12:45
*** shardy has quit IRC12:47
*** rdopiera has joined #tripleo12:48
EmilienMjaosorior: I'm curious to know why the rabbitmq/ipv6 patch does not pass nonha12:48
EmilienMon http://logs.openstack.org/64/292664/1/check-tripleo/gate-tripleo-ci-f22-nonha/b9477b7/console.html#_2016-03-15_07_39_51_05312:48
*** shardy has joined #tripleo12:48
EmilienMjaosorior: the curl is timeouting I guess12:49
openstackgerritJuan Antonio Osorio Robles proposed openstack-infra/tripleo-ci: TEST: overcloud SSL with netiso  https://review.openstack.org/29278712:49
*** colonwq has joined #tripleo12:50
*** dprince has joined #tripleo12:52
*** trown has joined #tripleo12:53
jaosoriorEmilienM: yeah, that curl seems to be the only thing that could time out there12:53
jaosoriorlets see12:53
derekhNot sure why this isn't merging, are things still stuck https://review.openstack.org/#/c/272210/12:54
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Basic beaker one node test.  https://review.openstack.org/28137612:54
*** Goneri has quit IRC12:54
*** morazi has joined #tripleo12:55
EmilienMderekh: you need to -A and +A again12:57
EmilienMto trigger Gerrit12:57
EmilienMit has +2 +A before the Depends-On remove12:57
EmilienManother way is to to "recheck" and +2 +A again12:58
jaosoriorEmilienM: I'm trying to find where that would be in the logs... but I haven't found anything that resembles that curl13:00
jaosoriorit just says the Exec of that Staging resource has timed out13:00
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: Change for configuring use_forwarded_for value for Nuage  https://review.openstack.org/29286813:01
dprinceslagle, derekh: would one of you be willing to run the meeting in an hour?13:01
EmilienMbnemec: the dependency merged on https://review.openstack.org/#/c/210405/13:01
* dprince is in a meeting today13:01
EmilienMjaosorior: this is where it timeouts. https://github.com/puppetlabs/puppetlabs-rabbitmq/blob/master/manifests/install/rabbitmqadmin.pp#L2013:01
EmilienMI think rabbitmq fails to start13:02
*** trown is now known as trown|ooobootcam13:02
*** lmiccini has joined #tripleo13:02
*** colonwq has quit IRC13:03
derekhdprince: ya, sure will do, happy to let slagle do it either,13:03
* derekh goes for food13:03
dprincederekh: thanks sir :)13:04
slaglederekh: go4it :)13:11
*** yamahata has joined #tripleo13:11
*** tzumainn has joined #tripleo13:14
jaosoriorEmilienM: Found it13:14
jaosoriorEmilienM: Responded in the CR13:15
jaosoriorEmilienM: but yeah, the issue was that you need to set that same parameter for the rabbitmq inclusion in the else clause below13:16
EmilienMjaosorior: nice catch13:16
*** akuznetsov has quit IRC13:17
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Normalize RabbitMQ IP for IPv6 deployments  https://review.openstack.org/29266413:17
EmilienMdprince: michchap and I are working on composable roles in puppet-tripleo, fyi13:19
*** colonwq has joined #tripleo13:19
dprinceEmilienM: sounds good13:19
dprinceEmilienM: I'd like to work out a high level plan as to who is doing what, and what order13:20
EmilienMdprince: michchap and I started to move out glance/keystone code from THT to puppet-tripleo13:21
EmilienMdprince: https://review.openstack.org/#/c/292712/13:22
*** greghaynes has quit IRC13:22
michchapdprince: I'm just testing the water with the keystone patch to start with while we argue out things like directory structure and coding style13:22
michchapEmilienM: let's bring the convo in here ok?13:22
dprinceEmilienM: I'd like all the pacemaker classes to be separate I think13:22
michchapdprince: That's how I did the keystone one, EmilienM's glance one is a single profile handling both cases13:23
dprinceEmilienM: and the best way to integrate that w/ Heat might actually to be to do the puppet-tripleo patch, then the "composable roles" patch in heat13:23
michchapdprince: I have profile/base/[service] and profile/pacemaker/[service]13:23
EmilienMdprince: I would do it in a iterative way, where we clean THT manifests everytime we patch puppet-tripleo13:24
dprincemichchap: excellent. I think I like them split better. Splitting them I think is where we want to end up but it will make the integration into tripleo-heat-templates a bit harder13:24
michchapdprince: so in theory you could swap out pacemaker for another system just by changing a string to load different classes13:24
EmilienMdprince: otherwise we'll have to rebase all the time13:24
dprinceEmilienM: yes, the rebase is a pain. I think I mentioned that to you :)13:24
michchapdprince: EmilienM: Right, doing it service by service is probably the only way we get anything merged13:24
EmilienMdprince: so if we managed to land puppet-tripleo patch, we need to land the THT patch, so we can iterate13:24
*** lblanchard has joined #tripleo13:25
*** chlong has joined #tripleo13:25
michchapEmilienM: yep. The second patch removing code from THT also has to be double checked against what's in puppet-tripleo to make sure we don't miss changes added after the profile was written13:25
*** ayoung has quit IRC13:26
dprinceEmilienM: exactly, so once you have your glance patch I would like to land this https://review.openstack.org/#/c/237370/13:26
michchapEmilienM: I'm also really worried about breaking vendor code in neutron because they plugin and mechanism selection is absolute spaghetti13:26
EmilienMdprince: no I'm not sure your patch is exactly what we want.13:27
EmilienMdprince: we want to move out Puppet code from THT13:27
dprinceEmilienM: I'm going to rebase it on your patch man13:27
dprinceEmilienM: ditching the puppet. It'll just be includes to puppet-tripleo13:27
michchapdprince: +113:27
dprincemichchap: hey, one idea you added to the spec was about keeping the "step" logic in the manifests vs. in t-h-t13:28
dprincemichchap: while I could see some benefits there I think perhaps the "step" logic belongs in t-h-t (where it currently is)13:28
michchapdprince: atm 'step' is just a parameter passed to each profile as I've written it13:29
EmilienMdprince: I don't think steps should be in THT13:29
dprincemichchap: correct, I'm concerned about the coupling it might cause between heat and puppet13:29
EmilienMI like michchap approach where he keeps it in manifests13:29
michchapdprince: Otherwise it's very hard to enable things selectively in later steps13:29
dprinceif we keep the steps in Heat we can add/subtract steps without changing puppet13:29
michchapdprince: I think there's an example in the keystone patch13:29
