| *** penick has quit IRC | 00:13 | |
| saneax | does tripleo.sh work on rhel 7.2? | 00:13 |
|---|---|---|
| *** penick has joined #tripleo | 00:14 | |
| saneax | I have gone through https://github.com/openstack/diskimage-builder/tree/master/elements/rhel and populated the relevant env variables ie DIB_CLOUD_IMAGES, DIB_RHSM_USER, DIB_RHSM_PASSWORD etc.. | 00:14 |
| saneax | still its unable to successfully run instack-virt-setup | 00:15 |
| *** dsariel_ has quit IRC | 00:17 | |
| *** dmacpher-afk has quit IRC | 00:20 | |
| *** thrash is now known as thrash|g0ne | 00:25 | |
| *** penick has quit IRC | 00:33 | |
| *** rbrady has quit IRC | 01:01 | |
| *** Ng has quit IRC | 01:01 | |
| *** bana_k has quit IRC | 01:07 | |
| *** Ng has joined #tripleo | 01:08 | |
| *** ChanServ sets mode: +v Ng | 01:08 | |
| *** tiswanso has joined #tripleo | 01:17 | |
| *** tiswanso has quit IRC | 01:18 | |
| *** tiswanso has joined #tripleo | 01:18 | |
| *** bana_k has joined #tripleo | 01:19 | |
| *** rbrady has joined #tripleo | 01:25 | |
| *** [1]cdearborn has quit IRC | 01:29 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements https://review.openstack.org/375997 | 01:37 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Re-enable ManageFirewall by default. https://review.openstack.org/381864 | 01:41 |
| *** fzdarsky_ has joined #tripleo | 01:46 | |
| *** fzdarsky|afk has quit IRC | 01:50 | |
| *** mbozhenko has joined #tripleo | 01:54 | |
| *** mbozhenko has quit IRC | 01:58 | |
| *** bana_k has quit IRC | 01:59 | |
| *** bana_k has joined #tripleo | 02:18 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: enable undercloud/ssl on multinode jobs https://review.openstack.org/382082 | 02:19 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Change rabbitmq queues HA mode from ha-all to ha-exactly https://review.openstack.org/379586 | 02:21 |
| *** ebalduf has joined #tripleo | 02:45 | |
| *** yamahata has quit IRC | 02:45 | |
| *** mbozhenko has joined #tripleo | 02:53 | |
| *** bana_k has quit IRC | 02:53 | |
| *** mbozhenko has quit IRC | 02:57 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Move the main template files for defalut services to new syntax generation https://review.openstack.org/381975 | 03:06 |
| *** rbrady has quit IRC | 03:16 | |
| *** rbrady has joined #tripleo | 03:19 | |
| *** rbrady is now known as rbrady-afk | 03:21 | |
| openstackgerrit | John Fulton proposed openstack/tripleo-heat-templates: Add option to specify Ceph Cluster Packages https://review.openstack.org/382132 | 03:29 |
| openstackgerrit | RedHat RDO CI proposed openstack/tripleo-heat-templates: GATE TEST, please ignore https://review.openstack.org/365449 | 03:30 |
| openstackgerrit | John Fulton proposed openstack/tripleo-heat-templates: Add option to specify Ceph Cluster Packages https://review.openstack.org/382132 | 03:31 |
| *** links has joined #tripleo | 03:45 | |
| *** sthillma has joined #tripleo | 03:46 | |
| *** fultonj has quit IRC | 03:52 | |
| *** sthillma has quit IRC | 03:54 | |
| *** tiswanso has quit IRC | 03:59 | |
| openstackgerrit | Merged openstack/diskimage-builder: Fix typo in extracting root partition https://review.openstack.org/377296 | 04:00 |
| openstackgerrit | Merged openstack/diskimage-builder: Updated from global requirements https://review.openstack.org/377431 | 04:06 |
| openstackgerrit | Merged openstack/diskimage-builder: Fix a command in Developer Documentation https://review.openstack.org/381833 | 04:06 |
| openstackgerrit | Merged openstack/puppet-tripleo: Add swift proxy for ceilometer middleware https://review.openstack.org/381250 | 04:21 |
| *** sudipto has joined #tripleo | 04:23 | |
| *** sudipto_ has joined #tripleo | 04:23 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Move the main template files for defalut services to new syntax generation https://review.openstack.org/382137 | 04:30 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Add generic template for custom roles. https://review.openstack.org/381587 | 04:37 |
| *** coolsvap has joined #tripleo | 04:45 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 04:48 |
| *** HenryG has quit IRC | 04:51 | |
| *** HenryG has joined #tripleo | 04:52 | |
| *** pgadiya has joined #tripleo | 04:53 | |
| *** skramaja has joined #tripleo | 05:02 | |
| saneax | /msg skramaja hi | 05:05 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements https://review.openstack.org/375997 | 05:05 |
| *** jaosorior has joined #tripleo | 05:07 | |
| *** akuznetsov has joined #tripleo | 05:08 | |
| openstackgerrit | Merged openstack/puppet-tripleo: Cleanup the firewall logic. https://review.openstack.org/381875 | 05:09 |
| openstackgerrit | Merged openstack/diskimage-builder: Move opensuse utils to zypper so they can be shared by SUSE-based distros https://review.openstack.org/379835 | 05:10 |
| *** akuznetsov has quit IRC | 05:12 | |
| *** penick has joined #tripleo | 05:18 | |
| *** saneax is now known as saneax-_-|AFK | 05:28 | |
| *** akshai has joined #tripleo | 05:32 | |
| *** sudswas__ has joined #tripleo | 05:34 | |
| *** jaosorior has quit IRC | 05:35 | |
| *** amoralej|off has quit IRC | 05:35 | |
| *** akrzos has quit IRC | 05:35 | |
| *** akrzos has joined #tripleo | 05:35 | |
| *** jaosorior has joined #tripleo | 05:35 | |
| *** sudipto has quit IRC | 05:36 | |
| *** akuznetsov has joined #tripleo | 05:36 | |
| *** dmanchad has quit IRC | 05:36 | |
| *** jpena|off has quit IRC | 05:36 | |
| *** sudipto has joined #tripleo | 05:37 | |
| *** sudipto_ has quit IRC | 05:37 | |
| *** amoralej has joined #tripleo | 05:41 | |
| *** dmanchad has joined #tripleo | 05:41 | |
| *** jpena|off has joined #tripleo | 05:43 | |
| *** tzumainn has quit IRC | 05:44 | |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: Change rabbitmq queues HA mode from ha-all to ha-exactly https://review.openstack.org/379586 | 05:50 |
| *** dsariel_ has joined #tripleo | 05:51 | |
| *** mbozhenko has joined #tripleo | 05:52 | |
| jaosorior | bandini: hey dude, are you around? | 05:52 |
| bandini | jaosorior: yes | 05:53 |
| bandini | marios: morning sir, could you re-abandon https://review.openstack.org/#/c/380979 ? | 05:53 |
| jaosorior | bandini: hey, how's your galera debugging? NEed help :/ | 05:53 |
| marios | bandini: lookin | 05:54 |
| bandini | jaosorior: it is fairly weak, but I will give it my best shot ;) Worse thing we can write up our findings and then harass mike/damien | 05:54 |
| jaosorior | bandini: bluejeans? | 05:54 |
| matbu | o/ | 05:55 |
| marios | bandini: so this *isn't* going to stable/newton? | 05:55 |
| marios | bandini: https://review.openstack.org/#/c/379586/ | 05:55 |
| marios | bandini: abandoned https://review.openstack.org/#/c/380979/ | 05:55 |
| marios | bandini: good morning :) | 05:56 |
| bandini | marios: correct, i ridiscussed with the rabbotmq guys and agreed to not push this now | 05:56 |
| bandini | marios: thanks ;) | 05:56 |
| bandini | jaosorior: yes 1 min | 05:56 |
| *** akshai has quit IRC | 05:56 | |
| bandini | yo matbu | 05:56 |
| jaosorior | bandini: I'll ping you in a bit, trying to redeploy. Will tell you when it's in a deployeable state again | 05:57 |
| bandini | jaosorior: ack I will be here | 05:57 |
| bandini | Any reviews for my last newton fix would be appreciated ;) https://review.openstack.org/#/c/381584/ | 06:00 |
| bandini | eternal glory for any reviewer! | 06:00 |
| *** pkovar has joined #tripleo | 06:00 | |
| jaosorior | bandini: alright | 06:03 |
| *** masco has joined #tripleo | 06:09 | |
| *** aufi has joined #tripleo | 06:09 | |
| *** akuznetsov has quit IRC | 06:11 | |
| *** mbozhenko has quit IRC | 06:17 | |
| *** rasca has joined #tripleo | 06:19 | |
| *** mbozhenko has joined #tripleo | 06:19 | |
| openstackgerrit | Adriano Petrich proposed openstack/tripleo-quickstart: WIP gate upgrade https://review.openstack.org/342161 | 06:29 |
| *** mbozhenko has quit IRC | 06:29 | |
| *** jprovazn has joined #tripleo | 06:31 | |
| *** mbozhenko has joined #tripleo | 06:32 | |
| *** shardy has joined #tripleo | 06:33 | |
| *** jtomasek_ has joined #tripleo | 06:36 | |
| *** milan has joined #tripleo | 06:36 | |
| *** oshvartz has joined #tripleo | 06:37 | |
| jaosorior | shardy: so I was able to replicate the mysql/galera issue on a new deployment | 06:41 |
| *** dciabrin has quit IRC | 06:41 | |
| jaosorior | for some reason it was working if I did that on a running overcloud | 06:42 |
| shardy | jaosorior: nice, that's good news | 06:42 |
| jaosorior | shardy: so... for some reason galera keeps waiting for the instances to start | 06:42 |
| jaosorior | and mysql seems not to run (even though the logs say otherwise) | 06:43 |
| shardy | is it because it's now trying to use a different network? | 06:43 |
| jaosorior | shardy: I don't know why still. My knowledge about how galera works is not that good... | 06:44 |
| jaosorior | shardy: trying to debug it with bandini. He's gonna try to reproduce it manually | 06:44 |
| jaosorior | shardy: but as far as I can tell, it should even be the same network as before... since the node names without the network default to the same address as internalapi | 06:44 |
| shardy | jaosorior: any hostname pointing to Mysql is now using internal_api, not ctlplane | 06:45 |
| shardy | MysqlNetwork: internal_api | 06:45 |
| shardy | jaosorior: the previous hostname was pointing at ctlplane wasn't it? | 06:45 |
| jaosorior | shardy: the previous hostnames for the wsrep_cluster_address (which is how galera determines the nodes in the cluster) where pointing to the node addresses without the network. | 06:46 |
| * bandini can reproduce | 06:46 | |
| *** pcaruana has joined #tripleo | 06:46 | |
| *** liverpooler has joined #tripleo | 06:46 | |
| shardy | ./manifests/profile/pacemaker/database/mysql.pp: $galera_node_names_lookup = hiera('mysql_node_names', hiera('galera_node_names', | 06:47 |
| shardy | jaosorior: that uses mysql_node_names, which now uses a different network | 06:47 |
| jaosorior | shardy: so it used to be wsrep_cluster_address = gcomm://overcloud-controller-0,overcloud-controller-1,overcloud-controller-2 ... and now it is... wsrep_cluster_address = gcomm://overcloud-controller-0.internalapi.localdomain,overcloud-controller-1.internalapi.localdomain,overcloud-controller-2.internalapi.localdomain | 06:48 |
| shardy | jaosorior: yup, and the old hostnames didn't point to the same network as they do now | 06:48 |
| shardy | (which was the problem we were trying to solve, right?) | 06:49 |
| jaosorior | indeed | 06:49 |
| jaosorior | yep | 06:49 |
| shardy | Ok, I've not had any coffee yet so just making sure :) | 06:49 |
| *** dciabrin has joined #tripleo | 06:49 | |
| bandini | coffee++ | 06:49 |
| * shardy goes to make coffee brb | 06:50 | |
| jaosorior | I'll go brew some | 06:50 |
| jaosorior | lol | 06:50 |
| *** jtomasek_ has quit IRC | 06:50 | |
| *** tremble has joined #tripleo | 06:53 | |
| *** rcernin has joined #tripleo | 06:53 | |
| *** yamahata has joined #tripleo | 06:57 | |
| *** gfidente has joined #tripleo | 06:59 | |
| *** panda|bbl is now known as panda | 07:04 | |
| *** bana_k has joined #tripleo | 07:04 | |
| jaosorior | bandini: is there a way to know which node was pacemaker choosing to bootstrap galera? | 07:05 |
| dciabrin | jaosorior, after bootstrap is finished? | 07:05 |
| jaosorior | dciabrin: I'm actually not sure if bootstrap is even finished in our case. | 07:06 |
| jaosorior | dciabrin: all the galera nodes seem to be marked as slaves | 07:06 |
| jaosorior | dciabrin: So here's the issue. we now specify the exact node names for the wsrep_cluster_address. And now we fail to start the galera cluster... | 07:07 |
| *** leanderthal|afk is now known as leanderthal | 07:07 | |
| *** tesseract- has joined #tripleo | 07:07 | |
| jaosorior | dciabrin: so it used to be wsrep_cluster_address = gcomm://overcloud-controller-0,overcloud-controller-1,overcloud-controller-2 ... and now it is... wsrep_cluster_address = gcomm://overcloud-controller-0.internalapi.localdomain,overcloud-controller-1.internalapi.localdomain,overcloud-controller-2.internalapi.localdomain | 07:07 |
| dciabrin | jaosorior, i'm just chatting with bandini about that right now | 07:08 |
| bandini | jaosorior: hop on our call | 07:08 |
| *** akuznetsov has joined #tripleo | 07:10 | |
| *** rcernin has quit IRC | 07:12 | |
| *** rcernin has joined #tripleo | 07:12 | |
| jaosorior | shardy: so this seems to be a pacemaker issue | 07:12 |
| *** numans has joined #tripleo | 07:13 | |
| *** b00tcat has joined #tripleo | 07:14 | |
| *** athomas has joined #tripleo | 07:15 | |
| jaosorior | shardy: sooo... it seems this is not gonna fly for now :/ | 07:16 |
| shardy | jaosorior: Ok, can you update the patch with details of the analysis, and we'll defer the bug until ocata? | 07:18 |
| jaosorior | shardy: lets do that | 07:19 |
| *** jpena|off is now known as jpena | 07:20 | |
| jaosorior | bandini: you probably have better understanding of this, can you fill in what we discussed there? This was the original issue we were trying to fix (before the rest of the stuff came up) https://bugs.launchpad.net/tripleo/+bug/1628521 | 07:20 |
| openstack | Launchpad bug 1628521 in tripleo "nodename list disregards network" [High,In progress] - Assigned to Juan Antonio Osorio Robles (juan-osorio-robles) | 07:20 |
| *** stevebaker has joined #tripleo | 07:20 | |
| shardy | please cut/paste the description into the patch and the bug, so folks will see why we abandoned/deferred things | 07:21 |
| bandini | jaosorior: sure thing | 07:21 |
| *** hogepodge has quit IRC | 07:21 | |
| jaosorior | shardy: so, it seems that we need a way to tell pacemaker that a node can have more than one name (and for this we need support for a mapping in the Resource Agent). And this is not yet possible... so we need to wait for pacemaker to have this capability. | 07:21 |
| bandini | it is more a galera resource agent thing (as opposed to pacemaker), but yes that is correct | 07:22 |
| jaosorior | bandini: which is why I know you're better suited to fill up the details, cause I'm probably getting this wrong haha | 07:23 |
| bandini | eheheh, on it | 07:23 |
| *** dbecker has joined #tripleo | 07:25 | |
| shardy | d0ugal: Hey can you review https://review.openstack.org/#/c/378740/ when you get a moment? | 07:25 |
| d0ugal | shardy: Sure, I'll look shortly | 07:26 |
| shardy | d0ugal: I know it's a partial solution, and during ocata we can wire in similar interfaces for plan create & get this better supported via the workflow interfaces | 07:26 |
| shardy | so it's something of a stopgap solution | 07:26 |
| shardy | (we really need it for Newton tho, or folks will have to copy all of the packaged tht) | 07:27 |
| bandini | jaosorior: done. | 07:29 |
| jaosorior | shardy: maybe it could work... but we would need to use a "short-name" mapping just for galera. So it gets exactly the same as it used to. While the other services get the correct names | 07:31 |
| jaosorior | bandini: thanks dude! | 07:31 |
| d0ugal | shardy: Sure, makes sense. | 07:31 |
| *** bana_k has quit IRC | 07:32 | |
| *** jpich has joined #tripleo | 07:33 | |
| *** shardy_ has joined #tripleo | 07:34 | |
| *** shardy has quit IRC | 07:36 | |
| *** akuznetsov has quit IRC | 07:37 | |
| *** akuznetsov has joined #tripleo | 07:37 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Use service-specific servernames for haproxy https://review.openstack.org/382177 | 07:39 |
| *** chem has joined #tripleo | 07:42 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Specify the Ceph packages to be installed https://review.openstack.org/382132 | 07:43 |
| *** mcornea has joined #tripleo | 07:44 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Specify the Ceph packages to be installed https://review.openstack.org/382132 | 07:44 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable usage of "short names" for Galera cluster https://review.openstack.org/382183 | 07:49 |
| *** chem has quit IRC | 07:53 | |
| *** chem has joined #tripleo | 07:53 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: j2 template per-role ServiceNetMapDefaults https://review.openstack.org/381902 | 07:54 |
| dciabrin | jaosorior, bandini: for tracking galera bootstrap on internal network -> https://bugzilla.redhat.com/show_bug.cgi?id=1381836 | 07:55 |
| openstack | bugzilla.redhat.com bug 1381836 in resource-agents "galera resource agent cannot bootstrap cluster when galera and pacemaker nodes' name differ" [Medium,New] - Assigned to dciabrin | 07:55 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable usage of "short names" for Galera cluster https://review.openstack.org/382183 | 07:56 |
| shardy_ | dciabrin: please can you raise a corresponding launchpad bug, and link it from the bz? | 07:56 |
| shardy_ | then we can add it to ocata to ensure the work is tracked | 07:56 |
| shardy_ | Or we can link it to https://bugs.launchpad.net/tripleo/+bug/1628521 I guess | 07:56 |
| openstack | Launchpad bug 1628521 in tripleo "nodename list disregards network" [High,In progress] - Assigned to Juan Antonio Osorio Robles (juan-osorio-robles) | 07:56 |
| dciabrin | shardy_, ok will do shortly | 07:56 |
| *** saneax-_-|AFK is now known as saneax | 07:57 | |
| *** shardy_ is now known as shardy | 07:57 | |
| dciabrin | shardy, oh yes, looks good to be. bandini ? ^ | 07:57 |
