| *** jistr has quit IRC | 00:00 | |
| *** jayg is now known as jayg|g0n3 | 00:00 | |
| *** jistr has joined #tripleo | 00:01 | |
| *** maeca1 has quit IRC | 00:21 | |
| *** fultonj has quit IRC | 00:35 | |
| *** limao has joined #tripleo | 00:40 | |
| *** fragatin_ has quit IRC | 01:12 | |
| *** fragatina has joined #tripleo | 01:12 | |
| *** bfournie has joined #tripleo | 01:22 | |
| *** limao has quit IRC | 01:32 | |
| *** limao has joined #tripleo | 01:33 | |
| *** mbound has joined #tripleo | 01:51 | |
| *** mbound has quit IRC | 01:52 | |
| openstackgerrit | Ade Lee proposed openstack/tripleo-heat-templates: Add Barbican to the overcloud - WIP https://review.openstack.org/352458 | 01:58 |
|---|---|---|
| *** [1]cdearborn has quit IRC | 02:07 | |
| openstackgerrit | Ryan Brady proposed openstack/tripleo-common: Adds parameters actions https://review.openstack.org/298682 | 02:07 |
| *** ayoung has quit IRC | 02:11 | |
| *** yamahata has joined #tripleo | 02:16 | |
| *** jdob has quit IRC | 02:30 | |
| *** coolsvap has joined #tripleo | 02:49 | |
| *** julim has quit IRC | 03:30 | |
| *** limao has quit IRC | 03:49 | |
| *** limao has joined #tripleo | 03:50 | |
| *** penick has quit IRC | 03:53 | |
| *** limao has quit IRC | 03:55 | |
| *** padkrish has joined #tripleo | 04:01 | |
| *** links has joined #tripleo | 04:01 | |
| openstackgerrit | Matt Young proposed openstack/tripleo-quickstart: Add check for virthost being defined but empty https://review.openstack.org/332872 | 04:13 |
| *** limao has joined #tripleo | 04:17 | |
| *** tzumainn has quit IRC | 04:20 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Configure keystone endpoints in service profiles https://review.openstack.org/352364 | 04:21 |
| *** masco has joined #tripleo | 04:27 | |
| *** fragatin_ has joined #tripleo | 04:36 | |
| *** fragatina has quit IRC | 04:39 | |
| *** fragatin_ has quit IRC | 04:40 | |
| *** limao_ has joined #tripleo | 04:48 | |
| *** coolsvap is now known as _coolsvap_ | 04:49 | |
| *** limao has quit IRC | 04:50 | |
| *** jaosorior has joined #tripleo | 04:55 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Handle redirects for keystone https://review.openstack.org/352956 | 05:00 |
| *** padkrish has quit IRC | 05:00 | |
| *** pcaruana has quit IRC | 05:01 | |
| *** skramaja has joined #tripleo | 05:01 | |
| *** _coolsvap_ is now known as coolsvap | 05:16 | |
| *** oshvartz has quit IRC | 05:18 | |
| *** numan_ has joined #tripleo | 05:20 | |
| -openstackstatus- NOTICE: zuul is being restarted to reload configuration. Jobs should be re-enqueued but if you're missing anything (and it's not on http://status.openstack.org/zuul/) please issue a recheck in 30min. | 05:26 | |
| *** skramaja has quit IRC | 05:38 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/python-tripleoclient: Add Barbican password to deployment https://review.openstack.org/353257 | 05:38 |
| *** skramaja has joined #tripleo | 05:38 | |
| *** jpeeler has quit IRC | 05:45 | |
| *** jpeeler has joined #tripleo | 05:49 | |
| *** penick has joined #tripleo | 05:51 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add Barbican to the overcloud - WIP https://review.openstack.org/352458 | 05:52 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: DO NOT MERGE - TESTING BARBICAN https://review.openstack.org/353262 | 05:54 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: DO NOT MERGE - TESTING BARBICAN https://review.openstack.org/353262 | 05:55 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add Barbican to the overcloud - WIP https://review.openstack.org/352458 | 05:55 |
| *** leanderthal|afk is now known as leanderthal | 06:03 | |
| *** rcernin has joined #tripleo | 06:04 | |
| *** akuznetsov has joined #tripleo | 06:13 | |
| *** bana_k has joined #tripleo | 06:17 | |
| *** saneax_AFK is now known as saneax | 06:19 | |
| *** aufi has joined #tripleo | 06:19 | |
| *** saneax is now known as Guest38498 | 06:19 | |
| *** tremble has joined #tripleo | 06:19 | |
| *** bootsha has joined #tripleo | 06:29 | |
| *** jprovazn has joined #tripleo | 06:34 | |
| *** akuznetsov has quit IRC | 06:35 | |
| *** fragatina has joined #tripleo | 06:37 | |
| *** oshvartz has joined #tripleo | 06:37 | |
| *** ebalduf has quit IRC | 06:41 | |
| *** fragatina has quit IRC | 06:42 | |
| *** tesseract- has joined #tripleo | 06:44 | |
| *** penick has quit IRC | 06:45 | |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart: Reproducing specific CI run https://review.openstack.org/340156 | 06:48 |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart: Devmode: rename delorean current so any existing repo is overwritten https://review.openstack.org/335953 | 06:48 |
| *** tosky has joined #tripleo | 06:49 | |
| *** bana_k has quit IRC | 06:53 | |
| *** liverpooler has joined #tripleo | 06:56 | |
| *** sshnaidm has quit IRC | 06:56 | |
| *** milan has joined #tripleo | 06:59 | |
| *** ccamacho has quit IRC | 07:01 | |
| *** radeks has joined #tripleo | 07:01 | |
| *** ccamacho has joined #tripleo | 07:04 | |
| ccamacho | morning o/ | 07:06 |
| matbu | o/ | 07:07 |
| openstackgerrit | Merged openstack/puppet-tripleo: Fix mistral sync_db profile steps https://review.openstack.org/352937 | 07:09 |
| *** cmyster_ is now known as cmyster | 07:09 | |
| *** cmyster has quit IRC | 07:09 | |
| *** cmyster has joined #tripleo | 07:09 | |
| openstackgerrit | Merged openstack/puppet-tripleo: Fix use of bootstrap_node in cinder base profile https://review.openstack.org/352365 | 07:09 |
| *** stendulker has joined #tripleo | 07:15 | |
| *** jpena|off is now known as jpena | 07:16 | |
| *** jistr has quit IRC | 07:18 | |
| *** crinkle has quit IRC | 07:19 | |
| *** greghaynes has quit IRC | 07:20 | |
| *** greghaynes has joined #tripleo | 07:20 | |
| *** athomas has joined #tripleo | 07:21 | |
| *** crinkle has joined #tripleo | 07:21 | |
| *** pcaruana has joined #tripleo | 07:24 | |
| *** mcornea has joined #tripleo | 07:24 | |
| *** ifarkas_afk is now known as ifarkas | 07:25 | |
| *** jistr has joined #tripleo | 07:25 | |
| *** skramaja has quit IRC | 07:26 | |
| *** Guest38498 is now known as saneax_AFK | 07:27 | |
| *** skramaja has joined #tripleo | 07:27 | |
| *** akuznetsov has joined #tripleo | 07:27 | |
| *** jprovazn has quit IRC | 07:29 | |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Add a dynamic ansible inventory script https://review.openstack.org/348305 | 07:29 |
| *** sshnaidm has joined #tripleo | 07:29 | |
| *** florianf has joined #tripleo | 07:31 | |
| *** ccamacho is now known as ccamacho|afk | 07:33 | |
| *** bootsha has quit IRC | 07:35 | |
| *** ebarrera has joined #tripleo | 07:35 | |
| *** shardy has joined #tripleo | 07:35 | |
| openstackgerrit | Xiang Chen proposed openstack/diskimage-builder: Explain difference between 'DIB_OFFLINE' and 'DIB_DEBIAN_USE_DEBOOTSTRAP_CACHE' https://review.openstack.org/345935 | 07:36 |
| matbu | marios: jistr gfidente|afk hi guys, i would like to understand this commit : https://github.com/openstack/tripleo-heat-templates/commit/cacf12c855d18f952c6915e5629fb26acdeddd21 | 07:37 |
| matbu | why the default has been removed | 07:37 |
| marios | matbu: so it would error if it wasn't set | 07:37 |
| matbu | marios: ya | 07:38 |
| matbu | and it does :) | 07:38 |
| marios | gfidente|afk: can explain better matbu, but it was set for new deployments by the client but not for updates | 07:38 |
| marios | matbu: so he wants to make sure it is set for update and not just the default | 07:38 |
| *** bootsha has joined #tripleo | 07:38 | |
| marios | matbu: i think was added to the client side to generate but again this is all just from looking at those/discussing with gfide | 07:38 |
| marios | matbu: possibly the client side didnt' get to mitaka? yet? | 07:39 |
| matbu | marios: k, but it's only set w/ a ceph deployment ? | 07:40 |
| marios | matbu: hmmm i think so but i am not 100% | 07:40 |
| *** hewbrocca-afk is now known as hewbrocca | 07:40 | |
| jaosorior | The OVB gate jobs seem to be queueing pretty badly. Somebody knows what's up? | 07:40 |
| *** abehl has joined #tripleo | 07:40 | |
| *** ramishra has quit IRC | 07:41 | |
| matbu | marios: k, i'll check, but in the case of a minimal deploy or without ceph, the default is required i guess, or at least could be overriden with an extra template file | 07:41 |
| jaosorior | shardy hey dude | 07:41 |
| marios | matbu: ack yeah i guess just add to your network_env.yaml or something for now | 07:41 |
| jaosorior | shardy: I pinged you yesterday with some thoughts about the haproxy service. Did you get time to read it? | 07:42 |
| *** jpich has joined #tripleo | 07:43 | |
| *** abehl has quit IRC | 07:44 | |
| shardy | jaosorior: Sorry I missed it, checking it now | 07:44 |
| matbu | marios: ack | 07:44 |
| *** abehl has joined #tripleo | 07:44 | |
| shardy | jaosorior: tbh I'm not keen on special-casing any services if we can avoid it | 07:44 |
| shardy | jaosorior: have you checked out how the composable firewall rules were done via hiera? | 07:44 |
| shardy | wondering if the same pattern can be reused | 07:44 |
| *** ramishra has joined #tripleo | 07:45 | |
| *** dtantsur|afk is now known as dtantsur | 07:45 | |
| shardy | jaosorior: also, I'm working on an approach that will provide a list of IPs running every enabled service | 07:45 |
| jaosorior | shardy: Haven't checked, can you point to that? | 07:45 |
| shardy | jaosorior: does that help? | 07:45 |
| shardy | https://review.openstack.org/#/c/348974/2/overcloud.yaml | 07:46 |
| jaosorior | shardy: maybe. I'm not entirely sure. I would like to stop hardcoding those ports in haproxy and the services and just fetch them from t-h-t | 07:46 |
| shardy | jaosorior: that is the composable _node_ips patch | 07:46 |
| jaosorior | I mean the firewall settings via hiera | 07:46 |
| shardy | so that will allow us to generate the IPs for all services without any hard-coded stuff in overcloud.yaml | 07:46 |
| shardy | yeah, one sec | 07:46 |
| shardy | https://github.com/openstack/puppet-tripleo/commit/31278c995f3149a29958d1205001c8ce7aba5884 | 07:47 |
| shardy | that's the puppet-tripleo part | 07:47 |
| * shardy gets the t-h-t patch | 07:48 | |
| jaosorior | shardy: I do see some firewall settings are in each specific manifest. And that's pretty cool. However, that works cause we need to set up those firewall settings for that node where the service is being deployed | 07:48 |
| shardy | git log --oneline | grep firewall | 07:48 |
| jaosorior | we don't have that assurance for haproxy as a composable service | 07:48 |
| jaosorior | the way things are heading, we could have one node type only deploy haproxy, and another node type contain the openstack services | 07:49 |
| jaosorior | so then, how would the haproxy manifest fetch that configuration? | 07:49 |
| shardy | https://review.openstack.org/#/c/344860/ | 07:49 |
| openstackgerrit | Babu Shanmugam proposed openstack/puppet-tripleo: Add ML2/OVN support for neutron profile https://review.openstack.org/314875 | 07:50 |
| shardy | jaosorior: lets take a step back for a second, what data related to the services does haproxy need? Lets define it exactly | 07:50 |
| jaosorior | sure | 07:50 |
| jaosorior | it needs the ports for all the services it points to | 07:50 |
| jaosorior | it needs which services are enabled | 07:50 |
| *** ioggstream has joined #tripleo | 07:50 | |
| jaosorior | it needs which node runs which service | 07:50 |
| jaosorior | maybe it needs even the fqdn of those nodes (if TLS is enabled internally) | 07:51 |
| shardy | Ok, thanks, lets take those one by one: | 07:51 |
| jaosorior | sure | 07:51 |
| shardy | The ports are in the EndpointMap | 07:51 |
| jaosorior | right | 07:52 |
| shardy | The EnabledServices can be created e.g see https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud.yaml#L1112 | 07:52 |
| shardy | the ServiceChains give us that | 07:52 |
| jaosorior | that we can get at the moment (but not as EndpointMap is planning to be changed) | 07:52 |
| shardy | jaosorior: why is EndpointMap planning to be changed? | 07:52 |
| jaosorior | shardy: Wasn't the plan to pass the information from the endpoint map to the specific service yaml files? | 07:53 |
| shardy | jaosorior: Yeah, but I think it's probably something we'll have to defer, the chicken/egg problem is kind of hard to break | 07:53 |
| jaosorior | sure | 07:53 |
| jaosorior | then lets assume we have that one | 07:54 |
| jaosorior | that's cool | 07:54 |
| jaosorior | ok, now to EnabledServices | 07:54 |
| *** akuznetsov has quit IRC | 07:54 | |
| jaosorior | shardy: That's done in the output of overcloud.yaml, haproxy needs that, how can we make it available? | 07:54 |
| shardy | jaosorior: we can just write it for all nodes in hiera | 07:54 |
| jaosorior | shardy: sure, I'm fine with that | 07:55 |
| shardy | https://review.openstack.org/#/c/348974/2/overcloud.yaml | 07:55 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Do not merge, checking composable mistral. https://review.openstack.org/352947 | 07:55 |
| jaosorior | shardy: cause right now, people enable their service in the haproxy service yaml, even though that service is not enabled... so we have a bunch of ports open and not listening to anything | 07:55 |
| shardy | so if you look there, we can have $role_names, $role_ips, $service_ips and enabled_services | 07:55 |
| *** ccamacho|afk is now known as ccamacho | 07:56 | |