EmilienMsteps are supposed to be pretty static13:30
michchapdprince: I'm going to put it out there that I think 'step' is something we'll replace at some point anyway13:30
dprinceI'd rather have a more well defined variable than just step I think13:30
EmilienMinstead of steps, I would rather try to use anchors13:31
michchapdprince: it has very overloaded meanings but it's really stating orchestration dependencies13:31
dprince'database_initialization' might be a better name for one of them13:31
EmilienMmichchap: what do you think about anchors?13:31
michchapdprince: yeah exactly, we're saying resource X depends on task Y13:31
dprinceetc. I'd rather see us come up with well defined associated meanings for these things13:31
michchapEmilienM: You don't need anchors, you need resource_fail13:31
michchapEmilienM: so that you can make something not happen if you don't want it to13:32
michchapEmilienM: anchor can't stop something from being included13:32
michchapEmilienM: the only way I found to do that was to have a dummy resource and force-fail it, then have things depend on it13:32
michchapEmilienM: then you have to start ignoring puppet's error codes though13:32
*** tosky has quit IRC13:33
michchapEmilienM: it's a rabbit hole I don't think we can go down right now13:33
EmilienMmichchap: I don't like ignore puppet's error codes13:33
EmilienMI have the feeling we need further discussion on all of this13:33
michchapEmilienM: it means that 'error' actually means 'I didn't do everything'13:33
dprinceregardless of how we implement steps, vs anchors, etc. (I like where these ideas are going) We still desire Heat as the uber orchestration tool to be able to control the flow between steps13:33
michchapEmilienM: you can build a system around that where you converge to a good state13:33
michchapEmilienM: but I think that's way off13:34
michchapdprince: yeah I agree, I think we're getting ahead of ourselves13:34
dprinceso long as we keep that capability in Heat I'm okay with iterating the implementation13:34
michchapdprince: in any case I think passing step to profiles makes sense for now because it means we only need 1 profile per service, rather than needing 1 per step per service13:35
michchapdprince: the data interface between THT and the manifests is always going to be tightly coupled to both, I think this is a first step to actually defining it13:35
dprincemichchap: ++13:36
michchapdprince: Right now we can move all the hiera() calls into profile parameters, and over time we can remove the hiera calls altogether and replace with data mappings in hiera13:36
EmilienMmichchap and I have to first move out all Puppet code from THT with exact feature parity13:37
michchapdprince: that leaves us with a stable data model that could in theory be consumed by something that isn't even puppet.13:37
dprincemichchap: yeah, we never intended to have as many hiera() calls as we have now. It has gotten a bit out of hand13:37
michchapdprince: which I think would probably make some people quite happy :)13:37
*** greghaynes has joined #tripleo13:37
michchapEmilienM: right. I am a little worried about some parts, particularly neutron as I said13:38
dprincemichchap: yes, in doing all this we very much want to keep the t-h-t interfaces somewhat pluggable13:38
michchapEmilienM: because test coverage is not great13:38
EmilienMmichchap: let's start simple, and do keystone + glance13:38
michchapEmilienM: yep13:38
EmilienMmichchap: are you going to change something in your patch?13:38
dprincemichchap: I think vendors will step up and help validate the new structures13:38
thrashd0ugal: marios need an opinion. jistr has already weighed in on this. mburned you too... the 'stack upgrade' command has three phases. pre, perform, and post.13:38
michchapEmilienM: I need to write param docs and spec tests still13:39
michchapdprince: It will make their lives easier in the long run13:39
thrashquestion is, before this gets set in stone, as it were, what should that be? flags of '--pre|perform|post' or distinct commands 'start|perform|finish'13:39
thrashor something else entirely?13:40
michchapdprince: quick q: what does this resource model? pacemaker::resource::ocf { 'openstack-core':13:40
thrashanyone else that has an opinion on this feel free to weigh in.13:40
michchapdprince: I wasn't sure if it should be in the keystone profile.13:40
*** panda has quit IRC13:40
d0ugalthrash: I prefer commands13:40
dprincemichchap: late for you now no? What time would be the best time for us to chat on these things (perhaps a couple times a week)13:40
*** pradk has joined #tripleo13:40
d0ugalthrash: I am not a fan of perform - just an idea: prepare,start,finish13:40
*** ayoung has joined #tripleo13:40
michchapdprince: yeah it's late but I'm a night owl. Now is fine and I'll try to be around for a couple of hours around this time.13:40
*** panda has joined #tripleo13:41
dprincemichchap: okay, I'll look for you early in my morning then13:41
mburnedthrash: no strong opinion for me13:41
d0ugalthrash: but as for flags vs commands, I'd pick commands to make a choice and move on, but I don't mind really.13:41
dprincemichchap: looking at the keystone patch13:41
michchapdprince: is this too early for you?13:41
mburnedbut d0ugal makes sense13:41
dprincemichchap: no, but I'm in a meeting this week13:41
michchapdprince: are you west or east coast?13:42
dprincemichchap: east13:42
michchapdprince: ah easy, best if I stay up then. I'm bad at mornings.13:42
thrashd0ugal: I do like prepare start finish13:42
thrashjistr: what do you think of that?^^^13:42
jistrthrash, d0ugal: +1 from me13:42
dprincemichchap: I think we are about as far apart as we can be... :)13:42
thrashlet's go with that then.13:42
jistrthrash: also +1 on commands over flags13:42
thrashthat's easy enough.13:43
d0ugalGreat. bbiab, dog walk.13:43
thrashI'll W-1 until done13:43
EmilienMmichchap: can you also do a patch in THT that test it? so we do the "service by service" thing13:43
michchapdprince: Yeah, the rest of the OPNFV team is also on the east coast so it's normal for me to do work around this time13:43
thrashjistr: good thing is that it doesn't change the tripleo-common code. :)13:43
EmilienMmichchap: using Depends-On13:43
*** Goneri has joined #tripleo13:43
michchapEmilienM: so what I was thinking was doing a catalog diff, since I honestly have 0 clue how to do tripleo testing.13:44
michchapEmilienM: but if you have an actual method for testing then show me and I'll copy that.13:44