| bandini | dciabrin: yeah! | 07:58 |
| shardy | ack, sounds good | 07:58 |
| shardy | perhaps we can enumerate the pieces that need fixing in the LP bug so we can keep track of them | 07:58 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Select per-network hostnames for service_node_names https://review.openstack.org/378764 | 07:59 |
| shardy | https://review.openstack.org/#/c/381754/ and https://review.openstack.org/#/c/381757/ are newton backports ready for review & passing CI | 07:59 |
| jaosorior | shardy: done | 08:00 |
| shardy | thanks! | 08:01 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Create the neutron-sanity-check validations https://review.openstack.org/381118 | 08:01 |
| jtomasek | d0ugal: Hi, do you know if there is a way to make a workflow input default to None? https://review.openstack.org/#/c/372628/5/workbooks/baremetal.yaml | 08:04 |
| jtomasek | d0ugal: this seems that it should be doable much simpler https://review.openstack.org/#/c/372628/5/tripleo_common/utils/nodes.py | 08:04 |
| *** fzdarsky_ is now known as fzdarsky | 08:05 | |
| jaosorior | mcornea: hey dude, I submitted a fix for the issue you found with haproxy having the server names hardcoded to be the controllers https://review.openstack.org/#/c/382177/ | 08:06 |
| shardy | jtomasek: can't you just set it in the workflow, e.g foo: null ? | 08:06 |
| jtomasek | shardy: that is probably what am I asking. I need to set the input value in a way where it ends up being None rather than 'None' in the action | 08:07 |
| jtomasek | shardy: so in the tag_node input I can do role: null ? | 08:07 |
| shardy | jtomasek: my understanding of the mistral dsl is that you can specify an optional default via foo: thedefault | 08:08 |
| shardy | so yes, I believe so | 08:08 |
| jpich | jtomasek: Yup, that would do it | 08:08 |
| jtomasek | shardy, jpich: ok thanks, I remember now that jpich was dealing with that too, right? | 08:08 |
| d0ugal | jtomasek: see the queue_name as an example of setting a default | 08:09 |
| jpich | jtomasek: Yep, you can see it for kernel_name etc in that file :) "None" becomes a string but null is None | 08:09 |
| mcornea | jaosorior: hey, thanks. I'll check it out, currently building an environment | 08:09 |
| jtomasek | d0ugal: I know how to set the default, I just didn't know what to set it to to make the value end up being None | 08:09 |
| *** radeks has joined #tripleo | 08:09 | |
| jtomasek | jpich: perfect | 08:10 |
| mcornea | jaosorior: is it dependent on https://review.openstack.org/#/c/378764/ ? | 08:10 |
| jaosorior | mcornea: it isn't | 08:10 |
| d0ugal | jtomasek: oh, I see. looks like you got an answer :) I think pyyaml accepts multiple names for null/None | 08:11 |
| jaosorior | mcornea: it doesn't necessarily give the right names... but I guess that will be fine without internal SSL | 08:11 |
| mcornea | jaosorior: cool | 08:11 |
| mcornea | jaosorior: do we have internal ssl for newton? | 08:11 |
| jaosorior | mcornea: right, so for newton it will be fine. | 08:11 |
| jtomasek | d0ugal, jpich: thanks! | 08:11 |
| *** hogepodge has joined #tripleo | 08:11 | |
| jpich | No problems! | 08:12 |
| shardy | mcornea: Hi! FYI I decided to stick with the original -r roles_data_override.yaml in https://review.openstack.org/#/c/378740/ | 08:12 |
| shardy | mcornea: but I've moved that to Partial-Bug and added the other use-cases we discussed to the bug | 08:12 |
| shardy | I'll work up patches for those but it may not be until next week - at least this allows testing to proceed with a copy of roles_data.yaml | 08:13 |
| *** zoli_gone-proxy is now known as zoliXXL | 08:13 | |
| mcornea | shardy: Hello! alright, I'll keep track of it. Thanks for letting me know | 08:14 |
| shardy | mcornea: the reason for my decision is we need to rework some stuff in the mistral workflow interfaces to enable this to be implemented for plan create | 08:14 |
| shardy | so it's probably best to also do the roles_data merging stuff inside mistral too, and that will take a little more time | 08:14 |
| *** ccamacho has joined #tripleo | 08:16 | |
| mcornea | shardy: yes, agree. as long as we don't have to copy the entire tht directory I think we should be fine for testing at this moment | 08:16 |
| *** akuznetsov has quit IRC | 08:18 | |
| *** ohamada has joined #tripleo | 08:18 | |
| mcornea | jaosorior: https://review.openstack.org/#/c/382177/ closes bug 1629098 right? | 08:19 |
| openstack | bug 1629098 in tripleo "The server names in haproxy configs do not match the ip addresses" [High,Triaged] https://launchpad.net/bugs/1629098 | 08:19 |
| jaosorior | mcornea: lol, thanks for pointing it out, wrong LP | 08:20 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Use service-specific servernames for haproxy https://review.openstack.org/382177 | 08:21 |
| *** spredzy has quit IRC | 08:22 | |
| *** spredzy has joined #tripleo | 08:22 | |
| *** dtantsur|afk is now known as dtantsur | 08:26 | |
| openstackgerrit | Markos Chandras proposed openstack/diskimage-builder: Add zypper-minimal element https://review.openstack.org/381575 | 08:31 |
| openstackgerrit | Markos Chandras proposed openstack/diskimage-builder: Move the opensuse mkinitrd script to the zypper element https://review.openstack.org/381574 | 08:31 |
| openstackgerrit | Markos Chandras proposed openstack/diskimage-builder: Add opensuse-minimal element https://review.openstack.org/381576 | 08:31 |
| *** hewbrocca-afk is now known as hewbrocca | 08:34 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Select per-network hostnames for service_node_names https://review.openstack.org/378764 | 08:37 |
| openstackgerrit | Markos Chandras proposed openstack/diskimage-builder: Add opensuse-minimal element https://review.openstack.org/381576 | 08:40 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 08:41 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 08:41 |
| *** egafford has joined #tripleo | 08:43 | |
| jaosorior | ccamacho: hey dude, when you get a chance can you re-visit https://review.openstack.org/#/c/366548/ and take a look at https://review.openstack.org/#/c/356430/ ? | 08:44 |
| ccamacho | jaosorior, yeahp I was going to start checking https://review.openstack.org/#/c/356430/ | 08:45 |
| *** derekh has joined #tripleo | 08:46 | |
| dtantsur | rook, shardy, a email thread would help a lot. ironic still somewhat lacks this kind of operations feedback | 08:47 |
| ccamacho | jaosorior, one more thing after checking the code I will like to deploy it locally and check the status of the endpoints and also some sanity checks, mind to add to the blog post this info? Something like if you want to try TLS everywhere you can try this.. :P | 08:48 |
| jaosorior | ccamacho: it's in my TODO list, but sure! I can try writing that up today. | 08:49 |
| shardy | ccamacho: Hey, good morning | 08:49 |
| jaosorior | ccamacho: do you deploy with quickstart or with instack? | 08:49 |
| shardy | ccamacho: I wanted to quickly sync up re the remaining composable roles patches | 08:49 |
| shardy | other than reviews/testing, is there anything I can do to help? | 08:50 |
| shardy | seems pretty close, hopefully we can get the remaining pieces landed today | 08:50 |
| ccamacho | jaosorior, yahpo I mean just to do some local testing "instack" | 08:50 |
| ccamacho | shardy morning!!! | 08:50 |
| *** jaosorior is now known as jaosorior_lunch | 08:50 | |
| ccamacho | shardy I thing what we need is already pushed | 08:51 |
| ccamacho | :) so I think we are close to land all | 08:51 |
| ccamacho | we might sync about the pending tasks to be done there | 08:51 |
| ccamacho | when having some time | 08:51 |
| shardy | ccamacho: ack, Ok thanks - I'll pull your latest patches and test, then add review feedback | 08:51 |
| ccamacho | or in the next dfg meeting | 08:51 |
| ccamacho | thanks! | 08:52 |
| *** kaslcrof_ has joined #tripleo | 08:53 | |
| ccamacho | quick question, for newton GA (https://dashboards.rdoproject.org/rdo-dev) we should have all landed in master and backported to stable/newton, right? | 08:54 |
| shardy | ccamacho: yup | 08:54 |
| ccamacho | ack, thanks | 08:54 |
| shardy | so we probably need to land all remaining patches to master today, and all backports tomorrow latest | 08:55 |
| shardy | then we can tag the final release late thursday or Friday | 08:55 |
| cmyster | can't wait :) | 09:01 |
| *** kaslcrof_ has quit IRC | 09:03 | |
| *** kaslcrof has joined #tripleo | 09:03 | |
| *** dsariel_ has quit IRC | 09:07 | |
| *** rawanh_ has joined #tripleo | 09:07 | |
| rawanh_ | Hi, is there a way to pass environment yaml parameter defaults to the overcloud nodes on some file? | 09:08 |
| cmyster | rawanh_: can you please give an example of whats missing? | 09:10 |
| *** nyechiel has joined #tripleo | 09:14 | |
| rawanh_ | cmyster, for example, I want to run some python script in the heat templates and it requires checking the parameter_defaults in the passed environment file in the deploy overcloud command | 09:15 |
| *** dbecker has quit IRC | 09:17 | |
| *** akrivoka has joined #tripleo | 09:18 | |
| cmyster | ok, so have an environment file with parameter_defaults in them (there are enough examples online for those) and add the file with -e | 09:18 |
| cmyster | thats if I understand your original request correctly | 09:19 |
| rawanh_ | cmyster, I already do that, but i can't find any example on how to pass parameters to python scripts in heat templates | 09:20 |
| *** abehl has joined #tripleo | 09:20 | |
| *** dsariel_ has joined #tripleo | 09:20 | |
| rawanh_ | so i thought about like if the environment yaml file is uploaded somewhere locally on the overcloud nodes, is it ? | 09:21 |
| cmyster | wait, you have scripts that you want to run during deployment ? | 09:21 |
| cmyster | by the deployment? | 09:22 |
| rawanh_ | yes, a python script that i set to run as a first boot script, but it needs values from the environemnt yaml file | 09:22 |
| cmyster | in this case I'd probably do something like write the file or parts of it somewhere on the target node and hen let the script read from that, since the node has no access to the original file | 09:24 |
| cmyster | not sure if it is possible though | 09:25 |
| cmyster | shardy? ^ | 09:25 |
| rawanh_ | aha I see, i thought about it too like in a shell script, but i wanted to see if there is an easier way to do it , maybe like passing them to some hiera file on the overcloud node and then refering to it | 09:26 |
| openstackgerrit | Merged openstack/puppet-tripleo: Fix the timeout for pacemaker systemd resources https://review.openstack.org/381584 | 09:31 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Add Select per-network hostnames for service_node_names to role.role.j2.yaml https://review.openstack.org/382217 | 09:31 |
| rook | dtantsur: hey, yeah, i need to get on it -- i have been preparing for a all day meeting today... | 09:34 |
| shardy | rawanh_, cmyster: Yes it's possible, e.g see these examples: | 09:34 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/firstboot/userdata_example.yaml#L52 | 09:34 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/extraconfig/all_nodes/neutron-ml2-cisco-nexus-ucsm.yaml#L229 | 09:35 |
| shardy | you first define the parameter (in the parameters block of the template) which corresponds to the key you set in parameter_defaults | 09:35 |
| shardy | then you pass that value via get_param into a OS::Heat::SoftwareConfig input via get_param | 09:36 |
| rawanh_ | but what if i'm using get_file: { script_name.py }, how can I pass the parameter ? are there python examples ? | 09:37 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Adds Environment File for Removing Sahara during M/N upgrade https://review.openstack.org/375517 | 09:38 |
| *** jlinkes has joined #tripleo | 09:39 | |
| *** ramishra has quit IRC | 09:39 | |
| *** cylopez has joined #tripleo | 09:40 | |
| openstackgerrit | Steven Hardy proposed openstack/python-tripleoclient: Add optional overcloud deploy roles_data.yaml override https://review.openstack.org/382224 | 09:40 |
| *** mbozhenko has quit IRC | 09:40 | |
| *** ramishra has joined #tripleo | 09:41 | |
| shardy | rawanh_: I just provided a python example | 09:41 |
| shardy | rawanh_: you pass the values via inputs to the software config resource | 09:41 |
| shardy | so you can still use get_file | 09:41 |
| marios | jistr: updated https://review.openstack.org/#/c/375517/5 and comment/commit msg for the context appreciate thoughts since you were already +2 there and after chat yesterday and before about this | 09:42 |
| marios | tosky not about yet it seems | 09:42 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/extraconfig/all_nodes/neutron-ml2-cisco-nexus-ucsm.yaml#L300 | 09:42 |
| jistr | sure thing, looking | 09:42 |
| shardy | rawanh_: ^^ there's an example of passing an input parameter into a SoftwareDeployment | 09:42 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/extraconfig/all_nodes/neutron-ml2-cisco-nexus-ucsm.yaml#L229 | 09:43 |
| shardy | which gets passed to the python script via that input | 09:43 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/extraconfig/all_nodes/neutron-ml2-cisco-nexus-ucsm.yaml#L242 | 09:43 |
| shardy | it's substituted before the script runs | 09:43 |
| ccamacho | shardy im getting this in some CI jobs 2016-10-05 09:25:52.039208 | Exception updating plan: Error parsing template https://192.0.2.2:13808/v1/AUTH_6abc651b20fe4317b4e404c0dbc30a4c/overcloud/overcloud.yaml Template format version not found. | 09:44 |
| jistr | marios: so you mean sahara included and enabled by default even on converge, and if you don't want it, you'll pass major-upgrade-remove-sahara for *both* the controller upgrade and the converge? | 09:44 |
| *** paramite has joined #tripleo | 09:44 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Select per-network hostnames for service_node_names https://review.openstack.org/378764 | 09:45 |
| ccamacho | like not using latest version from Heat? | 09:45 |
| shardy | ccamacho: Ok - we may need to add something to the ci job which downloads the plan files so we can see exactly what was rendered | 09:45 |
| marios | jistr: right. since now/v4... it was enabled on controller upgrade by default. but on converge you had to do something special | 09:45 |
| marios | jistr: instead jsut make the default all the way through to be keep sahara. and if you want to remove, include the env | 09:45 |
| shardy | ccamacho: it's more likely something broke with the rendering so we passed an empty file or something | 09:46 |
| ccamacho | mmmm ack | 09:46 |
| marios | jistr: same result, just thought it may be cleaner frmo the operator/docs point of view | 09:46 |
| shardy | as that error means there was no version, not that the version was wrong | 09:46 |
| ccamacho | shardy^ | 09:46 |
| ccamacho | I see | 09:46 |
| *** jaosorior_lunch is now known as jaosorior | 09:46 | |
| shardy | I'd see if you can reproduce locally, then swift download overcloud overcloud.yaml to see what's up | 09:46 |
| marios | jistr: i updated with another comment about something i changed refresh | 09:46 |
| *** paramite has quit IRC | 09:47 | |
| *** paramite has joined #tripleo | 09:47 | |
| *** pgadiya has quit IRC | 09:49 | |
| jistr | marios: ack +2 | 09:52 |
| marios | jistr: thanks lets see if tosky is happier this way too :) | 09:53 |
| marios | jistr: enjoy lunch! | 09:53 |
| jistr | thanks :) | 09:55 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 09:56 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Add generic template for custom roles. https://review.openstack.org/381587 | 09:56 |
| *** mbozhenko has joined #tripleo | 09:56 | |
| ccamacho | shardy, just rebased the submissions and working fine locally | 09:56 |
| *** pgadiya has joined #tripleo | 10:01 | |
| shardy | https://review.openstack.org/#/c/381902/ is ready for review if anyone has a moment | 10:03 |
| shardy | ccamacho: ack, thanks, going to make a coffee then try them myself | 10:03 |
| ccamacho | shardy ack | 10:03 |
| *** mbozhenko has quit IRC | 10:03 | |
| jaosorior | shardy: can you check this out? https://review.openstack.org/#/c/382177/ | 10:04 |
| rawanh_ | shardy, thanks I will try it | 10:08 |
| openstackgerrit | Isaac Beckman proposed openstack/diskimage-builder: Enable ssh password authentication for devuser https://review.openstack.org/376572 | 10:11 |
| *** tosky has joined #tripleo | 10:18 | |
| *** zoliXXL is now known as zoli|lunch | 10:20 | |
| shardy | jaosorior: lgtm - we should probably try a WIP patch removing controller_node_names and see what else breaks | 10:22 |
| *** dbecker has joined #tripleo | 10:23 | |
| *** rawanh_ has quit IRC | 10:28 | |
| *** tzumainn has joined #tripleo | 10:30 | |
| jaosorior | shardy: that sounds like a reasonable next step | 10:31 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-common: Support node untagging https://review.openstack.org/372628 | 10:31 |
| jaosorior | shardy: right now I'm working on something else, so if someone else can check that up, it would be cool | 10:31 |
| *** cylopez has quit IRC | 10:31 | |
| jaosorior | mcornea: hey dude, let me know when you try out the puppet-tripleo patch for haproxy | 10:31 |
| chem | hi, how can I include swap.j2.yaml, I tried OS::TripleO::AllNodesExtraConfig: /usr/share/openstack-tripleo-heat-templates/extraconfig/all_nodes/swap.yaml, but it fails with Could not fetch contents for file://... | 10:31 |