| shardy | combining that data gives you what you need I think? | 07:56 |
| jaosorior | shardy: sure, if we can get that for all the node types | 07:56 |
| shardy | jaosorior: Ok, well rebasing that patch is on my todo list for today, if I add the enabled_services as another patch, would you OK with pulling, testing and seeing if it can be wired in to haproxy to do what you need? | 07:57 |
| jaosorior | shardy: Sure, will do | 07:57 |
| jaosorior | shardy: just point me to the stuff to pull and I'll test it out | 07:57 |
| jaosorior | shardy: In the meantime, can you review this patch? https://review.openstack.org/#/c/352364/ it does the keystone endpoint provisioning from each service manifest in the bootstrap node | 07:58 |
| jaosorior | been testing that out in my machine. And also this patch tests in in CI in the nonha job https://review.openstack.org/#/c/352705/ | 07:59 |
| *** bootsha has quit IRC | 07:59 | |
| jaosorior | shardy: Ah! One more thing. I've been working on making the CA deployment be based on puppet/hiera instead of the bash-based current approach. EmilienM asked me to do it as a composable service, and I'm fine with that | 08:01 |
| jaosorior | only thing I'm afraid of, is that people will end up not adding the CA service to the service list | 08:01 |
| jaosorior | and stuff will break and that will be a mess to debug | 08:01 |
| shardy | jaosorior: how does this play with the os-cloud-config stuff run by tripleoclient? Does it silently ignore the stuff already created? | 08:01 |
| jaosorior | with BZ such as (TLS doesn't work) | 08:01 |
| shardy | I'm wondering how we prove it works in CI, unless we have a dependent patch removing/disabling the os-cloud-config logic | 08:01 |
| jaosorior | let me re-check that | 08:02 |
| shardy | d0ugal: ^^ FYI this may simplify things for the switch to mistral for deployment | 08:02 |
| d0ugal | shardy: reading. | 08:03 |
| shardy | jaosorior: the patch looks fine, but I'd prefer to see proof all endpoints are fine with os-cloud-config disabled | 08:03 |
| shardy | I can test it locally, but a CI pass would be even better so folks can refer to it | 08:03 |
| d0ugal | Neat, this does seem useful :) | 08:03 |
| jaosorior | shardy: So what happens is that the tripleoclient goes ahead and iterates the service list | 08:04 |
| jaosorior | if it doesn't find the service in keystone, then it tries to add it | 08:04 |
| *** sanjayu|AFK has quit IRC | 08:04 | |
| jaosorior | if it finds it, then it ignores it | 08:04 |
| *** sanjayu|AFK has joined #tripleo | 08:04 | |
| shardy | jaosorior: Ok, so we have no way of knowing if your patch missed some services and os-cloud-config filled in the gaps ;) | 08:05 |
| *** bootsha has joined #tripleo | 08:05 | |
| jaosorior | shardy: Not at the moment | 08:05 |
| jaosorior | shardy: What I could do, is add an output to t-h-t that tells tripleoclient not to manage the endpoints | 08:05 |
| shardy | jaosorior: this is important because it would be good to drop os-cloud-config in the transition to using mistral for deployment | 08:05 |
| jaosorior | if the client notices this, then it should skip the postconfig entirely | 08:06 |
| *** lucas-dinner is now known as lucasagomes | 08:06 | |
| jaosorior | this is nice in the sense that it allows for backward compatibility | 08:06 |
| shardy | jaosorior: +1 that sounds like a good idea | 08:06 |
| jaosorior | alright | 08:07 |
| jaosorior | I'll code that then | 08:07 |
| *** jprovazn has joined #tripleo | 08:08 | |
| *** chem has joined #tripleo | 08:10 | |
| hewbrocca | os-cloud-config.die.die.die | 08:10 |
| d0ugal | lol | 08:11 |
| *** jayg|g0n3 has quit IRC | 08:12 | |
| *** jeckersb_gone has quit IRC | 08:12 | |
| *** radez has quit IRC | 08:16 | |
| *** openstackgerrit has quit IRC | 08:18 | |
| *** openstackgerrit has joined #tripleo | 08:18 | |
| *** social has quit IRC | 08:20 | |
| jaosorior | hewbrocca: I'm on it :P | 08:22 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Configure keystone endpoints in service profiles https://review.openstack.org/352364 | 08:23 |
| jaosorior | shardy, gfidente: just removed the Depends-On ^^ | 08:23 |
| jaosorior | cause i'll add a t-h-t patch that depends on that commit. So this way we avoid cycles | 08:23 |
| *** amoralej|off is now known as amoralej | 08:26 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add flag to assert that puppet manages the keystone endpoints https://review.openstack.org/353307 | 08:27 |
| *** saneax_AFK is now known as saneax | 08:27 | |
| *** saneax is now known as Guest25838 | 08:28 | |
| *** electrofelix has joined #tripleo | 08:36 | |
| *** akuznetsov has joined #tripleo | 08:37 | |
| *** skramaja has quit IRC | 08:38 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/python-tripleoclient: Skip post-config based on t-h-t flag https://review.openstack.org/353316 | 08:40 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/python-tripleoclient: Skip post-config based on t-h-t flag https://review.openstack.org/353316 | 08:41 |
| jaosorior | Done | 08:41 |
| *** skramaja has joined #tripleo | 08:41 | |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Add a dynamic ansible inventory script https://review.openstack.org/348305 | 08:42 |
| *** shardy has quit IRC | 08:44 | |
| *** shardy has joined #tripleo | 08:45 | |
| *** ohamada has joined #tripleo | 08:46 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: DO NOT MERGE - TESTING BARBICAN https://review.openstack.org/353262 | 08:47 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add Barbican to the overcloud - WIP https://review.openstack.org/352458 | 08:47 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Retain existing ComputeHostnameFormat when upgrading older envs https://review.openstack.org/305705 | 08:48 |
| jaosorior | d0ugal: so... I think in RDO mistral still hasn't been promoted :/ | 08:48 |
| *** jayg|g0n3 has joined #tripleo | 08:50 | |
| *** jeckersb_gone has joined #tripleo | 08:50 | |
| *** gfidente|afk is now known as gfidente | 08:50 | |
| *** pgadiya has joined #tripleo | 08:51 | |
| gfidente | matbu, marios hey that's all correct | 08:51 |
| gfidente | it works fine with roles but not in mitaka | 08:51 |
| gfidente | https://bugs.launchpad.net/tripleo/+bug/1610258 | 08:52 |
| openstack | Launchpad bug 1610258 in tripleo "CephAdminKey is required for upgrade" [High,In progress] - Assigned to Giulio Fidente (gfidente) | 08:52 |
| jaosorior | shardy: oh, by the way, what do you think about the CA stuff that I mentioned earlier? | 08:55 |
| matbu | gfidente: ha great thanks for the lp (and review) | 08:57 |
| *** akuznetsov has quit IRC | 08:57 | |
| d0ugal | jaosorior: damn, any idea why? | 08:57 |
| jaosorior | d0ugal: nope | 08:57 |
| shardy | jaosorior: we've got no way to express dependencies between services, so I think it's probably just like if someone decides to remove e.g Keystone? | 08:58 |
| d0ugal | I guess if Mistral is going to keep breaking this I'll need to learn about it... | 08:58 |
| shardy | jaosorior: there is a heat spec I raised which aims to introduce capabilities and dependencies between nested resources, but the implementation has stalled | 08:58 |
| shardy | we could introduce validation via mistral using the capabilities_map I guess | 08:58 |
| jaosorior | shardy: I was thinking to just use it the same as tripleo::packages and tripleo::firewall | 08:59 |
| jaosorior | is stuff that you assume gets ran on every node anyway | 08:59 |
| shardy | jaosorior: So you wouldn't want pluggable backends? | 08:59 |
| *** shardy is now known as shardy_mtg | 09:00 | |
| jaosorior | shardy: not for this. this only grabs a CA cert (or several) and adds it to the trusted ones in the node | 09:00 |
| jaosorior | that's it | 09:00 |
| dtantsur | hey folks, please review a one-liner https://review.openstack.org/#/c/342117/ passes both gates in 2 rechecks :) | 09:01 |
| dtantsur | we need to start collecting LLDP for some topology-awareness work | 09:01 |
| jaosorior | I want plugable backends for certificates issued by certmonger. not for trusting CAs. Those are two very different operations | 09:01 |
| *** social has joined #tripleo | 09:01 | |
| jaosorior | shardy: the tricky thing is that this needs to be done in all the nodes. Else we will start getting TLS issues | 09:02 |
| jaosorior | and those can be a damn hard thing to debug, cause it's a very generic error message that openssl gives | 09:03 |
| jaosorior | I'm actually not even sure why I put those manifests for the CA injection under the certmonger directory. They don't even interact with certmonger | 09:04 |
| *** nijaba has quit IRC | 09:04 | |
| *** akuznetsov has joined #tripleo | 09:04 | |
| jaosorior | let me fix that | 09:04 |
| *** nijaba has joined #tripleo | 09:07 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Add manifests to inject and trust CA certificates https://review.openstack.org/353021 | 09:11 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add deployment of CAs via hieradata https://review.openstack.org/353037 | 09:14 |
| *** limao_ has quit IRC | 09:23 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/python-tripleoclient: Skip post-config based on t-h-t flag https://review.openstack.org/353316 | 09:25 |
| *** limao has joined #tripleo | 09:26 | |
| *** bootsha has quit IRC | 09:26 | |
| *** oshvartz has quit IRC | 09:26 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add flag to assert that puppet manages the keystone endpoints https://review.openstack.org/353307 | 09:26 |
| *** paramite has joined #tripleo | 09:28 | |
| openstackgerrit | Merged openstack/tripleo-docs: Fix environment-specific limiting for upgrade docs https://review.openstack.org/352800 | 09:33 |
| *** akuznetsov has quit IRC | 09:36 | |
| *** radeks has quit IRC | 09:38 | |
| *** mgould|afk is now known as mgould | 09:39 | |
| *** akuznetsov has joined #tripleo | 09:41 | |
| *** jubapa has joined #tripleo | 09:42 | |
| *** mbound has joined #tripleo | 09:44 | |
| shadower | dtantsur: can haz a review in turn? https://review.openstack.org/#/c/352278/ it's just a docs thing | 09:48 |
| *** radeks has joined #tripleo | 09:50 | |
| openstackgerrit | Merged openstack/instack-undercloud: Enable LLDP collection during introspection if inspection_extras is on https://review.openstack.org/342117 | 09:51 |
| *** jubapa has quit IRC | 09:53 | |
| *** gchamoul is now known as gchamoul|afk | 09:57 | |
| *** gchamoul|afk is now known as gchamoul | 09:57 | |
| *** jubapa has joined #tripleo | 09:59 | |
| *** jubapa has quit IRC | 10:00 | |
| *** akrivoka has joined #tripleo | 10:02 | |
| *** limao has quit IRC | 10:03 | |
| *** adarazs is now known as adarazs_lunch | 10:05 | |
| *** akuznetsov has quit IRC | 10:07 | |
| jaosorior | marios, shardy: I just ran a deployment with these commits https://review.openstack.org/352364 https://review.openstack.org/353307 https://review.openstack.org/353316 which basically configures the keystone endpoints in puppet and skips the post-config from the client. | 10:11 |
| jaosorior | HA deployment with net-iso, and it looks like this: http://paste.openstack.org/show/553521/ | 10:11 |
| jaosorior | pingtest just passed and now I'm running tempest on it | 10:12 |
| jaosorior | currently cinderv3 is being provisioned to localhost, but that's quite easy to add | 10:14 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add support for cinder v3 endpoint https://review.openstack.org/353364 | 10:19 |
| *** snecklifter has joined #tripleo | 10:23 | |
| snecklifter | hi folks, anyone played with the max_concurrent_builds setting for baremetal? | 10:24 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add hieradata for Manila v1 and v2 endpoints https://review.openstack.org/352353 | 10:24 |
| shardy_mtg | Can someone approve https://review.openstack.org/#/c/345650/ please? | 10:25 |
| *** shardy_mtg is now known as shardy | 10:25 | |
| openstackgerrit | Merged openstack/tripleo-validations: Autogenerate the list of existing validations https://review.openstack.org/352278 | 10:26 |
| *** bootsha has joined #tripleo | 10:26 | |
| jaosorior | shardy: done | 10:27 |
| shardy | thanks! | 10:27 |
| jaosorior | no problem | 10:27 |
| *** kullmar has joined #tripleo | 10:27 | |
| *** yamahata has quit IRC | 10:29 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add support for cinder v3 endpoint https://review.openstack.org/353364 | 10:30 |
| *** kullmar has left #tripleo | 10:31 | |
| *** mbound has quit IRC | 10:33 | |
| *** bootsha has quit IRC | 10:33 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Remove the ControllerCount constraint https://review.openstack.org/345650 | 10:34 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 10:38 |
| jaosorior | shardy: Let me know when I should start testing the patches you mentioned | 10:43 |
| *** rajinir has quit IRC | 10:45 | |
| *** gchamoul is now known as gchamoul|afk | 10:45 | |
| *** apetrich has joined #tripleo | 10:46 | |
| jaosorior | Hey guys, small review passing CI https://review.openstack.org/#/c/352956/ | 10:47 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Migrate the deploy command to use the Mistral workflows https://review.openstack.org/342261 | 10:47 |
| *** gchamoul|afk is now known as gchamoul | 10:49 | |
| *** bootsha has joined #tripleo | 10:49 | |
| jaosorior | hey guys, I'm trying to add barbican as an optional service for the overcloud. Tried a deployment and I'm seeing stuff like this: | 10:56 |
