| *** beagles has joined #tripleo | 00:07 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add IMAGE_ELEMENT_YAML https://review.openstack.org/335265 | 00:23 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Making element overriding explicit https://review.openstack.org/334785 | 00:23 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Convert pkg-map and svc-map copies to explicit variables https://review.openstack.org/335308 | 00:23 |
| *** limao has joined #tripleo | 00:32 | |
| *** maeca2 has quit IRC | 00:34 | |
| *** noslzzp_ has joined #tripleo | 00:35 | |
| *** noslzzp has quit IRC | 00:37 | |
| *** akshai has joined #tripleo | 00:39 | |
| *** noslzzp_ has quit IRC | 00:40 | |
| *** jlinkes has quit IRC | 00:43 | |
| *** bank_ has quit IRC | 00:45 | |
| *** akshai has quit IRC | 00:48 | |
| *** jlinkes has joined #tripleo | 00:50 | |
| *** saneax is now known as saneax-_-|AFK | 00:52 | |
| *** noslzzp has joined #tripleo | 00:53 | |
| *** chlong has quit IRC | 01:03 | |
| *** akshai has joined #tripleo | 01:03 | |
| *** jlinkes_ has joined #tripleo | 01:15 | |
| *** jlinkes has quit IRC | 01:18 | |
| *** skramaja_ has joined #tripleo | 01:19 | |
| *** skramaja_ has quit IRC | 01:19 | |
| *** noslzzp_ has joined #tripleo | 01:23 | |
| *** noslzzp has quit IRC | 01:23 | |
| *** noslzzp_ has quit IRC | 01:29 | |
| *** noslzzp has joined #tripleo | 01:29 | |
| *** noslzzp_ has joined #tripleo | 01:40 | |
| *** openstack has joined #tripleo | 01:43 | |
| *** noslzzp_ has quit IRC | 01:46 | |
| *** noslzzp has joined #tripleo | 01:49 | |
| *** noslzzp has quit IRC | 01:53 | |
| *** noslzzp_ has joined #tripleo | 01:53 | |
| *** thrash is now known as thrash|g0ne | 02:01 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Disable sahara tests in tempest https://review.openstack.org/361936 | 02:06 |
| *** bkopilov has quit IRC | 02:10 | |
| *** bkopilov_ has quit IRC | 02:10 | |
| *** chlong has joined #tripleo | 02:18 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: Implement undercloud upgrade job - Mitaka -> Newton https://review.openstack.org/346995 | 02:20 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario001: deploy Ceph https://review.openstack.org/366810 | 02:20 |
| *** yamahata has quit IRC | 02:21 | |
| *** maeca1 has joined #tripleo | 02:22 | |
| *** maeca1 has quit IRC | 02:22 | |
| *** jlinkes_ has quit IRC | 02:32 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: haproxy/glance-api: set 90m client/server https://review.openstack.org/367055 | 02:43 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario001: deploy Ceph https://review.openstack.org/366810 | 02:44 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: scenario001: test when ceph is installed https://review.openstack.org/366812 | 02:44 |
| *** jlinkes has joined #tripleo | 02:46 | |
| *** bana_k has joined #tripleo | 02:53 | |
| *** bana_k has quit IRC | 02:54 | |
| *** Ryjedo_ has joined #tripleo | 03:14 | |
| *** Ryjedo has quit IRC | 03:14 | |
| *** Ryjedo_ is now known as Ryjedo | 03:14 | |
| *** jlinkes has quit IRC | 03:26 | |
| *** bkopilov_ has joined #tripleo | 03:28 | |
| *** bkopilov has joined #tripleo | 03:28 | |
| *** jlinkes has joined #tripleo | 03:33 | |
| *** tzumainn has quit IRC | 03:34 | |
| *** jlinkes has quit IRC | 04:20 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Move element-info a standard entry-point https://review.openstack.org/330893 | 04:26 |
| *** jlinkes has joined #tripleo | 04:26 | |
| *** links has joined #tripleo | 04:28 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Move element-info a standard entry-point https://review.openstack.org/330893 | 04:36 |
| *** can8dnSix has joined #tripleo | 04:41 | |
| *** fragatina has joined #tripleo | 04:41 | |
| can8dnSix | hello. Does anyone know if it is possible to use TripleO to create 3 total bonds? + VLAN tagging? | 04:41 |
| can8dnSix | thanks! | 04:41 |
| *** fragatina has quit IRC | 04:45 | |
| *** pgadiya has joined #tripleo | 04:52 | |
| *** abregman has joined #tripleo | 04:52 | |
| can8dnSix | bump | 04:57 |
| *** akshai has quit IRC | 04:58 | |
| *** sshnaidm|afk has quit IRC | 05:07 | |
| *** can8dnSix has quit IRC | 05:10 | |
| *** jaosorior has joined #tripleo | 05:10 | |
| *** jlinkes_ has joined #tripleo | 05:12 | |
| *** chlong has quit IRC | 05:14 | |
| *** jlinkes has quit IRC | 05:15 | |
| *** redhatkj has joined #tripleo | 05:16 | |
| *** kjw3 has quit IRC | 05:19 | |
| *** abregman has quit IRC | 05:28 | |
| *** chlong has joined #tripleo | 05:30 | |
| *** links has quit IRC | 05:31 | |
| *** links has joined #tripleo | 05:46 | |
| *** abregman has joined #tripleo | 05:52 | |
| *** fragatina has joined #tripleo | 05:52 | |
| *** fragatina has quit IRC | 05:56 | |
| *** pkovar has joined #tripleo | 06:01 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 06:04 |
| *** jlinkes has joined #tripleo | 06:07 | |
| *** jlinkes_ has quit IRC | 06:07 | |
| *** oshvartz has joined #tripleo | 06:07 | |
| *** florianf has joined #tripleo | 06:10 | |
| *** chlong has quit IRC | 06:13 | |
| *** tremble has joined #tripleo | 06:14 | |
| *** tremble has joined #tripleo | 06:14 | |
| *** bana_k has joined #tripleo | 06:20 | |
| *** liverpooler has quit IRC | 06:21 | |
| *** chlong has joined #tripleo | 06:25 | |
| *** yamahata has joined #tripleo | 06:26 | |
| *** jprovazn has joined #tripleo | 06:29 | |
| *** saneax-_-|AFK is now known as saneax | 06:36 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 06:36 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Image URLs need update https://review.openstack.org/367115 | 06:37 |
| openstackgerrit | JiWei proposed openstack/os-net-config: Raise NotImplementedError instead of NotImplemented https://review.openstack.org/367116 | 06:39 |
| shadower | hey jaosorior: do you know if the ovb gate is fixed yet? | 06:43 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Migrate to using osc-lib https://review.openstack.org/335460 | 06:43 |
| jaosorior | marios: can you check this commit out? https://review.openstack.org/#/c/365585/ | 06:43 |
| jaosorior | shadower: I have no clue... it seems it might be :/ at least some stuff have passed recently | 06:44 |
| openstackgerrit | JiWei proposed openstack/os-net-config: Raise NotImplementedError instead of NotImplemented https://review.openstack.org/367123 | 06:44 |
| shadower | jaosorior: thanks. Guess I'll have to recheck and see | 06:44 |
| jaosorior | I did the same and it's been around an hour.... and my stuff hasn't crashed yet. So it seems promising :D | 06:45 |
| shadower | \o/ | 06:46 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 06:49 |
| marios | jaosorior: sure man, i already have this in my queue from yesterday https://review.openstack.org/#/c/357765 will get to them in bit | 06:50 |
| *** jpena|away is now known as jpena|off | 06:52 | |
| d0ugal | shadower: Yeah, I am seeing stuff pass here: http://tripleo.org/cistatus.html | 06:53 |
| d0ugal | so it looks promisng... | 06:53 |
| d0ugal | jtomasek: Are you around? | 06:54 |
| jtomasek | d0ugal: yep | 06:54 |
| d0ugal | jtomasek: Have you thought at all about the workflow for updating a deployment? | 06:55 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 06:55 |