| chem | any idea ? | 10:31 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Make keystone api network hiera composable https://review.openstack.org/381754 | 10:32 |
| *** ramishra has quit IRC | 10:32 | |
| chem | jistr: maybe you know :) ^ | 10:32 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Replace per role manifests with a common role manifest https://review.openstack.org/381757 | 10:32 |
| *** ramishra has joined #tripleo | 10:33 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Select per-network hostnames for service_node_names https://review.openstack.org/378764 | 10:34 |
| shardy | matbu: Hey my tripleoclient patch caused https://review.openstack.org/#/c/379547/ to merge conflict, want me to push a rebase, or are you planning another revision anyway? | 10:36 |
| *** ramishra has quit IRC | 10:37 | |
| *** fxpester has joined #tripleo | 10:37 | |
| fxpester | hi all | 10:37 |
| fxpester | maybe anyone messed with error: "grubby fatal error: unable to find a suitable template" when building rhel images with dib ? this is breaking grub.conf so image can`t boot. | 10:37 |
| matbu | shardy: i'll probably update to put a fixed name for /tmp/tht, if you prefer to have it fix | 10:37 |
| *** ramishra has joined #tripleo | 10:38 | |
| shardy | matbu: ack - yeah I think that may be best, as we had some bugs previously where /tmp/$random dirs/files were created and they recently got fixe | 10:38 |
| shardy | fixed | 10:38 |
| matbu | shardy: ack , so i'll rebase, i'm going to do it after lunch | 10:38 |
| shardy | I agree having the latest files around would be useful, but it's probably too much if we persist every deploy/update without cleaning up (IMHO) | 10:39 |
| shardy | matbu: great, thanks! | 10:39 |
| matbu | shardy: or do you prefer to test it now with the fixed dir ? | 10:39 |
| shardy | I'll test & provide review feedback after you push the update | 10:39 |
| shardy | matbu: I'm testing some other stuff now, so after lunch will be fine | 10:39 |
| matbu | shardy: k cool :) | 10:39 |
| matbu | thx | 10:39 |
| *** athomas has quit IRC | 10:40 | |
| *** mbozhenko has joined #tripleo | 10:41 | |
| shardy | fxpester: Not seen that one myself, you might want to ask in #openstack-dib | 10:42 |
| fxpester | thx, will try to ask them, in anycase there is a link for bug - https://bugs.launchpad.net/diskimage-builder/+bug/1477179 | 10:43 |
| openstack | Launchpad bug 1477179 in diskimage-builder "centos 6 image doesn't boot after creation" [Medium,Triaged] | 10:43 |
| jtomasek | honza, akrivoka: Hi, I've updated the tagging workflow patch slightly but it still needs work due to the fact that it does not update node counts and flavor parameters when nodes are untagged | 10:44 |
| dtantsur | shardy, oh, we got a new channel for DIB? | 10:45 |
| shardy | dtantsur: apparently so, I just found it ;) | 10:45 |
| dtantsur | heh | 10:45 |
| jtomasek | honza, akrivoka: and since the rc3 is just around the corner, I am going to update GUI so it allows user to set node counts on the role cards, which is the most important thing and then we can iterate more on making the tagging work as needed | 10:45 |
| *** ramishra has quit IRC | 10:47 | |
| *** ramishra has joined #tripleo | 10:48 | |
| akrivoka | jtomasek: sounds good, thanks! I'll look at the updated patches | 10:48 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: j2 template per-role ServiceNetMapDefaults https://review.openstack.org/381902 | 10:53 |
| *** saneax is now known as saneax-_-|AFK | 10:53 | |
| *** jlinkes has quit IRC | 10:54 | |
| *** thrash|g0ne is now known as thrash | 10:54 | |
| *** jlinkes has joined #tripleo | 10:57 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: j2 template per-role ServiceNetMapDefaults https://review.openstack.org/382273 | 10:57 |
| *** panda has quit IRC | 10:57 | |
| *** saneax-_-|AFK is now known as saneax | 10:58 | |
| openstackgerrit | Steven Hardy proposed openstack/python-tripleoclient: Add optional overcloud deploy roles_data.yaml override https://review.openstack.org/382224 | 10:59 |
| *** adarazs is now known as adarazs_lunch | 10:59 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: j2 template role config templates https://review.openstack.org/382276 | 11:00 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-ui: Save the result of direct action calls in Mistral https://review.openstack.org/382277 | 11:00 |
| *** radeks has quit IRC | 11:01 | |
| *** athomas has joined #tripleo | 11:03 | |
| *** pkovar has quit IRC | 11:05 | |
| jistr | chem: could it be this? https://bugs.launchpad.net/tripleo/+bug/1624727 | 11:07 |
| openstack | Launchpad bug 1624727 in tripleo "Could not fetch contents for file:///home/stack/tripleo-heat-templates/puppet/post.yaml" [Critical,In progress] - Assigned to mbu (mat-bultel) | 11:07 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-ui: Save the result of direct action calls in Mistral https://review.openstack.org/382277 | 11:07 |
| chem | jistr: nope, I've applied the patch already ... | 11:07 |
| mcornea | jaosorior: hey, yep. haven't finished my deployment yet | 11:08 |
| chem | jistr: but with the jinja2 template system, I've no idea what I'm doing really. Do you have a pointer doc ? | 11:09 |
| matbu | chem: do you have the review up to date ? | 11:11 |
| dtantsur | I'm testing the newton bits from RDO, and I want to note that the deploy command progress output has become much nicer. thanks whoever did it | 11:11 |
| jistr | chem: hmm i don't recall seeing a doc, but what i think happens is that we send the raw templates (unrendered j2) to mistral, which saves them into swift, and triggers the rendering and adds rendered files into swift too | 11:11 |
| chem | matbu: checking | 11:12 |
| jistr | and then i'm a bit fuzzy on what happens next -- if we trigger Heat deploy from Mistral too, or if we're not there yet and we're still downloading the templates and triggering the heat stack-create/update locally via tripleoclient | 11:12 |
| matbu | chem: there was a bug that i fixed yesterday | 11:12 |
| chem | matbu: so, yes I have the latest from this morning | 11:12 |
| d0ugal | jistr: The deploy happens in mistral | 11:12 |
| *** panda has joined #tripleo | 11:12 | |
| jistr | d0ugal: thanks | 11:12 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 11:13 |
| d0ugal | jistr: but things get a bit fuzzy, tripleoclient isn't totally disconnected yet, it hacks a bit to work around some limitations. | 11:13 |
| chem | jistr: because I can see the use of YAQL in mistral but I could not find any reference to jinja2 | 11:13 |
| matbu | chem: interesting | 11:14 |
| d0ugal | chem: Mistral itself doesn't support jinja2 yet (it will very soon, but not for Newton) | 11:14 |
| chem | d0ugal: so it's a custom workflow that render the template ? | 11:14 |
| jistr | chem: i guess this is getting called from mistral then https://github.com/openstack/tripleo-common/blob/e09f4689d813d3b8fbab059515f656ed9bdd8236/tripleo_common/actions/templates.py#L71 | 11:14 |
| d0ugal | chem: Yeah, the rendering is in our code. | 11:14 |
| *** ccamacho is now known as ccamacho|lunch | 11:15 | |
| rbrady-afk | d0ugal, chem: coming soon - https://review.openstack.org/#/c/374813/ | 11:15 |
| chem | d0ugal: ack, thanks. | 11:15 |
| *** rbrady-afk is now known as rbrady | 11:15 | |
| openstackgerrit | Brad P. Crochet proposed openstack/python-tripleoclient: Downloads templates from swift before processing update https://review.openstack.org/381899 | 11:15 |
| d0ugal | rbrady: I think that is actually a different usage of jinja2 than they want, but that is what I was thinking of too at first. | 11:15 |
| jistr | matbu: btw related to ^^, do you know why do we need to download the templates to local machine, if the Heat deploy happens via Mistral? | 11:16 |
| d0ugal | I think the line is a bit blurry when it comes to what Mistral does and what TripleO does with (or adds to) mistral | 11:16 |
| jistr | matbu: i'm a bit fuzzy on the purpose of your patch to tripleoclient | 11:16 |
| d0ugal | jistr: we download them because some rendered files are missing | 11:16 |
| chem | matbu: it's easy to test it, in a oooq env, play add http://paste.openstack.org/show/584425/ and run the overcloud-init step, you should have the error | 11:16 |
| d0ugal | jistr: and there needs to be template processing locally | 11:16 |
| d0ugal | jistr: this is one of those hacks :) | 11:17 |
| shardy | yeah, we've created a split-brain situation because we combine client side template processing with server side j2 rendering | 11:17 |
| jistr | d0ugal: yea i got as much, but what i'm wondering is what do we do locally still that needs those rendered files | 11:17 |
| shardy | jistr: it because the heatclient function which finds locally referenced files resolves everything in the tree | 11:17 |
| shardy | including all the rendered files | 11:18 |
| shardy | and it can't do that until the *.j2.yaml files are rendered | 11:18 |
| shardy | so chicken/egg problem | 11:18 |
| matbu | hehe thanks shardy , faster and clearer | 11:18 |
| shardy | I think during ocata we need to look at introducing a more contrained interface which removes the need for client side heatclient resolution | 11:18 |
| shardy | like --extra-files-dir or something | 11:19 |
| rbrady | shardy: +1 | 11:19 |
| shardy | larsks and I were chatting about it, and this would allow us to deprecate (and eventually remove) all the client side scary template processing | 11:19 |
| jistr | hmm yea but why do we call heatclient from tripleoclient... Is it to process the environment files into a single environment? Or something else? (Or that + something else?) | 11:19 |
| shardy | jistr: yea we call heatclient from tripleoclient, primarily to resolve all the stuff like get_file in nested templates (including those referenced from rendered j2.yaml files and their children) | 11:20 |
| shardy | so we can build a complete files map to pass to heat | 11:20 |
| shardy | (or, in this case, figure out what needs to be added to the plan in swift) | 11:20 |
| d0ugal | jistr: so, for example, if I do "overcloud deploy --templates -e my_env.yaml" - my_env.yaml could contain references to files anywhere locally - we need to process the files to find them | 11:21 |
| d0ugal | jistr: and better heatclient does that than we re-invent it | 11:21 |
| chem | d0ugal: matbu that's exactly the problem I got during with this http://paste.openstack.org/show/584425/, the swap.j2.yaml seems to not be included | 11:21 |
| jistr | ahh right i see... so the templates go to mistral for rendering, then we download them back and resolve get_file etc., and upload back to mistral, which in the end calls heat? | 11:21 |
| shardy | jistr: yeah, exactly | 11:22 |
| shardy | it's kind of messy but that's the situation right now | 11:22 |
| jistr | ok thanks now i see | 11:22 |
| chem | shardy: thanks | 11:22 |
| shardy | jistr: here's a concrete example of where things break: https://bugs.launchpad.net/tripleo/+bug/1625783 | 11:22 |
| openstack | Launchpad bug 1625783 in tripleo "unable to leverage swap-partition extraconfig" [High,Triaged] - Assigned to Steven Hardy (shardy) | 11:22 |
| d0ugal | jistr: Yeah, it's really weird :( | 11:22 |
| shardy | jistr: in that case we've got the option of using a relative path (then it's found inside the plan), but the heatclient processing breaks for all absolute paths | 11:23 |
| d0ugal | jistr: the --extra-files-dir would solve it as we get used to tell us where everything is | 11:23 |
| chem | matbu: d0ugal shardy so that's the bug I'm hitting, all right :) | 11:23 |
| openstackgerrit | Alexey Deryugin proposed openstack/puppet-pacemaker: [CI TEST] DO NOT MERGE https://review.openstack.org/382333 | 11:23 |
| shardy | so I think we need this workaround at least as an interim measure to allow backwards compatibility | 11:23 |
| d0ugal | chem: I think there is a patch up for that now | 11:23 |
| *** mburned_out is now known as mburned | 11:23 | |
| d0ugal | chem: it may have even landed this morning. | 11:23 |
| shardy | chem: sec, let me add a comment explaining one workaround | 11:24 |
| shardy | chem: comment added to https://bugs.launchpad.net/tripleo/+bug/1625783 | 11:26 |
| openstack | Launchpad bug 1625783 in tripleo "unable to leverage swap-partition extraconfig" [High,Triaged] - Assigned to Steven Hardy (shardy) | 11:26 |
| shardy | the final fix hasn't yet landed, but you can use a relative path to work around it for now | 11:26 |
| *** rbowen has joined #tripleo | 11:27 | |
| chem | shardy: ack, thanks | 11:28 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 11:30 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 11:30 |
| *** coolsvap is now known as coolsvap_ | 11:30 | |
| chem | shardy: matbu: so I tested with the patch and with relative path and it fails in the same way | 11:31 |
| chem | Could not fetch contents for file:///home/stack/extraconfig/all_nodes/swap.yaml | 11:32 |
| shardy | chem: Ok that is confusing as I tested the relative path workaround myself earlier today | 11:32 |
| shardy | chem: what tripleoclient version do you have? | 11:33 |
| shardy | chem: for the relative path workaround to work, https://review.openstack.org/#/c/365735/ will be needed | 11:33 |
| chem | shardy: oups ... wrong session, it's downstream : 5.2.0-1.el7ost @rhelosp-10.0-brew | 11:34 |
| shardy | that landed a couple of weeks ago | 11:34 |
| shardy | chem: ack, please check you've got that patch in the build? | 11:34 |
| chem | shardy: checking | 11:34 |
| dtantsur | do simple overcloud stack updates work now? or should I wait for https://bugs.launchpad.net/tripleo/+bug/1624727 to be fixed? | 11:34 |
| openstack | Launchpad bug 1624727 in tripleo "Could not fetch contents for file:///home/stack/tripleo-heat-templates/puppet/post.yaml" [Critical,In progress] - Assigned to mbu (mat-bultel) | 11:34 |
| shardy | dtantsur: updates should work, I think we need that fix for major version upgrades tho | 11:35 |
| dtantsur | thanks | 11:35 |
| *** shardy is now known as shardy_lunch | 11:35 | |
| matbu | chem: i'll test the case of swp.j2 | 11:36 |
| *** ohamada has quit IRC | 11:37 | |
| *** ohamada has joined #tripleo | 11:37 | |
| jpich | mandre: Hello!! Since you very unwisely offered to help with UI puppet problems... :-) Do you see an obvious solution to https://bugs.launchpad.net/tripleo/+bug/1630546 ? | 11:38 |
| openstack | Launchpad bug 1630546 in tripleo "Can't log in to the UI: wrong keystone endpoint URL" [Critical,Triaged] | 11:38 |
| jpich | mandre: Any tip or hiera debugging advice would be most welcome, though if nothing comes to mind no worries!! It's probably time I learn about all this anyhow | 11:39 |
| mandre | jpich: so it looks like we need to use a different hiera key | 11:43 |
| jpich | mandre: Is there a magical command or place where we could see all the existing keys/data? | 11:44 |
| mandre | jpich: if you know what value you're looking for, I find it easier to grep through /etc/puppet/hieradata to find the right key | 11:44 |
| jpich | mandre: Thank you for the tip! | 11:44 |
| mandre | jpich: sudo grep -R 5000 /etc/puppet/hieradata/ | 11:44 |
| jpich | mandre: So... specifically calling out to "keystone_auth_uri_v2" looks like | 11:45 |
| chem | shardy_lunch: well, the changelog isn't usefull, but reviewing the patch, it looks like it's in there. | 11:46 |
| jpich | mandre: Thank you for your help! I'll submit a patch. I didn't even know where to start looking for the hiera values | 11:46 |
| mandre | jpich: maybe you can give it a try with keystone_auth_uri, I don't see the reason why pinning it to v2 | 11:46 |
| jpich | mandre: v2 is what's recommended in the sample -> https://github.com/openstack/tripleo-ui/blob/master/dist/tripleo_ui_config.js.sample#L7 , I don't know if anyone has tested with v3 at all | 11:47 |
| jpich | mandre: Something to try in Ocata!! | 11:49 |
| mandre | jpich: don't you want to be the one? :) | 11:49 |
| jpich | mandre: Maybe not this time ;) | 11:49 |
| tzumainn | jpich, mandre, I would advise against it, we've been living that dream in ManageIQ :P | 11:49 |
| mandre | tzumainn: haha, so you're *the one*. Let's stick to v2 | 11:50 |
| openstackgerrit | mathieu bultel proposed openstack/python-tripleoclient: Download templates from swift before processing with heatclient https://review.openstack.org/379547 | 11:50 |
| EmilienM | hello | 11:51 |
| *** pgadiya has quit IRC | 11:51 | |
| thrash | o/ | 11:51 |
| hewbrocca | EmilienM: good morning! | 11:51 |
| jpich | mornin' | 11:51 |
| *** pkovar has joined #tripleo | 11:52 | |
| *** matbu is now known as matbu|lch | 11:53 | |
| *** zoli|lunch is now known as zoli | 11:53 | |
| *** mburned is now known as mburned_out | 11:55 | |
| openstackgerrit | Stig Telfer proposed openstack/tripleo-docs: Fix OpenStack client invocation https://review.openstack.org/382353 | 11:56 |
| *** rbowen has quit IRC | 12:00 | |
| *** zoli is now known as zoliXXL | 12:00 | |
| *** trown|outtypewww is now known as trown | 12:02 | |
| EmilienM | shardy_lunch: I see a lot of backports failing CI https://review.openstack.org/#/q/tripleo/rc3+status:open - I'm wondering if we should land them in a specific order | 12:02 |
| openstackgerrit | Julie Pichon proposed openstack/puppet-tripleo: Explicitly use Keystone v2 endpoint in the UI https://review.openstack.org/382359 | 12:03 |