| jaosorior | Error: Could not find class ::barbican for overcloud-controller-2.openstacklocal on node overcloud-controller-2.openstacklocal\u001 | 10:56 |
| jaosorior | should I be adding puppet-barbican to some image? | 10:56 |
| *** dprince has joined #tripleo | 10:59 | |
| jaosorior | dprince hey dude | 11:01 |
| *** ioggstream has quit IRC | 11:01 | |
| dprince | jaosorior: hey | 11:02 |
| jaosorior | dprince: trying to add barbican as an optional service for the overcloud. We have a patch in puppet-tripleo for the profile, one in the client for the password, and finally the one in t-h-t. But I get something like this: Error: Could not find class ::barbican for overcloud-controller-2.openstacklocal on node overcloud-controller-2.openstacklocal | 11:02 |
| jaosorior | dprince: Should I have added puppet-barbican to the image or something like that? | 11:03 |
| dprince | jaosorior: yes. 2 ways to get it. source or package | 11:03 |
| dprince | jaosorior: if you use the 'puppet-modules' element and set DIB_INSTALLTYPE_puppet_modules=source, you can add it to that element and get it via a git checkout | 11:04 |
| dprince | jaosorior: otherwise we'll need a puppet-barbican RPM and probably make openstack-puppet-modules require it | 11:04 |
| dprince | jaosorior: For local testing I use upload-puppet-modules (the script) and you can have whatever local checkouts you want then easily | 11:05 |
| jaosorior | dprince: What do we want in the long run? | 11:05 |
| dprince | jaosorior: I think the puppet-modules element already has barbican anyways | 11:06 |
| dprince | jaosorior: so the missing piece is probably the package. I think that is what CI uses | 11:06 |
| dprince | jaosorior: long term we want both | 11:06 |
| dprince | jaosorior: I'm one of (perhaps a few) that doesn't like installing any puppet modules into overcloud images via packages | 11:07 |
| jaosorior | ok, I'll see if we can get that package then | 11:07 |
| dprince | jaosorior: they aren't needed, and we have other (arguably better) ways for managing and deploying them via Swift now | 11:07 |
| jaosorior | dprince: Sure, but I also want that to work in CI though :P | 11:09 |
| *** dbecker has quit IRC | 11:09 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for keystone https://review.openstack.org/327029 | 11:10 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for heat https://review.openstack.org/327069 | 11:10 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for glance API and registry https://review.openstack.org/327473 | 11:10 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for cinder-api https://review.openstack.org/328859 | 11:10 |
| *** stendulker has quit IRC | 11:12 | |
| jaosorior | shardy: Have you been seeing some errors where the client stops having connection to heat? http://logs.openstack.org/07/352907/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/0a8cf67/console.html#_2016-08-10_10_02_20_655765 | 11:12 |
| *** adarazs_lunch is now known as adarazs | 11:14 | |
| jaosorior | bandini: In the next-gen HA arch, is rabbitmq also managed by pcmk? | 11:14 |
| *** dsariel has joined #tripleo | 11:14 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for RabbitMQ https://review.openstack.org/327482 | 11:15 |
| bandini | jaosorior: yes. these are the services managed by pcmk http://acksyn.org/files/tripleo/light-cib-nomongo.pdf | 11:16 |
| jaosorior | ccamacho: Will the alphabetical sorting of parameters be enforced for all parameters as a check? | 11:16 |
| ccamacho | jaosorior, thats one of the things that im trying to get to work on the puppet-lint checks | 11:16 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: DO NOT MERGE: testing ironic https://review.openstack.org/329872 | 11:17 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Composable service for configuring Nova compute with Ironic https://review.openstack.org/341433 | 11:17 |
| ccamacho | jaosorior All profiles parameters sorted following: "Mandatory params first sorted alphabetically then optional params sorted alphabetically." | 11:17 |
| jaosorior | ccamacho: I see. So then we will no longer be able to symbolically group parameters with a comment, right? | 11:18 |
| *** r-mibu has quit IRC | 11:19 | |
| *** r-mibu has joined #tripleo | 11:19 | |
| ccamacho | jaosorior, a comment can be added with the parameters details, right? | 11:20 |
| *** cmyster has quit IRC | 11:21 | |
| openstackgerrit | Babu Shanmugam proposed openstack/puppet-tripleo: Add ML2/OVN support for neutron profile https://review.openstack.org/314875 | 11:21 |
| *** lucasagomes is now known as lucas-hungry | 11:22 | |
| jaosorior | ccamacho: Not sure how that's gonna work. One instance where this is used commonly is for deprecated parameters https://github.com/openstack/puppet-keystone/blob/master/manifests/init.pp#L694 | 11:22 |
| hewbrocca | bandini: I was just asking jistr -- with HA-next-gen will we be able to deploy galera and rabbit on different machines (or, more importantly, run them on different machines) | 11:23 |
| hewbrocca | bandini: or is that coming later | 11:23 |
| *** social has quit IRC | 11:23 | |
| *** dbecker has joined #tripleo | 11:24 | |
| bandini | hewbrocca: so as things are today the services here http://acksyn.org/files/tripleo/light-cib-nomongo.pdf need to be deployed on nodes managed by pacemaker | 11:25 |
| bandini | hewbrocca: does that answer your question? | 11:25 |
| hewbrocca | Partially. And that is fine | 11:26 |
| hewbrocca | Imagine a case though where we wanted galera running on three nodes and rabbit running on 6 | 11:26 |
| hewbrocca | does that imply two separate pacemaker clusters by definition | 11:27 |
| hewbrocca | or doesn't it matter | 11:27 |
| *** social has joined #tripleo | 11:28 | |
| *** jaosorior has quit IRC | 11:28 | |
| bandini | today we could deploy six nodes with pacemaker. three of them having galera and six of them rabbits | 11:28 |
| *** jaosorior has joined #tripleo | 11:29 | |
| bandini | now andrew has a design to be more flexible there, but it will take quite a bit of work to bring it to tripleo | 11:29 |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates: Availability monitoring agents support https://review.openstack.org/254788 | 11:31 |
| *** pkovar has joined #tripleo | 11:33 | |
| *** jaosorior has quit IRC | 11:33 | |
| *** jaosorior has joined #tripleo | 11:33 | |
| ccamacho | jaosorior, gfidente, we might create then a sort of parameters categories from hiera, user-provided and/or deprecated.. But still need to figure why the puppet-lint its not working properly | 11:34 |
| jaosorior | huh? | 11:34 |
| *** cmyster has joined #tripleo | 11:35 | |
| jaosorior | sorry, my connection got flaky | 11:35 |
| *** fragatina has joined #tripleo | 11:36 | |
| jaosorior | any tripleo-quickstart core that wants to take a look at this? https://review.openstack.org/#/c/352227/ | 11:38 |
| hewbrocca | adarazs: ^^^ | 11:39 |
| jaosorior | gfidente: Hey dude, can you take a look at this one? https://review.openstack.org/#/c/352956/ | 11:39 |
| adarazs | jaosorior: looking | 11:39 |
| jaosorior | that configuration in haproxy is needed for federation | 11:40 |
| jaosorior | adarazs: thanks | 11:40 |
| *** fragatina has quit IRC | 11:41 | |
| *** ccamacho is now known as ccamacho|lunch | 11:42 | |
| *** bfournie has quit IRC | 11:42 | |
| adarazs | jaosorior: merged. | 11:42 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Migrate the deploy command to use the Mistral workflows https://review.openstack.org/342261 | 11:43 |
| d0ugal | shardy: Getting massively hacky, but almost there ^ | 11:43 |
| jaosorior | adarazs: thanks | 11:43 |
| jaosorior | d0ugal: hey dude, can you check this one out? https://review.openstack.org/#/c/353257/ | 11:44 |
| jaosorior | d0ugal: and this https://review.openstack.org/#/c/353316/ :D | 11:45 |
| d0ugal | jaosorior: LGTM | 11:46 |
| jaosorior | d0ugal: thanks dude | 11:46 |
| *** dbecker has quit IRC | 11:47 | |
| d0ugal | I have so much regret about the password code in tripleoclient | 11:47 |
| openstackgerrit | Merged openstack/tripleo-common: Prevent replacement of Port, Net and Subnet https://review.openstack.org/308381 | 11:47 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Migrate the deploy command to use the Mistral workflows https://review.openstack.org/342261 | 11:47 |
| * d0ugal goes to get lunch and walk the dog | 11:47 | |
| *** dbecker has joined #tripleo | 11:48 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: Enable options for auto-generated SSL certificates https://review.openstack.org/352227 | 11:48 |
| jaosorior | d0ugal: will that be replaced by a mistral workflow that reads the overcloud stack and gets the services from there? :O | 11:48 |
| jaosorior | d0ugal: enjoy | 11:48 |
| d0ugal | jaosorior: not sure, maybe! | 11:48 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Reinstall python-requests in bootstrap script https://review.openstack.org/350521 | 11:48 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Fix resetting /etc/hosts https://review.openstack.org/351917 | 11:48 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Raise ping test count after multinode setup https://review.openstack.org/351916 | 11:48 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Make $TRIPLEO_ROOT overrideable https://review.openstack.org/350539 | 11:48 |
| *** ohamada has quit IRC | 11:49 | |
| *** ohamada has joined #tripleo | 11:49 | |
| *** dbecker has quit IRC | 11:49 | |
| *** athomas has quit IRC | 11:52 | |
| *** dbecker has joined #tripleo | 11:53 | |
| dprince | d0ugal: why do you need to re-write paths there? | 11:53 |
| *** jcoufal has joined #tripleo | 11:54 | |
| *** morazi has joined #tripleo | 11:55 | |
| gfidente | jaosorior, commented on https://review.openstack.org/#/c/352956/3 | 11:56 |
| *** social has quit IRC | 11:56 | |
| *** oshvartz has joined #tripleo | 11:56 | |
| *** weshay has joined #tripleo | 11:56 | |
| *** social has joined #tripleo | 11:57 | |
| jaosorior | gfidente: Right, so that's pretty much copied from what exists at the moment for horizon | 11:57 |
| gfidente | guys, cinder-backup role https://review.openstack.org/#/c/304563 and puppet https://review.openstack.org/#/c/311218/ | 11:57 |
| *** jpena is now known as jpena|lunch | 11:57 | |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-common: Prevent replacement of Port, Net and Subnet https://review.openstack.org/353411 | 11:57 |
| *** athomas has joined #tripleo | 11:58 | |
| gfidente | jaosorior, so taking on a side the case insensitive thing and hdr(host) | 11:58 |
| gfidente | isn't psprep missing the condition? | 11:58 |
| *** maeca1 has joined #tripleo | 11:58 | |
| jaosorior | gfidente: It's rsprep not psprep | 11:59 |
| gfidente | jaosorior, yeah that one anyway :) | 12:00 |
| gfidente | my point is that the above X-Forwarded-Proto have the ssl_fc condition | 12:00 |
| jaosorior | gfidente | 12:01 |
| gfidente | should rsprep (or rspirep) do the same? | 12:01 |
| jaosorior | you are right | 12:01 |
| jaosorior | it is indeed missing that condition | 12:01 |
| gfidente | jaosorior, wow | 12:01 |
| jaosorior | but why case insensitive? | 12:01 |
| gfidente | I take the rest of the day off | 12:01 |
| *** links has quit IRC | 12:01 | |
| gfidente | ahahah | 12:01 |
| jaosorior | hahaha | 12:01 |
| jaosorior | awesome | 12:02 |
| gfidente | :( | 12:02 |
| jaosorior | alright | 12:02 |
| jaosorior | now | 12:02 |
| jaosorior | haha I meant that it was a good review, not awesome that you're taking a day off | 12:02 |
| jaosorior | I need you for more reviews haha | 12:02 |
| *** rhallisey has joined #tripleo | 12:04 | |
| *** social has quit IRC | 12:04 | |
| d0ugal | dprince: because we might have paths that lead outside t-h-t, like we document here: http://docs.openstack.org/developer/tripleo-docs/advanced_deployment/network_isolation.html#create-network-environment-file | 12:05 |
| *** social has joined #tripleo | 12:05 | |
| openstackgerrit | Merged openstack/tripleo-common: Remove the delorean-repo element from CentOS images https://review.openstack.org/336524 | 12:05 |
| openstackgerrit | Merged openstack/tripleo-common: Remove epel element from base yaml ramdisk agent image https://review.openstack.org/336551 | 12:09 |
| openstackgerrit | Merged openstack/tripleo-common: Change image names to match what the rest of the system expects https://review.openstack.org/336642 | 12:09 |
| *** dprince has quit IRC | 12:10 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Handle redirects for keystone https://review.openstack.org/352956 | 12:10 |
| *** dprince has joined #tripleo | 12:11 | |
| jaosorior | gfidente: Fixed it. Thanks for the review dude! | 12:11 |
| *** egafford has quit IRC | 12:12 | |
| *** ioggstream has joined #tripleo | 12:18 | |
| *** jcoufal has quit IRC | 12:18 | |
| *** bfournie has joined #tripleo | 12:19 | |
| *** snecklifter has quit IRC | 12:19 | |
| openstackgerrit | Julie Pichon proposed openstack/tripleo-common: Add boot option argument to the Node Registration Mistral workflow https://review.openstack.org/348875 | 12:20 |
| *** rlandy has joined #tripleo | 12:23 | |
| *** lucas-hungry is now known as lucasagomes | 12:29 | |
| *** social has quit IRC | 12:29 | |
| *** akuznetsov has joined #tripleo | 12:30 | |
| *** ccamacho|lunch is now known as ccamacho | 12:33 | |
| *** jdob has joined #tripleo | 12:34 | |
| *** jcoufal has joined #tripleo | 12:35 | |