| d0ugal | jtomasek: i.e. you deploy something, then you upload new templates and re-deploy | 06:55 |
| jtomasek | d0ugal: not entirely, I'd assume that the same deploy workflow would be used, although I am not very familiar if there are any special requirements so subsequent deployments | 06:56 |
| d0ugal | jtomasek: The main thing is making sure everything is in sync I guess | 06:57 |
| d0ugal | jtomasek: I'd have to manually delete the container contents and re-upload | 06:57 |
| d0ugal | and then what if the mistral env doesn't match now? | 06:57 |
| d0ugal | :/ | 06:57 |
| jtomasek | d0ugal: you need to recreate the container/plan with every subdequent deploy call, because of how cli works? | 06:58 |
| d0ugal | jtomasek: well, I need to update the plan somehow | 06:59 |
| d0ugal | I thought re-creating it might be easiest | 06:59 |
| d0ugal | but I can't do that, it wont let me delete a plan with a related stack | 06:59 |
| *** fzdarsky has joined #tripleo | 06:59 | |
| jtomasek | d0ugal: from the plan oriented deployment point of view, I'd expect, you just update the plan with additional templates and deploy again | 06:59 |
| d0ugal | jtomasek: Sure, that's fine | 07:00 |
| d0ugal | jtomasek: but the question is, how do you update the templates? :) | 07:00 |
| *** anshul has joined #tripleo | 07:00 | |
| jtomasek | d0ugal: by overriding them in the container by uploading file with the same 'path' | 07:01 |
| d0ugal | jtomasek: but what if the mistral env then doesn't match up? call create_plan again? | 07:01 |
| d0ugal | jtomasek: i.e. maybe there is a new root template | 07:02 |
| d0ugal | or root env | 07:02 |
| jtomasek | d0ugal: match up in what sense? Such as some selected environments don't exist in the plan any more? | 07:02 |
| jtomasek | yeah, that is the problem to solve | 07:02 |
| jtomasek | we've had a few discussions on that topic | 07:02 |
| d0ugal | oh, we have? :) | 07:03 |
| jtomasek | d0ugal: not we but 'we' hah | 07:03 |
| d0ugal | lol | 07:03 |
| d0ugal | and I think I need to delete everything from swift before uploading the new files - otherwise some old files not in the plan might get left behind | 07:03 |
| *** tesseract- has joined #tripleo | 07:04 | |
| d0ugal | Okay, so I guess I am going to delete the container contents, re-upload and hope the mistral environment matches for now. Then file a new bug to improve that and maybe add a "update_plan" | 07:04 |
| d0ugal | (which would update the env) | 07:05 |
| jtomasek | d0ugal: there is a bunch of such files from the start anyway, since not all templates are used for a deployment | 07:05 |
| *** hjensas has joined #tripleo | 07:05 | |
| *** hjensas has joined #tripleo | 07:05 | |
| d0ugal | jtomasek: Sure | 07:05 |
| jtomasek | d0ugal: ok, we'd need some kind of 'diffing' which would update the mistral environment after plan is updated | 07:05 |
| d0ugal | jtomasek: Yeah, I'll open a bug for that. | 07:06 |
| jtomasek | d0ugal: thanks | 07:06 |
| d0ugal | jtomasek: Thank you :) I had no idea what to do, I have a plan now. | 07:11 |
| *** mcornea has joined #tripleo | 07:14 | |
| *** aufi has joined #tripleo | 07:14 | |
| *** bana_k has quit IRC | 07:15 | |
| jtomasek | d0ugal: in general this is almost the same problem which the user is facing with the old cli. when he wants to update, he needs to make sure that he provides a correct set of environments and parameters (-e -p) | 07:17 |
| jtomasek | difference is that those are now stored in mistral environment | 07:18 |
| d0ugal | jtomasek: Yup, true. Making sure they have the same behaviour and flexibility with the new approach is tricky :) | 07:18 |
| jtomasek | d0ugal: yeah | 07:18 |
| *** liverpooler has joined #tripleo | 07:22 | |
| *** dsariel has joined #tripleo | 07:24 | |
| *** jlinkes has quit IRC | 07:24 | |
| *** bana_k has joined #tripleo | 07:25 | |
| *** zoli_gone-proxy is now known as zoliXXL | 07:25 | |
| *** abregman_ has joined #tripleo | 07:29 | |
| *** abregman has quit IRC | 07:32 | |
| *** ebarrera has joined #tripleo | 07:32 | |
| *** nyechiel_ has joined #tripleo | 07:33 | |
| *** ifarkas_afk is now known as ifarkas | 07:33 | |
| *** openstackgerrit has quit IRC | 07:33 | |
| *** openstackgerrit has joined #tripleo | 07:34 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Update Nodes listing https://review.openstack.org/365580 | 07:35 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Roles Listing using Mistral Action https://review.openstack.org/360569 | 07:36 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Update the plan on a re-deploy https://review.openstack.org/366541 | 07:37 |
| *** zoliXXL is now known as zoli_gone-proxy | 07:37 | |
| d0ugal | shadower: My ovb recheck failed :( | 07:38 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Roles Listing using Mistral Action https://review.openstack.org/360569 | 07:39 |
| *** dtantsur|afk is now known as dtantsur | 07:42 | |
| *** jlinkes has joined #tripleo | 07:42 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: ModalPanel component https://review.openstack.org/366615 | 07:42 |
| *** jlinkes_ has joined #tripleo | 07:44 | |
| *** zoli_gone-proxy is now known as zoliXXL | 07:44 | |
| *** jlinkes has quit IRC | 07:47 | |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Update plan files when re-deploying https://review.openstack.org/366541 | 07:50 |
| *** ohamada has joined #tripleo | 07:51 | |
| shadower | d0ugal: nooooo | 07:51 |
| *** jpich has joined #tripleo | 07:52 | |
| *** bana_k has quit IRC | 07:52 | |
| *** chlong has quit IRC | 07:52 | |
| jaosorior | shadower: for me ha failed and nonha passed. And ha failed cause of a node registration error... which is a sporadic error in CI | 07:53 |
| jaosorior | so it seems that ovb is not as broken as yesterday | 07:53 |
| *** jpena|off is now known as jpena | 07:54 | |
| d0ugal | lol | 07:54 |
| d0ugal | Progress! | 07:54 |
| shadower | okay, that's good to hear. My rechecks haven't finished yet | 07:54 |
| *** karthiks has joined #tripleo | 07:57 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fetch internal certificates for HAProxy based on network https://review.openstack.org/366548 | 07:58 |
| *** dbecker has joined #tripleo | 07:59 | |
| *** yamahata has quit IRC | 08:01 | |
| *** athomas has joined #tripleo | 08:01 | |
| *** pgadiya has quit IRC | 08:01 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove "type: direct" from workflows as it is the default https://review.openstack.org/341617 | 08:03 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Update the Mistral action names https://review.openstack.org/366519 | 08:07 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Hook internal TLS flag to apache-based services https://review.openstack.org/366075 | 08:07 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add HAProxy TLS handled by certmonger as composable service https://review.openstack.org/356430 | 08:07 |
| *** dbecker has quit IRC | 08:08 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add keystone networks for the different endpoints https://review.openstack.org/367176 | 08:13 |
| jaosorior | jistr: are you around yet? | 08:13 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: DONT MERGE: WIP: Nothing here https://review.openstack.org/365324 | 08:13 |