| EmilienM | bandini: I'm confused, so https://review.openstack.org/#/c/379586/ is not for newton anymore? | 12:03 |
| *** jprovazn has quit IRC | 12:03 | |
| *** rbowen has joined #tripleo | 12:03 | |
| *** mburned_out is now known as mburned | 12:03 | |
| *** jayg|g0n3 is now known as jayg | 12:04 | |
| *** [1]cdearborn has joined #tripleo | 12:04 | |
| *** ipsecguy has quit IRC | 12:04 | |
| *** ipsecguy has joined #tripleo | 12:05 | |
| *** adarazs_lunch is now known as adarazs | 12:05 | |
| EmilienM | marios: https://review.openstack.org/#/c/375517/ looks ok to me, can you push the update suggested by Sofer, so we can merge it today? | 12:06 |
| bandini | EmilienM: correct. I had a call with the rabbitmq guys yesterda and in the end we agreed to push it to ocata only | 12:06 |
| *** amoralej is now known as amoralej|lunch | 12:06 | |
| EmilienM | bandini: ok. Did we land anything in stable/newton already for that? | 12:06 |
| *** social has quit IRC | 12:06 | |
| *** sudipto has quit IRC | 12:06 | |
| *** sudswas__ has quit IRC | 12:06 | |
| bandini | EmilienM: no we did not, so we're good. So we addressed the concerns of doing it so late | 12:07 |
| EmilienM | bandini: ok | 12:07 |
| *** pgadiya has joined #tripleo | 12:07 | |
| EmilienM | can anyone approve this backport please? https://review.openstack.org/#/c/381568/ | 12:08 |
| trown | EmilienM: with RC2 I am seeing "[overcloud]: CREATE_FAILED Resource CREATE failed: Expression consumed too much memory" during deploy... known issue? | 12:08 |
| *** jcoufal has joined #tripleo | 12:09 | |
| EmilienM | jaosorior: approving https://review.openstack.org/#/c/382177/ - please backport it | 12:09 |
| EmilienM | trown: yes | 12:09 |
| EmilienM | trown: fixed by https://github.com/openstack/puppet-heat/commit/7c2489b23cc6fbaaff8c486a7fd139afccd9d1b4 | 12:09 |
| EmilienM | it's not yet released I'm working on it this morning | 12:09 |
| *** social has joined #tripleo | 12:09 | |
| EmilienM | it should have been released yesterday but as you know CI was down for some time | 12:09 |
| gfidente | EmilienM I think we'll want this guy too https://review.openstack.org/#/c/382132/ | 12:10 |
| gfidente | until we work out in the puppet module a mechanism to pull in different packages based on the repos enabled | 12:10 |
| trown | EmilienM: cool thanks, this is with the RDO release candidate, so I will make sure that makes it in | 12:10 |
| EmilienM | gfidente: reviewing it now. | 12:10 |
| tbarron | Jokke_: gfidente manila deploy with cephfs backend incorrectly leaves enabled_share_protocols set to default value 'NFS,CIFS' in manila.conf | 12:10 |
| tbarron | Jokke_: gfidente this breaks share creation with ceph backend | 12:11 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 12:11 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 12:11 |
| tbarron | Jokke_: gfidente we should set it to 'NFS,CIFS,CEPHFS' (easiest approach) or conditionally set it depending on the enabled backends (can do that later when more types of backend get enabled) | 12:12 |
| tbarron | Jokke_: gfidente can we fix this for rc3, should I raise a bug? | 12:12 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Include ceilometer in swift proxy pipeline https://review.openstack.org/371950 | 12:12 |
| EmilienM | gfidente: -1 on https://review.openstack.org/#/c/382131/ | 12:13 |
| jpich | EmilienM: Is puppet-tripleo matching the TripleO schedule in terms of cutting a rc3 tomorrow? I'm not sure if it belongs to the puppet project or to us? | 12:13 |
| honza | jtomasek: why does it not update the counts? what's the problem? | 12:13 |
| marios | EmilienM: ack will do tosky did you get a chance to check https://review.openstack.org/#/c/375517/ I updated it | 12:13 |
| *** coolsvap_ is now known as coolsvap | 12:13 | |
| gfidente | tbarron ack I think we can start by adding CEPHFS to the list | 12:14 |
| EmilienM | gfidente: and same https://review.openstack.org/#/c/382132/ | 12:14 |
| *** nyechiel has quit IRC | 12:14 | |
| jtomasek | honza: https://review.openstack.org/#/c/372628/5/workbooks/baremetal.yaml | 12:14 |
| EmilienM | gfidente: why don't we have ceph metapackage everywhere? | 12:14 |
| tosky | marios: yep, so basically did you manage to overcome the "we must specify sahara.yaml" problem? | 12:14 |
| EmilienM | jpich: yes, puppet-tripleo belongs to tripleo | 12:14 |
| jtomasek | honza: 'update_role_parameters' task | 12:14 |
| tbarron | gfidente: yes, that will work for now, having 'NFS,CIFS' in there for cephfs deploy won't do any harm the way the manila code is currently written | 12:14 |
| *** pgadiya has quit IRC | 12:14 | |
| tbarron | Jokke_: ^^ | 12:14 |
| tosky | marios: I guess it depends on how I read the " any subsequent stack updates" | 12:14 |
| EmilienM | jpich: https://github.com/openstack/releases/blob/master/deliverables/newton/puppet-tripleo.yaml#L3 | 12:15 |
| jtomasek | honza: one sec | 12:15 |
| tosky | marios: what happens if the subsequent stack update has no sahara.yaml? Is the service simply disabled until sahara.yaml is readded? | 12:15 |
| honza | jtomasek: no rush | 12:15 |
| tbarron | gfidente: Jokke_ so that will work both for the cephfs only and the cephfs plus generic and/or netapp cases | 12:15 |
| jpich | EmilienM: Awesome, thanks for the info and URL! I might send an infra patch to have gerrit patches to puppet-tripleo update the tripleo launchpad automatically too, then | 12:16 |
| marios | tosky: so there is a general case of needing to include -e sahara for stack update for ever and ever . this is not caused by or dealt with in this patch | 12:16 |
| marios | tosky: so yes if you do a stack update with out the sahara env, you will lose sahara | 12:16 |
| gfidente | tbarron but we should open a bug yes | 12:17 |
| tosky | marios: lose == lose data, or lose the service? | 12:17 |
| gfidente | not sure if you have time for it? | 12:17 |
| marios | tosky: lose = OS::Heat::None for the service so we don't actively go wiping data afaik | 12:17 |
| *** cylopez has joined #tripleo | 12:17 | |
| *** masco has quit IRC | 12:18 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Ceilometer Wsgi Mitaka->Newton upgrades https://review.openstack.org/360004 | 12:18 |
| openstackgerrit | Julie Pichon proposed openstack/puppet-tripleo: Explicitly use Keystone v2 endpoint in the UI https://review.openstack.org/382359 | 12:19 |
| jtomasek | honza: ok, so it sets the counts correctly, although set_count_and_flavor_params in utils/parametes.py needs to get updated. It needs to check if there are any nodes tagged with that role and if not, reset the flavor parameter to 'baremetal' | 12:19 |
| EmilienM | jpich: thanks | 12:19 |
| jtomasek | honza: which should be quick change. Can you do it please? | 12:19 |
| *** maticue has joined #tripleo | 12:20 | |
| honza | jtomasek: oh, i see --- i thought i introduced a bug somewhere | 12:20 |
| marios | tosky: the idea was that with v4 if you do nothing, on controller upgrade you keep service, but on converge you have to do something special to keep it (add the -e file) | 12:20 |
| honza | jtomasek: yes, i can do it | 12:20 |
| tosky | marios: ok, so it's still the same point raised by jistr yesterday: you always need -e sahara.yaml | 12:21 |
| marios | tosky: so the idea is to make is by default always keep sahara if you do nothing (so we can deal with the -e sahara env by adding it in the converge environment file) | 12:21 |
| tosky | because you need to converge at some point | 12:21 |
| jtomasek | honza: otherwise it is working as expected. The GUI patch needs slight change, because of how the form works selecting the nodes to tag/untag is not that simple | 12:21 |
| jtomasek | honza: I'll take a look into it now | 12:21 |
| *** pradk has joined #tripleo | 12:21 | |
| marios | tosky: if you want/need to ditch sahara, then you include the 'ditch sahara' env file for controller upgrade and converge | 12:21 |
| gfidente | tbarron and using NFS,CIFS,CEPHFS will work as well | 12:22 |
| marios | tosky: the 'you always need sahara' is an aside... as i said it isn't caused by or dealt with in a permanent way in this patch | 12:22 |
| gfidente | when not using ceph? | 12:22 |
| marios | tosky: i.e. you always have that because we switch from mitaka On to newton Off | 12:22 |
| gfidente | tbarron if so the change is a oneline | 12:22 |
| tosky | marios: I see, I misunderstood the last description on the review | 12:22 |
| tbarron | gfidente: it would be a big hack though, we should just leave the default when not using ceph unless that's too hard | 12:23 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add CEPHFS to the list of Manila enabled share protocols https://review.openstack.org/382372 | 12:23 |
| marios | tosky: we *can* deal with it temporarily by adding the sahara lines in the converge environment file so thats what we do... rsult is if you do nothing you keep your sahara from mitaka | 12:23 |
| *** jprovazn has joined #tripleo | 12:23 | |
| marios | tosky: we need a docs section to talk about sahara and say that for any stack update operations after upgrading from mitaka to newton ... include bla | 12:24 |
| tbarron | gfidente: Jokke_ I'll make a bug in a moment | 12:24 |
| marios | tosky: but for the upgrade we can deal with it | 12:24 |
| gfidente | tbarron so we should really join NFS,CIFS from generic with CEPHFS from cephfsnative | 12:24 |
| gfidente | did I get it right? | 12:24 |
| *** ccamacho|lunch is now known as ccamacho | 12:24 | |
| jaosorior | shardy_lunch: so the per-network service node names patch now passes the HA gate. https://review.openstack.org/#/c/378764/ MySQL now uses only the shortnames as a workaround to the issue we saw. Need to figure out what's wrong with the nonha gate though | 12:24 |
| tosky | marios: that's expected; I was simply worried about not having the data wiped out (i.e. sane default behavior, even if by mistake the service is disabled) | 12:25 |
| tbarron | gfidente: generic and netapp just use the default so they aren't actually setting it, but that's essentially correct. | 12:25 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-tripleoclient: Downloads templates from swift before processing update https://review.openstack.org/381899 | 12:25 |
| tbarron | gfidente: they implicitly have 'NFS,CIFS' as value | 12:25 |
| gfidente | yeah but I think we can use both geneirc and cephfs at the same time | 12:26 |
| tbarron | gfidente: we can make that explicit, or just set 'NFS,CIFS,CEPHFS' when cephfs backend is enabled | 12:26 |
| EmilienM | ccamacho: https://review.openstack.org/#/c/382137/ is failing, maybe because a dep was not backported? | 12:26 |
| tbarron | gfidente: right, the really proper thing to do would be to join elements into a string depending on what is enabled | 12:26 |
| EmilienM | ccamacho: I'm confused with all patches that you did with Steve, and the order to land it | 12:26 |
| ccamacho | shardy I think there is a chicken egg issue with https://review.openstack.org/#/c/381593 and https://review.openstack.org/#/c/381587/ mmmm the tripleo-common patch will create the template for the basic roles returning an error... how can we fix this? | 12:27 |
| ccamacho | EmilienM let me check | 12:27 |
| tbarron | gfidente: but I think it would be acceptable for now to just set 'NFS,CIFS,CEPHFS' iff cephfs is enabled since the manila code does not care if one has too many protocols enabled | 12:27 |
| *** shardy_lunch is now known as shardy | 12:28 | |
| tbarron | gfidente: that will work for the cephfs only and cephfs plus other backend cases | 12:28 |
| * tosky bbl | 12:28 | |
| gfidente | tbarron sec I think we can set it selectively | 12:28 |
| tbarron | gfidente: cool | 12:28 |
| *** tosky has quit IRC | 12:28 | |
| *** matbu|lch is now known as matbu | 12:28 | |
| *** pgadiya has joined #tripleo | 12:28 | |
| sai | beagles: yo | 12:28 |
| beagles | sai: hey | 12:29 |
| sai | beagles: not sure if u got a chance to see the msgs rook and i sent yesterday | 12:29 |
| sai | its regarding reserved_host_memory being set to 2048 on computes | 12:29 |
| sai | and with DVR each router created taking around 50MB on the compute | 12:29 |
| sai | this could lead to problems... | 12:29 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Include ceilometer in swift proxy pipeline https://review.openstack.org/371950 | 12:29 |
| ccamacho | EmilienM 382137 does not have any dependency as this is the first patch that needs to be applied to move the template names. I rechecked the patch a few mins ago lets wait for the result | 12:30 |
| sai | is there a way we could intelligently change that value only on DVR based deployments | 12:30 |
| EmilienM | ccamacho: before recheck, have you seen why it fails? | 12:30 |
| beagles | sai: haven't seen them yet... | 12:30 |
| *** [1]cdearborn has quit IRC | 12:31 | |
| beagles | sai: possibly... looking | 12:31 |
| sai | ah ok it was yesterday | 12:31 |
| ccamacho | EmilienM yeahp it fails before, in the delorean build step | 12:31 |
| EmilienM | ccamacho: ack | 12:31 |
| ccamacho | EmilienM quick question, I have crossed deps between https://review.openstack.org/#/c/381587/ and https://review.openstack.org/#/c/381593/ the THT submission uses the code from tripleo-common and it works fine, but the tripleo-common submission if not used with the THT patch will override the roles templates and will fail.. | 12:33 |
| jaosorior | gfidente: hey dude, any idea what's wrong with ceph in this commit? https://review.openstack.org/#/c/378764/ | 12:34 |
| *** eggmaster has joined #tripleo | 12:34 | |
| gfidente | jaosorior sec | 12:34 |
| gfidente | working on two more things :( | 12:34 |
| *** akrivoka has quit IRC | 12:35 | |
| EmilienM | ccamacho: you mean the 2 patches depend each other? | 12:35 |
| beagles | sai: wait a sec.. did you say that each router adds 50 MB to the DVR node? | 12:36 |
| ccamacho | EmilienM yeahp | 12:36 |
| beagles | sai: as in neutron router, not agent instances? | 12:36 |
| EmilienM | ccamacho: that's a code design problem where you break backward compatibility I presume. | 12:36 |
| openstackgerrit | Adriano Petrich proposed openstack/tripleo-quickstart: WIP gate upgrade https://review.openstack.org/342161 | 12:36 |
| beagles | sai: so 5 routers = 250 MB? | 12:36 |
| ccamacho | I can make the tripleo-common submission backwards compatibility | 12:36 |
| ccamacho | as a workarround to land it | 12:36 |
| marios | pradk: hey am testing the ceilo migration... rough estimate how long it took for you (i think it is hanging, can't quickly see errors yet) bandini did you run it fully yet? https://review.openstack.org/#/c/360004/ I mean | 12:36 |
| sai | beagles: right | 12:36 |
| ccamacho | ack | 12:37 |
| beagles | (if so, holy crap... that seems wrong) | 12:37 |
| sai | beagles: https://bugzilla.redhat.com/show_bug.cgi?id=1381610 | 12:37 |
| openstack | bugzilla.redhat.com bug 1381610 in openstack-neutron "Metadata Proxy Memory Growth" [Medium,New] - Assigned to amuller | 12:37 |
| *** rlandy has joined #tripleo | 12:37 | |
| pradk | marios, hanging where? | 12:37 |
| sai | that BZ was with legacy case, but ive seen similar behaviour with DVR | 12:37 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Set enabled_share_protocols based on enabled backends https://review.openstack.org/382386 | 12:37 |
| sai | its basically the metadata proxy | 12:37 |
| pradk | marios, the ceilo step is pretty quick actually | 12:37 |
| EmilienM | ccamacho: or we can merge them together | 12:37 |
| gfidente | tbarron https://review.openstack.org/382386 ? | 12:37 |
| *** akrivoka has joined #tripleo | 12:37 | |
| tbarron | gfidente: Jokke_ https://bugs.launchpad.net/tripleo/+bug/1630564 | 12:37 |
| openstack | Launchpad bug 1630564 in tripleo "manila enabled_share_protocols wrong for cephfs" [Undecided,New] | 12:37 |
| EmilienM | ccamacho: we're really running out of time now | 12:37 |
| beagles | sai: I presume there is actually metadata involved ? | 12:37 |
| tbarron | gfidente: looking, thanks | 12:37 |
| ccamacho | EmilienM yeahp they can be merged together | 12:38 |
| EmilienM | ccamacho: ah, it even doesn't pass multinode job, which are voting | 12:38 |
| beagles | sai: ie, it isn't just arbitrary? | 12:38 |
| marios | pradk: yeah ok then i must have an issue/other bug... hanging i mean update_in_progress for close to 45 mins | 12:38 |
| EmilienM | ccamacho: they can't. | 12:38 |
| EmilienM | so yes, you have to make the code backward compatible | 12:38 |
| ccamacho | no worries Ill make the change | 12:38 |
| sai | beagles: nope, just launching routers..nothing else | 12:38 |
| EmilienM | otherwise it won't land | 12:38 |
| ccamacho | easy change | 12:38 |
| beagles | sai, holy crap | 12:38 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Set enabled_share_protocols based on enabled backends https://review.openstack.org/382386 | 12:38 |
| bandini | marios, pradk: nope haven't tried yet. Will run an upgrade cycle and get back to you guys (~1 hour or so) | 12:38 |
| sai | but with DVR the compute has to have a gguest attached to the router for the router to be created on the node | 12:38 |
| marios | pradk: weird thing is i don't see the /etc/httpd/conf.d/10-ceilometer_wsgi.conf either | 12:39 |
| beagles | sai, yup | 12:39 |
| marios | bandini: thanks | 12:39 |
| *** [1]cdearborn has joined #tripleo | 12:39 | |