EmilienMmichchap: patch THT that include your profiles and use Depends-On the puppet-tripleo patch13:45
michchapEmilienM: oh right, I get what you mean13:45
michchapEmilienM: will do13:45
EmilienMmichchap: once it's done, i'll rebase on top of your patch, and continue your pattern13:46
*** tosky has joined #tripleo13:47
EmilienMmichchap: we need to keep the idea of one process == one profile, for containers13:47
michchapEmilienM: yeah I agree with that. I should have done something other than keystone to make it clear :)13:47
*** lucas-hungry is now known as lucasagomes13:48
slaglederekh: sounds like there are some issues with the gate queue, which is why the tripleo.sh patches havent merged13:48
EmilienMmichchap: actually keystone is good to start too13:48
dprincemichchap: I'm not the expert on the pacemaker OCF thing. It appears to be a convenience class around a Pacemaker OCF resource13:48
dprincemichchap: if you have questions on pacemaker specific bits please talk to jistr, gfidente, or marios I think13:48
*** paramite is now known as paramite|afk13:49
dprincemichchap: but like I said early I would like all pacemaker manifests separate. The vision I would like to see is us using *much* less pacemaker in our HA architecture and splitting things out would be a way to move towards that13:49
openstackgerritMichael Chapman proposed openstack/tripleo-heat-templates: Use keystone and db sync profiles  https://review.openstack.org/29289113:49
EmilienMmichchap: you missed the Depends-On13:50
michchapdprince: I'm totally with you on that.13:50
michchapEmilienM: derp I'm retarded13:50
EmilienMno you work by night :-P13:51
EmilienMusually people sleep at your time13:51
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Add a service provider.  https://review.openstack.org/28612413:51
openstackgerritMichael Chapman proposed openstack/tripleo-heat-templates: Use keystone and db sync profiles  https://review.openstack.org/29289113:52
michchapEmilienM: is that right? I don't think I've ever actually done a depends-on...13:52
*** paramite|afk is now known as paramite13:53
EmilienMmichchap: yes13:53
michchapEmilienM: cheers.13:53
michchapEmilienM: so one thing I was thinking once we have the profiles is we could use them for puppet ci independent of heat.13:55
EmilienMmichchap: from where? Puppet OpenStack CI ?13:55
michchapEmilienM: Yeah. Just a thought. I know you already have a working system.13:55
EmilienMmichchap: we already have https://github.com/openstack/puppet-openstack-integration/tree/master/manifests13:55
michchapEmilienM: I know, but the more code shared the better etc. It's just an idea.13:56
EmilienMmichchap: I agree with you13:56
EmilienMmichchap: but our manifests in p-o-i are very simple & basic13:56
EmilienMwe use default values for a maximum of things13:57
jristdprince: did you see my email about tripleosphinx?13:57
michchapEmilienM: yep.13:57
dprincejrist: yes, I will reply13:57
jristI can update the readme with your reply13:57
EmilienMmichchap: I see a value of sharing code, but first we need to focus on getting composable roles in tripleo13:58
michchapEmilienM: yep13:58
EmilienMif we see that our profiles are flexible enough, ok let's use-them in p-o-i13:58
EmilienMbut that's a future iteration13:58
michchapEmilienM: yep agree.13:58
derekhMeeting time14:00
jristderekh: which? :)14:01
openstackgerritKeith Schincke proposed openstack/tripleo-heat-templates: Add support for deploying RGW on controllers.  https://review.openstack.org/28902714:01
*** saneax_AFK is now known as saneax14:01
derekhjrist: tripleo #openstack-meeting-alt14:02
jristderekh: thanks, wasn't on the calendar14:02
jristthanks daylight savings14:02
*** Guest41345 has quit IRC14:03
*** Marga_ has quit IRC14:08
*** morazi has quit IRC14:10
*** bvandenh has joined #tripleo14:14
*** sshnaidm has quit IRC14:23
*** tosky has quit IRC14:23
*** sshnaidm has joined #tripleo14:25
*** tosky has joined #tripleo14:25
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Manage keystone initialization directly in t-h-t manifests  https://review.openstack.org/24416214:27
*** Guest91794 has quit IRC14:31
*** miles has joined #tripleo14:32
*** bvandenh has quit IRC14:33
*** mgould has quit IRC14:34
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Basic beaker one node test.  https://review.openstack.org/28137614:42
*** yamahata has quit IRC14:44
michchapEmilienM: dprince as far as who is doing what for the puppet roles/profiles, I have half-done neutron so I can probably get an initial cut of that done tomorrow. EmilienM feel free to take whatever one you want and I'll catch you same time tomorrow.14:44
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Upgrades: Add finish stage to 'overcloud upgrade'  https://review.openstack.org/28872814:45
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Upgrades: Add 'stack upgrade' command  https://review.openstack.org/28660614:45
EmilienMmichchap: did you send new patchset?14:45
EmilienMmichchap: good night man, ttyl14:46
michchapEmilienM: of what?14:46
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Basic beaker one node test.  https://review.openstack.org/28137614:46
dprincemichchap: thanks, sounds good14:46
EmilienMmichchap: of puppet-tripleo14:46
michchapEmilienM: no I haven't pushed anything in the last hour14:46
EmilienMmichchap: what needs to be done on this patch except doc?14:46
michchapEmilienM: The keystone one needs spec and doc14:46
EmilienMif nothing, I can do doc this afternoon, so these stuffs are ready to land14:46
michchapEmilienM: and it needs to pass CI :D14:46
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Upgrades: Add finish stage to 'overcloud upgrade'  https://review.openstack.org/28872814:46
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Upgrades: Add 'stack upgrade' command  https://review.openstack.org/28660614:46
*** oshvartz has quit IRC14:46
michchapEmilienM: feel free to take it over14:47
EmilienMmichchap: ok I'll write unit tests & doc14:47
EmilienMso hopefully tomorrow it pass CI14:47
michchapyeah go for it - I'll be asleep for a few hours anyway14:47
*** rpothier has joined #tripleo14:47
EmilienMmichchap: sweet dreams man14:47
michchapI'll ping you when I'm awake and I can take it over when you finish up if it still needs work14:48
michchapEmilienM:  night14:48