| *** sshnaidm|afk has joined #tripleo | 08:14 | |
| *** sshnaidm|afk is now known as sshnaidm | 08:14 | |
| shadower | mandre: | 08:21 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Make Nova VNC Proxy service name match service net map https://review.openstack.org/367200 | 08:23 |
| *** bvandenh has joined #tripleo | 08:25 | |
| shadower | mandre: thinking about the ssh_copy_keys on the undercloud. I'm trying to make the undercloud validations not depend on it | 08:26 |
| *** shardy has joined #tripleo | 08:26 | |
| jaosorior | shardy: hey dude, seems I found an issue the yaql stuff we're using. | 08:27 |
| shadower | mandre: but won't we need it eventually for HA undercloud anyway? Not that it's a big concern now. | 08:27 |
| jaosorior | shardy: so this applies for my patch adding the networks for the services. And also for the vip-config stuff that adds the nodes that host a service and the VIP for that service | 08:28 |
| shadower | mandre: but maybe just running the copy_ssh_workflow would be more robust | 08:28 |
| shadower | mandre: what do you think? | 08:28 |
| shardy | jaosorior: Ok, what is the issue? | 08:29 |
| jaosorior | shardy: seems that that the service name for nova vnc proxy has a mismatch. And so an extra capital letter in the network service map makes the lower case key "nova_vnc_proxy". While the service name we set is nova_vncproxy | 08:29 |
| jaosorior | shardy: trying to fix it here https://review.openstack.org/#/c/367200/ | 08:29 |
| shardy | jaosorior: ack - thanks, yeah I had the same issue with some other services but missed that one | 08:30 |
| jaosorior | shardy: yeah, I stumbled upon keystone too | 08:30 |
| *** dsariel has quit IRC | 08:30 | |
| jaosorior | shardy: and another issue is the heat APIs | 08:30 |
| jaosorior | shardy: heat-api by itself works. But cloudwatch and cfn are problematic too | 08:30 |
| *** cwolferh has quit IRC | 08:31 | |
| jaosorior | shardy: right now I'm going around that by fetching the network from heat-api for all three (I'm testing TLS everywhere for haproxy). | 08:31 |
| jaosorior | shardy: but... I' | 08:31 |
| shardy | jaosorior: ack, OK well I guess we'll have to fix them one by one until everything lines up | 08:31 |
| jaosorior | I'm not sure what to do | 08:31 |
| jaosorior | shardy: should we add those entries to the servicenetmap? | 08:31 |
| shardy | jaosorior: Yes, if we want the hiera to be generated | 08:31 |
| jaosorior | shardy: or do we always assume that they will run with heat-api?> | 08:31 |
| jaosorior | OK | 08:31 |
| jaosorior | I'll do it then | 08:31 |
| shardy | jaosorior: I don't think we should assume that | 08:32 |
| *** abregman_ has quit IRC | 08:32 | |
| shardy | ++ thanks | 08:32 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 08:32 |
| *** anshul has quit IRC | 08:33 | |
| *** paramite has quit IRC | 08:34 | |
| *** pblaho has joined #tripleo | 08:35 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add Heat's Cfn and Cloudwatch networks to ServiceNetMap https://review.openstack.org/367208 | 08:36 |
| *** jaosorior is now known as jaosorior_lunch | 08:40 | |
| *** cwolferh has joined #tripleo | 08:41 | |
| *** abregman_ has joined #tripleo | 08:44 | |
| *** snecklifter has quit IRC | 08:45 | |
| *** anshul has joined #tripleo | 08:49 | |
| *** apetrich has quit IRC | 08:50 | |
| *** pkovar has quit IRC | 08:52 | |
| sshnaidm | bnemec, slagle, please ping me when you're around | 08:53 |
| *** derekh has joined #tripleo | 08:55 | |
| derekh | sshnaidm: just saw your email about the mirror server, adding you now | 08:55 |
| sshnaidm | derekh, oh, thanks | 08:55 |
| mandre | shadower: you're exact, this will be required for HA undercloud | 08:55 |
| sshnaidm | derekh, promotion didn't work, want to see what is going on there | 08:55 |
| derekh | sshnaidm: added | 08:56 |
| derekh | centos@66.187.229.139 | 08:56 |
| sshnaidm | derekh, thanks | 08:56 |
| shadower | mandre: yea thought so. Thanks | 08:59 |
| sshnaidm | derekh, /root/.promoterc: No such file or directory - do you know what should be in this file? | 08:59 |
| sshnaidm | derekh, I suppose it's $RDO_PROMOTE_TOKEN , but what is it.. | 09:01 |
| *** jbadiapa has joined #tripleo | 09:01 | |
| derekh | yup, forgot about it, trown|outtypewww sent it to me in a mail, I'll send it to you now | 09:01 |
| shardy | sshnaidm: Hi, did we make any progress yesterday debugging the AFS mirror issues? | 09:02 |
| d0ugal | shardy: Hey, can you take another look at https://review.openstack.org/#/c/366541/ when you get a chance? | 09:02 |
| shardy | still seem to be a lot of jobs failing with repo related issues | 09:02 |
| d0ugal | shardy: I figured out a new approach which I think it better (and the plan deletion had problems anyway) | 09:02 |
| openstackgerrit | Merged openstack/python-tripleoclient: Replace agent elements with package python-heat-agent-puppet https://review.openstack.org/366406 | 09:03 |
| sshnaidm | shardy, no progress yet, yesterday we worked on making cloud stable and environments clean up | 09:03 |
| derekh | sshnaidm: sent, ounce you add that to the mirror server you should be able to just run the promote script instead of waiting for cron | 09:04 |
| shardy | sshnaidm: Ok, thanks | 09:04 |
| sshnaidm | shardy, so it could be possible that network issue will go with all these problems, but seems it didn't | 09:04 |
| shardy | d0ugal: I like the update_plan part, seem logical | 09:04 |
| sshnaidm | derekh, sure, thanks | 09:04 |
| shardy | d0ugal: and I assume that's a workflow where we can add process_templates to do the j2 templating? | 09:04 |
| shardy | d0ugal: one question, we reset_parameters, but then the deploy re-creates them in the mistral env, right? | 09:05 |
| sshnaidm | shardy, we'll try to figure out what is the problem today | 09:05 |
| shardy | sshnaidm: Ok, thanks | 09:05 |
| d0ugal | shardy: Yeah, my concern is a parameter will be left over from the previous deploy but not included in the new one | 09:05 |
| shardy | we've got a huge backlog of patches for RC1 so getting CI working again is critical | 09:05 |
| d0ugal | shardy: i.e. you remove one to use the default. | 09:05 |
| shardy | d0ugal: that won't work anyway because we do patch updates with heat | 09:06 |
| shardy | although IIRC we did pass clear_parameters in some cases | 09:06 |
| d0ugal | shardy: oh, good point. | 09:06 |
| shardy | probably needs some testing around that | 09:06 |
| shardy | d0ugal: otherwise lgtm | 09:06 |
| d0ugal | shardy: hmm, if we use patch deploy is there any reason for us to store parameters at all? | 09:07 |
| d0ugal | it just seems like we will always risk being out of sync with something else. | 09:07 |
| shardy | d0ugal: Not really, we've got an API for heat now which can retrieve the actual heat environment from the running overcloud | 09:07 |
| shardy | I guess we can perhaps optimize that later if storing them in mistral is workinG OK atm | 09:07 |
| d0ugal | shardy: Yeah, sounds good | 09:08 |
| * d0ugal wants to avoid changing as much as possible :) | 09:08 | |
| openstackgerrit | Merged openstack/tripleo-common: Allow for building specific images https://review.openstack.org/363095 | 09:08 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 09:16 |