| beagles | sai, perf testing ftw | 12:39 |
| *** mburned is now known as mburned_out | 12:39 | |
| sai | with dvr i launched 50 guest on node(each connected to a router) | 12:39 |
| sai | metadata proxy grew to 1GB beagles | 12:39 |
| honza | jtomasek: sorry, the more i think about it, the more i don't understand why we're hardcoding the "baremetal" flavor --- can you elaborate? | 12:40 |
| jtomasek | honza: baremetal is used for nodes which aren't tagged | 12:40 |
| pradk | marios, addressed your comments btw .. i left the bootstrap check as is. extra check shouldnt hurt i think | 12:41 |
| marios | pradk: ack thanks | 12:41 |
| *** jpena is now known as jpena|lunch | 12:41 | |
| jpich | EmilienM: Would it be possible to include https://review.openstack.org/#/c/382359/ for rc3 consideration? It's a one liner change with a huge usability impact on the UI. I do realise it's coming in late :/ | 12:42 |
| honza | jtomasek: so, after the untagging operation we need to update that role's count, and the baremetal count? | 12:43 |
| tbarron | gfidente: that logic looks right - I can redeploy against ceph and against netapp with that change now that you've taught me how the artifacts tricks | 12:43 |
| EmilienM | jpich: ok | 12:43 |
| sai | beagles: DVR with 50 routers onthe compute http://pix.toile-libre.org/upload/original/1475671409.png | 12:43 |
| jpich | EmilienM: Thank you, will update topic etc | 12:44 |
| tbarron | gfidente: and I'll close out that bug and re-open it against puppet-tripleo :) | 12:44 |
| beagles | sai: memory is getting released when the coimputes are deleted? | 12:44 |
| jtomasek | honza: yes, the role count parameter should match the number of nodes tagged to that role -> whis is working correctly afaik. And if we have nodes tagged to a role, then we need to update role flavor parameter to match its tag | 12:44 |
| sai | beagles: yup released when routers are deleted | 12:44 |
| *** Goneri has joined #tripleo | 12:45 | |
| sai | that test u saw beagles was rally launching routers | 12:45 |
| sai | so memor growth until it reaches the intended number of routers and then rally starts to delete so we see decline in memory | 12:45 |
| jtomasek | honza: this should also already work. Only thing missing is that when there is no node tagged to that role any more, then we need to set that role flavor back to baremetal | 12:45 |
| *** fultonj has joined #tripleo | 12:45 | |
| beagles | sai, well that's something at least (not leaking) | 12:45 |
| sai | beagles: the concern is with us reserving only 2048 MB on the computes for the host itself | 12:46 |
| sai | works ok with Legacy | 12:46 |
| sai | but game changes with DVR with routers on computes too | 12:46 |
| honza | jtomasek: ugh, this is so complicated, but i think i got it, thanks | 12:48 |
| *** ohamada has quit IRC | 12:48 | |
| *** ohamada has joined #tripleo | 12:48 | |
| beagles | sai, probably the best thing to do for the moment is modify the environment file to increase this amount. What's the amount you are suggesting? | 12:49 |
| *** radeks has joined #tripleo | 12:50 | |
| beagles | sai, we'll might be able to do something more intelligent in Ocata, but I'm a bit leery of trying to be too clever | 12:50 |
| openstackgerrit | Honza Pokorny proposed openstack/tripleo-common: Support node untagging https://review.openstack.org/372628 | 12:50 |
| beagles | "we'll might" .. jeez... | 12:51 |
| honza | jtomasek: ^^^ pushed | 12:51 |
| openstackgerrit | Merged openstack/puppet-tripleo: Change rabbitmq queues HA mode from ha-all to ha-exactly https://review.openstack.org/379586 | 12:51 |
| *** milan is now known as milan|bbl | 12:52 | |
| jtomasek | honza: thanks, I'll check it in a minute | 12:53 |
| *** mburned_out is now known as mburned | 12:53 | |
| sai | beagles: doing the math, around 50MB per router..boils down to how may routers we expect to support on a compute | 12:54 |
| beagles | sai: right ... so there really isn't a proper value, just an expected floor | 12:55 |
| sai | right | 12:56 |
| openstackgerrit | Merged openstack/diskimage-builder: Default to http://ftp.us.debian.org/debian for debian-minimal https://review.openstack.org/378038 | 12:56 |
| sai | so maybe a kbase article or something advising customers to change the env files if deploying dvr? | 12:56 |
| beagles | sai: yeah. I'll can patch the environment file so there is something in it for Ocata, but whether it is there or not for Newton, we are going to need good docs around it | 12:57 |
| sai | beagles: im going to file a BZ just so we keep track | 12:58 |
| beagles | sai: I expected *some* extra memory consumption, but I gotta say I'm surprised at the amount | 12:58 |
| beagles | sai: cool.. +1 | 12:58 |
| sai | beagles: ajo also has an upstream BZ regarding metadata proxy mem growth(its linked in the bz i sent u) | 12:58 |
| beagles | sai: ack | 12:58 |
| *** pradk has quit IRC | 12:59 | |
| *** pgadiya has quit IRC | 12:59 | |
| *** jcoufal_ has joined #tripleo | 13:03 | |
| *** jcoufal has quit IRC | 13:03 | |
| chem | matbu: to have the upgrade working (not using the ansible role), at convergence step (upstream) I had to remove num_engine_workers: 1 for heat from the deploy_env.yaml file. Do you do something similar in the role ? | 13:04 |
| *** morazi has joined #tripleo | 13:05 | |
| *** sudswas__ has joined #tripleo | 13:06 | |
| *** sudipto has joined #tripleo | 13:06 | |
| matbu | chem: afaik i don't use the deploy_env.yaml | 13:06 |
| chem | matbu: good call :) | 13:07 |
| matbu | chem: i only set the worker on the undercloud | 13:07 |
| chem | matbu: yep, I've seen the sed, but was wondering. | 13:07 |
| matbu | chem: i mean the heat worker only | 13:07 |
| matbu | yep | 13:07 |
| chem | matbu: thanks | 13:07 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Specify the Ceph packages to be installed https://review.openstack.org/382132 | 13:09 |
| *** zoliXXL is now known as zoli|relocating | 13:09 | |
| *** panda is now known as panda|afk | 13:10 | |
| *** tiswanso has joined #tripleo | 13:10 | |
| *** morazi has quit IRC | 13:11 | |
| *** lblanchard has joined #tripleo | 13:12 | |
| *** zoli|relocating is now known as zoli_gone-proxy | 13:14 | |
| *** sudipto has quit IRC | 13:16 | |
| *** sudswas__ has quit IRC | 13:16 | |
| *** ccamacho has quit IRC | 13:16 | |
| *** dprince has joined #tripleo | 13:16 | |
| *** amoralej|lunch is now known as amoralej | 13:21 | |
| *** mbozhenko has quit IRC | 13:22 | |
| openstackgerrit | Merged openstack/puppet-tripleo: Use service-specific servernames for haproxy https://review.openstack.org/382177 | 13:22 |
| openstackgerrit | Merged openstack/puppet-tripleo: Clean out UI httpd configuration file https://review.openstack.org/381568 | 13:22 |
| openstackgerrit | Miles Gould proposed openstack/python-tripleoclient: [WIP] Add DRAC RAID configuration to baremetal ready state command https://review.openstack.org/299221 | 13:23 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable usage of "short names" for Ceph cluster https://review.openstack.org/382408 | 13:23 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 13:23 |
| *** rhallisey has joined #tripleo | 13:23 | |
| jtomasek | honza: works great! | 13:24 |
| rook | sai beagles: https://bugs.launchpad.net/tripleo/+bug/1630583 | 13:24 |
| openstack | Launchpad bug 1630583 in tripleo "RFE : If DVR is enabled increase reserved_host_memory_mb" [Undecided,New] | 13:24 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Integrate node tagging workflow https://review.openstack.org/367562 | 13:25 |
| *** morazi has joined #tripleo | 13:25 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Select per-network hostnames for service_node_names https://review.openstack.org/378764 | 13:25 |
| sai | ok downstream rook https://bugzilla.redhat.com/show_bug.cgi?id=1381994 | 13:26 |
| openstack | bugzilla.redhat.com bug 1381994 in openstack-tripleo "RFE: Increase reserved_host_memory on nova computes when deploying with DVR" [Unspecified,New] - Assigned to jslagle | 13:26 |
| * honza is relieved | 13:27 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for gnocchi https://review.openstack.org/377645 | 13:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for ceilometer https://review.openstack.org/377644 | 13:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for aodh https://review.openstack.org/365457 | 13:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for keystone https://review.openstack.org/327029 | 13:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable internal TLS for MySQL https://review.openstack.org/378471 | 13:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 13:27 |
| tbarron | gfidente: syntax issue when trying to deploy with https://review.openstack.org/#/c/382386 | 13:30 |
| chem | shardy: I've got the impression that hieradatas persit somewhere. I try to remove heat::config::heat_config: ... num_engine_worker from a template and it keeps on being recreated in hieradata/controller_extraconfig.yaml. Is that expected, and if so how can I delete it ? | 13:31 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Adds Environment File for Removing Sahara during M/N upgrade https://review.openstack.org/375517 | 13:33 |
| shardy | chem: so you've removed it from the ControllerExtraConfig parameter, then run an update? | 13:33 |
| *** nyechiel has joined #tripleo | 13:33 | |
| marios | jistr: tosky: chem please revote when you can i updated for chem suggestion thanks https://review.openstack.org/#/c/375517/ | 13:33 |
| *** akshai has joined #tripleo | 13:34 | |
| *** ccamacho has joined #tripleo | 13:34 | |
| chem | shardy: yep | 13:34 |
| shardy | chem: the data does persist inside heat, but if you pass a new ControllerExtraConfig parameter, it should override the old one (and that parameter is mapped into the controller_extraconfig.yaml file) | 13:34 |
| EmilienM | shardy: what patch do we need backported into stable/newton to have https://review.openstack.org/#/c/382137/ passing? | 13:35 |
| chem | shardy: that's the problem, I didn't reuse the deploy_env where it is defined, so it was not overwritten | 13:35 |
| *** jaosorior has quit IRC | 13:36 | |
| larsks | I'm hitting: Error rendering template puppet/config.role.j2.yaml : 'role' is undefined | 13:36 |
| chem | shardy: can I just delete it from heat with some command then ? or should I re-use the template for it to being overwritten ? | 13:36 |
| larsks | I'm in the middle of trying to recover from a bad rebase, so it's possible I broke something. | 13:36 |
| larsks | Has anyone else seen that? | 13:36 |
| shardy | chem: ah, so you'll need to define a new environment with a new definition of ControllerExtraConfig not containing the stuff you don't want | 13:36 |
| *** jaosorior has joined #tripleo | 13:37 | |
| shardy | chem: we use PATCH updates to heat, so parameters previously passed are sticky | 13:37 |
| *** abregman has joined #tripleo | 13:37 | |
| chem | shardy: oki, thanks :) | 13:37 |
| shardy | chem: yes, there are ways to clear a parameter via heatclient too, but I've not tested this particular case | 13:37 |
| chem | shardy: don't bother, will overwrite with a template | 13:37 |
| shardy | chem: "heat stack-update -x -c ControllerExtraConfig overcloud" might work | 13:38 |
| chem | shardy: thanks, going to try then | 13:38 |
| shardy | EmilienM: looking, I don't think that has dependencies as it's just renaming some files | 13:38 |
| *** jeckersb_gone is now known as jeckersb | 13:39 | |
| *** bnemec has joined #tripleo | 13:40 | |
| gfidente | jaosorior commenting on the submission | 13:40 |
| gfidente | not trivial to fix it seems :( | 13:40 |
| openstackgerrit | John Trowbridge proposed openstack/python-tripleoclient: Open python-tripleoclient for Ocata https://review.openstack.org/382418 | 13:41 |
| *** panda|afk is now known as panda | 13:41 | |
| jaosorior | gfidente: right, so what I did was fall back to ceph to keep using the short-names for the nodes (same as I did for mysql) | 13:41 |
| jaosorior | and we'll figure it out in Ocata X_x | 13:42 |
| trown | EmilienM: do we have a handy list of repos for tripleo so that I can do https://review.openstack.org/#/c/382418/ for all of them? without that we have newer release tag on stable/newton than master | 13:43 |
| panda | bnemec: are SSL deployments working with network isolation ? | 13:43 |
| jaosorior | panda: they should. What's up? | 13:43 |
| gfidente | jaosorior right cause short does not include .network. subdomain right? | 13:44 |
| bnemec | I haven't done one in a long time, so I wouldn't know. | 13:44 |
| jaosorior | panda: we're not doing SSL in the internal network (yet) and that's pretty much what I've been working on. | 13:44 |
| gfidente | jaosorior also looking at ceph.conf I am wondering why both public_network and cluster_network are set to the same subnet | 13:44 |
| gfidente | jaosorior which seems wrong | 13:44 |
| jaosorior | gfidente: I have no idea | 13:44 |
| EmilienM | trown: what is this patch doing? | 13:44 |
| shardy | EmilienM: https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L109 | 13:44 |
| larsks | shardy, deploying with current tht master, I am getting "Exception updating plan: Error rendering template puppet/config.role.j2.yaml : 'role' is undefined". | 13:44 |
| shardy | the log says it failed there, but I'm not sure why | 13:45 |
| shardy | the image build seemed to work | 13:45 |
| larsks | (that's with unmodified master) | 13:45 |
| shardy | looking for more clues | 13:45 |
| trown | EmilienM: it makes `python setup.py --version` return 6.x.x instead of 5.0.0 on master | 13:45 |
| jaosorior | gfidente: well, ceph::mon uses $::hostname as the default, which will end up as the short name, that has no network. | 13:45 |
| jaosorior | gfidente: that's in puppet-ceph | 13:45 |
| trown | EmilienM: we tagged 5.2.0 on stable/newton, so packages built on master will have lower version than newton right now | 13:45 |
| shardy | larsks: have you also pulled the latest tripleo-common and reloaded the mistral actions? | 13:45 |
| shardy | that error sounds like you're running old tripleo-common | 13:45 |
| larsks | shardy, I have not. Trying now. | 13:46 |
| trown | EmilienM: it is a pbr trick to not have to create a "fake" release on master to have the correct version there | 13:46 |
| EmilienM | trown: ok | 13:46 |
| EmilienM | trown: no I don't have a trick | 13:46 |
| trown | EmilienM: k, I will look at release repo, thanks | 13:46 |
| *** social has quit IRC | 13:46 | |
| EmilienM | shardy: recheck? | 13:47 |
| panda | jaosorior: I'm trying to move things around to add IPv6 coverage without adding more gate jobs. A first suggestion was that all HA will go with IPv6, and IPv4 net-iso will be tested on non-ha, which are doing SSL too | 13:47 |
| *** ayoung has quit IRC | 13:47 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Set enabled_share_protocols based on enabled backends https://review.openstack.org/382386 | 13:48 |
| *** social has joined #tripleo | 13:48 | |
| jaosorior | panda: right now there is no reason for SSL not to work with net-iso. Since right now we're only doing so for the public endpoints in HAProxy. | 13:48 |
| panda | jaosorior: I remember trying to test net-iso with the current SSL configuration in SSL and failing. I'll test it again now. | 13:49 |
| shardy | EmilienM: Yeah, I think so | 13:50 |
| shardy | we're not logging all the tripleo.sh output, so my best guess is the image build worked, but the upload to glance didn't | 13:50 |
| shardy | but there's no errors in the glance log either | 13:50 |
| openstackgerrit | John Trowbridge proposed openstack/instack-undercloud: Open instack-undercloud for Ocata https://review.openstack.org/382422 | 13:51 |
| jaosorior | panda: that's what my deployment is using right now. | 13:51 |
| *** jpena|lunch is now known as jpena | 13:54 | |
| openstackgerrit | John Trowbridge proposed openstack/os-collect-config: Open os-collect-config for Ocata https://review.openstack.org/382424 | 13:54 |
| *** mbozhenko has joined #tripleo | 13:54 | |
| *** yamahata has quit IRC | 13:55 | |
| *** yamahata has joined #tripleo | 13:55 | |
| bandini | marios: tested the ceilometer stuff, there is one more issue, which I have noted in the review | 13:55 |
| panda | jaosorior: what are you using to deploy ? | 13:56 |
| bnemec | shardy: Image build output goes to a separate file, and any errors should be in there. It's called something like image-build.txt in the undercloud's /var/log directory. | 13:56 |
| shardy | bnemec: Yeah, I looked in there and AFAICS the image build worked | 13:57 |
| shardy | http://logs.openstack.org/37/382137/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/2decf99/console.html#_2016-10-05_12_54_14_786748 | 13:58 |
| shardy | https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L109 | 13:58 |
| openstackgerrit | John Trowbridge proposed openstack/os-net-config: Open os-net-config for Ocata https://review.openstack.org/382429 | 13:58 |
| jaosorior | panda: so I'm using the network isolation default files that come with oooq, then you need to specify a CA (that also the undercloud trusts) and a certificate as pointed out here http://tripleo.org/advanced_deployment/ssl.html#overcloud-ssl | 13:58 |