| EmilienM | good morning folks ! | 12:38 |
| ccamacho | EmilienM morning! | 12:43 |
| *** bootsha has quit IRC | 12:43 | |
| jaosorior | Hey guys... seems heat is having issues | 12:45 |
| jaosorior | I've seen a bunch of | 12:45 |
| jaosorior | Unable to establish connection to https://192.0.2.2:13004/v1/705f00862f5c4a69aa172faf4c9c177b/stacks/overcloud/ca4dcaf2-ccce-4d9c-a1d7-bc1a0c64e9b3 | 12:45 |
| jaosorior | lately | 12:45 |
| jaosorior | and yeah, basically it seems heat restarts in the middle of the deploy | 12:46 |
| *** bootsha has joined #tripleo | 12:46 | |
| shardy | jaosorior: heat, on the undercloud, restarts? | 12:46 |
| shardy | unless you run out of memory that seems unlikely | 12:47 |
| jaosorior | shardy: By the looks of this, so far I think it's heat http://logs.openstack.org/38/352338/9/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/0c83270/logs/undercloud/var/log/heat/heat-api.txt.gz#_2016-08-10_12_04_22_985 | 12:47 |
| *** jayg|g0n3 is now known as jayg | 12:47 | |
| jaosorior | shardy: And actually that's what I think it's going on. Running out of memory | 12:47 |
| shardy | so not exactly a heat problem then | 12:47 |
| jaosorior | right | 12:47 |
| *** thrash|g0ne is now known as thrash | 12:48 | |
| jaosorior | the point is | 12:48 |
| jaosorior | it seems that it's becoming a recurring problem | 12:48 |
| *** pradk has quit IRC | 12:48 | |
| *** rbrady has joined #tripleo | 12:48 | |
| *** fultonj has joined #tripleo | 12:49 | |
| *** amoralej is now known as amoralej|lunch | 12:51 | |
| jaosorior | shardy: We're reducing the number of workers for some undercloud services in the CI, right? | 12:51 |
| shardy | jaosorior: Yes, to limit ram usage, but I don't see anything get killed in that run? | 12:51 |
| EmilienM | jaosorior, shardy: can you look https://review.openstack.org/#/c/353090/ please? | 12:51 |
| shardy | and that auth_token caching warning seems like something we need to address, but it's not actually broken heat-api? | 12:51 |
| jaosorior | shardy: Maybe I misinterpretted what's going on | 12:51 |
| shardy | One thing I notice is we're repeatedly reloading the policy.json which is probably another bug or misconfiguration | 12:52 |
| jaosorior | shardy: so, for some reason the overcloud deploy fails with "Unable to establish connection to ..." | 12:52 |
| gfidente | slagle, so the reason why we don't get the broken image is that it isn't promoted? | 12:52 |
| jaosorior | which is what I posted above | 12:52 |
| jaosorior | shardy: It did seem strange to me that heat got restarted there though | 12:53 |
| jaosorior | thought it might have been an OOM issue | 12:53 |
| shardy | jaosorior: AFAICS heat didn't get restarted | 12:53 |
| slagle | gfidente: yea, the promote only happens if the all 3 periodic jobs pass | 12:53 |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: CI test: try puppet lint job https://review.openstack.org/353452 | 12:54 |
| EmilienM | ccamacho: ^ | 12:54 |
| slagle | or actually it's just the ha job right now | 12:54 |
| EmilienM | let's try | 12:54 |
| slagle | gfidente: then we'd be using the newer repo in all the check-tripleo jobs | 12:54 |
| slagle | gfidente: but what i was saying is that none of the jobs are using cached overcloud images right now | 12:54 |
| jaosorior | then I don't understand why the policy is being reloaded again and again | 12:54 |
| gfidente | slagle, and the code for the cached image which I swear I merged | 12:54 |
| jaosorior | thought that would just happen in the beginning of the server run | 12:54 |
| gfidente | slagle, right that was next question, where is that now? | 12:55 |
| slagle | yea and we had to turn it off due to switching to packaged puppet modules | 12:55 |
| gfidente | ah | 12:55 |
| slagle | but we could probably re-enable it now | 12:55 |
| shardy | jaosorior: it is happening every API call | 12:55 |
| gfidente | ok taking images on a side though | 12:55 |
| gfidente | things *are* working as intended | 12:55 |
| shardy | so that's a problem, but I don't think heat-api gets restarted | 12:56 |
| gfidente | so what do we do with the actual cinder issue? | 12:56 |
| *** fultonj has quit IRC | 12:56 | |
| jaosorior | Ffunky | 12:56 |
| slagle | gfidente: we either update cinder.conf to work around the issue, or we wait for cinder to fix | 12:57 |
| jaosorior | shardy: Alright, well, still need to figure out why that "Unable to establish connection" error is happening... | 12:57 |
| *** fultonj has joined #tripleo | 12:57 | |
| *** tzumainn has joined #tripleo | 12:57 | |
| gfidente | slagle, ok I want to try the workaround and see if we can make it a general feature for managing volume types | 12:59 |
| jaosorior | d0ugal: Do you know if we're deploying mistral over httpd? | 13:01 |
| shardy | jaosorior: No, we're not | 13:06 |
| *** social has joined #tripleo | 13:06 | |
| shardy | ps ax | grep mistral shows it running independent processes for engine, API and executor | 13:06 |
| jaosorior | I see | 13:07 |
| jaosorior | thanks | 13:07 |
| *** weshay is now known as weshay_brb | 13:08 | |
| rbrady | shardy: I attempted to run mistral over httpd. https://bugs.launchpad.net/mistral/+bug/1581649 | 13:12 |
| openstack | Launchpad bug 1581649 in Mistral "Action Execution Response Timeout" [High,New] - Assigned to Ryan Brady (rbrady) | 13:12 |
| *** weshay_brb is now known as weshay | 13:13 | |
| rbrady | shardy: there was a fix that went in about a week ago for something else and I was told it may mitigate the timeout issue seen in the bug above, buit haven't had time to test yet | 13:13 |
| jaosorior | rbrady: Would be great to have it deployed in httpd though :D | 13:14 |
| rbrady | jaosorior: it will eventually, it's just a time/resource issue at the moment | 13:15 |
| *** jpena|lunch is now known as jpena | 13:15 | |
| shardy | rbrady: ack, thanks for the update | 13:17 |
| *** sshnaidm has quit IRC | 13:17 | |
| *** pgadiya has quit IRC | 13:22 | |
| *** apetrich has quit IRC | 13:22 | |
| *** julim has joined #tripleo | 13:24 | |
| slagle | gfidente: if you're going to work on a fix for tripleo, can you assign the tripleo bug to yourself? https://bugs.launchpad.net/bugs/1610073 | 13:24 |
| openstack | Launchpad bug 1610073 in Cinder "Create volume without volume type failed" [High,In progress] - Assigned to Karthik Prabhu Vinod (karthik-prabhu-vinod) | 13:24 |
| gfidente | slagle, ack | 13:25 |
| *** dbecker has quit IRC | 13:25 | |
| *** alee has joined #tripleo | 13:27 | |
| jaosorior | EmilienM: Hey man, alee is working on getting barbican support for the overcloud. We, however, need a puppet-barbican RPM. Where can we submit the spec? | 13:27 |
| EmilienM | indeed | 13:28 |
| EmilienM | jaosorior: let me handle it | 13:28 |
| *** jeckersb_gone is now known as jeckersb | 13:28 | |
| jaosorior | EmilienM: Thanks man. Let us know if you need any help | 13:28 |
| alee | EmilienM, awesome - thanks! ditto on the help. | 13:29 |
| *** tzumainn has quit IRC | 13:29 | |
| *** tzumainn has joined #tripleo | 13:29 | |
| jaosorior | alee: Now the last thing to figure out I think is how the barbican endpoint looks like | 13:29 |
| jaosorior | alee: I've seen deployments that are versioned and versionless... I don't remember if there were any issues with either | 13:29 |
| alee | jaosorior, ah - you mean versioned vs unversioned .. | 13:30 |
| gfidente | jaosorior, was thinking about the endpoints | 13:30 |
| gfidente | we make it in step 5 but all nodes run it | 13:31 |
| gfidente | should a bootstrap_nodeid be useful there? | 13:31 |
| alee | jaosorior, so in the tempest test, I think we needed to put a versioned form in nova.conf , cinder.conf .. getting link .. | 13:32 |
| gfidente | shardy, what was our goal for bootstrap_nodeid | 13:33 |
| gfidente | to always carry the name of a "controller" or the index 0 of a node in the same resource group? | 13:34 |
| shardy | gfidente: it's now set for all roles | 13:34 |
| gfidente | so it carries novacompute-0 on the novacompute group? | 13:35 |
| *** akuznets_ has joined #tripleo | 13:35 | |
| shardy | https://review.openstack.org/#/c/337346/ | 13:35 |
| shardy | gfidente: Yup, we just extract the first name and IP from each ResourceGroup | 13:35 |
| shardy | and deploy it via the AllNodesDeployment for each role | 13:35 |
| gfidente | cool so I can get one node of the group deploying a role to do things | 13:35 |
| shardy | gfidente: yup, the idea was it now works exactly the same on any Role | 13:36 |
| gfidente | ack | 13:36 |
| *** akuznetsov has quit IRC | 13:36 | |
| alee | jaosorior, https://github.com/openstack/puppet-openstack-integration/blob/master/manifests/nova.pp#L95 | 13:37 |
| *** bootsha has quit IRC | 13:37 | |
| alee | jaosorior, https://github.com/openstack/puppet-openstack-integration/blob/master/manifests/cinder.pp#L67 | 13:37 |
| *** bootsha has joined #tripleo | 13:38 | |
| alee | jaosorior, so how about both :) | 13:38 |
| *** [1]cdearborn has joined #tripleo | 13:39 | |
| jaosorior | alee: We could... but it's just gonna be messy in the code | 13:39 |
| jaosorior | lets try what there's at the moment, the versionless one | 13:39 |
| alee | jaosorior, yes - no - I think it should be versionless | 13:39 |
| jaosorior | and see if that works | 13:39 |
| jaosorior | alee: I agree | 13:39 |
| jaosorior | but some components don't play well with that | 13:40 |
| alee | jaosorior, like cinder in this case .. | 13:40 |
| jaosorior | alee does cinder fail if we don't put a version there? | 13:40 |
| jaosorior | At least that should work on the barbicanclient side | 13:40 |
| alee | jaosorior, I believe so yes -- which is why we needed to put it in there | 13:40 |
| alee | jaosorior, but we could do a quick test in the tempest test and change to verionless in cinder .. | 13:42 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Do not merge, checking composable mistral. https://review.openstack.org/352947 | 13:42 |
| *** sshnaidm has joined #tripleo | 13:43 | |
| *** amoralej|lunch is now known as amoralej | 13:44 | |
| jaosorior | alee: Can you do that? | 13:45 |
| alee | jaosorior, yup - will do | 13:45 |
| jaosorior | awesome | 13:46 |
| jaosorior | gfidente: Not all the nodes run it | 13:46 |
| jaosorior | I actually already use bootstrap_nodeid | 13:46 |
| gfidente | jaosorior, ack | 13:46 |
| jaosorior | gfidente: For instance https://review.openstack.org/#/c/352364/8/manifests/profile/base/aodh/api.pp | 13:47 |
| *** egafford has joined #tripleo | 13:48 | |
| *** lblanchard has joined #tripleo | 13:51 | |
| sshnaidm | slagle, shardy hi, do you know what promotes and updates delorean current-tripleo repo? | 13:51 |
| slagle | sshnaidm: yes if the periodic jobs pass, it triggers a cron job on the mirror server | 13:52 |
| slagle | which runs the promote script from tripleo-ci | 13:52 |
| sshnaidm | slagle, so after each successful periodic job it sets current delorean repo as tripleo-current, right? | 13:53 |
| sshnaidm | weshay, ^^ | 13:53 |
| *** ayoung has joined #tripleo | 13:53 | |
| slagle | yes, more or less | 13:54 |
| weshay | slagle, and that creates http://buildlogs.centos.org/centos/7/cloud/x86_64/rdo-trunk-master-tripleo/ | 13:54 |
| weshay | or a different repo | 13:54 |
| *** jubapa has joined #tripleo | 13:55 | |
| slagle | weshay: yes | 13:55 |
| sshnaidm | weshay, afaik current-tripleo points to this one, it's a link now | 13:56 |
| *** rodrigods has quit IRC | 13:56 | |
| slagle | right now it is 7 days old and blocked on https://bugs.launchpad.net/tripleo/+bug/1610073 | 13:56 |
| openstack | Launchpad bug 1610073 in Cinder "Create volume without volume type failed" [High,In progress] - Assigned to Karthik Prabhu Vinod (karthik-prabhu-vinod) | 13:56 |
| *** rodrigods has joined #tripleo | 13:56 | |
| slagle | gfidente said he was going to work on a fix | 13:56 |
| slagle | the cinder patch doesnt honestly look that close to merging | 13:57 |
| *** jubapa has quit IRC | 13:57 | |
| gfidente | slagle, yep have first step coming | 13:57 |
| amoralej | gfidente, basic doubt about fix for the cinder | 13:57 |
| amoralej | if we move to create volume-type in cinder | 13:58 |
| amoralej | that has any impact on migrations? | 13:58 |
| amoralej | sorry, i don't know too much about migrations with tripleo | 13:58 |
| jaosorior | thrash: Hey dude, can you check out this commit? https://review.openstack.org/#/c/353257/ | 13:58 |
| thrash | jaosorior: sure. | 13:59 |
| jaosorior | d0ugal: Is it necessary to have the mistral endpoint versioned? | 14:00 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a volume type for each backend enabled in Cinder https://review.openstack.org/353499 | 14:00 |
| gfidente | amoralej, indeed first I am adding the types alone | 14:00 |
| gfidente | amoralej, actually can you check https://review.openstack.org/353499 to see if it makes sense to you as well? | 14:01 |
| slagle | lucasagomes: hi. what do I need to do to get Ironic to create the config drive partition? | 14:01 |
| slagle | lucasagomes: i'm booting with --config-drive True, but that doesnt seem to do anything | 14:01 |
| EmilienM | gfidente: see my comment | 14:02 |
| jaosorior | rbrady: ping | 14:02 |
| EmilienM | I'm not sure it's puppet3 compliant | 14:02 |
| rbrady | jaosorior: yo | 14:02 |
| slagle | lucasagomes: nm, that did work. i was looking in the wrong place :) | 14:02 |
| gfidente | EmilienM, the each() thing? | 14:02 |
| lucasagomes | slagle, hi there | 14:02 |