| openstackgerrit | Dmitry Tantsur proposed openstack/instack-undercloud: Disable unsupported drivers and fix DRAC dependency https://review.openstack.org/367237 | 09:20 |
| dtantsur | this is somewhat opinionated, but I'd prefer we merge that ^^^ | 09:20 |
| dtantsur | otherwise we enable drivers that upstream ironic no longer supports and plan on removing | 09:21 |
| *** lucasagomes is now known as lucas-relocate | 09:21 | |
| mandre | can I get an ack on https://review.openstack.org/#/c/362701/? it's a simple patch that adds unit tests | 09:22 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 09:24 |
| shardy | mandre: why are we creating a default plan in instack-undercloud? | 09:24 |
| shardy | shouldn't the client decide what plan to create on deployment? | 09:25 |
| shardy | (tripleoclient in fact does this now) | 09:25 |
| *** bvandenh has quit IRC | 09:25 | |
| shardy | d0ugal: ^^ how will this work with updating the plan | 09:25 |
| shardy | what if the updated plan is highly customized and incompatible with the default one? | 09:26 |
| mandre | shardy: not sure, akrivoka would know since she's the one who wrote the original patch | 09:27 |
| d0ugal | shardy: hmm, thinking. | 09:27 |
| shardy | mandre: Ok - sounds like it's for the UI, which is fine, but I don't want to break the CLI with it | 09:27 |
| mandre | shardy: https://review.openstack.org/#/c/349532/ | 09:28 |
| openstackgerrit | Merged openstack/puppet-tripleo: Add class to write overcloud VIPs into /etc/hosts https://review.openstack.org/357762 | 09:29 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Create entries for overcloud VIPs in /etc/hosts https://review.openstack.org/357765 | 09:30 |
| d0ugal | shardy: Yeah, that is a good point. I guess we need an update_plan version of the create_plan action. | 09:30 |
| d0ugal | or plan.create as it is called now :) | 09:30 |
| *** mbound has joined #tripleo | 09:30 | |
| d0ugal | https://github.com/openstack/tripleo-common/blob/master/tripleo_common/actions/plan.py#L59 | 09:30 |
| d0ugal | shardy: I think the default plan is there to make the GUI experience nicer. | 09:31 |
| shardy | d0ugal: sure, that's fine - will it break the CLI with your new update strategy? | 09:31 |
| d0ugal | shardy: It wont break it, but I think I need to add a new extra step | 09:31 |
| d0ugal | shardy: at the moment it would break with a different root template for example | 09:31 |
| shardy | e.g we could end up with a mix of the default plan and e.g my --templates /foo/super-custom/tht | 09:31 |
| d0ugal | shardy: https://github.com/openstack/tripleo-common/blob/master/tripleo_common/actions/plan.py#L83-L87 | 09:32 |
| d0ugal | shardy: That is never updated after the first create | 09:32 |
| shardy | I personally want my CLI deployments clean - I don't want my files merged with an existing plan | 09:32 |
| d0ugal | shardy: The update strategy I added in that patch doesn't merge - it deletes the existing files and uploads the new | 09:32 |
| shardy | d0ugal: all of the existing files? | 09:33 |
| d0ugal | shardy: all of them :) | 09:33 |
| shardy | or only those that exist in the new plan? | 09:33 |
| shardy | Ok, so it's a delete, but not a delete ;) | 09:33 |
| d0ugal | Yeah | 09:33 |
| shardy | fair enough | 09:33 |
| shardy | thanks for clarifying | 09:33 |
| *** r-mibu has quit IRC | 09:33 | |
| d0ugal | shardy: I can't delete the plan, because the action wont let me while the stack exists... | 09:33 |
| shardy | ah, yeah | 09:33 |
| d0ugal | shardy: but I didn't want to risk stale files being left behind | 09:33 |
| shardy | I ran into that yesterday | 09:33 |
| *** r-mibu has joined #tripleo | 09:33 | |
| shardy | cool, sounds OK then | 09:33 |
| d0ugal | shardy: however, there is still a potential issue with the mistral env - I'll write a patch for that. | 09:34 |
| shardy | d0ugal: did you see https://bugs.launchpad.net/tripleo/+bug/1620747 ? | 09:35 |
| openstack | Launchpad bug 1620747 in tripleo "Plan creation fails due to duplicate name error" [High,Triaged] | 09:35 |
| d0ugal | shardy: Yeah | 09:35 |
| shardy | that's what happens if you delete the swift plan but not the mistral env | 09:35 |
| shardy | very non-obvious until you figure it out :) | 09:35 |
| d0ugal | shardy: indeed | 09:35 |
| d0ugal | shardy: I'll try and do that too :) | 09:36 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Fix promote script and add logs https://review.openstack.org/367244 | 09:36 |
| *** anshul has quit IRC | 09:38 | |
| d0ugal | shardy: or actually jpich might have a look at it | 09:40 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: set up quickstart to replace instack-virt-setup https://review.openstack.org/358089 | 09:40 |
| jpich | d0ugal, shardy: Will see if I can bring up a better error message | 09:41 |
| *** apetrich has joined #tripleo | 09:41 | |
| *** pgadiya has joined #tripleo | 09:43 | |
| shardy | jpich: thanks! | 09:43 |
| *** sshnaidm is now known as sshnaidm|lnch | 09:44 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: [WIP] Undercloud reboot https://review.openstack.org/367253 | 09:48 |
| *** anshul has joined #tripleo | 09:49 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove "type: direct" from workflows as it is the default https://review.openstack.org/341617 | 09:50 |
| *** social_ has quit IRC | 09:50 | |
| shardy | Can we get another review on https://review.openstack.org/#/c/354016/ so we can declare ironic-integration complete for Newton? | 09:56 |
| dtantsur | +1 to the request ^^^ :) | 09:58 |
| *** pkovar has joined #tripleo | 10:02 | |
| *** lucas-relocate is now known as lucasagomes | 10:05 | |
| *** akrivoka has joined #tripleo | 10:07 | |
| *** maeca1 has joined #tripleo | 10:08 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fix and improve the ansible-lint gate job https://review.openstack.org/366758 | 10:13 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: WIP: Fix ansible-lint errors in all the repo https://review.openstack.org/366837 | 10:13 |
| *** adarazs is now known as adarazs_lunch | 10:13 | |
| *** bvandenh has joined #tripleo | 10:18 | |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 10:20 |
| *** sshnaidm|lnch is now known as sshnaidm | 10:21 | |
| flepied | sshnaidm: do you keep https://etherpad.openstack.org/p/tripleo-ci-status up to date? I re-worked a little bit the form to be dashboard friendly | 10:22 |
| sshnaidm | flepied, yes.. finally I know who did this :) | 10:23 |
| sshnaidm | flepied, however I'm not sure it will stay so, I plan to do it automatically updated | 10:24 |
| flepied | sshnaidm: I modified the dashboard to parse it so now there is a link to this etherpad and the number of issues is reported | 10:24 |
| flepied | sshnaidm: ok but could you keep the format? | 10:24 |
| sshnaidm | flepied, thanks, and did you see my pull request for tempest job in dashboard? | 10:24 |
| flepied | sshnaidm: yes and I wanted to discuss it with you | 10:25 |
| sshnaidm | flepied, ok | 10:25 |
| flepied | sshnaidm: I don't think it makes sense to have a new tile because that's not a new stage | 10:25 |
| flepied | sshnaidm: the dashboard needs to be synthetic with the minimum number of info and some links if we need more | 10:26 |
| flepied | sshnaidm: so I feel the link to the tempest results should be more on the etherpad or a link in the dashboard but not a new tile | 10:27 |