| shardy | http://logs.openstack.org/37/382137/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/2decf99/logs/undercloud/var/log/image_build.txt.gz#_2016-10-05_12_54_11_000 | 13:58 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: restore missing fluentd client functionality https://review.openstack.org/382430 | 13:59 |
| jaosorior | so I'm passing -e ~/ssl-heat-templates/environments/enable-tls.yaml -e ~/ssl-heat-templates/environments/tls-endpoints-public-dns.yaml -e ~/cloudname.yaml -e ~/ssl-heat-templates/environments/inject-trust-anchor.yaml | 13:59 |
| marios | thanks bandini | 14:00 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-image-elements: Open tripleo-image-elements for Ocata https://review.openstack.org/382431 | 14:00 |
| bandini | marios: I will test the obvious fix later on | 14:00 |
| panda | jaosorior: fom where are you taking ~/ssl-heat-templates/ ? | 14:01 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-common: Open tripleo-common for Ocata https://review.openstack.org/382432 | 14:02 |
| jaosorior | panda: I have copies the t-h-t templates to my home directory. Either way, you need to modify the enable-tls.yaml and inject-trust-anchor.yaml environments, since you need to fill in the cert, key and CA cert. | 14:02 |
| jaosorior | panda: also, you might want to use ~/ssl-heat-templates/environments/tls-endpoints-public-ip.yaml since that way you don't need to set the cloudname | 14:03 |
| panda | jaosorior: tripleo-ci is using always the same file it seems, I don't see them being generated each run | 14:03 |
| bnemec | shardy: Oh, I wonder if this is because of the glance v2 change: http://logs.openstack.org/37/382137/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/2decf99/console.html#_2016-10-05_12_54_14_708367 | 14:04 |
| jaosorior | panda: it is indeed the same file. We don't distribute that to people | 14:04 |
| jaosorior | works for testing | 14:04 |
| *** pradk has joined #tripleo | 14:05 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-heat-templates: Open tripleo-heat-templates for Ocata https://review.openstack.org/382439 | 14:06 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-puppet-elements: Open tripleo-puppet-elements for Ocata https://review.openstack.org/382441 | 14:07 |
| *** absubram has quit IRC | 14:08 | |
| *** dprince has quit IRC | 14:10 | |
| trown | EmilienM: ok I think I got everything in https://review.openstack.org/#/q/topic:open-ocata ... note this effects upgrade testing because the version of packages built on master is lower than the version on stable/newton and our yum plugin priorities trick does not work for upgrades | 14:11 |
| ccamacho | EmilienM shardy both https://review.openstack.org/#/c/381593/ and https://review.openstack.org/#/c/382137/ are broken because of the same reason http://logs.openstack.org/37/382137/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/f2f718a/logs/undercloud/var/log/mistral/executor.txt.gz | 14:11 |
| ccamacho | overcloud not found | 14:11 |
| *** oshvartz has quit IRC | 14:11 | |
| d0ugal | ccamacho: overcloud not found? | 14:12 |
| ccamacho | dOugal yeahp | 14:12 |
| ccamacho | but not sure why | 14:13 |
| d0ugal | ccamacho: I don't think they are the errors that you are looking for :) | 14:13 |
| EmilienM | trown: +2 | 14:14 |
| jtomasek | EmilienM: shardy: do you think we could get a chance of getting this in for rc3 https://bugs.launchpad.net/tripleo/+bug/1625264 ? The fix is up for review here https://review.openstack.org/#/c/372628/7 | 14:14 |
| openstack | Launchpad bug 1625264 in tripleo "Node tagging workflow doesn't support untagging" [High,In progress] - Assigned to Honza Pokorny (hpokorny) | 14:14 |
| jaosorior | shardy: well, I gotta go. I submitted the patch with some dependencies to two workarounds. one for mysql and one for ceph. | 14:14 |
| d0ugal | ccamacho: The mistral workflow checks to see if the container and mistral env exist before creating them. That is all that is. | 14:14 |
| ccamacho | d0ugal thanks then :P | 14:14 |
| jtomasek | EmilienM: it brings a critical feature for the GUI | 14:14 |
| pradk | marios, bandini, are you running the ceilo upgrade before or after init | 14:14 |
| *** links has quit IRC | 14:14 | |
| d0ugal | ccamacho: https://github.com/openstack/tripleo-common/blob/master/workbooks/plan_management.yaml#L129-L143 | 14:14 |
| pradk | marios, as i mentioned depending on what you ahve the hiera attribute is different | 14:14 |
| jtomasek | EmilienM: I'd rather say fixes critical bug in GUI.... | 14:15 |
| d0ugal | ccamacho: but they do look scary in the log, we should improve that. | 14:15 |
| EmilienM | jtomasek: ok | 14:15 |
| jtomasek | EmilienM: thank you! | 14:15 |
| EmilienM | jtomasek: when it pass CI, I'll review it | 14:15 |
| *** jaosorior has quit IRC | 14:16 | |
| ccamacho | d0ugal: Pretty red for me :) | 14:16 |
| matbu | chem: di you figure out your issue with the swap.J2.yaml ? | 14:16 |
| bandini | pradk: after init | 14:16 |
| bandini | pradk: has anything changed in that regard? i.e. should it be moved before init? | 14:17 |
| matbu | chem: if not you can use the following path : /tmp/tripleo-heat-templates/extraconfig/all_nodes/swap.yaml | 14:18 |
| bandini | pradk: am happy to switch it wherever, we just need to be 100% clear where it goes. Want me to try before? | 14:18 |
| pradk | bandini, i have been doing it before as it dint really make much difference either way .. so thought that would keep newton variable out of the way. I'll let you guys correct me, as yuou know more about the upgrade workflow and right place to run these. | 14:19 |
| dmsimard | slagle, EmilienM: python-openstackclient-3.2.0-2 just made it to buildlogs (newton testing repositories) - it's 3.2.0 with RDO carrying the patch from https://review.openstack.org/#/c/372712/ | 14:19 |
| bandini | pradk: my hunch would be "wherever it makes more sense and is more solid", but I will let marios comment on that | 14:20 |
| EmilienM | dmsimard: thanks a ton for that | 14:20 |
| pradk | marios, wdyt ? run the pre upgrade before or after init .. based on that we either have mongo_node_ips or mongodb_node_ips | 14:21 |
| marios | pradk: before init - do you have time to join scrum in 10 mins? | 14:21 |
| shardy | bnemec: aha, are we missing a newton backport wrt glance v2? | 14:21 |
| marios | pradk: at least that was the understanding we reached last time we discussed it | 14:21 |
| pradk | marios, sure .. hmm i might have a conflict, lemme check | 14:21 |
| marios | pradk: k | 14:21 |
| pradk | marios, correct | 14:21 |
| pradk | bandini, ^^ | 14:21 |
| bandini | ack | 14:22 |
| chem | matbu: no, this is strange, the relative path work on upstream, but not on downstream, both with your patch applied and the patch mentionned by shardy applied. Something else must be missing downstream. | 14:25 |
| chem | matbu: going to try with hardcoded tmp path | 14:25 |
| matbu | chem: k | 14:27 |
| *** cschwede has joined #tripleo | 14:28 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 14:29 |
| *** radeks has quit IRC | 14:29 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Fixing resources path in OpenDaylight https://review.openstack.org/376007 | 14:30 |
| *** rodrigods has quit IRC | 14:32 | |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Create new docker command hook. https://review.openstack.org/312723 | 14:32 |
| *** rodrigods has joined #tripleo | 14:32 | |
| *** akshai has quit IRC | 14:33 | |
| *** radeks has joined #tripleo | 14:37 | |
| chem | matbu: No such file or directory: '/tmp/tmpEwW3R_/tripleo-heat-templates//user-files/2b3fb492414d856054b43750b731ffad-tht-swap.yaml | 14:38 |
| *** mbozhenko has quit IRC | 14:38 | |
| *** nyechiel has quit IRC | 14:38 | |
| *** tiswanso has quit IRC | 14:38 | |
| *** fultonj has quit IRC | 14:38 | |
| *** eggmaster has quit IRC | 14:38 | |
| *** maticue has quit IRC | 14:38 | |
| *** hogepodge has quit IRC | 14:38 | |
| *** mcornea has quit IRC | 14:38 | |
| *** b00tcat has quit IRC | 14:38 | |
| *** tesseract- has quit IRC | 14:38 | |
| *** fxpester has quit IRC | 14:38 | |
| *** milan|bbl has quit IRC | 14:38 | |
| *** jpena has quit IRC | 14:38 | |
| *** akrzos has quit IRC | 14:38 | |
| *** ebalduf has quit IRC | 14:38 | |
| *** Ng has quit IRC | 14:38 | |
| *** tobias-fiberdata has quit IRC | 14:38 | |
| *** pmannidi has quit IRC | 14:38 | |
| *** davidlenwell has quit IRC | 14:38 | |
| *** fultonj has joined #tripleo | 14:39 | |
| *** mbozhenko has joined #tripleo | 14:39 | |
| *** pmannidi has joined #tripleo | 14:39 | |
| *** tobias-fiberdata has joined #tripleo | 14:39 | |
| *** mcornea has joined #tripleo | 14:39 | |
| *** nyechiel has joined #tripleo | 14:39 | |
| *** tesseract- has joined #tripleo | 14:39 | |
| *** maticue has joined #tripleo | 14:39 | |
| *** fxpester has joined #tripleo | 14:39 | |
| *** milan|bbl has joined #tripleo | 14:39 | |
| *** b00tcat has joined #tripleo | 14:39 | |
| *** b00tcat has quit IRC | 14:39 | |
| *** b00tcat has joined #tripleo | 14:39 | |
| *** tiswanso has joined #tripleo | 14:39 | |
| *** hogepodge has joined #tripleo | 14:39 | |
| *** bnemec has quit IRC | 14:39 | |
| *** akrzos has joined #tripleo | 14:39 | |
| *** jpena has joined #tripleo | 14:39 | |
| *** davidlenwell has joined #tripleo | 14:40 | |
| *** ChanServ sets mode: +v davidlenwell | 14:40 | |
| trozet | shardy: hi | 14:41 |
| *** eggmaster has joined #tripleo | 14:41 | |
| *** akshai has joined #tripleo | 14:41 | |
| *** akshai has quit IRC | 14:42 | |
| *** bnemec has joined #tripleo | 14:44 | |
| *** akshai has joined #tripleo | 14:44 | |
| matbu | chem: i'm trying to reproduce | 14:44 |
| *** saneax is now known as saneax-_-|AFK | 14:44 | |
| matbu | chem: the tht-swap.yaml is a file that you include ? | 14:44 |
| *** penick has quit IRC | 14:44 | |
| shardy | trozet: hi | 14:45 |
| trozet | shardy: so I got a hold of the setup related to the opendaylight_api_node_ips missing | 14:45 |
| chem | matbu: hum you can use /usr/share/openstack-tripleo-heat-templates/extraconfig/all_nodes/swap.yaml | 14:45 |
| trozet | shardy: i see opendaylight_api listed as an enabled service.I also see the node names set for odl, so my suspicion is something is wrong with the ServiceNetMap | 14:46 |
| trozet | shardy: https://paste.fedoraproject.org/444179/ | 14:46 |
| *** Ng has joined #tripleo | 14:46 | |
| *** ChanServ sets mode: +v Ng | 14:46 | |
| matbu | chem: can you paste your deploy command ?, cause i have the swap.yaml into swift | 14:47 |
| shardy | trozet: have you added OpendaylightApi to the ServiceNetMap as an input parameter? | 14:47 |
| chem | matbu: I think upstream is working, I'm running it (with your patch) | 14:47 |
| matbu | chem: ha yep your are downstream | 14:47 |
| matbu | i'll try to dpeloy dwonstream | 14:48 |
| chem | matbu: http://paste.openstack.org/show/584476/ | 14:48 |
| shardy | trozet: as mentioned in https://bugs.launchpad.net/tripleo/+bug/1629408 I think it's missing in the ServiceNetMapDefaults? | 14:48 |
| openstack | Launchpad bug 1629408 in tripleo "OpenDaylight missing Controller IP list - deploys always fail" [High,In progress] - Assigned to Tim Rozet (trozet) | 14:48 |
| shardy | so we either need to add it, or you need to pass it via ServiceNetMap in an environment file | 14:48 |
| chem | matbu: downstream with your patch doesn't work (and shardy workaround with patch doesn't work) | 14:49 |
| matbu | chem: k good to know, there is probably something missing | 14:49 |
| *** yamahata has quit IRC | 14:49 | |
| trozet | shardy: I see it in there https://paste.fedoraproject.org/444184/14756789/ | 14:49 |
| *** yamahata has joined #tripleo | 14:49 | |
| shardy | trozet: Aha! | 14:51 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/opendaylight-api.yaml#L66 | 14:51 |
| chem | matbu: hum ... I can do a swift list overcloud to have the template | 14:51 |
| shardy | the service_name is opendaylight_api | 14:51 |
| chem | matbu nice | 14:51 |
| shardy | but the ServiceNetMap entry is OpenDaylightApi | 14:51 |
| shardy | should be OpendaylightApiNetwork | 14:51 |
| shardy | or the camel to snake conversion won't work | 14:51 |
| tbarron | gfidente: when you get a chance, pls. look at result reported in https://review.openstack.org/#/c/382386/3 | 14:51 |
| shardy | you could also use opendaylight_api_network | 14:51 |
| shardy | which is what we would prefer to use, except backwards compatibility prevents switching right now | 14:52 |
| trozet | shardy: camels and snakes :) ? | 14:53 |
| trozet | shardy: you linked line 66, the service_name? | 14:53 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Fix OpendaylightApiNetwork key naming https://review.openstack.org/382480 | 14:54 |
| shardy | trozet: try that ^^ :) | 14:54 |
| shardy | trozet: we convert OpenDaylightApiNetwork to open_daylight_api_network | 14:54 |
| trozet | shardy: oh now i see... | 14:54 |
| *** zoli_gone-proxy is now known as zoliXXL | 14:54 | |
| trozet | LOL | 14:54 |
| shardy | which doesn't match the service name | 14:54 |
| shardy | it'd be good to add some validation to this during ocata | 14:55 |
| trozet | shardy: man what a nuance | 14:55 |
| *** Lokesh_Jain has joined #tripleo | 14:56 | |
| shardy | trozet: ideally we'd have everything in ServiceNetMap exactly match the service_names | 14:56 |
| trozet | shardy: so its all converted to lower and split with _ on camel case | 14:56 |
| shardy | but it's going to take a while to get to that | 14:56 |
| shardy | trozet: yeah | 14:56 |
| *** tosky has joined #tripleo | 14:56 | |
| shardy | maybe we should add a comment explaining that, or add it to the decription for the ServiceNetMap parameter | 14:57 |
| trozet | shardy: i blame ODL for joing 2 words into 1 for their project name | 14:57 |
| trozet | joining | 14:57 |
| trozet | shardy: yeah it should be in the description, or some dev ref guide | 14:57 |
| trozet | shardy: let me try that | 14:57 |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-common: Remove references to overcloud-without-mergepy https://review.openstack.org/382486 | 14:59 |
| *** ebalduf has joined #tripleo | 14:59 | |
| *** jistr is now known as jistr|call | 15:01 | |
| *** r-mibu has quit IRC | 15:04 | |
| *** r-mibu has joined #tripleo | 15:04 | |
| *** mbozhenko has quit IRC | 15:05 | |
| thrash | d0ugal: ^^^ | 15:06 |
| yolanda | hi, i started deploying tripleo on master, and i'm getting this error: os-refresh-config: error: unrecognized arguments: --timeout 14400 | 15:06 |
| *** ebalduf has quit IRC | 15:06 | |
| yolanda | i have os-refresh-config 0.1.11, that don't seem to support --timeout flag, is that a known issue? | 15:06 |
| social | I'm the only one hitting MissingSchema: Invalid URL 'x/v1/images/detail?limit=20': No schema supplied. Perhaps you meant http://x/v1/images/detail?limit=20? on openstack overcloud image upload --update-existing | 15:06 |
| *** radeks has quit IRC | 15:06 | |
| *** zoliXXL is now known as zoli_gone-proxy | 15:08 | |
| *** zoli_gone-proxy is now known as zoliXXL | 15:10 | |
| shardy | social: No, it seems to be happening in stable/newton CI, I'm not yet sure what the fix is | 15:10 |
| larsks | So...after updating to tht master, python-tripleoclient master, and tripleo-common master, running a deployment seems to be failing a mistral action with: Failed to find action [action_name=tripleo.process_templates]. But there exists in 'mistral actions-list' an action "tripleo.templates.process"... | 15:10 |
| larsks | (also I ran mistral-db-manage populate) | 15:11 |
| *** rcernin has quit IRC | 15:12 | |
| shardy | larsks: gah, yeah, I hit that issue recently too - the only workaround I found was to drop the mistral DB, recreate an empty one, then re-run mistral-db-manage populate | 15:12 |
| shardy | basically we renamed some actions, and mistral seems to have no way to purge old actions :( | 15:12 |
| shardy | it should only be an issue for devs upgrading from pre-release code tho | 15:13 |
| *** pcaruana has quit IRC | 15:13 | |
| shardy | d0ugal: ^^ unless you can offer any less drastic workaround? | 15:13 |
| larsks | Man, developing on master is harsh :). Thanks, I'll try that out... | 15:13 |
| *** jistr|call is now known as jistr | 15:13 | |
| *** dmacpher has joined #tripleo | 15:16 | |
| d0ugal | shardy: Sorry, I don't have any good ideas. | 15:16 |
| d0ugal | shardy: Isn't the problem that some code is looking for the old names? | 15:16 |
| d0ugal | shardy: because there shouldn't be an issue with dupe names, so I'm a bit unsure about it | 15:17 |
| shardy | d0ugal: some code that's ended up inside the mistral DB I think | 15:19 |
| shardy | as I wasn't able to get past it without dropping | 15:19 |
| shardy | d0ugal: the main issue seems to be that the populate command injects actions in a manner that means they can never be removed | 15:19 |
| shardy | and the action-delete CLI doesn't work for them, even as admin | 15:19 |
| shardy | I expect there's a flag in the db which could be toggled to get past that | 15:20 |
| shardy | but I took the drop-everything approach instead ;) | 15:20 |
| larsks | shardy, after dropping the mistral db, re-creating, re-running populate, I get: Workflow not found [workflow_identifier=tripleo.plan_management.v1.create_deployment_plan] | 15:20 |
| shardy | http://paste.openstack.org/show/584488/ | 15:21 |
| shardy | larsks: you'll need to reload them from /usr/share/openstack-tripleo-common/workbooks | 15:21 |
| shardy | as in that script | 15:21 |