| jaosorior | rbrady: Hey dude, do you know if it's necessary for the mistral endpoint to be versioned? | 14:02 |
| EmilienM | gfidente: yes | 14:02 |
| gfidente | EmilienM, sigh we need that to give set_volume a different value for each item | 14:02 |
| EmilienM | I know | 14:02 |
| gfidente | hmm | 14:02 |
| slagle | lucasagomes: hi. i figured it out, it was pebkac | 14:02 |
| lucasagomes | slagle, oh right, yeah you either pass --config-drive or you can also set a nova configuration (will grab the name) which forces all instances to have a config drive | 14:03 |
| EmilienM | that's why I'm pushing on puppet4 thing | 14:03 |
| lucasagomes | slagle, cool np | 14:03 |
| jaosorior | thrash: thanks for the review dude | 14:03 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a volume type for each backend enabled in Cinder https://review.openstack.org/353499 | 14:03 |
| rbrady | jaosorior: not sure. I can find out for you. Or d0ugal may know | 14:03 |
| d0ugal | jaosorior: good question, I am not sure. | 14:03 |
| gfidente | EmilienM, ok I had to fix something there anyway, so I updated it, let's see if it passes I read it should be there form puppet 3.2 ? | 14:03 |
| rbrady | d0ugal: lol | 14:03 |
| lucasagomes | slagle, just as a FYI, the nova configuration is [DEFAULT]/force_config_drive=True (then you don't need to pass --config-drive True to the nova boot command) | 14:03 |
| jaosorior | lol | 14:03 |
| d0ugal | :) | 14:03 |
| jaosorior | happens | 14:03 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: [WIP] Add fluentd service https://review.openstack.org/353506 | 14:04 |
| EmilienM | gfidente: I think we're running 3.6.2 iirc | 14:04 |
| EmilienM | but we need to check | 14:04 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/puppet-tripleo: [WIP] updates to fluentd support https://review.openstack.org/353507 | 14:04 |
| thrash | jaosorior: np | 14:05 |
| gfidente | amoralej, now assuming the volume types do the job | 14:05 |
| gfidente | we need to pick a default | 14:05 |
| gfidente | concerns about upgrades, is it only affecting new volumes? | 14:06 |
| *** dsariel has quit IRC | 14:06 | |
| gfidente | or will default_volume_type cause issues for pre-existing volumes as well? | 14:06 |
| amoralej | my concern is, if we are running in a system without volume-type | 14:06 |
| amoralej | and we upgrade, will be the volume-type created? | 14:06 |
| *** pradk has joined #tripleo | 14:06 | |
| gfidente | amoralej, oh yes it will | 14:06 |
| amoralej | and i'm not sure if that can have any impact on pre-existing volumes | 14:07 |
| amoralej | i remember some issue in the past | 14:07 |
| amoralej | but i'm trying to remember if only happened if we didn't set a separated backend initially | 14:08 |
| openstackgerrit | Merged openstack/python-tripleoclient: Add Barbican password to deployment https://review.openstack.org/353257 | 14:08 |
| amoralej | which wouldn't affect us, as we allways create a separated backend section in cinder.conf even if we don't create the volume-type, right? | 14:08 |
| *** jbadiapa has joined #tripleo | 14:08 | |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart: Expand templated bash to allow for doc generation https://review.openstack.org/347592 | 14:08 |
| jaosorior | gfidente: Hey dude, got a sec for a quick CR? https://review.openstack.org/#/c/353364/ | 14:09 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: WIP: Upgrade scripts to migrate aodh alarm data https://review.openstack.org/353042 | 14:09 |
| *** bootsha has quit IRC | 14:10 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Remove NeutronPassword parameter from puppet/controller.yaml https://review.openstack.org/352907 | 14:11 |
| gfidente | amoralej, from cinder code looks like it is really only consumed in create_volume | 14:12 |
| *** Ryjedo has joined #tripleo | 14:12 | |
| gfidente | amoralej, https://github.com/openstack/cinder/blob/master/cinder/volume/flows/api/create_volume.py#L407-L411 ? | 14:13 |
| gfidente | it is the only reference I could find to it | 14:13 |
| amoralej | ok, i'm going to do a quick test, but i think the issue was when moving backend configuration from default section to a separated section | 14:13 |
| gfidente | amoralej, ok that's good news as we always have multiple backends, even when using single backend | 14:13 |
| gfidente | so let's try to pick a default | 14:13 |
| amoralej | yes, that's a good idea | 14:13 |
| amoralej | yeap | 14:14 |
| amoralej | we can have any backend combination, right? | 14:14 |
| amoralej | backends combination | 14:15 |
| gfidente | yeah | 14:15 |
| amoralej | we can define a priorities order for the different types? | 14:15 |
| sshnaidm | slagle, and how is regular delorean repo is updated? | 14:16 |
| gfidente | amoralej, ehe so my problem at this time is that, because of composable roles, our api service has no clue which backends will be enabled in -volume | 14:17 |
| gfidente | ^_^ | 14:18 |
| gfidente | which is what we wanted | 14:18 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Upgrade scripts to migrate aodh alarm data https://review.openstack.org/353042 | 14:18 |
| gfidente | :P | 14:18 |
| *** tzumainn has quit IRC | 14:18 | |
| amoralej | mmmmm, that's a problem... | 14:19 |
| gfidente | amoralej, ahaha or a feature :P | 14:19 |
| amoralej | :D | 14:19 |
| gfidente | maybe we can use cinder_config in volume.pp | 14:20 |
| amoralej | with composable roles, the api and volumes are created in different puppet runs? or in the same one? | 14:20 |
| gfidente | amoralej, if on same node, same one | 14:20 |
| gfidente | but it's "pure luck" | 14:21 |
| amoralej | but if -api server is running in a different server as -volume, using cinder_config in volume.pp will not run it in the right server, right? | 14:23 |
| gfidente | right | 14:23 |
| gfidente | so we'd have it set only on -volume | 14:24 |
| gfidente | we need it only in -api instead? | 14:24 |
| amoralej | gfidente, can't the list of backends passed to the -api somehow?, from heat i mean | 14:24 |
| *** ebalduf has joined #tripleo | 14:25 | |
| *** jprovazn has quit IRC | 14:25 | |
| gfidente | not really, we build the list in puppet | 14:25 |
| amoralej | yes | 14:25 |
| amoralej | default_volume_type is only in cinder::api | 14:26 |
| *** tzumainn has joined #tripleo | 14:26 | |
| EmilienM | do we know why we didn't have a promotion for 1 week? | 14:26 |
| EmilienM | sshnaidm: ^ | 14:26 |
| sshnaidm | EmilienM, right now - cinder bug, before this - mistral bug | 14:27 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Upgrade scripts to migrate aodh alarm data https://review.openstack.org/353042 | 14:27 |
| sshnaidm | EmilienM, mistral bugS - plural | 14:28 |
| EmilienM | what is the cinder bug? | 14:28 |
| sshnaidm | EmilienM, Create volume without volume type failed https://bugs.launchpad.net/tripleo/+bug/1610073 | 14:29 |
| openstack | Launchpad bug 1610073 in Cinder "Create volume without volume type failed" [High,In progress] - Assigned to Karthik Prabhu Vinod (karthik-prabhu-vinod) | 14:29 |
| EmilienM | ah the one gfidente is working | 14:29 |
| EmilienM | wait, is it something that was done in cinder recently? | 14:29 |
| amoralej | yes EmilienM | 14:30 |
| EmilienM | ok https://review.openstack.org/#/c/353120/ will fix it | 14:30 |
| EmilienM | dang ! | 14:30 |
| EmilienM | can't we promote the repo to a commit before the patch that broke cinder? | 14:30 |
| amoralej | you mean pin cinder version? | 14:31 |
| EmilienM | yeah | 14:31 |
| EmilienM | but let me see gfidente's patch again | 14:31 |
| EmilienM | his code won't work I think | 14:31 |
| amoralej | no, pining again..... | 14:31 |
| sshnaidm | EmilienM, if it will be merged, I think we can run periodic job without waiting for tomorrow | 14:31 |
| EmilienM | gfidente: I tested it, it doesn't work | 14:32 |
| EmilienM | it's a puppet4 thing | 14:32 |
| EmilienM | gfidente: you're on the edge ! | 14:32 |
| *** athomas has quit IRC | 14:32 | |
| EmilienM | sshnaidm: we don't need cinder patch, we can fix it in tripleo | 14:32 |
| EmilienM | sshnaidm: let me help on it | 14:32 |
| sshnaidm | EmilienM, you mean when creating volume? | 14:33 |
| EmilienM | volume type | 14:33 |
| *** tzumainn has quit IRC | 14:33 | |
| amoralej | EmilienM, the patch that gfindente sent is only the first step | 14:33 |
| EmilienM | https://review.openstack.org/#/c/353499/ | 14:33 |
| amoralej | to create the volume-type | 14:33 |
| EmilienM | it won't work ^ | 14:33 |
| EmilienM | because it's a puppet4 syntax | 14:33 |
| EmilienM | let me see how we can do | 14:33 |
| EmilienM | gfidente: I'm going to create a patch for puppet3 | 14:34 |
| amoralej | then we need to add parameter the default_volume_type what seems harder that one (I) would think | 14:34 |
| hewbrocca | gfidente: would you please stop with the broken code | 14:34 |
| hewbrocca | marios: ^^^ you see what we're up against here? | 14:34 |
| EmilienM | hewbrocca: lol | 14:34 |
| EmilienM | hewbrocca: his code is actually awesome | 14:34 |
| EmilienM | it's just too early for us, we are still on puppet3 | 14:34 |
| EmilienM | but we'll move soon | 14:35 |
| gfidente | EmilienM, ack if you have an alternative to each() it's fine | 14:35 |
| gfidente | I see it should work with parser = future | 14:35 |
| gfidente | but I am not sure about the consequences of doing that so if we can do without each() is also good | 14:36 |
| EmilienM | gfidente: yeah but AFIK we don't enable it | 14:36 |
| EmilienM | gfidente: right, me neither | 14:36 |
| EmilienM | gfidente: afik it's experimental | 14:36 |
| EmilienM | and the last thing we want now is to add untested things into tripleo | 14:36 |
| slagle | sshnaidm: what are you asking? just how delorean current itself gets updated? | 14:36 |
| gfidente | wait | 14:36 |
| slagle | collaborate | 14:37 |
| sshnaidm | slagle, yeah, who does build package when new commit is merged into some openstack repo | 14:38 |
| *** jbadiapa has quit IRC | 14:38 | |
| marios | hewbrocca: in scrum call reading back in bit | 14:38 |
| slagle | sshnaidm: delorean | 14:38 |
| gfidente | EmilienM, looks like we can create a define in the manifest | 14:38 |
| slagle | sshnaidm: it's a long running service | 14:38 |
| tbarron | marios: bswartz, egafford, Jokke_ vkmc and I are working on plugins and are thinking we should move the ManilaGeneric stuff that is currently in puppet/services/base.yaml to enviornments/manila_generic.yaml or some such | 14:38 |
| gfidente | EmilienM, let me try that | 14:39 |
| EmilienM | no need | 14:39 |
| slagle | sshnaidm: looking for new commits, building packages, updating repos | 14:39 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Create a volume type for each backend enabled in Cinder https://review.openstack.org/353535 | 14:39 |
| hewbrocca | marios: it's no big deal it's just gfidente behaving badly | 14:39 |
| EmilienM | gfidente: ^ | 14:39 |
| sshnaidm | slagle, are the some links/docs I can look at them about it? | 14:39 |
| tbarron | marios: we are discussing this on #tripleo-netapp-manila-plugin | 14:39 |
| slagle | sshnaidm: possibly, check in #rdo. they maintain it | 14:39 |
| tbarron | marios: if you could join there, we would like to make sure that our plans are sane | 14:40 |
| sshnaidm | slagle, ok | 14:40 |
| tbarron | marios: we'll try not to bother you too much :) | 14:40 |
| gfidente | EmilienM, so that's happening while we don't know if api is available yet though | 14:40 |
| EmilienM | gfidente: a define would work also | 14:40 |
| gfidente | right, let me push the define as well so we cover all cases | 14:41 |
| EmilienM | yeah, we need to set some orchestration | 14:41 |
| EmilienM | gfidente: ok, go ahead | 14:41 |
| openstackgerrit | Julie Pichon proposed openstack/tripleo-specs: Add 'bug tagging' policy https://review.openstack.org/352852 | 14:42 |
| EmilienM | gfidente: why don't you use this type? https://github.com/openstack/puppet-openstack-integration/blob/master/manifests/cinder.pp#L134-L137 | 14:42 |
| EmilienM | gfidente: nevermind, it's same result | 14:43 |
| *** athomas has joined #tripleo | 14:43 | |
| EmilienM | gfidente: for orchestration, please look https://github.com/openstack/puppet-cinder/blob/master/lib/puppet/type/cinder_type.rb#L23 | 14:43 |
| EmilienM | do don't worry, API should be up when we create the type | 14:44 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a volume type for each backend enabled in Cinder https://review.openstack.org/353499 | 14:44 |
| gfidente | yeah but in your case it's also doing it on each volume node | 14:44 |
| gfidente | so even if one api is , we get all volume to try to create the type | 14:44 |
| EmilienM | gfidente: we need to do it on api node | 14:45 |
| jaosorior | did we change something in puppet lately? | 14:45 |
| gfidente | nope on volume | 14:45 |
| jaosorior | I just saw this error output | 14:45 |
| jaosorior | http://logs.openstack.org/64/352364/8/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/02c136e/logs/postci.txt.gz#_2016-08-10_14_33_23_000 | 14:45 |
| marios | tbarron: Jokke_ o/ my apologies for not being able to move faster on the netap review fixup. sure we can use #tripleo-netapp-manila-plugin but tbh i have enough trouble monitoring irc channels so pinging me here works fine too. | 14:45 |
| EmilienM | create the type? | 14:45 |
| jaosorior | with a BUNCH of deprecation messages | 14:45 |
| EmilienM | we need to create the volume type where cinder-api runs | 14:45 |