| sshnaidm | flepied, do you want to include it into existing "squares"? | 10:27 |
| flepied | sshnaidm: yes that would be better but I didn't think on how to do it | 10:27 |
| flepied | sshnaidm: also in term of design the lower row is for Mitaka so it doesn't make sense to have the square for tempest there anyway | 10:28 |
| *** ramishra has quit IRC | 10:29 | |
| sshnaidm | flepied, as I understood from weshay it's for visibility of "tempest-ready" of upstream master, so I'm not sure that including just a link could solve this | 10:29 |
| sshnaidm | flepied, also I don't have a link to tempest jobs, it's not jenkins | 10:30 |
| flepied | sshnaidm: perhaps that's not the right vehicle to carry this tempest readiness indicator. we need to brainstorm a little bit more. | 10:31 |
| sshnaidm | flepied, ok, I'd prefer to include weshay in this discussion | 10:31 |
| flepied | sshnaidm: yes of course | 10:31 |
| *** ramishra has joined #tripleo | 10:32 | |
| *** limao has quit IRC | 10:33 | |
| sshnaidm | flepied, where is your etherpad parser? | 10:33 |
| flepied | sshnaidm: it is in feed-dashboard.sh in rdo-dashboard | 10:34 |
| sshnaidm | flepied, the etherpad parser? | 10:34 |
| flepied | sshnaidm: yes that's just a couple of shell commands | 10:34 |
| *** zoliXXL is now known as zoli|lunch | 10:39 | |
| *** zoli|lunch is now known as zoli_gone-proxy | 10:40 | |
| *** mburned_out is now known as mburned | 10:44 | |
| *** apetrich has quit IRC | 10:45 | |
| sshnaidm | derekh, this patch should solve pip install problems (hope so) https://review.openstack.org/#/c/366687/ I see more and more such errors.. | 10:48 |
| *** jcoufal has quit IRC | 10:50 | |
| *** sshnaidm is now known as sshnaidm|afk | 10:54 | |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 10:55 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Move AllNodesValidationDeployments into jinja template loop https://review.openstack.org/337587 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert overcloud.yaml to support jinja2 templating https://review.openstack.org/315679 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert allNodesConfig properties to composable jinja2 https://review.openstack.org/365794 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Move AllNodesDeployments into jinja template loop https://review.openstack.org/337267 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Move role ResourceGroups inside the jinja2 loop https://review.openstack.org/365793 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Move role deployment steps into puppet/post.yaml https://review.openstack.org/365763 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Remove *ExtraConfig parameters from overcloud.yaml https://review.openstack.org/365792 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Move per-role NetIpListMap's into jinja template loop https://review.openstack.org/364749 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert SwiftDevicesAndProxyConfig to composable format https://review.openstack.org/364748 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert deploy steps to jinja2 loop https://review.openstack.org/365796 | 10:58 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert UpdateWorkflow to support composable roles https://review.openstack.org/367282 | 10:58 |
| shardy | jistr, marios: hey would appreciate any feedback on https://review.openstack.org/367282 | 10:58 |
| shardy | I modified the UpgradeWorkflow stuff to support composable roles | 10:59 |
| *** pgadiya has quit IRC | 10:59 | |
| *** jpena is now known as jpena|lunch | 10:59 | |
| *** bkopilov has quit IRC | 11:01 | |
| sshnaidm|afk | derekh, why do we build images? I think we switched to cached | 11:02 |
| *** bkopilov_ has quit IRC | 11:02 | |
| *** pkovar has quit IRC | 11:02 | |
| marios | shardy: thanks very much will do | 11:02 |
| tbarron | marios: when you get a chance, pls. look over https://etherpad.openstack.org/p/manila-overcloud-deploy-with-netapp-notes and let me know where I'm going wrong: I'm attempting to build overcloud images with the changes from 354014, 366760, and 354019 but the overcloud images' /etc/puppet/modules/tripleo/... doesn't have the changes, | 11:03 |
| *** adarazs_lunch is now known as adarazs | 11:04 | |
| marios | tbarron: will do | 11:05 |
| tbarron | marios: thanks, hopefully I'll actually learn this stuff :) | 11:06 |
| *** pgadiya has joined #tripleo | 11:07 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-docs: Documentation for installing and using Ironic in overcloud https://review.openstack.org/354016 | 11:08 |
| dtantsur | trown|outtypewww, updated ^^ | 11:08 |
| b00tcat | in a default isntallation of an overcloud, should all the tempest tests pass? because I'm getting some FAILED messages in the output :-/ | 11:09 |
| derekh | sshnaidm|afk: I've already +2's the pip retry patch | 11:09 |
| derekh | sshnaidm|afk: which patch did use the image cache? | 11:09 |
| marios | tbarron: thanks ... sorry it is so involved to test, esp. with the puppet-tripleo side needing the image build... getting there slowly :) | 11:09 |
| *** bvandenh has quit IRC | 11:10 | |
| *** maeca1 has quit IRC | 11:12 | |
| jistr | shardy: spotted a superfluous endfor, otherwise looks good :) There's probably no use in amending the rest of the upgrades until they are adapted for composable services. | 11:12 |
| openstackgerrit | Merged openstack/instack-undercloud: Unit tests for _create_default_plan() https://review.openstack.org/362701 | 11:15 |
| *** dsariel has joined #tripleo | 11:17 | |
| shardy | jistr: thanks! | 11:17 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert UpdateWorkflow to support composable roles https://review.openstack.org/367282 | 11:19 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert AllNodesExtraConfig to support composable roles https://review.openstack.org/367295 | 11:19 |
| shardy | interested in feedback re the AllNodesExtraConfig patch | 11:19 |
| shardy | atm I can't see any way to avoid breaking backwards compatibility | 11:20 |
| *** coolsvap is now known as _coolsvap_ | 11:21 | |
| *** paramite has joined #tripleo | 11:21 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: WIP: Fix ansible-lint errors in all the repo https://review.openstack.org/366837 | 11:22 |
| *** bvandenh has joined #tripleo | 11:23 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fix all ci-scripts to use $LOCATION https://review.openstack.org/367300 | 11:27 |
| marios | shardy: so i had a look at https://review.openstack.org/#/c/367282/ - i guess that depends on a whole bunch of other stuff right? | 11:30 |
| *** pkovar has joined #tripleo | 11:30 | |
| *** bvandenh has quit IRC | 11:30 | |
| *** egafford has quit IRC | 11:30 | |
| marios | shardy: it looks OK and doesn't change existing behaviour (I mean frmo the user point of view) so we could land it but I haven't been following the roles reviews (yeah just checked shortlog on https://review.openstack.org/gitweb?p=openstack/tripleo-heat-templates.git;a=shortlog;h=f4b808823e973b06f5e9922d900e0e156c7ff64c ) | 11:31 |
| *** bvandenh has joined #tripleo | 11:31 | |