*** akuznetsov has joined #tripleo14:49
*** bvandenh has joined #tripleo14:49
*** aufi has quit IRC14:50
*** aufi has joined #tripleo14:52
openstackgerritRyan Hallisey proposed openstack/tripleo-heat-templates: Debugging for the container gate  https://review.openstack.org/29292814:54
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: Split out controller manifest steps  https://review.openstack.org/23624314:56
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: Configure ControllerServices via resource chains  https://review.openstack.org/25956814:56
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891514:56
openstackgerritGiacomo Tenaglia proposed openstack/puppet-pacemaker: Enable pcsd_mode for RHEL >= 6.6  https://review.openstack.org/29293114:57
*** links has quit IRC14:59
-openstackstatus- NOTICE: Launchpad OpenID SSO is currently experiencing issues preventing login. The Launchpad team is working on the issue15:00
*** ChanServ changes topic to "Launchpad OpenID SSO is currently experiencing issues preventing login. The Launchpad team is working on the issue"15:00
*** bvandenh has quit IRC15:00
*** jaosorior has quit IRC15:01
d0ugalIf anyone has time for reviews, here is a small tht patch that needs them https://review.openstack.org/#/c/28886915:04
*** rwsu has quit IRC15:04
*** xinwu has joined #tripleo15:05
*** tesseract has joined #tripleo15:06
*** tesseract is now known as Guest7146715:07
openstackgerritBrad P. Crochet proposed openstack/tripleo-common: Upgrades: Add post-upgrade stack update  https://review.openstack.org/28874415:07
openstackgerritBrad P. Crochet proposed openstack/tripleo-common: Upgrades: Add initialization to StackUpgradeManager  https://review.openstack.org/29293815:07
*** rwsu has joined #tripleo15:08
derekhslagle: can you remove you A+ from this and add it again https://review.openstack.org/#/c/27221015:09
*** paramite is now known as paramite|afk15:09
*** morazi has joined #tripleo15:11
*** paramite|afk is now known as paramite15:11
openstackgerritGiacomo Tenaglia proposed openstack/puppet-pacemaker: Add operation parameter for pacemaker::resource::ip class.  https://review.openstack.org/29294315:12
slaglederekh: i did, but i still dont see it entering the gate queue15:19
slaglein fact, i see nothing in that queue at all15:19
ayoungderekh, when Tripleo sets up Rabbit MQ, is sets: rabbit_userid=guest and rabbit_password=guest by default.  Is there some way to change that?15:20
*** miles is now known as mgould15:23
openstackgerritMerged openstack-infra/tripleo-ci: Move tripleo.sh into tripleo-ci repo  https://review.openstack.org/27221015:23
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Upgrades: Add initialization stage to 'overcloud upgrade'  https://review.openstack.org/29295515:23
slaglederekh: k, it's in the gate now15:23
*** yamahata has joined #tripleo15:23
slaglei had to toggle +A again15:23
derekhslagle: thanks, weird15:24
derekhayoung: off the top of my head I'm not sure, looking15:24
ayoungderekh, thanks...I need to learn me some Tripleo navigation, but I'm not there yet15:25
bnemecUgh, I thought we fixed the rabbit default.  Apparently not.15:26
bnemecayoung: You can change it via Heat parameters.15:26
derekhayoung: RabbitUserName and RabbitPassword15:27
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891515:27
ayoungbnemec, OK, and to affect that change, Tripleo Quickstart needs to generate a config file I assume?15:27
bnemecNo idea.  I don't use it.15:28
openstackgerritMerged openstack/tripleo-docs: Update the location of tripleo.sh  https://review.openstack.org/29284515:28
pradkslagle, jistr, Is there anything we're waiting on to +w aodh patches?15:28
openstackgerritMerged openstack/tripleo-heat-templates: Ensure access to Redis is password protected  https://review.openstack.org/21040515:28
ayoungbnemec, well, whatever calls openstack overcloud install, right?15:28
bnemecayoung: Yes, that's where it needs to be passed.15:29
*** akuznetsov has quit IRC15:30
ayoungbnemec, so the call I am triggering is : exec openstack overcloud deploy --templates --libvirt-type qemu --control-flavor oooq_control --compute-flavor oooq_compute --ceph-storage-flavor oooq_ceph --timeout 60  \15:30
ayoung    ${DEPLOY_ENV_YAML:+-e $DEPLOY_ENV_YAML}15:30
openstackgerritMerged openstack/tripleo-common: Remove tripleo.sh (now in tripleo-ci repo)  https://review.openstack.org/29284215:30
ayoungI assume that the change would be  in the   $DEPLOY_ENV_YAML faile15:31
slaglederekh: BOOM, tripleo.sh moved15:31
slaglepradk: yep, we were waiting on passing CI on the depends-on15:31
slaglepradk: i'd gather they won't pass on their own, given how they've failed15:32
slaglepradk: but i havent had a chance to look into if that's the case yet15:32
pradkslagle, ok, i'm not sure why the tripleclient patch is failing ci.. i'm looking at the logs and dont think its realted to aodh15:32
*** ccamacho has quit IRC15:33
-openstackstatus- NOTICE: Launchpad SSO is back to normal - happy hacking15:33
*** ChanServ changes topic to "TripleO | CI status: http://tripleo.org/cistatus.html | Docs: http://tripleo.org/"15:33
*** ccamacho has joined #tripleo15:33
*** chlong has quit IRC15:33
ayoungbnemec, there is a file in the same dir called tripleo-overcloud-passwords.  Should it be as easy as adding an additional password in there?  Or is that just the output from some other process?15:34
*** xinwu has quit IRC15:34
bnemecayoung: No, that file is only for passwords generated by the client.15:34
bnemecWhich this should be, but apparently it's not.15:35
ayoungbnemec, in this case, when you say client, is the the openstack client, or some deeper library?15:36
bnemecayoung: tripleoclient15:36
*** saneax is now known as saneax_AFK15:40
openstackgerritMerged openstack-infra/tripleo-ci: Rename the ceph to upgrades  https://review.openstack.org/29235915:41
openstackgerritMerged openstack/tripleo-puppet-elements: overcloud-controller: install python-networking-odl  https://review.openstack.org/26678115:43
openstackgerritMerged openstack/python-tripleoclient: Remove keystone init deprecation message  https://review.openstack.org/29057015:45
*** yamahata has quit IRC15:45
*** rpothier has left #tripleo15:46
openstackgerritDan Prince proposed openstack-infra/tripleo-ci: Metrics tracking for TripleO deployment tasks  https://review.openstack.org/29139315:46
openstackgerritDan Prince proposed openstack-infra/tripleo-ci: Add common bash functions to help track metrics.  https://review.openstack.org/29139215:46