| openstackgerrit | Merged openstack/os-net-config: Fix Linux bond/slave ifup ordering https://review.openstack.org/353072 | 14:45 |
| marios | tbarron: Jokke_ how about this, i will try and put out a rough sketch/review on the netapp review itself by tomorrow? | 14:46 |
| gfidente | EmilienM, well it shouldn't matter which node is creating it right? | 14:46 |
| gfidente | this is same stuff in puppet3 https://review.openstack.org/#/c/353499/3/manifests/profile/base/cinder/volume.pp | 14:46 |
| openstackgerrit | Brent Eagles proposed openstack/tripleo-heat-templates: Add parameter for enabling purging configuration files https://review.openstack.org/353543 | 14:46 |
| EmilienM | gfidente: it does | 14:47 |
| EmilienM | gfidente: https://github.com/openstack/puppet-cinder/blob/master/lib/puppet/type/cinder_type.rb#L23 | 14:47 |
| gfidente | oh so the type resource is pulling in api locally? | 14:47 |
| gfidente | why should it? | 14:47 |
| EmilienM | gfidente: https://github.com/openstack/puppet-cinder/blob/master/lib/puppet/provider/cinder_type/openstack.rb#L47 | 14:48 |
| EmilienM | it's a provider | 14:48 |
| EmilienM | it's using openstack client to call cinder API | 14:48 |
| EmilienM | so if you want to make sure cinder-api is running, you better run the provider on the node where cinder-api is running | 14:48 |
| EmilienM | you can run it anywhere but you'll never be sure cinder-api is running | 14:48 |
| EmilienM | so yeah feel free to do it on volume node, I just tell you we'll have problems | 14:48 |
| *** tzumainn has joined #tripleo | 14:48 | |
| *** yamahata has joined #tripleo | 14:49 | |
| gfidente | EmilienM, so that's why your patch will fail, mine won't because of the step | 14:49 |
| gfidente | :P | 14:49 |
| tbarron | marios: i think bswartz is going to put up a new patch and abandon the rhefner netapp plugin patch, so we don't need review yet | 14:49 |
| tbarron | ^^ bswartz can correct | 14:49 |
| EmilienM | gfidente: ok cool. So last thing is to add the define in a sep file please | 14:49 |
| EmilienM | and I think we're goo | 14:49 |
| EmilienM | good* | 14:49 |
| gfidente | EmilienM, wait we need to set a default type too | 14:50 |
| EmilienM | yep | 14:50 |
| tbarron | marios: we are using #tripleo-netapp-manila-plugin to figure out the best approach for such plugins | 14:50 |
| *** Guest25838 is now known as saneax_AFK | 14:51 | |
| bswartz | tbarron marios: yes for netapp I intent to abandon rhefner's patch and make a new one | 14:53 |
| dtantsur | folks, is there a way to enable debug logging for overcloud services? | 14:53 |
| * rhefner wakes up | 14:53 | |
| rhefner | bswartz: I approve :) | 14:53 |
| bswartz | tbarron marios: however I indent to use the generic driver a a model and if it's not done correctly currently then we should fix that | 14:53 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 14:53 |
| rhefner | also marios: thanks so much for getting manila merged! you're awesome! | 14:54 |
| rhefner | dustins was singing praises the other day | 14:54 |
| *** pblaho has joined #tripleo | 14:54 | |
| shardy | dtantsur: http://paste.openstack.org/show/553594/ something like that should do it | 14:55 |
| adarazs | dtantsur: I see options for each service. | 14:55 |
| dtantsur | shardy, fantastic.. do you know why we don't do it in gate btw? | 14:55 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: WIP - Add puppet-lint 2.0.0 to check puppet-tripleo manifests https://review.openstack.org/352847 | 14:55 |
| dtantsur | adarazs, I was thinking more about what shardy provided. I guess I have to make my ironic manifests support that as well | 14:56 |
| lucasagomes | dprince, BOOM around? Quick question, are you still using the iboot driver in Ironic for your local tests right? | 14:56 |
| adarazs | dtantsur: yeah, I forgot to add "in the THT parameters" :) | 14:56 |
| EmilienM | ccamacho: puppet-tripleo doesn't use puppet-openstack_spec_helper FYI | 14:57 |
| adarazs | dtantsur: just reading back how uninformative that sounds :D | 14:57 |
| lucasagomes | dprince, dtantsur raised a good question here: https://review.openstack.org/#/c/273684/7/ironic_staging_drivers/iboot/__init__.py (see L59). Does it work without the mgmt interface? | 14:57 |
| ccamacho | EmilienM ack thanks | 14:57 |
| sshnaidm | dtantsur, I see debug messages in gate logs | 14:57 |
| shardy | dtantsur: Yeah turning it on would be good, but we'd need to see what the performance impact is as we're close to the infra timeout on some jobs | 14:58 |
| dprince | lucasagomes: yes | 14:58 |
| dtantsur | sshnaidm, hmm, not for keystone at least... | 14:58 |
| *** oshvartz has quit IRC | 14:58 | |
| lucasagomes | dprince, cool, just wanted to check! Thanks | 14:58 |
| dprince | lucasagomes: that driver works great for me | 14:58 |
| dtantsur | shardy, yeah... anyway, would be handy. usual dsvm jobs have debug enabled | 14:58 |
| dprince | lucasagomes: FWIW I tried switching to AMT. AMT is total crap | 14:58 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud: Pre-install packages before running puppet https://review.openstack.org/353121 | 14:58 |
| marios | tbarron: ok thanks... bswartz hi, why do you want to abandon rhefner review? It may makes sense but I was going to have a go at fixing that one up first anyway. big +1 to whether we have landed the generic in a pluggable way. I was hoping a first pass at the netapp review would surface those issues | 14:59 |
| marios | rhefner: o/ hey :) | 14:59 |
| marios | rhefner: we did it in the end :) | 14:59 |
| marios | rhefner: v 91 | 14:59 |
| lucasagomes | dprince, right, yeah both (iboot and amt) will be soon out of the tree unfortunately. So I'm trying to get it merged into ironic-staging-drivers | 14:59 |
| dprince | lucasagomes: having external power switches (iboot) working is awesome. That driver always works for me | 14:59 |
| lucasagomes | dprince, should be there soon, I will fix the few comments in the patch | 14:59 |
| lucasagomes | dprince, cool, glad to know! | 14:59 |
| tbarron | rhefner++ | 14:59 |
| tbarron | marios++ | 14:59 |
| dprince | lucasagomes: ack, thanks for the heads up | 14:59 |
| bswartz | marios: do you want to rebase it? if so you're more than welcome. I tried a rebase and found almost nothing in common between the old parent commit and HEAD | 14:59 |
| lucasagomes | dprince, np! Are you going to be around in barcelona? | 15:00 |
| dprince | lucasagomes: yes | 15:00 |
| * marios joining call biab | 15:00 | |
| * lucasagomes will look for a bar with a pool table | 15:00 | |
| dprince | lucasagomes: :) | 15:00 |
| lucasagomes | dprince, ^ we gotta settle it this time | 15:00 |
| *** ayoung has quit IRC | 15:01 | |
| *** d0ugal has quit IRC | 15:01 | |
| *** fultonj has quit IRC | 15:01 | |
| *** fultonj has joined #tripleo | 15:01 | |
| dtantsur | oh, I see folks already making detailed plans for the summit :D | 15:02 |
| *** masco has quit IRC | 15:03 | |
| lucasagomes | dtantsur, :-) | 15:03 |
| *** tremble has quit IRC | 15:03 | |
| *** radeks has quit IRC | 15:04 | |
| openstackgerrit | Merged openstack/tripleo-common: Add support to image build yaml input to handle env vars https://review.openstack.org/318087 | 15:04 |
| *** mcornea has quit IRC | 15:07 | |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart: Expand templated bash to allow for doc generation https://review.openstack.org/347592 | 15:11 |
| *** david-lyle has quit IRC | 15:12 | |
| *** ayoung has joined #tripleo | 15:12 | |
| *** david-lyle has joined #tripleo | 15:12 | |
| marios | bswartz: yeah ok i will have a go and see how it works out. indeed there will be quite a rebase from what's there but i'd like to think about m oving all the confign, particularly from the overcloud_controller.pp/controller_pacemaker.pp into the puppet-tripleo side (likely) appropriate manila services. on the tripleo heat templates side the extraconfig/pre_deploy/controller/manila-netapp.yaml can also be m | 15:16 |
| marios | oved into the manila-* services under puppet/servcies. as i said i will try spend some time to look at it tomorrow | 15:16 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: WIP - Add puppet-lint 2.0.0 to check puppet-tripleo manifests https://review.openstack.org/352847 | 15:16 |
| *** fragatina has joined #tripleo | 15:16 | |
| *** fragatina has quit IRC | 15:16 | |
| marios | fwiw, bswartz if you haven't already , ccamacho has a good writeup on how the composable services works in general (puppet-tripleo vs tripleo-heat-templates) which may be helpful http://docs.openstack.org/developer/tripleo-docs/developer/tht_walkthrough/tht_walkthrough.html | 15:17 |
| *** fragatina has joined #tripleo | 15:17 | |
| *** radez has joined #tripleo | 15:21 | |
| *** leanderthal is now known as leanderthal|afk | 15:21 | |
| *** jrist has quit IRC | 15:22 | |
| *** jrist has joined #tripleo | 15:23 | |
| *** ebarrera has quit IRC | 15:26 | |
| vkmc | 224977 | 15:27 |
| openstackgerrit | Merged openstack/tripleo-ui: Use inline-source-map to enable source maps in karma tests https://review.openstack.org/351557 | 15:28 |
| openstackgerrit | Merged openstack/tripleo-ui: Environments Listing and Updating using Mistral Actions https://review.openstack.org/350586 | 15:28 |
| EmilienM | bnemec: hey! could you please look https://review.openstack.org/#/c/351330/ ? I did a last recheck, but I think the patch is good to go | 15:30 |
| EmilienM | bnemec: the update worked http://logs.openstack.org/30/351330/7/experimental/gate-tripleo-ci-centos-7-nonha-multinode-updates-nv/87a16e9/console.html#_2016-08-10_04_19_44_317525 | 15:30 |
| *** gregwork has joined #tripleo | 15:31 | |
| EmilienM | slagle: if you have some time also ^ | 15:31 |
| *** jubapa has joined #tripleo | 15:31 | |
| *** weshay is now known as weshay_mtg | 15:32 | |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Align rabbitmq hiera key with service name https://review.openstack.org/353582 | 15:34 |
| *** ifarkas is now known as ifarkas_afk | 15:36 | |
| *** rwsu has quit IRC | 15:36 | |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates: Remove Liberty->Mitaka migration functions https://review.openstack.org/353597 | 15:45 |
| sshnaidm | pabelanger, hi, can you please look at this in your time? https://review.openstack.org/#/c/352880/ thanks | 15:47 |
| shardy | EmilienM: Hi, I have a question re rabbit_hosts in the puppet modules | 15:56 |
| shardy | https://github.com/openstack/puppet-oslo/blob/master/manifests/messaging/rabbit.pp#L180 | 15:56 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Create a volume type for each backend enabled in Cinder https://review.openstack.org/353499 | 15:56 |
| shardy | would it be acceptable to add a type check there, so you can optionally pass a json serialized list, and have puppet use parsejson to unpack it to a list? | 15:56 |
| *** jpena is now known as jpena|away | 15:56 | |
| EmilienM | wow we never did that before | 15:57 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/all-nodes-config.yaml#L317 | 15:57 |
| *** numan_ has quit IRC | 15:57 | |
| shardy | EmilienM: I want to replace all those hard coded yaml references with hiera interpolation inside the service templates | 15:57 |
| shardy | which I think means we have to pass the list as a string | 15:57 |
| shardy | (which is actually slightly easier for the heat templates anyway) | 15:57 |
| dtantsur | folks, how do I get (from EndpointMap) keystone URL without /v2.0 suffix? | 15:58 |
| EmilienM | so this value in puppet right now only accepts an array | 15:58 |
| shardy | EmilienM: I could put the translation in puppet-tripleo, but it would be cleaner if we could just support it in puppet-oslo | 15:58 |
| EmilienM | and you want to give a string, json formated, right? | 15:58 |
| shardy | https://github.com/puppetlabs/puppetlabs-stdlib/blob/master/lib/puppet/parser/functions/parsejson.rb | 15:58 |
| shardy | EmilienM: Yes, I want the option to pass a json string instead of a list | 15:58 |
| shardy | it could be a new parameter I guess | 15:58 |
| EmilienM | it would be a super cool feature | 15:58 |
| EmilienM | I can hack something | 15:59 |
| EmilienM | let me finish the cinder patch that we are doing to unblock promotion and I can start looking to it | 15:59 |
| shardy | EmilienM: If you have time that would be great, if not I can have a try (your puppet will be much quicker than mine I expect :) | 15:59 |
| EmilienM | shardy: I love the idea | 16:00 |
| EmilienM | shardy: let me try something today, I'll let you know | 16:00 |
| shardy | EmilienM: great, thanks! :) | 16:00 |
| EmilienM | shardy: can you look https://review.openstack.org/#/c/353090/ please? | 16:01 |
| *** rcernin has quit IRC | 16:01 | |
| *** jprovazn has joined #tripleo | 16:01 | |
| shardy | dtantsur: https://github.com/openstack/tripleo-heat-templates/blob/master/network/endpoints/endpoint_map.yaml#L1233 | 16:01 |
| shardy | dtantsur: you can use uri_no_suffix | 16:01 |
| shardy | EmilienM: sure! | 16:02 |
| bswartz | marios: whatever you can do will be appreciated -- just let me know if you make any progress | 16:02 |
| dtantsur | shardy, thanks! | 16:02 |