| marios | shardy:so I don't feel confident enough to +2 that with a straight face :) I'll have to dig a bit further on some of those parent reviews, but tomorrow shardy | 11:32 |
| marios | shardy: 'it looks OK' i mean the templating so we don't have to repeat the script delivery is cool i like it | 11:33 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 11:34 |
| marios | shardy: does the jinja processing happen before this is fed into heat engine (I mean, for debug purposes etc you get the actual qualified variable names right?) | 11:34 |
| *** apetrich has joined #tripleo | 11:35 | |
| *** bvandenh_ has joined #tripleo | 11:35 | |
| *** fragatina has joined #tripleo | 11:36 | |
| *** noslzzp has joined #tripleo | 11:37 | |
| *** noslzzp_ has quit IRC | 11:38 | |
| *** bvandenh has quit IRC | 11:39 | |
| *** fragatina has quit IRC | 11:41 | |
| *** apetrich has quit IRC | 11:46 | |
| *** apetrich has joined #tripleo | 11:46 | |
| *** jcoufal has joined #tripleo | 11:46 | |
| EmilienM | hello | 11:48 |
| *** lucasagomes is now known as lucas-hungry | 11:48 | |
| *** links has quit IRC | 11:52 | |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 11:53 |
| *** jaosorior_lunch is now known as jaosorior | 11:56 | |
| *** apetrich has quit IRC | 11:56 | |
| *** thrash|g0ne is now known as thrash | 11:56 | |
| *** rhallisey has joined #tripleo | 11:57 | |
| *** apetrich has joined #tripleo | 11:58 | |
| marios | tbarron: see etherpad | 11:59 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: haproxy/glance-api: set 90m client/server https://review.openstack.org/367055 | 12:00 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenarios: set Debug to True https://review.openstack.org/366896 | 12:00 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario001: deploy Ceph https://review.openstack.org/366810 | 12:00 |
| *** sshnaidm|afk is now known as sshnaidm | 12:00 | |
| *** zoli_gone-proxy is now known as zoliXXL | 12:02 | |
| sshnaidm | derekh, here we switched to cached images: https://review.openstack.org/#/c/359481/ , I wonder why we build them now | 12:02 |
| *** maeca1 has joined #tripleo | 12:02 | |
| marios | tbarron: hope that makes sense? for puppet-tripleo i think there was some misunderstanding about the 'DIB_INSTALLTYPE' | 12:02 |
| EmilienM | slagle: I WIPed https://review.openstack.org/#/c/367055/ | 12:03 |
| tbarron | marios: ok, i'll use gerrit as direct source of truth for the puppet-tripleo stuff instead of cloning & cherry-picking locally. But shouldn't the local approach work too, theoretically, so that I must be executing something wrong? | 12:03 |
| EmilienM | slagle: see my comment but it looks like image upload hangs forever when using ceph backend | 12:03 |
| tbarron | marios: I do understand what you are saying about tht though, thanks for clearing that up. | 12:04 |
| *** apetrich has quit IRC | 12:05 | |
| marios | tbarron: i think the DIB_REPOLOCATION_ was intended for pulling from git/review.openstack etc ... i am not sure about getting that to use a local source but i could be wrong. | 12:05 |
| jaosorior | jistr: Hey man, can you check this commit out? https://review.openstack.org/#/c/367200/1 It's functionality can be verified by checking (for example in the nonha job) that nova_vnc_proxy_network is indeed written to all-nodes-config.yaml | 12:05 |
| tbarron | marios: ok, i can find out more about that later, i'll use gerrit directly as you have indicated, thanks! | 12:06 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: Add TripleO scenarios jobs in status dashboard https://review.openstack.org/367324 | 12:06 |
| marios | tbarron: been a while since i've gone looking for dib :) https://github.com/openstack/diskimage-builder/tree/master/elements/source-repositories#override-per-source | 12:07 |
| *** dprince has joined #tripleo | 12:09 | |
| tbarron | marios: yeah, probably my syntax is wrong for the local repo from what I see there. | 12:09 |
| *** trown|outtypewww is now known as trown | 12:09 | |
| derekh | sshnaidm: was this before this morning promote ? we had a new mirror server with no cached images so had to build them | 12:09 |
| tbarron | marios: thank you for being so tactful, but you just made a good point. My puzzle is about DIB, not about OOO proper. | 12:09 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 12:10 |
| sshnaidm | derekh, right, missed this.. | 12:11 |
| marios | tbarron: haha np honestly man i had to go looking it has been a while ;) | 12:11 |
| jistr | jaosorior: neat, lgtm | 12:11 |
| jaosorior | shardy, jistr on the same tracks (and actually part of the chain) can you guys review this one? https://review.openstack.org/#/c/367176/1 | 12:12 |
| derekh | sshnaidm: ya, the old mirror server vanished, I think the clean up script might have cleaned it up by accident | 12:12 |
| *** bfournie has joined #tripleo | 12:12 | |
| *** maeca1 has quit IRC | 12:12 | |
| openstackgerrit | Martin André proposed openstack/instack-undercloud: Introduce 'enable_validations' option https://review.openstack.org/322893 | 12:12 |
| openstackgerrit | Martin André proposed openstack/instack-undercloud: Deploy SSH keys to overcloud in post config https://review.openstack.org/362194 | 12:12 |
| openstackgerrit | Martin André proposed openstack/instack-undercloud: Deploy validations SSH key in post config https://review.openstack.org/362194 | 12:16 |
| pabelanger | sshnaidm: derekh: I've added mirror.regionone.tripleo-test-cloud-rh1.openstack.org to cacti.o.o, if we are having networking issues, hopefully cacti will expose some data: http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=352 | 12:19 |
| *** jpena|lunch is now known as jpena | 12:21 | |
| *** jprovazn has quit IRC | 12:22 | |
| sshnaidm | pabelanger, thanks, what do mean "errors" there? | 12:22 |
| *** pradk has joined #tripleo | 12:22 | |
| sshnaidm | pabelanger, and what is host/IP of AFS PyPi mirror? | 12:23 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-heat-templates: Update capabilities-map.yaml https://review.openstack.org/364842 | 12:24 |
| rbrady | shardy: if you have time, I need a bit more info on the bug you raised about stacks and plans being coupled together. https://bugs.launchpad.net/tripleo/+bug/1609454 | 12:25 |
| openstack | Launchpad bug 1609454 in tripleo "Workflows Assuming a Single Plan is Associated with a Single Stack" [Medium,In progress] - Assigned to Ryan Brady (rbrady) | 12:25 |
| trown | shardy: if you put your +2 back on https://review.openstack.org/354016 I think we can merge it | 12:26 |
| rbrady | shardy: specifically, "If an operator creates a stack named "overcloud" with a given plan, do we need to attempt to keep them from deleting the plan used for a current running stack?" | 12:26 |
| pabelanger | sshnaidm: "error" there? | 12:26 |
| jaosorior | jistr: should we merge this commit? https://review.openstack.org/#/c/359131/1 the stuff that it actually affects passed | 12:27 |
| pabelanger | sshnaidm: mirror.regionone.tripleo-test-cloud-rh1.openstack.org is the DNS, which resolves to the current ip address | 12:27 |
| sshnaidm | pabelanger, http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=6862&rra_id=all | 12:27 |
| *** bkopilov_ has joined #tripleo | 12:27 | |
| *** dsariel has quit IRC | 12:28 | |