| shardy | openstack undercloud install does this for you, but you can do it manually | 15:21 |
| larsks | I see. Well, now I know.... | 15:22 |
| shardy | (the sleeps in there are probably not all needed) | 15:22 |
| d0ugal | shardy: oh, yeah, I need to look into this. | 15:22 |
| shardy | larsks: there is some more detail in https://github.com/openstack/tripleo-common/blob/master/doc/source/tripleo-mistral.rst#using-workflows | 15:23 |
| shardy | but hopefully the above is enough to get things working for you | 15:24 |
| *** tesseract- has quit IRC | 15:24 | |
| trozet | shardy: i see the variable now, deploy still failed so debugging that now | 15:25 |
| *** weshay has quit IRC | 15:25 | |
| larsks | shardy, that seems to have gotten things moving. Thanks again... | 15:25 |
| shardy | trozet: ack, Ok sounds like progress :) | 15:26 |
| shardy | larsks: great, np! | 15:26 |
| trozet | shardy: thanks for the help | 15:26 |
| *** cylopez has quit IRC | 15:28 | |
| shardy | dtantsur: Hey can you pls check out https://bugs.launchpad.net/tripleo/+bug/1630203 when you get a moment | 15:30 |
| openstack | Launchpad bug 1630203 in tripleo "Tripleo needs to support full overcloud disk images" [Undecided,New] | 15:30 |
| shardy | I'd like to ensure that makes sense wrt TripleO ironic integration before assigning to the ocata backlog | 15:30 |
| dtantsur | shardy, in a meeting, gimme a few minutes | 15:30 |
| shardy | dtantsur: ack, thanks, not urgent | 15:30 |
| dtantsur | the title makes sense | 15:30 |
| dtantsur | :) | 15:30 |
| *** weshay has joined #tripleo | 15:30 | |
| *** lucasagomes is now known as lucas-hungry | 15:35 | |
| *** eglynn has joined #tripleo | 15:36 | |
| yolanda | hi shardy , dtantsur , so i am testing this Tripleo full overcoud support. So far i got progress with it, as soon as i bypassed the tripleo limitation | 15:36 |
| yolanda | so the blockers i had, is that tripleo was asking for initrd and vmlinuz, and adding this as properties for overcloud-full.qcow2 | 15:36 |
| *** numans has quit IRC | 15:36 | |
| yolanda | as soon as i provided an image without these properties, ironic picked it and deployed properly | 15:36 |
| dtantsur | yolanda, the same qcow2 image? this is strange, to be honest | 15:37 |
| *** dsariel_ has quit IRC | 15:37 | |
| *** jaosorior has joined #tripleo | 15:37 | |
| yolanda | dtantsur, not the same qcow2 image, but a full disk image | 15:37 |
| yolanda | so i generated a full disk image using vm element instead of baremetal | 15:37 |
| dtantsur | yolanda, correct, it's not clear from your bug | 15:38 |
| dtantsur | could you please update it to explicitly say that we need a new way of building images for that? | 15:38 |
| yolanda | dtantsur, i need to step out for an hour, do you want that we talk later so i clarify you? | 15:38 |
| *** sudipto has joined #tripleo | 15:38 | |
| *** sudswas__ has joined #tripleo | 15:38 | |
| *** weshay has quit IRC | 15:38 | |
| dtantsur | yolanda, I understand that, but random people reading the bug probably won't :) so please update the bug when you're back | 15:38 |
| yolanda | sure i'll do | 15:39 |
| jaosorior | shardy: so this is now green :D https://review.openstack.org/#/c/378764/ | 15:39 |
| dtantsur | shardy, modulo building a suitable image, this ^^ makes sense to me | 15:39 |
| jaosorior | depends on two workarounds (one for mysql and one for ceph) though | 15:39 |
| *** dprince has joined #tripleo | 15:39 | |
| shardy | dtantsur, yolanda: ack thanks - and yes please if someone can enumerate the work items on the bug that would be great | 15:40 |
| shardy | jaosorior: ack, nice, will check it out :) | 15:40 |
| dtantsur | shardy, I'll defer it to yolanda, as I've personally never tried whole disk images | 15:41 |
| *** tremble has quit IRC | 15:41 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for gnocchi https://review.openstack.org/377645 | 15:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for ceilometer https://review.openstack.org/377644 | 15:41 |
| shardy | dtantsur: ack, yeah me neither hence asking for a second opinion :) | 15:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for aodh https://review.openstack.org/365457 | 15:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable TLS in the internal network for keystone https://review.openstack.org/327029 | 15:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Enable internal TLS for MySQL https://review.openstack.org/378471 | 15:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 15:41 |
| openstackgerrit | Martin André proposed openstack/tripleo-heat-templates: Have docker start script honor configuration https://review.openstack.org/366138 | 15:41 |
| openstackgerrit | Martin André proposed openstack/tripleo-heat-templates: Add steps to containerized compute deployment https://review.openstack.org/346927 | 15:41 |
| openstackgerrit | Martin André proposed openstack/tripleo-heat-templates: WIP: Containerized Services for Composable Roles https://review.openstack.org/330659 | 15:41 |
| openstackgerrit | Martin André proposed openstack/tripleo-heat-templates: Bind mount files to run DiD in latest atomic host https://review.openstack.org/347218 | 15:41 |
| openstackgerrit | Martin André proposed openstack/tripleo-heat-templates: Be more inclusive in insecure registry regex https://review.openstack.org/366111 | 15:41 |
| Slower | if anyone wants to help us shorten that patch stack.. feel free to review | 15:44 |
| Slower | a lot of the depends-on patches are quite small | 15:44 |
| jaosorior | ccamacho: are you around? | 15:44 |
| ccamacho | yeahp | 15:44 |
| *** weshay has joined #tripleo | 15:44 | |
| ccamacho | jaosorior ^ | 15:45 |
| Slower | dprince: shardy ^ | 15:45 |
| dprince | Slower: ack | 15:47 |
| shardy | Slower: ack, as soon as we get Newton tagged I'll be happy to help land them | 15:47 |
| Slower | shardy: that's happening soon? | 15:48 |
| shardy | Slower: Yes, by the end of this week | 15:48 |
| Slower | cool | 15:48 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 15:48 |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci: DO NOT MERGE testing switch to Ocata DLRN instance https://review.openstack.org/382521 | 15:49 |
| *** milan|bbl is now known as milan | 15:49 | |
| jaosorior | ccamacho: hey dude, so I'm finishing up the blog post for testing TLS everywhere (or at least the stuff I have until now): http://jaormx.github.io/2016/testing-out-the-tls-everywhere-patches-for-tripleo/ let me know what you think | 15:50 |
| trown | EmilienM: ^ test we talked about in RDO meeting | 15:50 |
| EmilienM | trown: thx a ton | 15:50 |
| ccamacho | jaosorior sure thing man, just let me finish to land the tripleo-common submission thats giving me some pain and ill finish the rest of the reviews :) | 15:51 |
| *** cylopez has joined #tripleo | 15:51 | |
| jaosorior | ccamacho: for testing out the HAProxy parts you don't need the last environment I mention in the blog post | 15:52 |
| jaosorior | that'll come if you wanna test apache and mysql | 15:52 |
| ccamacho | perfect :) | 15:52 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-docs: Recommend using only metadata cleaning with Ironic in virtual overcloud https://review.openstack.org/382524 | 15:53 |
| *** sudipto has quit IRC | 15:53 | |
| *** sudswas__ has quit IRC | 15:53 | |
| jaosorior | ccamacho: thanks dude | 15:57 |
| ccamacho | jaosorior np man | 15:57 |
| jaosorior | ccamacho: now... unfortunately there was a merge conflict... and the votes on the puppet part are gone :( | 15:57 |
| *** aufi has quit IRC | 15:58 | |
| openstackgerrit | Gonéri Le Bouder proposed openstack/instack-undercloud: ironic: turn on ipxe timout https://review.openstack.org/382528 | 15:58 |
| jaosorior | mcornea: ping | 15:59 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Fix OpendaylightApiNetwork key naming https://review.openstack.org/382480 | 15:59 |
| mcornea | jaosorior: hey | 15:59 |
| jaosorior | mcornea: sooo.... did the commit work? :D (sorry to poke you so much) | 16:00 |
| mcornea | jaosorior: it's currently deploying, sorry it took so long but I hit issues while building the env | 16:00 |
| *** jpich has quit IRC | 16:00 | |
| mcornea | jaosorior: checking haproxy.cfg it seems to be ok. | 16:01 |
| mcornea | jaosorior:http://paste.openstack.org/show/584497/ | 16:02 |
| jaosorior | mcornea: so you added a custom role then? | 16:03 |
| jaosorior | for what services? | 16:03 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 16:03 |
| mcornea | jaosorior: yes, for the api services mostly: http://paste.openstack.org/show/584499/ | 16:03 |
| *** ccamacho is now known as ccamacho|willbeb | 16:05 | |
| *** ccamacho|willbeb is now known as ccamacho|willbbc | 16:05 | |
| jaosorior | mcornea: great! so that looks fine | 16:05 |
| *** ccamacho|willbbc is now known as ccamacho|wllbbck | 16:05 | |
| mcornea | jaosorior: yep, it looks good | 16:05 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Use service-specific servernames for haproxy https://review.openstack.org/382534 | 16:06 |
| jaosorior | EmilienM, shardy can you check this out? ^^ | 16:06 |
| jaosorior | backport | 16:06 |
| openstackgerrit | Tim Rozet proposed openstack/tripleo-heat-templates: Renames OpenDaylight to OpenDaylightApi and splits out OVS configuration https://review.openstack.org/381160 | 16:07 |
| mcornea | jaosorior: nice, overcloud deploy complete and all services are up in haproxy | 16:10 |
| *** milan has quit IRC | 16:15 | |
| openstackgerrit | Tim Rozet proposed openstack/tripleo-heat-templates: Fixing resources path in OpenDaylight https://review.openstack.org/382541 | 16:15 |
| EmilienM | jaosorior: yes | 16:17 |
| *** rajinir has joined #tripleo | 16:17 | |
| *** dtantsur is now known as dtantsur|afk | 16:19 | |
| *** rbrady is now known as rbrady-food | 16:21 | |
| jaosorior | mcornea: awwww yeah | 16:25 |
| jaosorior | mcornea: thanks for checking it out dude | 16:25 |
| *** ayoung has joined #tripleo | 16:28 | |
| *** akuznetsov has joined #tripleo | 16:31 | |
| hewbrocca | mcornea: woooooooooooo | 16:33 |
| *** fzdarsky is now known as fzdarsky|afk | 16:33 | |
| *** lucas-hungry is now known as lucasagomes | 16:35 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Include ceilometer in swift proxy pipeline https://review.openstack.org/371950 | 16:37 |
| *** abregman has quit IRC | 16:39 | |
| *** chem has quit IRC | 16:40 | |
| *** chem has joined #tripleo | 16:41 | |
| *** jaosorior has quit IRC | 16:42 | |
| *** rasca has quit IRC | 16:43 | |
| *** yamahata has quit IRC | 16:44 | |
| *** jlinkes has quit IRC | 16:47 | |
| *** trown is now known as trown|lunch | 16:47 | |
| *** bana_k has joined #tripleo | 16:51 | |
| *** chem has quit IRC | 16:51 | |
| *** chem has joined #tripleo | 16:51 | |
| *** jlinkes has joined #tripleo | 16:51 | |
| EmilienM | jpena, trown|lunch: it doesn't seem to work, we're still installing Newton: https://review.openstack.org/#/c/382521/ | 16:52 |
| jpena | EmilienM, I think it's https://review.openstack.org/#/c/382521/1/scripts/te-broker/te-broker.sh <- the first curl points to buildlogs | 16:53 |
| *** lucasagomes is now known as lucas-brb | 16:55 | |
| *** chem has quit IRC | 16:56 | |
| *** chem has joined #tripleo | 16:56 | |
| *** hewbrocca is now known as hewbrocca-afk | 16:57 | |
| bandini | pradk, marios: so if I do the ceilometer migration before init, I still do it with the newton THT+patch in review, right? because in that case I still get the exact same error about mongo_node_ips missing in hiera | 16:59 |
| pradk | bandini, shouldnt you be have the mitaka hiera data at this point or you have newton? | 17:00 |
| *** derekh has quit IRC | 17:01 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Set enabled_share_protocols based on enabled backends https://review.openstack.org/382386 | 17:02 |
| EmilienM | jpena: can you update it? I'm not sure about the URL | 17:02 |
| bandini | pradk: I have mongodb_* entries, so I assume that no matter what the newton hiera gets pulled in as soon as we do the ceilometer step (no matter if before or after init). does that make sense marios? | 17:03 |
| bandini | unless we want to patch mitaka templates and do it with those | 17:03 |
| openstackgerrit | Javier Peña proposed openstack-infra/tripleo-ci: DO NOT MERGE testing switch to Ocata DLRN instance https://review.openstack.org/382521 | 17:04 |
| pradk | bandini, i'm a bit confused.. so you have mongo_* in your hiera or mongodb_* ? | 17:07 |
| *** dbecker has quit IRC | 17:07 | |
| bandini | pradk: I have mongodb_* entries | 17:08 |
| pradk | bandini, if its db_ then you ahve newton code already updated.. if you have mongo_* you have mitaka code | 17:08 |
| pradk | i'm fine with just updating the code to mongodb_ and let it run where you have | 17:08 |
| bandini | pradk: exactly. which means that even when doing the ceilometer migration *before* init the new hieradata gets pulled in | 17:08 |
| bandini | I will give it a shot with mongodb everywhere | 17:09 |
| bandini | in the pupet migration code I mean | 17:09 |
| pradk | ok lets just move forward with that.. and see if rest looks good | 17:09 |
| pradk | bandini, i'll update the s/mongo/mongdb .. will wait for you to confirm if rest looks ok | 17:10 |
| *** egafford has quit IRC | 17:11 | |
| openstackgerrit | Merged openstack/os-net-config: Add options attribute to NFVSwitch https://review.openstack.org/375763 | 17:11 |
| *** lucas-brb is now known as lucasagomes | 17:11 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Ceilometer Wsgi Mitaka->Newton upgrades https://review.openstack.org/360004 | 17:11 |
| bandini | pradk: ack I will report in the review (will take a bit as I need to go out for a bit now) | 17:11 |
| pradk | bandini, cool, i updated the review with the switch | 17:12 |
| bandini | ack | 17:13 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: restore missing fluentd client functionality https://review.openstack.org/382430 | 17:14 |
| *** larsks has left #tripleo | 17:14 | |
| *** larsks has joined #tripleo | 17:14 | |
| *** bana_k has quit IRC | 17:15 | |
| bandini | pradk: let's assume it works (hopefully I am not jinxing it). what do I need to run to test things are okay? would you have a oneliner to verify it? | 17:16 |
| pradk | bandini, just make sure apache is up and port 8777 is up .. once you have it up if you want i can hop on and check too | 17:17 |
| pradk | bandini, if you want to pass on the remote session link, i can keep an eye | 17:17 |
| larsks | Hello all...when rebasing my fluentd service patch on shardy's mistral/jinja/etc changes, we lost a bunch of functionality because I didn't fully understand the changes introduced by the templated roles. https://review.openstack.org/#/c/382430/ corrects this. | 17:17 |
| *** chem has quit IRC | 17:18 | |
| *** rbrady-food is now known as rbrady | 17:20 | |
| *** yamahata has joined #tripleo | 17:22 | |
| thrash | d0ugal: rbrady have y'all had a chance to test anymore? I can't seem to get the update to finish once it reaches the breakpoint. | 17:22 |
| *** ccamacho|wllbbck is now known as ccamacho | 17:23 | |
| *** chem has joined #tripleo | 17:24 | |
| *** kaslcrof has quit IRC | 17:26 | |
| *** chem has quit IRC | 17:28 | |
| *** dsariel_ has joined #tripleo | 17:29 | |
| rbrady | thrash: working on another bug at the moment | 17:30 |
| shardy | larsks: lgtm but can you please raise a bug and update the commit message? | 17:31 |
| shardy | larsks: we'll need a bug to backport to stable/newton | 17:31 |
| larsks | shardy, sure, will do. | 17:31 |
| *** akrivoka has quit IRC | 17:31 | |
| *** shardy is now known as shardy_afk | 17:32 | |
| *** yamahata has quit IRC | 17:34 | |
| *** ohamada has quit IRC | 17:34 | |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: restore missing fluentd client functionality https://review.openstack.org/382430 | 17:35 |
| *** bana_k has joined #tripleo | 17:36 | |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: restore missing fluentd client functionality https://review.openstack.org/382430 | 17:39 |
| *** athomas has quit IRC | 17:45 | |
| *** jpena is now known as jpena|off | 17:45 | |
| *** karthiks has quit IRC | 17:46 | |
| *** tosky has quit IRC | 17:47 | |
| *** mlupton has joined #tripleo | 17:47 | |
| openstackgerrit | Deepak proposed openstack/tripleo-validations: Changed the home-page link https://review.openstack.org/382586 | 17:49 |
| *** amoralej is now known as amoralej|off | 17:50 | |
| *** abregman has joined #tripleo | 17:51 | |
| *** pkovar has quit IRC | 17:55 | |
| *** mcornea has quit IRC | 17:56 | |
| openstackgerrit | Deepak proposed openstack/tripleo-common: Changed the home-page link https://review.openstack.org/382590 | 17:57 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Adds Environment File for Removing Sahara during M/N upgrade https://review.openstack.org/375517 | 18:01 |
| *** akuznetsov has quit IRC | 18:01 | |
| *** akuznetsov has joined #tripleo | 18:02 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 18:02 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 18:02 |
| openstackgerrit | Deepak proposed openstack/tripleo-specs: Changed the home-page link https://review.openstack.org/382591 | 18:03 |
| *** yamahata has joined #tripleo | 18:04 | |
| *** trown|lunch is now known as trown | 18:04 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 18:06 |