| bswartz | marios: my intuition is that addition of a new driver should really be as simple as copy-pasting some boilerplate stuff and changing the names, then replacing the list of config options | 16:03 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 16:04 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Fix resetting /etc/hosts https://review.openstack.org/351917 | 16:05 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Raise ping test count after multinode setup https://review.openstack.org/351916 | 16:05 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Make $TRIPLEO_ROOT overrideable https://review.openstack.org/350539 | 16:05 |
| *** panda is now known as panda|bbl | 16:06 | |
| marios | bswartz: ack thanks i will try get to that tomorrow somepoint | 16:06 |
| marios | bswartz: eod here | 16:07 |
| *** weshay_mtg is now known as weshay | 16:11 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: DO NOT MERGE: testing ironic https://review.openstack.org/329872 | 16:13 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Composable service for configuring Nova compute with Ironic https://review.openstack.org/341433 | 16:13 |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-image-elements: Restore rdo-release element https://review.openstack.org/353624 | 16:13 |
| bnemec | slagle: EmilienM shardy: ^ is breaking the stable/liberty jobs right now | 16:14 |
| openstackgerrit | Merged openstack/instack-undercloud: Update hieradata parameters to match recent changes https://review.openstack.org/353090 | 16:15 |
| *** yamahata has quit IRC | 16:15 | |
| *** penick has joined #tripleo | 16:17 | |
| shardy | bnemec: to clarify, why is stable CI using master t-i-e? | 16:18 |
| bnemec | shardy: I assume we don't branch it, like diskimage-builder. | 16:19 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Configure Cinder default volume type https://review.openstack.org/353626 | 16:19 |
| shardy | bnemec: No, we do branch it | 16:19 |
| shardy | Oh wait | 16:19 |
| EmilienM | shardy: you want me to remove +A ? | 16:20 |
| shardy | No, sorry you're right, I was looking at t-p-e | 16:20 |
| bnemec | Yeah, only an icehouse branch: https://review.openstack.org/#/admin/projects/openstack/tripleo-image-elements,branches | 16:20 |
| shardy | EmilienM: No, the revert is fine, but perhaps we should consider branching the repo for newton | 16:20 |
| EmilienM | shardy: ack, I can do it if you want this week | 16:21 |
| bnemec | We should probably just deprecate the whole repo. I don't think we want to be using it anymore anyway. | 16:21 |
| shardy | bnemec: Yeah obviously that would be even better | 16:21 |
| bnemec | Oh, but we do still use a few elements from there, don't we? | 16:22 |
| bnemec | os-*-config and such. | 16:22 |
| shardy | yeah there's a few things we'd have to move elsewhere | 16:23 |
| *** abehl has quit IRC | 16:23 | |
| *** ebarrera has joined #tripleo | 16:24 | |
| *** pcaruana has quit IRC | 16:29 | |
| *** abehl has joined #tripleo | 16:29 | |
| jaosorior | /exit/ | 16:29 |
| *** jaosorior has quit IRC | 16:29 | |
| *** hewbrocca is now known as hewbrocca-afk | 16:31 | |
| *** milan has quit IRC | 16:31 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud: Pre-install packages before running puppet https://review.openstack.org/353121 | 16:32 |
| *** mgould is now known as mgould|afk | 16:32 | |
| *** milan has joined #tripleo | 16:34 | |
| sshnaidm | slagle, bnemec, maybe we need to consider increasing memory for undercloud and controllers on rh1? https://bugs.launchpad.net/tripleo/+bug/1611842 | 16:35 |
| openstack | Launchpad bug 1611842 in tripleo "CI: puppet memory fails" [Undecided,New] | 16:35 |
| *** amoralej is now known as amoralej|off | 16:35 | |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Align mongodb hiera key with service name https://review.openstack.org/353582 | 16:35 |
| bnemec | sshnaidm: Yeah, it seems like 5 GB is not enough for the overcloud nodes anymore. | 16:36 |
| sshnaidm | bnemec, agree | 16:36 |
| bnemec | We probably need to bump the RAM and/or get the service matrix working. | 16:36 |
| bnemec | I don't know the status on the latter though. | 16:37 |
| shardy | dprince: Hey if you get time could you pls revisit https://review.openstack.org/#/c/345510 ? | 16:37 |
| shardy | I need to rebase that series and that one is ready to land I think | 16:37 |
| sshnaidm | bnemec, I discovered that we have swap now (I didn't see it before), but it's not used, seems like swappiness is not configured. | 16:39 |
| sshnaidm | bnemec, but I'd vote for increasing memory, swap will make our jobs much longer.. | 16:39 |
| sshnaidm | slagle, can you please push it https://review.openstack.org/#/c/353103/ , or there is need for additional approvements? | 16:42 |
| *** skramaja has quit IRC | 16:42 | |
| *** florianf has quit IRC | 16:43 | |
| *** sshnaidm is now known as sshnaidm|afk | 16:43 | |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates: Availability monitoring agents support https://review.openstack.org/254788 | 16:45 |
| paramite | jistr, slagle, EmilienM, shardy: Hi guys, following patch is ready to for review. Whenever you will find some free time, I would be really glad for a review. Thx | 16:47 |
| paramite | https://review.openstack.org/#/c/254788 | 16:47 |
| *** tosky has quit IRC | 16:50 | |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Align hiera keys with service names https://review.openstack.org/353582 | 16:50 |
| *** lucasagomes is now known as lucas-dinner | 16:52 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Align node_ips hiera keys with the service name. https://review.openstack.org/353637 | 16:52 |
| *** jpich has quit IRC | 16:53 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a default volume type https://review.openstack.org/353499 | 16:53 |
| *** rlandy is now known as rlandy|afk | 16:54 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a default volume type https://review.openstack.org/353499 | 16:55 |
| *** yamahata has joined #tripleo | 16:56 | |
| *** dsariel has joined #tripleo | 16:56 | |
| *** paramite has quit IRC | 16:57 | |
| *** dsariel has quit IRC | 16:58 | |
| *** ohamada has quit IRC | 17:02 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Configure Cinder default volume type https://review.openstack.org/353644 | 17:03 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: cinder api: set a default volume type https://review.openstack.org/353645 | 17:03 |
| slagle | bnemec: sshnaidm|afk : let's just bump the memory for now, we should have plenty | 17:03 |
| slagle | bnemec: i think it's just a matter of updating the flavors | 17:03 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Align node_ips hiera keys with the service name. https://review.openstack.org/353637 | 17:03 |
| bnemec | We can go in small increments too. Start out adding just 512 MB for now? | 17:03 |
| slagle | wfm | 17:04 |
| bnemec | Okay, will do. | 17:05 |
| openstackgerrit | Merged openstack/tripleo-image-elements: Restore rdo-release element https://review.openstack.org/353624 | 17:06 |
| *** ayoung has quit IRC | 17:06 | |
| *** tbarron is now known as tbarron|lunch | 17:09 | |
| openstackgerrit | Giulio Fidente proposed openstack-infra/tripleo-ci: Pass a volume_type for the pingtest https://review.openstack.org/353654 | 17:10 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci: Bump baremetal vm memory to 5.5 GB https://review.openstack.org/353655 | 17:13 |
| bnemec | ^ is live on rh1 | 17:13 |
| bnemec | Oh, I should mark that as closing the bug though. | 17:14 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci: Bump baremetal vm memory to 5.5 GB https://review.openstack.org/353655 | 17:14 |
| bnemec | There | 17:14 |
| *** milan has quit IRC | 17:15 | |
| bnemec | real 17m50.730s | 17:18 |
| bnemec | Not bad. That's _almost_ got the undercloud install back down to where it was a year and a half ago. :-) | 17:18 |
| *** dtantsur is now known as dtantsur|afk | 17:19 | |
| *** ayoung has joined #tripleo | 17:20 | |
| *** ebarrera has quit IRC | 17:22 | |
| *** jpena|away is now known as jpena|off | 17:23 | |
| *** penick has quit IRC | 17:24 | |
| beagles | EmilienM, I'm seeing something unexpected with that purge_config stuff. Maybe I'm mistaken, but I figured that with neutron::purge_config set to true, if I were to edit the neutron.conf file and add a configuration value and rerun "overcloud deploy", it would rewrite the configuration file. Or maybe it is isn't doing so because I didn't actually *change* anything in the heat templates between runs? | 17:26 |
| beagles | EmilienM, as it is, the value I manually added is still there afterwards | 17:27 |
| EmilienM | o_O | 17:27 |
| EmilienM | let me finish a task I'll look | 17:27 |
| beagles | EmilienM, k.. I'll try again in the meantime | 17:28 |
| *** rbrady is now known as rbrady-food | 17:33 | |
| *** numan_ has joined #tripleo | 17:34 | |
| *** jubapa has quit IRC | 17:35 | |
| *** penick has joined #tripleo | 17:36 | |
| *** ioggstream has quit IRC | 17:39 | |
| *** bana_k has joined #tripleo | 17:45 | |
| *** electrofelix has quit IRC | 17:49 | |
| *** dbecker has joined #tripleo | 17:49 | |
| *** pkovar has quit IRC | 17:51 | |
| *** akuznets_ has quit IRC | 17:53 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Stop Validation https://review.openstack.org/336561 | 17:56 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Triggering validations and messaging https://review.openstack.org/336560 | 17:56 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: List Validations and Validation results https://review.openstack.org/334770 | 17:56 |
| *** bana_k has quit IRC | 18:07 | |
| *** fragatina has quit IRC | 18:07 | |
| *** tesseract- has quit IRC | 18:13 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 18:15 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements https://review.openstack.org/348695 | 18:20 |
| *** aufi has quit IRC | 18:22 | |
| *** dbecker has quit IRC | 18:26 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Remove deprecated net_ip_subnet_map output https://review.openstack.org/346055 | 18:27 |
| *** bana_k has joined #tripleo | 18:27 | |
| *** dbecker has joined #tripleo | 18:28 | |
| *** fultonj has quit IRC | 18:31 | |
| *** ayoung has quit IRC | 18:31 | |
| *** fultonj has joined #tripleo | 18:32 | |
| *** rbrady-food is now known as rbrady | 18:32 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Switch rabbit_hosts list to use hiera interpolation https://review.openstack.org/353680 | 18:35 |
| *** shardy is now known as shardy_afk | 18:35 | |
| dtrainor | hi. I'm testing out some installs and when I try to pxeboot another system on the network, I see dnsmasq-dhcp saying: DHCPDISCOVER(tap....) <client mac> no addresses available. The only thing I've changed in the Undercloud was the 'local_interface' statement. | 18:35 |
| dtrainor | As a result, the client never gets to pxeboot and I can't provision | 18:38 |
| dprince | dtrainor: I believe you are hitting the DHCP wildcard | 18:39 |
| dprince | dtrainor: perhaps running your PXE boots on a different network? Or disable the introspection feature of the undercloud? | 18:40 |
| dprince | dtrainor: when you say 'pxeboot another system' what do you mean exactly? | 18:41 |
| dtrainor | i mean i'm trying to provision an overcloud system from the undercloud. the undercloud has a nic on the vlan that the overcloud systems are expected to pxeboot from | 18:42 |
| dtrainor | ugh. this is like monday all over again. i see what's going on. | 18:43 |
| *** fragatina has joined #tripleo | 18:44 | |
| *** dbecker has quit IRC | 18:44 | |
| dtrainor | sorry | 18:44 |
| dprince | dtrainor: np, glad you found it | 18:44 |
| *** rlandy|afk is now known as rlandy | 18:47 | |
| dtrainor | if i wanted to go the wildcard route, i'd want to look at AHC with discovery etc correct? | 18:47 |
| *** rbrady has quit IRC | 18:47 | |
| *** tbarron|lunch is now known as tbarron | 18:52 | |
| *** liverpooler has quit IRC | 18:52 | |
| *** dsariel has joined #tripleo | 18:56 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add cinder-backup service as role https://review.openstack.org/304563 | 18:59 |
| *** numan_ has quit IRC | 18:59 | |
| *** sshnaidm|afk has quit IRC | 19:00 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add cinder-backup service as role https://review.openstack.org/304563 | 19:02 |
| *** rbrady has joined #tripleo | 19:02 | |
| *** bana_k has quit IRC | 19:04 | |
| jeckersb | what sort of expectations should i have with quickstart + devmode? | 19:16 |
| jeckersb | i've tried it off and on a few times and it seems like it always fails for some reason when updating the image | 19:17 |
| *** panda|bbl is now known as panda | 19:17 | |
| larsks | jeckersb: depending on the failures you are seeing, https://review.openstack.org/#/c/352112/ may help. | 19:27 |
| openstackgerrit | Pradeep Kilambi proposed openstack/puppet-tripleo: Move ceilometer api to run under apache wsgi https://review.openstack.org/352476 | 19:27 |
| *** oshvartz has joined #tripleo | 19:27 | |
| jeckersb | larsks: hm i'll give it a try and see if it helps | 19:30 |
| jeckersb | i think it's just something going wrong with virt-customize | 19:30 |
| larsks | Okay, if that's the case, this won't help at all. | 19:30 |
| jeckersb | it throws an error somewhere but the output is nearly impossible to read | 19:31 |
| * jeckersb will reformat and try to make sense of it | 19:31 | |
| larsks | Can you paste the output somewhere? | 19:31 |
| jeckersb | larsks: https://fedorapeople.org/~jeckersb/quickstart.log | 19:32 |
| larsks | jeckersb: fyi, it helps if you run that through 'ansifilter' before pasting (strips out all the formatting codes), but I will manage... | 19:33 |