| pabelanger | sshnaidm: believe that will display errors related to network traffic on the specific interface | 12:28 |
| *** toure has joined #tripleo | 12:28 | |
| pabelanger | sshnaidm: not sure reported from where | 12:28 |
| EmilienM | jaosorior: hold on | 12:29 |
| jistr | jaosorior: personally i like to merge greener, but not blocking it if you want to go ahead and +A it :) i think you understand the effects better than me anyway | 12:29 |
| pabelanger | sshnaidm: I haven't actually seen any data there before on other hosts | 12:29 |
| EmilienM | jistr: right | 12:29 |
| EmilienM | jistr, jaosorior: and I fail to find in logs the results of undercloud pingtest | 12:29 |
| EmilienM | I'm still looking for it | 12:29 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 12:29 |
| *** bkopilov has joined #tripleo | 12:29 | |
| EmilienM | ok I got it http://logs.openstack.org/31/359131/1/check/gate-tripleo-ci-centos-7-undercloud/37fcf7a/logs/var/log/undercloud_install.txt.gz#_2016-09-08_09_24_03_000 | 12:30 |
| EmilienM | jaosorior: +2 on my side, let's do recheck and see how it works | 12:30 |
| jaosorior | EmilienM: right, and there we can see all the SSL warnings | 12:31 |
| openstackgerrit | Martin André proposed openstack/instack-undercloud: Deploy validations SSH key in post config https://review.openstack.org/362194 | 12:31 |
| jaosorior | EmilienM: sure. But this doesn't even touch the other jobs... so I don't really see a point | 12:31 |
| jaosorior | jistr, EmilienM: but yeah in that log Emilien showed we can see all the SSL warnings, so at least there we know it's working | 12:31 |
| jaosorior | I should probably add a filter for those warnings. They are pretty annoying | 12:31 |
| EmilienM | jaosorior: I agree | 12:32 |
| *** akshai has joined #tripleo | 12:32 | |
| sshnaidm | pabelanger, are you sure mirror.regionone.tripleo-test-cloud-rh1.openstack.org is DNS? http://paste.openstack.org/show/569187/ | 12:32 |
| EmilienM | jaosorior: I +2ed, and recheck. Let's see next round of OVB checks, otherwise we'll land it. | 12:32 |
| jaosorior | EmilienM: but it doesn't touch OVB. Or is it just to be safe? | 12:33 |
| EmilienM | jaosorior: just to be safe | 12:33 |
| EmilienM | jaosorior: we'll merge it today no worries | 12:33 |
| jaosorior | fair enough | 12:33 |
| *** honza has joined #tripleo | 12:34 | |
| pabelanger | sshnaidm: yes: mirror.regionone.tripleo-test-cloud-rh1.openstack.org. 3579 IN A 66.187.229.135 | 12:34 |
| *** honza is now known as Guest43969 | 12:34 | |
| pabelanger | http://mirror.regionone.tripleo-test-cloud-rh1.openstack.org/pypi/ | 12:34 |
| pabelanger | is pypi mirror | 12:34 |
| sshnaidm | pabelanger, well, so mirror.regionone.. is not DNS server | 12:35 |
| *** akshai_ has joined #tripleo | 12:36 | |
| pabelanger | sshnaidm: right, not a DNS server. but the DNS for the IP | 12:36 |
| sshnaidm | the hostname | 12:37 |
| d0ugal | shardy: got a moment for a question? | 12:37 |
| pabelanger | sshnaidm: sounds right | 12:38 |
| *** lucas-hungry is now known as lucasagomes | 12:38 | |
| *** akshai has quit IRC | 12:39 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Fix promote script and add logs https://review.openstack.org/367244 | 12:40 |
| openstackgerrit | Michele Baldessari proposed openstack/tripleo-heat-templates: WIP Add a function to upgrade from full HA to NG HA https://review.openstack.org/358626 | 12:40 |
| sshnaidm | pabelanger, maybe I miss something, but I don't see we use this pypi mirror in our CI scripts. Do we? | 12:43 |
| *** dsariel has joined #tripleo | 12:43 | |
| pabelanger | sshnaidm: nodepool currently sets the mirrors: http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/scripts/configure_mirror.sh | 12:46 |
| *** fultonj has joined #tripleo | 12:46 | |
| *** fultonj has quit IRC | 12:46 | |
| *** apetrich has joined #tripleo | 12:47 | |
| shardy | d0ugal: Hi, sure | 12:47 |
| d0ugal | shardy: I wrote it up as a bug: https://bugs.launchpad.net/tripleo/+bug/1621462 | 12:47 |
| openstack | Launchpad bug 1621462 in tripleo "Provide a way to update a plan after the templates are changed in swift" [Critical,Confirmed] - Assigned to Dougal Matthews (d0ugal) | 12:47 |
| *** rlandy has joined #tripleo | 12:47 | |
| d0ugal | shardy: I am trying to figure out how to update the mistral environment after the plan files are updated, I see 3 approaches that I listed in the bug. | 12:48 |
| shardy | d0ugal: ack, will check it out | 12:48 |
| d0ugal | shardy: Thanks - I think I might update my plan update patch to delete the plan if it isn't deployed - that is cleaner. | 12:49 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: DONT MERGE: WIP: Nothing here https://review.openstack.org/365324 | 12:49 |
| *** fultonj has joined #tripleo | 12:49 | |
| *** fultonj has quit IRC | 12:50 | |
| *** jaosorior has quit IRC | 12:51 | |
| *** egafford has joined #tripleo | 12:51 | |
| *** jaosorior has joined #tripleo | 12:52 | |
| *** rhallisey has quit IRC | 12:52 | |
| matbu | bandini: hello, did you figure out the issue with the converge ? | 12:53 |
| matbu | bandini: i reproduce it | 12:53 |
| matbu | just now | 12:53 |
| *** fultonj has joined #tripleo | 12:53 | |
| thrash | jaosorior: I'm at a loss | 12:53 |
| thrash | jaosorior: Don't know enough about haproxy | 12:54 |
| bandini | matbu: ah you did. I am glad I am not crazy ;) | 12:54 |
| bandini | Ng: ^ | 12:54 |
| *** pkovar1 has joined #tripleo | 12:54 | |
| Ng | aha | 12:54 |
| bandini | matbu: I tried bisecting it but I ended up on a commit that worked once and failed another time, so I am not sure yet what is the cause | 12:55 |
| bandini | matbu: my findings so far are here https://bugs.launchpad.net/tripleo/+bug/1620696 | 12:55 |
| openstack | Launchpad bug 1620696 in tripleo "M/N upgrades UPDATE_FAILED .enabled_services.list_join: Incorrect arguments to "list_join" should be: "list_join" : [ " ", [ "str1", "str2"]]" [Undecided,New] | 12:55 |
| openstackgerrit | Imre Farkas proposed openstack/python-tripleoclient: Update baremetal ready state command https://review.openstack.org/367346 | 12:55 |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-heat-templates: Composable Zaqar services https://review.openstack.org/331682 | 12:55 |
| Ng | we put some extra debugging in the list_join function and it kinda looked like all of the arguments to enabled_services were resolving to None | 12:55 |
| d0ugal | shardy: I also add a new comment here: https://bugs.launchpad.net/tripleo/+bug/1620932 | 12:56 |
| openstack | Launchpad bug 1620932 in tripleo "openstack overcloud deploy will use the old plan and wont update on a second deploy" [Critical,In progress] - Assigned to Dougal Matthews (d0ugal) | 12:56 |
| d0ugal | shardy: They are related, but still different bugs I think | 12:56 |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-heat-templates: Composable Mistral services https://review.openstack.org/323436 | 12:56 |
| *** pkovar has quit IRC | 12:56 | |
| matbu | bandini: Ng interesting | 12:56 |
| jaosorior | thrash: yeah dude, spent a bunch of time trying to figure it out and didn't make it :/ | 12:57 |
| jaosorior | that's why I asked for help :/ | 12:57 |
| matbu | bandini: Ng i'll try to debug a bit my env | 12:57 |
| Ng | awesome :) | 12:57 |
| *** shardy is now known as shardy_mtg | 12:57 | |