*** absubram has joined #tripleo15:50
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Basic beaker one node test.  https://review.openstack.org/28137615:52
*** shivrao has joined #tripleo15:57
*** links has joined #tripleo15:58
rasca_Hi all, can you please help me to understand why while deploying mitaka with tripleo and having a validated instackenv.json if I run openstack baremetal import --json ~/instackenv.json I get just "baremetal" as output with exit code 1 and no host imported into ironic?15:58
*** rasca_ is now known as rasca15:58
rascanote that the same instackenv.json used with ospd7 and 8 works15:59
*** cwolferh has joined #tripleo16:06
rascathis is the output of the command http://pastebin.test.redhat.com/35687416:07
*** chem has joined #tripleo16:08
*** ramishra has quit IRC16:08
*** xinwu has joined #tripleo16:10
jistrhmm so client_manager doesn't have the baremetal attribute defined16:10
*** ramishra has joined #tripleo16:10
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Properly setup DNS for the container CI job  https://review.openstack.org/29302216:11
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Use Fedora 23 atomic in container gate  https://review.openstack.org/29302316:11
jistri wonder if that's specific to tripleoclient, or if that's something provided by openstackclient itself. d0ugal or thrash might know more about this16:11
jistrrasca: just guessing here, do you have python-ironicclient installed?16:11
jistron the machine where you run the `openstack` command16:12
d0ugaljrist: the baremetal client was added in python-tripleoclient16:13
d0ugalbut there is a patch to remove it in favour of the python-ironicclient one, I am not sure if that has landed.16:13
*** aufi has quit IRC16:13
d0ugaljrist: Yeah, it has: https://github.com/openstack/python-tripleoclient/commit/000f67a4c7150f983f5f4b71dc20850e5e460e0316:14
chemEmilienM: commented, that's a bit long....16:14
chemEmilienM: on  https://review.openstack.org/#/c/292664/16:14
d0ugalso, yeah, now baremetal is dependany on the python-ironicclient16:14
* jrist giggles at d0ugal 16:15
jistrjrist: :)16:15
mariosjrist: its the scotich accent16:15
jristit's called a brogyue16:15
EmilienMchem: the problem is rabbitmq is not binded on localhost16:16
d0ugalmarios: Yup, I'll blame that. It isn't me being lazy and careless at typing at all.16:16
EmilienMchem: so not sure it will work16:16
jistrrasca: so perhaps you could verify that python-ironicclient is installed, and try the import again16:16
chemEmilienM: it was before the patch I mentioned in the comment16:16
chemEmilienM: everything was localhost16:16
thrashjistr: the ironic client would need to provide that.16:17
EmilienMchem: "In this patch 93ba60204689e7391de8496d852572f82df8477f in OPM" the link does not work16:17
EmilienMso I don't know which commit you're talking about16:17
rascajistr, I've got that package python-ironicclient-0.8.1-1.el7.noarch16:17
rascajistr, and it was installed while I tried the test16:18
thrashrasca: the ironicclient osc plugin is in a bit of a state of flux. There isn't a whole lot there yet.16:18
thrashrasca: I missed the issue.16:18
*** snecklifter has joined #tripleo16:18
chemEmilienM: I find you the link 2s16:19
jistrrasca: hmm that seems old version though16:19
rascathrash,  http://pastebin.test.redhat.com/35687416:19
thrashrasca: looking16:19
jistrfor dev setups with mitaka, it's best to use tripleo.sh to set up the repositories16:19
chemEmilienM:https://github.com/redhat-openstack/openstack-puppet-modules/commit/93ba60204689e7391de8496d852572f82df8477f but anyway that really the import of the rabbitmq patch that matters16:20
*** ramishra has quit IRC16:21
*** snecklifter has quit IRC16:21
*** snecklifter has joined #tripleo16:21
rascajistr, yep, I've cloned the repo right now but it does not contain the tripleo.sh!16:22
thrashrasca: jistr that's it. 0.8.1 doesn't have the osc plugin in it.16:22
thrashrasca: I think it moved...16:22
thrashto tripleo-ci16:22
rascathrash, will give it a try16:22
thrashrasca: https://github.com/openstack-infra/tripleo-ci16:22
*** ramishra has joined #tripleo16:23
thrashrasca: https://raw.githubusercontent.com/openstack-infra/tripleo-ci/master/scripts/tripleo.sh16:23
EmilienMchem: tripleo master does not use opm16:23
thrashalthough you might need the rest of the repo.16:23
*** mbound has quit IRC16:23
EmilienMchem: so https://github.com/puppetlabs/puppetlabs-rabbitmq/commit/dac8de9d95c5771b7ef7596b73a59d4108138e3a. ?16:23
chemEmilienM: ack, then it just imported part of the new pupppetlabs-rabbitmq16:24
rascathrash, I'm givin it a try, but do I need to redeploy also the undercloud?16:24
EmilienMI think doing a curl in puppet manifest is the most horrible thing I've seen lately16:24
thrashrasca: you shouldn't..16:24
thrashjust update python-ironicclient16:24
EmilienMchem: ^16:24
rascaok then16:25
rascaafter the repo command I've got plenty of packages to update16:25
chemEmilienM:yes that's the one (the link in comment doesn't work as it include the dot at the end!)16:25
EmilienMchem: during your ipv6 tests with jpena, did you get it work?16:25
rascathrash, jistr let's see what happens16:25
chemEmilienM: we were working on osp-8 were this rabbitmq patch was not there16:25
EmilienMchem: https://github.com/puppetlabs/puppetlabs-rabbitmq/commit/dac8de9d95c5771b7ef7596b73a59d4108138e3a ?16:26
chemEmilienM: yes, that's the one :)16:26
*** thrash is now known as thrash|lunch16:27
EmilienMchem: how localhost can work if we have ipv6 binding? (localhost is ipv4 no?)16:27
openstackgerritDougal Matthews proposed openstack/tripleo-docs: [WIP] Deploy and Scale Swift  https://review.openstack.org/29171116:28
adarazsEmilienM: on most distros there's ::1 associated with localhost in /etc/hosts, but not sure if that matters.16:29
*** ramishra has quit IRC16:30
EmilienMadarazs, chem: did you test with localhost?16:30
chemEmilienM: ?16:30
EmilienMchem: your review inline, with localhost, did you test it on ipv6 networks?16:30
*** mikelk has quit IRC16:31
EmilienMbecause I'm not sure it will work16:31
*** ramishra has joined #tripleo16:31
adarazsnot sure what we're talking about exactly, just saying that localhost can be resolved to something ipv6 :)16:31
*** ohamada has quit IRC16:31
*** shivrao has quit IRC16:32
chemEmilienM: it's just a stuff that changed between working ipv6 and non working ipv6, but didn't took the time to make the ipv6 test (and I've lost my beaker instance, so that won't happen today)16:32