| jeckersb | i see... virt-customize: update_image-2149357464.sh: command exited with an error | 19:33 |
| jeckersb | but i don't see the actual error so easily | 19:33 |
| jeckersb | i'm sure it's in there somewhere :) | 19:34 |
| jeckersb | oh i see conflicts with swift packages | 19:35 |
| panda | jeckersb: it's in the /tmp/builder.log in the image | 19:35 |
| jeckersb | https://paste.fedoraproject.org/405746/14708577/ | 19:35 |
| jeckersb | basically | 19:35 |
| larsks | It looks like it's failing in the middle of a yum operation. How much memory to have available? | 19:36 |
| panda | seems to be a dependency problem | 19:36 |
| jeckersb | yeah | 19:37 |
| larsks | panda: where do you see that? | 19:37 |
| jeckersb | maybe i just have bad timing on attempting this :) | 19:37 |
| larsks | I see the abort in the middle of the yum update. | 19:37 |
| jeckersb | virthost has 32G in this case | 19:37 |
| larsks | Hmm, should be okay. | 19:37 |
| beagles | EmilienM, I wonder... if this "purge_config" not working is a consequence of how the providers in the puppet-neutron module are written | 19:37 |
| EmilienM | beagles: ok let me look now | 19:38 |
| beagles | EmilienM, they are different than say.. cinder .. in that the don't define self.file_path in the providers | 19:38 |
| beagles | EmilienM, ie it's not a static method (or whatever you call it in ruby ;)) | 19:38 |
| * beagles is now officially way out of his depth with the puppet/ruby thing | 19:39 | |
| EmilienM | wait I'm looking | 19:39 |
| jeckersb | i wonder if just adding --skip-broken to yum will make this mostly work for my purposes | 19:41 |
| * jeckersb really doesn't care about swift | 19:41 | |
| panda | larsks: Error: Package: openstack-swift-proxy-2.7.1-0.20160613173619.d75e6d3.el7.centos.noarch (@delorean)\n Requires: openstack-swift = 2.7.1-0.20160613173619.d75e6d3.el7.centos\n Removing: openstack-swift-2.7.1-0.20160613173619.d75e6d3.el7.centos.noarch (@delorean)\n openstack-swift = 2.7.1-0.20160613173619.d75e6d3.el7.centos\n Obsoleted By: | 19:41 |
| panda | python-swift-2.9.1-0.20160802144313.d819ae0.el7.centos.noarch (delorean-tripleo)\n | 19:41 |
| larsks | panda: oh, I see that. That actually looks familiar. jeckersb, are you modifying the set of available repositories at all? | 19:41 |
| larsks | (It's possible also that delorean is just broken at the moment, of course.) | 19:42 |
| EmilienM | I'm reading doc https://docs.puppet.com/puppet/latest/reference/type.html#resources-attribute-purge | 19:42 |
| jeckersb | larsks: nope just pasted devmode from the quickstart doc | 19:42 |
| panda | jeckersb: what you're seen is a failed attempt to update the starting image we use for under/overcloud from tripleo-ci contents | 19:42 |
| panda | we're updating with latest delorean, so if that is broken or in the midlle of something, the update fails | 19:43 |
| EmilienM | beagles: let me try something in puppet CI | 19:43 |
| EmilienM | beagles: I'll get back to you | 19:43 |
| *** ayoung has joined #tripleo | 19:43 | |
| *** bana_k has joined #tripleo | 19:43 | |
| EmilienM | beagles: oh wait, ci is super busy today, I'm going to reproduce it in my env | 19:43 |
| jeckersb | actually i think it might just be timing out, ignoring the dependency errors | 19:44 |
| beagles | EmilienM, k | 19:44 |
| jeckersb | setup/undercloud : upload repos on image and run update --------------- 958.74s | 19:44 |
| jeckersb | that is set to timeout at 900s wit SIGTERM | 19:44 |
| jeckersb | yeah i think so | 19:46 |
| myoung | jeckersb: when looking at virt-customize output from big yum operations, it's handy to check out /tmp/builder.log from the image being operated on (e.g. https://github.com/redhat-openstack/ansible-role-tripleo-image-build/blob/master/tasks/main.yml#L52) | 19:46 |
| jeckersb | Updating : 1:python-neutron-lbaas-9.0.0-0.20160802010804.f9e791b. 115/429virt-customize: update_image-2149357464.sh: command exited with an error | 19:46 |
| *** jprovazn has quit IRC | 19:46 | |
| jeckersb | it only got through 115 of 429 packages before it hit the timeout | 19:46 |
| panda | jeckersb: yeah, 900 is the soft timeout, 1200 is the hard | 19:46 |
| myoung | jeckersb: the ansible output with all the \n is enough to make eyes bleed. It's true! | 19:47 |
| jeckersb | yep :) | 19:47 |
| jeckersb | does anyone know offhand if libguestfs would be writing the 'yum update' output to the console on the appliance as it's going? | 19:48 |
| jeckersb | just wondering where the bottleneck is | 19:48 |
| jeckersb | i'm guessing slow to download and then running out of time during install | 19:48 |
| panda | jeckersb: /tmp/builder.log on the image | 19:49 |
| jeckersb | guess maybe i need a squid cache | 19:49 |
| panda | jeckersb: on the console directly, I don't know | 19:49 |
| jeckersb | panda: can i watch that in real time as it's updating though? | 19:49 |
| jeckersb | i don't think it logs how long each package took to download | 19:50 |
| jeckersb | maybe it does | 19:50 |
| * jeckersb looks at the damn log | 19:50 | |
| panda | jeckersb: hm, I don't think so | 19:52 |
| jeckersb | hm builder.log has... | 19:55 |
| jeckersb | Total 1.4 MB/s | 153 MB 01:53 | 19:55 |
| jeckersb | so i guess only 1m53s of that was download time | 19:55 |
| jeckersb | instead the package install is taking too long | 19:56 |
| * jeckersb tries to watch the console... | 19:56 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add support for cinder v3 endpoint https://review.openstack.org/353364 | 19:59 |
| *** radeks has joined #tripleo | 20:00 | |
| *** d0ugal has joined #tripleo | 20:02 | |
| *** sshnaidm|afk has joined #tripleo | 20:02 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: Fix InvocationError during ansible-lint run. https://review.openstack.org/352957 | 20:09 |
| *** rwsu has joined #tripleo | 20:15 | |
| *** dprince has quit IRC | 20:16 | |
| *** rwsu has quit IRC | 20:20 | |
| *** radeks has quit IRC | 20:22 | |
| *** jayg is now known as jayg|g0n3 | 20:29 | |
| *** julim has quit IRC | 20:30 | |
| *** shardy_afk has quit IRC | 20:31 | |
| *** coolsvap has quit IRC | 20:31 | |
| *** lblanchard has quit IRC | 20:32 | |
| *** bana_k has quit IRC | 20:32 | |
| *** jeckersb is now known as jeckersb_gone | 20:36 | |
| *** rwsu has joined #tripleo | 20:37 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Use updated overcloud-full image name https://review.openstack.org/353729 | 20:37 |
| slagle | EmilienM: multinode jobs failing. fix is this ^ | 20:37 |
| EmilienM | slagle: +2 | 20:38 |
| EmilienM | slagle: gate-tripleo-ci-centos-7-nonha-multinode-nv was red | 20:38 |
| EmilienM | so yeah, having it voting will really help | 20:38 |
| EmilienM | so people will actually look at it | 20:38 |
| *** beagles is now known as beagles_biab | 20:41 | |
| slagle | EmilienM: yea i just updated the project-config patch | 20:41 |
| slagle | it will be in the gate queue for the puppet projects you migrated over as well | 20:41 |
| slagle | so please have a look | 20:42 |
| *** rbrady has quit IRC | 20:42 | |
| EmilienM | ok looking | 20:45 |
| EmilienM | slagle: you masterize yaml bro | 20:46 |
| slagle | i'm in ur project. config'ing your yaml | 20:46 |
| EmilienM | slagle: ok +1, we'll see how painful it is from now :) | 20:47 |
| *** shardy_afk has joined #tripleo | 20:48 | |
| *** rwsu has quit IRC | 20:48 | |
| *** shardy_afk has quit IRC | 20:58 | |
| *** akrivoka has quit IRC | 20:58 | |
| *** jubapa has joined #tripleo | 20:59 | |
| *** jubapa has quit IRC | 21:00 | |
| dmsimard | Where can we see the results of the periodic tripleo ci promotion job ? | 21:00 |
| *** jubapa has joined #tripleo | 21:00 | |
| weshay | dmsimard, http://status-tripleoci.rhcloud.com/ | 21:04 |
| weshay | look for master, then periodic | 21:04 |
| weshay | dmsimard, or here http://tripleo.org/cistatus-periodic.html | 21:05 |
| weshay | but it's not updated | 21:05 |
| dmsimard | weshay: I don't see master on http://tripleo.org/cistatus-periodic.html ? | 21:07 |
| weshay | ya.. not updated | 21:07 |
| weshay | sshnaidm|afk, created http://status-tripleoci.rhcloud.com/ for his own use.. and turns out.. works pretty well :) | 21:07 |
| *** fragatin_ has joined #tripleo | 21:08 | |
| pabelanger | dmsimard: weshay: you should see about added them to http://status.openstack.org/openstack-health/#/ | 21:11 |
| pabelanger | not sure how that is done | 21:11 |
| dmsimard | pabelanger: magic | 21:11 |
| pabelanger | logstash I think | 21:11 |
| weshay | rockin | 21:11 |
| dmsimard | pabelanger: but seriously: tl;dr is making sure tempest subunit file is at the root of the logdir | 21:12 |
| weshay | srsly.. pabelanger magic | 21:12 |
| *** fragatina has quit IRC | 21:12 | |
| dmsimard | pabelanger: mtreinish's magic picks up from there | 21:12 |
| dmsimard | stackviz is pretty cool too. | 21:12 |
| pabelanger | dmsimard: Yup, I am working on that too. So stackviz will graph the jobs at the end of runs | 21:12 |
| dmsimard | pabelanger: ah, automatically ? | 21:12 |
| pabelanger | https://review.openstack.org/#/c/352961/ | 21:12 |
| pabelanger | still need to tweak it | 21:13 |
| *** bana_k has joined #tripleo | 21:13 | |
| *** bana_k has quit IRC | 21:13 | |
| dmsimard | pabelanger: ah, just for tripleo | 21:14 |
| dmsimard | pabelanger: thought you were implementing magic stackviz like openstack-health :) | 21:14 |
| pabelanger | dmsimard: Ya, tripleo-ci first | 21:14 |
| *** jeckersb_gone is now known as jeckersb | 21:22 | |
| *** panda is now known as panda|zZ | 21:26 | |
| *** bana_k has joined #tripleo | 21:33 | |
| *** bana_k has quit IRC | 21:35 | |
| openstackgerrit | imain proposed openstack/tripleo-common: Create new docker command hook. https://review.openstack.org/312723 | 21:36 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Move Nova settings into nova-* services https://review.openstack.org/353748 | 21:37 |
| *** shardy has joined #tripleo | 21:40 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Create a default volume type https://review.openstack.org/353499 | 21:42 |
| *** gfidente has quit IRC | 21:44 | |
| *** dbecker has joined #tripleo | 21:45 | |
| openstackgerrit | imain proposed openstack/tripleo-common: Create new docker command hook. https://review.openstack.org/312723 | 21:47 |
| *** ioggstream has joined #tripleo | 21:49 | |
| *** jcoufal has quit IRC | 21:49 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Switch rabbit_hosts list to use hiera interpolation https://review.openstack.org/353680 | 21:53 |
| openstackgerrit | imain proposed openstack/tripleo-common: Create new docker command hook. https://review.openstack.org/312723 | 21:59 |
| shardy | EmilienM: ^^ Hey FYI the interpolation part of that seems to be working now, but puppet is getting stuck running step3 | 22:00 |
| shardy | need some sleep but feel free to update the tht patch if you spot any issue | 22:00 |
| EmilienM | shardy: what? my stuff works? | 22:00 |
| EmilienM | yeah, go sleep it's late | 22:01 |
| EmilienM | we can cotninue tomorrow | 22:01 |
| shardy | EmilienM: I'm not certain yet, but the interpolation does work as noted in the comments | 22:01 |
| EmilienM | cool | 22:01 |
| shardy | EmilienM: thanks for your help with it, looks like it should work modulo some more debugging | 22:02 |
| EmilienM | shardy: there is a hier aproblem | 22:03 |
| EmilienM | http://logs.openstack.org/80/353680/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/d482b9a/logs/postci.txt.gz#_2016-08-10_20_18_44_000 | 22:03 |
| EmilienM | found character that cannot start any token while scanning for the next token at line 36 column 27 | 22:03 |
| EmilienM | its a bad lookup | 22:03 |
| shardy | EmilienM: see patchset 2, I think I fixed that | 22:03 |
| shardy | I needed to double quote the hiera lookup in the service templates | 22:04 |
| EmilienM | indeed | 22:04 |
| EmilienM | yeah | 22:04 |
| EmilienM | wow awesome | 22:04 |
| EmilienM | so slagle is the yaml master | 22:04 |
| EmilienM | and we are the json master | 22:04 |
| openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add specs dir https://review.openstack.org/336109 | 22:05 |
| *** ioggstream has quit IRC | 22:05 | |
| *** [1]cdearborn has quit IRC | 22:06 | |
| *** shardy has quit IRC | 22:07 | |
| *** yamahata has quit IRC | 22:08 | |
| *** yamahata has joined #tripleo | 22:09 | |
| *** yamahata has quit IRC | 22:15 | |
| *** fragatin_ has quit IRC | 22:26 | |
| *** fragatina has joined #tripleo | 22:27 | |
| *** beagles_biab is now known as beagles | 22:33 | |
| *** egafford has quit IRC | 22:33 | |
| *** yamahata has joined #tripleo | 22:37 | |
| *** myoung is now known as myoung|afk | 22:39 | |
| *** pradk has quit IRC | 22:47 | |
| *** dbecker has quit IRC | 22:56 | |
| *** rwsu has joined #tripleo | 23:04 | |
| *** maeca1 has quit IRC | 23:07 | |
| *** saneax_AFK is now known as saneax | 23:13 | |
| *** saneax is now known as Guest7941 | 23:14 | |
| *** bfournie has quit IRC | 23:15 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Use updated overcloud-full image name https://review.openstack.org/353729 | 23:28 |
| *** ayoung has quit IRC | 23:30 | |
| *** chlong|mtg has quit IRC | 23:31 | |
| *** jubapa has quit IRC | 23:37 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Allow map_replace substitution of network names https://review.openstack.org/345510 | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!