| *** akuznetsov has quit IRC | 18:06 | |
| openstackgerrit | Brent Eagles proposed openstack/tripleo-heat-templates: Increase reserved memory for computes when enabling DVR https://review.openstack.org/382594 | 18:08 |
| *** yolanda has quit IRC | 18:08 | |
| *** abehl has quit IRC | 18:09 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 18:15 |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart: Update libvirt role to identify flavor correctly https://review.openstack.org/382601 | 18:15 |
| *** yolanda has joined #tripleo | 18:15 | |
| openstackgerrit | Deepak proposed openstack/tripleo-incubator: Changed the home-page link https://review.openstack.org/382602 | 18:19 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci: [WIP] Add IPv6 coverage using existing jobs https://review.openstack.org/382515 | 18:20 |
| openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo: Add ceph profile rspec testing https://review.openstack.org/382603 | 18:23 |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart: Update libvirt role to identify flavor correctly https://review.openstack.org/382601 | 18:24 |
| *** jlinkes has quit IRC | 18:24 | |
| openstackgerrit | Deepak proposed openstack/python-tripleoclient: Changed the home-page link https://review.openstack.org/382609 | 18:27 |
| *** anshul has joined #tripleo | 18:27 | |
| openstackgerrit | Deepak proposed openstack/tripleo-quickstart: Changed the home-page link https://review.openstack.org/382612 | 18:29 |
| *** dbecker has joined #tripleo | 18:29 | |
| *** oshvartz has joined #tripleo | 18:31 | |
| *** bana_k has quit IRC | 18:31 | |
| *** rbowen has quit IRC | 18:32 | |
| *** zoliXXL is now known as zoli|gone | 18:35 | |
| *** zoli|gone is now known as zoli_gone-proxy | 18:36 | |
| *** dsariel_ has quit IRC | 18:38 | |
| *** eglynn has quit IRC | 18:49 | |
| *** egafford has joined #tripleo | 18:52 | |
| *** mlupton has quit IRC | 18:55 | |
| *** mlupton has joined #tripleo | 18:56 | |
| *** pkovar has joined #tripleo | 18:59 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 19:06 |
| trown | EmilienM: still getting newton packages on https://review.openstack.org/382521 :( | 19:13 |
| trown | I don't see anywhere else in the tripleo-ci we are setting repos... maybe it is because we have cached images? | 19:13 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci: Make the ovb-updates job work again https://review.openstack.org/374406 | 19:15 |
| rbrady | slagle: do you have a moment to talk about https://bugs.launchpad.net/tripleo/+bug/1626736 ? | 19:19 |
| openstack | Launchpad bug 1626736 in tripleo " Unable to delete overcloud node" [Critical,Triaged] - Assigned to Ryan Brady (rbrady) | 19:19 |
| slagle | rbrady: yes | 19:19 |
| rbrady | slagle: looking at https://github.com/openstack/tripleo-common/blob/master/tripleo_common/scale.py#L59-L93 and confused a bit by the stack params | 19:20 |
| rbrady | slagle: it looks like it's getting the params by first getting the resources from the stack and then removing some | 19:21 |
| rbrady | slagle: before calling heat. my question is what to do with the parameters in the plan (stored in mistral environment) | 19:22 |
| openstackgerrit | Alex Schultz proposed openstack/instack-undercloud: Pin beaker due to ruby requirement https://review.openstack.org/382628 | 19:22 |
| openstackgerrit | Alex Schultz proposed openstack/instack-undercloud: Open instack-undercloud for Ocata https://review.openstack.org/382422 | 19:22 |
| rbrady | slagle: does it matter if the two collections of parameters are the same or not? | 19:23 |
| slagle | rbrady: shouldn't the parameters in the plan match what's in heat? | 19:24 |
| *** dsariel_ has joined #tripleo | 19:24 | |
| slagle | rbrady: the only parameters that this seems to be changing is *Count and *RemovalPolicies | 19:26 |
| slagle | where * is the role name | 19:26 |
| slagle | rbrady: so it figures out the Count parameters based on the # of nodes have been requested to be deleted, and any previously failed deleted nodes | 19:29 |
| slagle | then sets that value as the new param value | 19:29 |
| slagle | and sets the RemovalPolicies param as well to the node ids that were requested to be deleted | 19:30 |
| slagle | i think you'd want to updated the plan to those new values | 19:31 |
| rbrady | slagle: ack, thanks | 19:31 |
| slagle | then do the stack update using the plan | 19:31 |
| slagle | also taking into consideration what was passed on the cli (--templates and -e's), just like for a normal stack update | 19:31 |
| slagle | jrist: did we confirm that a workflow for scale down is a newton requirement? | 19:32 |
| *** jprovazn has quit IRC | 19:36 | |
| *** bkreitch has quit IRC | 19:37 | |
| *** lazy_prince has quit IRC | 19:37 | |
| *** bkreitch has joined #tripleo | 19:37 | |
| *** lazy_prince has joined #tripleo | 19:39 | |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci: DO NOT MERGE testing switch to Ocata DLRN instance https://review.openstack.org/382521 | 19:51 |
| trown | EmilienM: trying to run it as periodic so we make sure we get new content ^, I set upload to 0 so hopefully that means it wont upload images at the end... | 19:52 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci: Redirect image build stderr to stdout https://review.openstack.org/382641 | 19:53 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 19:57 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 19:57 |
| *** mbozhenko has joined #tripleo | 20:01 | |
| *** lucasagomes is now known as lucas-afk | 20:02 | |
| *** mbozhenko has quit IRC | 20:06 | |
| *** jpich has joined #tripleo | 20:08 | |
| *** mcornea has joined #tripleo | 20:14 | |
| trown | could I get some more eyes on https://review.openstack.org/#/q/topic:open-ocata ... upgrades will not work from newton to master because the version is lower on master atm | 20:15 |
| *** dbecker has quit IRC | 20:16 | |
| *** jayg is now known as jayg|g0n3 | 20:25 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-common: Add support to create role main template file based in role.role.j2.yaml https://review.openstack.org/381593 | 20:27 |
| *** penick has joined #tripleo | 20:35 | |
| jpich | Wondering if perhaps https://review.openstack.org/#/c/382359/ could get some attention as well? It's a one line puppet update with a major impact for the UI | 20:39 |
| *** leanderthal is now known as leanderthal|afk | 20:41 | |
| *** mlupton has quit IRC | 20:43 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 20:46 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 20:46 |
| trown | the updates job passing on https://review.openstack.org/#/c/382521/3 makes me highly suspicious... but the nonha-multinode looks like a legit pass using "ocata" packages | 20:46 |
| slagle | i can tell you that the undercloud-upgrades job doesn't actually test upgrades until my patches merge :) | 20:48 |
| *** trown is now known as trown|outtypewww | 20:48 | |
| slagle | it just installs the same thing twice as-is | 20:48 |
| trown|outtypewww | well it is doing a good job of that at least :) | 20:48 |
| *** jeckersb is now known as jeckersb_gone | 20:50 | |
| *** lblanchard has quit IRC | 20:53 | |
| *** anshul has quit IRC | 20:55 | |
| EmilienM | trown|outtypewww: sorry I was afk | 20:56 |
| *** abregman has quit IRC | 20:56 | |
| EmilienM | slagle: can you approve https://review.openstack.org/#/q/topic:open-ocata please? | 20:56 |
| EmilienM | trown|outtypewww: I think we run ocata now | 20:57 |
| *** Goneri has quit IRC | 21:00 | |
| slagle | EmilienM: done | 21:00 |
| EmilienM | slagle: thx | 21:01 |
| mwhahaha | ocata merge won't land until https://review.openstack.org/#/c/382628/ gets merged | 21:01 |
| mwhahaha | yay beaker | 21:01 |
| mwhahaha | for instack-undercloud | 21:01 |
| EmilienM | slagle: so https://review.openstack.org/#/c/382521/ is passing multinode with ocata, what do we do? | 21:01 |
| EmilienM | we just wait for the automatic switch right? | 21:02 |
| EmilienM | mwhahaha: what? | 21:02 |
| EmilienM | mwhahaha: sigh | 21:02 |
| mwhahaha | EmilienM: puppet syntax jobs are red on instack-undercloud | 21:03 |
| mwhahaha | cause beaker 3.1.0 requires ruby 2.2.x | 21:03 |
| EmilienM | mwhahaha: puppet CI is good right? | 21:03 |
| EmilienM | slagle: also, can you approve https://review.openstack.org/#/c/382628/ please? | 21:03 |
| mwhahaha | yea cause we're still pinned from the 3.0.0 stuff | 21:03 |
| EmilienM | mwhahaha: we'll need to backport it | 21:03 |
| *** dciabrin has quit IRC | 21:04 | |
| openstackgerrit | Ben Nemec proposed openstack/python-tripleoclient: Remove auth_required=False from upload image command https://review.openstack.org/382656 | 21:04 |
| EmilienM | mwhahaha: ok I'm approving it | 21:05 |
| EmilienM | mwhahaha: maybe we could switch to puppet-openstack_spec_helper one day? | 21:05 |
| mwhahaha | k i'll cherry-pick when it merges | 21:05 |
| EmilienM | k thx | 21:05 |
| mwhahaha | EmilienM: probably | 21:05 |
| EmilienM | yeah that sounds safe | 21:05 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Checking stable/newton https://review.openstack.org/382657 | 21:06 |
| larsks | EmilienM, do you know what is behind the ci failures on https://review.openstack.org/#/c/382056/? Looks like missing packages in ci environment and some ssh errors. Wondering if a recheck is worth it at this point... | 21:07 |
| EmilienM | larsks: let me look | 21:08 |
| openstackgerrit | Ben Nemec proposed openstack/python-tripleoclient: Remove auth_required=False from upload image command https://review.openstack.org/382658 | 21:08 |
| mwhahaha | larsks: you'll need to rebase on top of 382628 for the puppet syntax job | 21:08 |
| EmilienM | larsks: sounds like an infra issue, you can recheck also | 21:09 |
| larsks | EmilienM, it sounds like I need a rebase first, I guess. | 21:10 |
| *** dprince has quit IRC | 21:10 | |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/instack-undercloud: correctly spell yaql_limit_iterators https://review.openstack.org/382056 | 21:11 |
| EmilienM | pradk: you can do recheck to the swift thing | 21:11 |
| EmilienM | I fixed it | 21:11 |
| *** ayoung has quit IRC | 21:12 | |
| *** rhallisey has quit IRC | 21:13 | |
| pradk | EmilienM, ty sir, will do | 21:14 |
| *** tiswanso has quit IRC | 21:15 | |
| bnemec | Heads up for anyone interested: we need https://review.openstack.org/382658 to unblock stable/newton. | 21:17 |
| openstackgerrit | Alex Schultz proposed openstack/instack-undercloud: Switch to use Puppet Openstack spec helper https://review.openstack.org/382662 | 21:20 |
| EmilienM | bnemec: approved | 21:20 |
| bnemec | EmilienM: Should probably do the master patch too then. | 21:22 |
| EmilienM | bnemec: I +2 the master, I'll let someone else approve | 21:23 |
| EmilienM | I thought the master patch was merged, my bad | 21:23 |
| bnemec | No, I just figured out what was wrong. | 21:24 |
| bnemec | I'm not entirely sure why master isn't broken too, but I suspect it has to do with some osc-lib version weirdness. | 21:25 |
| *** jpich has quit IRC | 21:33 | |
| ccamacho | EmilienM do you think that this line 2016-10-05 19:37:49.304 30093 ERROR ironic_inspector.main Unauthorized: The request you have made requires authentication. (HTTP 401) (Request-ID: req-b25fa091-972e-4d92-911c-cb0fe466369d) Can be related to the newton gate errors? | 21:35 |
| ccamacho | http://logs.openstack.org/37/382137/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/bf5a189/logs/undercloud/var/log/ironic-inspector/ironic-inspector.txt.gz | 21:35 |
| EmilienM | ccamacho: no idea | 21:35 |
| *** jcoufal_ is now known as jcoufal | 21:36 | |
| EmilienM | bnemec: is it why newton jobs are failing? | 21:37 |
| EmilienM | bnemec: your patch I mean | 21:38 |
| *** b00tcat has quit IRC | 21:39 | |
| *** bnemec has quit IRC | 21:39 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Include ceilometer in swift proxy pipeline https://review.openstack.org/371950 | 21:40 |
| EmilienM | pradk: done ^ | 21:40 |
| pradk | cool | 21:41 |
| *** tiswanso has joined #tripleo | 21:43 | |
| openstackgerrit | Merged openstack/os-net-config: Open os-net-config for Ocata https://review.openstack.org/382429 | 21:44 |
| *** akshai has quit IRC | 21:46 | |
| *** tiswanso has quit IRC | 21:47 | |
| tdasilva | EmilienM: hi, is this defining the order in which the pipeline will be built? https://review.openstack.org/#/c/371950/13/puppet/services/swift-proxy.yaml@90 | 21:47 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Open tripleo-heat-templates for Ocata https://review.openstack.org/382439 | 21:47 |
| openstackgerrit | Merged openstack/os-collect-config: Open os-collect-config for Ocata https://review.openstack.org/382424 | 21:47 |
| openstackgerrit | Merged openstack/tripleo-puppet-elements: Open tripleo-puppet-elements for Ocata https://review.openstack.org/382441 | 21:48 |
| *** dsavineau has quit IRC | 21:48 | |
| openstackgerrit | Merged openstack/tripleo-common: Open tripleo-common for Ocata https://review.openstack.org/382432 | 21:49 |
| *** bana_k has joined #tripleo | 21:52 | |
| ccamacho | EmilienM both https://review.openstack.org/#/c/381593 and https://review.openstack.org/#/c/381587/ are passing CI just need to land them in order to submit fix for the chicken egg that I had.. | 21:52 |
| ccamacho | can they be merged at the same time? | 21:52 |
| *** rbrady is now known as rbrady-afk | 21:52 | |
| openstackgerrit | Tom Barron proposed openstack/tripleo-heat-templates: Set proper ceph config path for manila https://review.openstack.org/382671 | 21:59 |
| *** penick has quit IRC | 22:01 | |
| *** mbozhenko has joined #tripleo | 22:02 | |
| *** jcoufal has quit IRC | 22:03 | |
| *** mcornea has quit IRC | 22:03 | |
| *** penick has joined #tripleo | 22:04 | |
| *** mbozhenko has quit IRC | 22:07 | |
| *** mburned is now known as mburned_out | 22:07 | |
| jrist | slagle: sorry for the slow reply, but the answer is "not from a UI perspective" - maybe from a CLI perspective though | 22:11 |
| jrist | slagle: can you refer me to the bug again? | 22:12 |
| *** [1]cdearborn has quit IRC | 22:22 | |
| *** morazi has quit IRC | 22:23 | |
| *** pkovar has quit IRC | 22:34 | |
| Slower | hmmm | 22:41 |
| Slower | so I need to update tripleo-common because I'm running into this error: | 22:41 |
| Slower | Exception creating plan: Error rendering template puppet/config.role.j2.yaml : 'role' is undefined | 22:41 |
| Slower | is there something more I have to do besides build and install it? | 22:41 |
| *** cylopez has quit IRC | 22:54 | |
| *** yamahata has quit IRC | 22:57 | |
| openstackgerrit | Ian Main proposed openstack/tripleo-heat-templates: Have docker start script honor configuration https://review.openstack.org/366138 | 23:02 |
| openstackgerrit | Ian Main proposed openstack/tripleo-heat-templates: Add steps to containerized compute deployment https://review.openstack.org/346927 | 23:02 |
| openstackgerrit | Ian Main proposed openstack/tripleo-heat-templates: WIP: Containerized Services for Composable Roles https://review.openstack.org/330659 | 23:02 |
| openstackgerrit | Ian Main proposed openstack/tripleo-heat-templates: Bind mount files to run DiD in latest atomic host https://review.openstack.org/347218 | 23:02 |
| openstackgerrit | Ian Main proposed openstack/tripleo-heat-templates: Be more inclusive in insecure registry regex https://review.openstack.org/366111 | 23:02 |
| *** nyechiel_ has joined #tripleo | 23:04 | |
| *** yamahata has joined #tripleo | 23:05 | |
| *** nyechiel has quit IRC | 23:05 | |
| openstackgerrit | Merged openstack/instack-undercloud: Pin beaker due to ruby requirement https://review.openstack.org/382628 | 23:06 |
| openstackgerrit | Merged openstack/instack-undercloud: Open instack-undercloud for Ocata https://review.openstack.org/382422 | 23:06 |
| openstackgerrit | Merged openstack/puppet-tripleo: Explicitly use Keystone v2 endpoint in the UI https://review.openstack.org/382359 | 23:06 |
| openstackgerrit | Alex Schultz proposed openstack/instack-undercloud: Pin beaker due to ruby requirement https://review.openstack.org/382686 | 23:14 |
| *** penick has quit IRC | 23:15 | |
| openstackgerrit | Alex Schultz proposed openstack/instack-undercloud: Switch to use Puppet Openstack spec helper https://review.openstack.org/382662 | 23:16 |
| *** pradk has quit IRC | 23:18 | |
| openstackgerrit | Merged openstack/python-tripleoclient: Remove auth_required=False from upload image command https://review.openstack.org/382658 | 23:18 |
| *** tiswanso has joined #tripleo | 23:28 | |
| *** tiswanso has quit IRC | 23:28 | |
| *** tiswanso has joined #tripleo | 23:29 | |
| openstackgerrit | Paul Belanger proposed openstack/diskimage-builder: Create (md5|sha256) checksum files for images https://review.openstack.org/380577 | 23:35 |
| slagle | jrist: https://bugs.launchpad.net/bugs/1626736 | 23:44 |
| openstack | Launchpad bug 1626736 in tripleo " Unable to delete overcloud node" [Critical,Triaged] - Assigned to Ryan Brady (rbrady) | 23:44 |
| *** tiswanso has quit IRC | 23:44 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Do Not Merge - Test Undercloud upgrade mitaka -> newton https://review.openstack.org/381309 | 23:48 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Undercloud upgrade for mitaka and newton https://review.openstack.org/381286 | 23:48 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