EmilienMadarazs: we're talking about https://review.openstack.org/#/c/292664/16:32
EmilienMadarazs: your bug16:32
EmilienMadarazs: do you have CI in place where we can use Depends-On to test the THT patch?16:33
adarazsEmilienM: yes. https://review.openstack.org/289445 -- if you add a depends on for this it should test it with the -ceph job.16:33
EmilienMadarazs: test ipv6?16:33
EmilienMlet me kick something then16:33
* adarazs meanwhile tries to debug the other two jobs there :)16:34
*** paramite has quit IRC16:34
*** jpena has joined #tripleo16:35
EmilienMchem: see jpena last comment if we set localhost, rabbitmq will be binded on localhost and clustering will fail16:35
EmilienMI think16:35
*** Marga_ has joined #tripleo16:35
*** rcernin has quit IRC16:35
*** eil397 has joined #tripleo16:38
*** andrearosa has joined #tripleo16:39
*** andrearosa has quit IRC16:39
*** andrearosa has joined #tripleo16:39
openstackgerritMerged openstack/tripleo-heat-templates: Change for configuring use_forwarded_for value for Nuage  https://review.openstack.org/29286816:40
jpenaEmilienM: yes, I think that'd be the case due to https://github.com/redhat-openstack/openstack-puppet-modules/blob/master/rabbitmq/manifests/config.pp#L61-L6416:40
EmilienMchem: so your proposal won't work either16:40
chemEmilienM: no, what jpena is saying is that bracket in the conf of the rabbitmq server don't work16:41
*** andrearosa has left #tripleo16:41
chemEmilienM: as for my proposal, well, rabbitmq would be listening on localhost behind the load-balacer, ... that may works :)16:41
jpenachem: but if we set node_ip_address='localhost', then rabbitmq-env.conf would contain NODE_IP_ADDRESS=localhost, binding rabbit only to
EmilienMand clustering config will fail16:42
chemEmilienM: ack16:42
*** andrearosa has joined #tripleo16:42
*** Guest71467 has quit IRC16:42
*** andrearosa has left #tripleo16:43
chemso localhost won't do and it really depends on the looks of the /etc/hosts (and of command idiosyncrasy) for it to have it working for ipv6 addr16:43
EmilienMchem: we need to fix https://github.com/puppetlabs/puppetlabs-rabbitmq/commit/dac8de9d95c5771b7ef7596b73a59d4108138e3a16:44
snecklifterdprince: ping, would it help if i abandoned https://review.openstack.org/#/c/291243/16:44
snecklifterto progress https://review.openstack.org/#/c/291322/316:44
EmilienMchem: imho, nothing has to be patched in tripleo16:45
*** morazi has quit IRC16:45
EmilienMchem: we need to fix puppetlabs-rabbitmq to drop brackets, wdyt?16:45
chemEmilienM: to drop ?16:45
EmilienMchem: we need to make curl happy, right?16:46
chemEmilienM: to add you mean ?16:46
*** links has quit IRC16:47
EmilienMchem: in your review, you say brackets does not work either16:47
chemEmilienM: that's down the rabbit hole: the thing is that even with the bracket curl send a error if 1: globbing in on (look for globoff in curl's man) ; 2: bracket are not escaped like this "\[::1\]"16:48
*** xinwu has quit IRC16:49
EmilienMchem: so what do you suggest?16:50
EmilienMchem: https://github.com/puppetlabs/puppetlabs-rabbitmq/search?utf8=%E2%9C%93&q=node_ip_address16:50
EmilienMsetting node_ip_address to localhost is a bad idea, because it will break rabbitmq binding on multi node setups.16:51
chemEmilienM:aggree, I missed the conf in the template16:51
chemEmilienM: to make thing right, as it is it would require : patch in rabbitmq and patch in staging::file16:52
chemEmilienM: at least that all I'm thinking about right now.16:53
EmilienMwhy staging::file?16:53
chemEmilienM: because it's the one actually running the curl command16:53
chemEmilienM: so it would be responsible to add the escape around the ipv616:53
EmilienMchem: ok, so we need to add the right options for ipv6?16:53
*** dprince has quit IRC16:54
EmilienMchem: this one ? https://github.com/voxpupuli/puppet-staging16:54
chemEmilienM: no16:54
chemEmilienM: that one https://github.com/nanliu/puppet-staging16:55
EmilienMchem: nanliu is a fork16:55
chemEmilienM: yep but it's the one used by puppetlab16:55
EmilienMwe need to use voxpupuli, which is the official namespace for Puppet community16:55
*** mbound has joined #tripleo16:55
EmilienMoh dear16:56
EmilienMwe need to fix that too16:56
chemthat's down the rabbit hole...16:56
EmilienMchem: I'm working on a puppetlabs-rabbitmq to change staging upstream repo16:57
*** olap has quit IRC16:58
openstackgerritDougal Matthews proposed openstack/tripleo-docs: Deploy and Scale Swift  https://review.openstack.org/29171116:59
*** tzumainn has quit IRC17:03
openstackgerrityolanda.robla proposed openstack/diskimage-builder: Generate fedora-atomic images using dib  https://review.openstack.org/28716717:04
adarazsderekh: can I ask your help for debugging the IPv6 (non-ha) gate? I'm really stuck. :/17:04
adarazsI see the "Service Unavailable (HTTP 503)" in the logs, but there's nothing in the glance logs on the overcloud controller and I don't find any more info either.17:04
adarazsderekh: if you can show me how I can gain access and debug a machine that is running/was running the job, that would be helpful.17:05
adarazsI could go and look around and try the same command and see what's wrong.17:05
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Deploy ceph in the HA job  https://review.openstack.org/29305117:06
*** Slower has joined #tripleo17:08
adarazsderekh: also I'm trying to be able to be more helpful in general in tripleo-ci, so that wouldn't be useless anyway. :)17:08
*** shivrao has joined #tripleo17:08
derekhadarazs: Once the ci tests are finished the instances are gone, can you point me at one of the runs your looking at and I'll take a quick look17:09
derekhadarazs: are the API enpoints v6 ip addresses ?17:09
*** thrash|lunch is now known as thrash17:09
adarazsderekh: http://logs.openstack.org/45/289445/9/check-tripleo/gate-tripleo-ci-f22-nonha/6d00d11/console.html -- yes, they do look like ipv6 addresses17:09
derekhadarazs: looking17:10
adarazsderekh: I know that they are gone, but not completely. I saw that we can get into them until they get recycled. that could be useful. :)17:10
*** leanderthal is now known as leanderthal|afk17:13
*** openstackgerrit has quit IRC17:14
*** openstackgerrit_ is now known as openstackgerrit17:14
*** openstackgerrit has quit IRC17:15
*** akrivoka has quit IRC17:15
*** openstackgerrit_ has joined #tripleo17:15
*** openstackgerrit_ is now known as openstackgerrit17:16
*** dshulyak has quit IRC17:16
*** dshulyak has joined #tripleo17:16
*** openstackgerrit_ has joined #tripleo17:16
*** tzumainn has joined #tripleo17:16
openstackgerritKeith Schincke proposed openstack/tripleo-heat-templates: Add support for deploying RGW on controllers.  https://review.openstack.org/28902717:17
derekhadarazs: I'll have to run one manually if you want to poke around it, I'll do that now and give you access, it will be ready to look at in about 90 minutes, I'll be gone but if you ant I'll poke around on it later tonight and see if I can find the problem17:18
*** oshvartz has joined #tripleo17:18
adarazsderekh: yeah, I'll have to get going before that. will that instance stay up? can I look at it tomorrow morning?17:19
derekhadarazs: yes, I'll kick it off now and check it later if it reproduces the problem I'll let you know17:20
adarazsderekh: okay! thank you for the help :)17:20
derekhadarazs: no prob17:20
adarazsderekh: have a nice afternoon/evening. :)17:20
*** xinwu has joined #tripleo17:23
*** rwsu has quit IRC17:24
*** pcaruana has quit IRC17:24
openstackgerritBen Nemec proposed openstack/instack-undercloud: Set glance::registry::auth_uri  https://review.openstack.org/29306017:24
*** shivrao has quit IRC17:25
openstackgerritBen Nemec proposed openstack/instack-undercloud: Set glance::registry::auth_uri  https://review.openstack.org/29306017:27
*** MaxPC has quit IRC17:30
*** dprince has joined #tripleo17:33
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Test overcloud SSL  https://review.openstack.org/28198817:35
*** mannidi_ has joined #tripleo17:35
*** pcaruana has joined #tripleo17:36
EmilienMderekh, adarazs: I sent an update about RabbitMQ IPv6 thing - things are blocked now, I'm trying to find a workaround in the meantime17:36
EmilienMderekh: or we could pin puppetlabs-rabbitmq17:38
*** panda has quit IRC17:40
derekhEmilienM: we could test adarazs's patch with a pin if it works then add a pin temporarily sounds ok to me17:40
EmilienMderekh: can I patch tripleo-common/scripts/tripleo.sh?17:40
*** rwsu has joined #tripleo17:41
EmilienMor did it move to somewhere else?17:41
*** morazi has joined #tripleo17:41
derekhEmilienM: its now in tripleo-ci17:41
derekhEmilienM: but the process is the same17:41
*** panda has joined #tripleo17:41
*** akuznetsov has joined #tripleo17:42
*** shivrao has joined #tripleo17:42
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: pin puppetlabs-rabbitmq  https://review.openstack.org/29307417:45
EmilienMderekh: ^17:45
derekhEmilienM: does that need to be pinned in the overcloud aswell ?17:46
EmilienMderekh: everywhere17:46
EmilienMI need to add it17:46
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: pin puppetlabs-rabbitmq  https://review.openstack.org/29307417:46
derekhEmilienM: needs to also go into the overcloud_images functions17:46
*** jpena is now known as jpena|off17:47
EmilienMI'll rebase Attila on top of mine.17:47
EmilienMadarazs: fyi17:47
*** akuznetsov has quit IRC17:48
derekhadarazs, when EmilienM has finished that I'll restart my manual job so I can poke around if its still failing17:49
derekhgotta go for now will kick it off in a bit17:49
*** derekh has quit IRC17:49
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: WIP: Use IPv6 on the gate jobs a test  https://review.openstack.org/28944517:52
EmilienMadarazs: rebased ^17:52
*** akrivoka has joined #tripleo17:54
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add undercloud idempotency test  https://review.openstack.org/27921817:55
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Test overcloud SSL  https://review.openstack.org/28198818:00
*** akrivoka has quit IRC18:03
openstackgerritimain proposed openstack/tripleo-heat-templates: Remove hack the pulls latest docker  https://review.openstack.org/25862518:05
*** lucasagomes is now known as lucas-dinner18:10
openstackgerritRyan Hallisey proposed openstack/tripleo-heat-templates: Remove hack the pulls latest docker  https://review.openstack.org/25862518:11
*** mannidi_ has quit IRC18:15
*** jistr has quit IRC18:15
openstackgerritMiles Gould proposed openstack/python-tripleoclient: Use Ironic API v1.11 to support ENROLL state  https://review.openstack.org/27220618:18
thrashd0ugal: https://review.openstack.org/#/c/29247118:18
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Use Fedora 23 atomic in container gate  https://review.openstack.org/29302318:19
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891518:19
ayoungHey, what is the simplest smoke test for an overcloud deployment?18:20
*** akrivoka has joined #tripleo18:22
colonwqyou can write a script to create user, upload glance image, configure network, start image(s), access image(s) via ssh on the various networks, create and attache cinder.18:22
bnemecayoung: tripleo.sh --overcloud-pingtest18:25
*** MaxPC has joined #tripleo18:26
ayoungbnemec, awesome thanks.18:26
toskyayoung: test_minimum_basic from tempest18:26
*** eil397 has left #tripleo18:33
*** mgould has quit IRC18:35
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Use Fedora 23 atomic in container gate  https://review.openstack.org/29302318:35
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891518:35
*** mkovacik has quit IRC18:40
*** jdob1 has joined #tripleo18:43
*** akrivoka has quit IRC18:50
*** electrofelix has quit IRC18:53
openstackgerritDan Prince proposed openstack/tripleo-docs: Network isolation for virtualized environments  https://review.openstack.org/29309118:54
*** jdob1 has quit IRC18:55
dprincedsneddon: we are too the point where I would like to invest the time in a focussed functional test suite for os-net-config18:58
dsneddondprince, I agree. I've been thinking about how to do that. Perhaps with dummy interfaces?18:58
dprincedsneddon: we don't have enough TripleO CI scenario's to test all of this I think18:58
dprincedsneddon: I'm thinking of a different test harness that uses perhaps more the devstack-gate type of testing18:58
dsneddonOh, I see, you mean an integrated functional test where it is tested inside TripleO?18:58
dprincedsneddon: but yes, with dummy interfaces18:58
dprincedsneddon: we should be able to cover quite a bit of this testing with dummy interfaces, even some bonding18:59
dsneddondprince, Agreed18:59
dprincedsneddon: probably won't be able to cover the finer points (LACP, etc) but that is okay18:59
ryansbHow does changing the OS_PASSWORD undercloud work? I see the stackrc gets it from hiera, is there a way to change it post-undercloud install?19:00
dprincedsneddon: anyways, this would be just os-net-config focussed I think. Not directly related to TripleO heat templates19:00
*** bnemec has quit IRC19:03
*** shivrao has quit IRC19:04
*** tosky has quit IRC19:13
*** bnemec has joined #tripleo19:13
openstackgerritKeith Schincke proposed openstack/tripleo-heat-templates: Add support for deploying RGW on controllers.  https://review.openstack.org/28902719:14
*** bnemec has quit IRC19:16
*** dprince has quit IRC19:18
*** bnemec has joined #tripleo19:18
*** shivrao has joined #tripleo19:18
*** trozet has joined #tripleo19:20
*** akrivoka has joined #tripleo19:30
*** dprince has joined #tripleo19:31
*** athomas has quit IRC19:32
*** trozet has quit IRC19:32
openstackgerritSagi Shnaidman proposed openstack/instack-undercloud: Use installed tempest version  https://review.openstack.org/29310119:32
openstackgerritSagi Shnaidman proposed openstack/instack-undercloud: Use installed tempest version  https://review.openstack.org/29310119:37
*** athomas has joined #tripleo19:40
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Use Fedora 23 atomic in container gate  https://review.openstack.org/29310819:44
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891519:46
*** dprince has quit IRC19:48
openstackgerritRyan Hallisey proposed openstack/tripleo-heat-templates: Remove hack the pulls latest docker  https://review.openstack.org/29311019:49
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the continer job to run again  https://review.openstack.org/28891519:50
*** ebalduf has joined #tripleo19:52
*** Goneri has quit IRC19:57
*** zaneb has joined #tripleo19:58
*** MaxPC has quit IRC19:58
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Test overcloud SSL  https://review.openstack.org/28198820:08
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add undercloud idempotency test  https://review.openstack.org/27921820:10
bnemecWhy must everything have moved since I started my CI patches? :-P20:11
*** jayg is now known as jayg|g0n320:15
*** Marga_ has quit IRC20:19
rhalliseybnemec, I'm in the same boat :/20:29
rhalliseyhad to redo a bunch of patches20:29
*** bswartz has quit IRC20:29
*** bswartz has joined #tripleo20:30
*** shardy has quit IRC21:10
*** lblanchard has quit IRC21:15
*** jtomasek has quit IRC21:16
*** jtomasek has joined #tripleo21:18
openstackgerritMichele Baldessari proposed openstack/python-tripleoclient: Clarify error when a template is missing  https://review.openstack.org/29314321:20
michchapEmilienM: morning21:21
EmilienMmichchap: hey, no progress on my side yet21:21
EmilienMI was in meetings and also stuck on a rabbitmq/ipv6 stuff21:22
openstackgerritMerged openstack/os-refresh-config: Updated from global requirements  https://review.openstack.org/26848321:22
openstackgerritMerged openstack/os-apply-config: Updated from global requirements  https://review.openstack.org/26848021:22
*** yamahata has joined #tripleo21:22
EmilienMmichchap: if you don't work on it, I'll probably do it tonight, after my bugfix for ipv621:22
michchapEmilienM: no worries. It failed ha CI so I'll have to double check that.21:22
EmilienMmichchap: ok. I'll let you know when I'm on it again21:23
michchapEmilienM: looks like the db sync HA code isn't correct21:24
EmilienMwrong copy paste?21:25
*** trown|ooobootcam has quit IRC21:27
openstackgerritMerged openstack/tripleo-common: Fix typos in code  https://review.openstack.org/28970721:28
*** jtomasek has quit IRC21:28
openstackgerritMerged openstack/tripleo-heat-templates: Fix typos  https://review.openstack.org/26512621:29
michchapError: Could not prefetch mysql_user provider 'mysql': Execution of '/usr/bin/mysql -NBe SELECT CONCAT(User, '@',Host) AS User FROM mysql.user' returned 1: ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock'21:29
michchapEmilienM: ^21:29
openstackgerritMerged openstack/tripleo-heat-templates: Install mongodb client package  https://review.openstack.org/27839421:30
EmilienMmichchap: sounds like galera did not start?21:30
michchapEmilienM: ah, it's running on multiple nodes21:30
michchapEmilienM: missing the check for pacemaker master21:30
openstackgerritMerged openstack-infra/tripleo-ci: Archive all of the delorean logs  https://review.openstack.org/27141621:31
*** rdopiera has quit IRC21:31
openstackgerritMerged openstack/tripleo-image-elements: py26 is no longer supported by Infra's CI  https://review.openstack.org/26170221:31
*** jprovazn has quit IRC21:32
openstackgerritMichael Chapman proposed openstack/puppet-tripleo: Add keystone and db sync profiles  https://review.openstack.org/29271221:33
*** colonwq has quit IRC21:33
openstackgerritMerged openstack/python-tripleoclient: Allow node import via yaml not only csv/json  https://review.openstack.org/25522821:33
openstackgerritDan Prince proposed openstack-infra/tripleo-ci: Metrics tracking for TripleO deployment tasks  https://review.openstack.org/29139321:34
openstackgerritMerged openstack/tripleo-puppet-elements: py26 is no longer supported by Infra's CI  https://review.openstack.org/26170321:35
*** pcaruana has quit IRC21:38
*** pradk has quit IRC21:38
*** panda has quit IRC21:40
*** panda has joined #tripleo21:40
*** pcaruana has joined #tripleo21:52
*** Marga_ has joined #tripleo21:56
*** yamahata has quit IRC22:00
*** kbyrne has quit IRC22:01
*** akrivoka has quit IRC22:09
*** rbrady has quit IRC22:10
*** david-lyle has quit IRC22:13
*** r-mibu has quit IRC22:20
*** r-mibu has joined #tripleo22:20
*** absubram has quit IRC22:21
*** dshulyak has quit IRC22:22
*** MaxPC has joined #tripleo22:30
*** shakamunyi has joined #tripleo22:32
*** weshay has quit IRC22:48
*** tzumainn has quit IRC22:49
*** mbound has quit IRC22:50
*** saneax_AFK is now known as saneax22:56
*** MaxPC has quit IRC23:00
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: pin puppetlabs-rabbitmq  https://review.openstack.org/29307423:04
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: WIP: Use IPv6 on the gate jobs a test  https://review.openstack.org/28944523:04
*** saneax is now known as saneax_AFK23:08
*** Goneri has joined #tripleo23:09
*** chlong has joined #tripleo23:27
*** xinwu has quit IRC23:34
*** tzumainn has joined #tripleo23:36
*** mbound has joined #tripleo23:51
*** mbound has quit IRC23:56
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: WIP: Use IPv6 on the gate jobs a test  https://review.openstack.org/28944523:58
*** xinwu has joined #tripleo23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!