| jaosorior | ayoung: hey dude, got time to help me out debug something? | 12:57 |
| ayoung | jaosorior, always | 12:57 |
| ayoung | jaosorior, what is the issue? | 12:58 |
| jaosorior | ayoung: I got TLS terminated everywhere with HAProxy and am having keystone issues | 12:58 |
| *** liverpooler has quit IRC | 12:58 | |
| jaosorior | ayoung: they might be reflecting wrong haproxy configuration... I still need to determine that | 12:58 |
| jaosorior | ayoung: but when trying to do any operation with keystone. the catalog ends up messed up | 12:58 |
| jaosorior | for instance | 12:59 |
| jaosorior | the management endpoint is set as | 12:59 |
| jaosorior | https://:35357/v2.0 | 12:59 |
| EmilienM | thrash: I have 2 infos for you | 12:59 |
| *** pabelanger has quit IRC | 12:59 | |
| jaosorior | however, the public endpoint is fine and I can get tokens | 12:59 |
| ayoung | tokens usually come prior to Service catalog | 12:59 |
| EmilienM | thrash: 1) scenario jobs are now voting and 2) scenario jobs are not triggered on tripleo-ci project too, when you try to patch the scenario heat template and/or the pingtest | 12:59 |
| ayoung | the order of ops is that a user provides the Auth URL, which is used to get a token and the service catalog, which is then used for all follow on operations on Keystone | 13:00 |
| Ng | matbu: is it possible to get heat to dump out a representation of what it thinks the templates all resolve to, so we can see for sure what it's seeing there? | 13:00 |
| ayoung | so, no surprise when tokens work and other things don't | 13:00 |
| thrash | EmilienM: ack | 13:00 |
| ayoung | jaosorior, make sense? | 13:00 |
| jaosorior | ayoung: it does | 13:00 |
| thrash | EmilienM: did you mean "now triggered"? | 13:00 |
| EmilienM | thrash: yes, sorry | 13:01 |
| thrash | EmilienM: ok.. just checking. :) | 13:01 |
| *** lblanchard has joined #tripleo | 13:01 | |
| *** liverpooler has joined #tripleo | 13:01 | |
| EmilienM | thrash: which means (and I need to think about it): your THT patch for zaqar will need to land BEFORE the tripleo-ci patch that adds the zaqar resources | 13:01 |
| EmilienM | thrash: if you see what I mean | 13:01 |
| ayoung | jaosorior, as far as how Endpoints are set up in Tripleo, I thought that there was an Endpoint map variable that collects all of the Endpoints and sends them to Keystone | 13:01 |
| thrash | EmilienM: i do. | 13:02 |
| EmilienM | cool | 13:03 |
| EmilienM | thrash: if the workflow sucks or if you have ideas, please let me kow | 13:03 |
| EmilienM | thrash: I'm still experimenting | 13:03 |
| jaosorior | ayoung: yeah, it's an endpoint catalog issue | 13:04 |
| jaosorior | gotta fix that, don't know what the actual problem is | 13:04 |
| thrash | EmilienM: which probably means the tripleo-ci patch should depend on the other patch. | 13:04 |
| thrash | But, that doesn't exactly solve it. | 13:04 |
| jaosorior | ayoung: this is how it looks like http://paste.openstack.org/show/569196/ | 13:04 |
| *** Guest43969 is now known as honza | 13:05 | |
| *** jcoufal has quit IRC | 13:05 | |
| openstackgerrit | Brad P. Crochet proposed openstack-infra/tripleo-ci: Add Zaqar to scenario002 https://review.openstack.org/365026 | 13:05 |
| thrash | EmilienM: I'll think about it a bit. | 13:06 |
| ayoung | jaosorior, is the question "Where did the hostname in adminURL get to?" | 13:06 |
| *** jcoufal has joined #tripleo | 13:07 | |
| shadower | rbrady, d0ugal: the post-deployment validations need to run a copy_ssh_keys workflow after overcloud is deployed -- but only if enable_validations is set in undercloud.conf/hiera | 13:08 |
| *** ohamada has quit IRC | 13:08 | |
| shadower | rbrady, d0ugal: does mistral have access to the hiera data? | 13:08 |
| *** ohamada has joined #tripleo | 13:08 | |
| EmilienM | slagle: I reported the glance upload issue: https://bugs.launchpad.net/tripleo/+bug/1621467 | 13:09 |
| openstack | Launchpad bug 1621467 in tripleo "all-in-one overcloud with Glance + RBD: image upload hangs forever" [High,Triaged] - Assigned to Emilien Macchi (emilienm) | 13:09 |
| d0ugal | shadower: Only due to the fact it is running on the undercloud which has hiera... | 13:09 |
| jaosorior | ayoung: I think I know the issue now | 13:09 |
| ayoung | cool | 13:09 |
| jaosorior | and I think it might be fault haha | 13:09 |
| *** yamahata has joined #tripleo | 13:09 | |
| d0ugal | shadower: I don't think there is any specific support for it, you'd just have to access it like you would in Python "normally" | 13:09 |
| shadower | d0ugal: ah right. Do you think that'd be okay in this situation or could we do it differently? | 13:10 |
| d0ugal | shadower: I can't think of any other way. | 13:10 |
| shadower | d0ugal: so shell out to hiera? (or did you mean reading the hieradata directly from the filesystem?) | 13:11 |
| d0ugal | shadower: I don't know :) Do we do this anywhere else from Python? | 13:11 |
| *** Goneri has joined #tripleo | 13:11 | |
| *** dsariel has quit IRC | 13:11 | |
| shadower | d0ugal: I checked tripleo-common and couldn't find anything there. But maybe we do it elsewhere | 13:11 |
| *** jraju has joined #tripleo | 13:12 | |
| d0ugal | shadower: oh, we did do it in tripleoclient at one point... | 13:12 |
| d0ugal | shadower: https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/utils.py#L452-L462 | 13:12 |
| d0ugal | shadower: some dreadful person wrote that, seems to work tho' | 13:12 |
| *** cdearborn has joined #tripleo | 13:12 | |
| shadower | haha | 13:12 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Switch management endpoint to use actual network name https://review.openstack.org/367362 | 13:13 |
| jaosorior | ayoung: alright, redeploying with the fix, lets see if that does the trick | 13:13 |
| openstackgerrit | Dan Prince proposed openstack/python-tripleoclient: Deploy the undercloud with Heat https://review.openstack.org/351351 | 13:14 |
| *** shardy_mtg is now known as shardy | 13:14 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: DONT MERGE: WIP: Nothing here https://review.openstack.org/365324 | 13:15 |
| d0ugal | shadower: haha, it isn't even used now! | 13:15 |
| shadower | d0ugal: all right, I'll whip something up and have you and Ryan tear it apart in gerrit | 13:16 |
| shardy | shadower: can't you just install the ssh keys via cloud-init instead of an independent workflow? | 13:16 |
| shardy | or via one of the post-deploy ExtraConfig interfaces if it needs to happen after the services are up? | 13:16 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Remove the get_hiera_key function https://review.openstack.org/367367 | 13:17 |
| shadower | shardy: so mandre has written that and I don't remember the details. But: iirc this is supposed to be optional (only when you enable validations) | 13:18 |
| shardy | shadower: sure, so you could have an environment -e enable_validations.yaml which configures things to deploy an additional key | 13:18 |
| shadower | it would make our lives much easier if we could just always copy the keys in | 13:18 |
| *** jprovazn has joined #tripleo | 13:19 | |
| shardy | shadower: well we do always copy *a* key in, so can you reuse that by putting the private key on the undercloud? | 13:19 |
| *** pgadiya has quit IRC | 13:19 | |
| shardy | I know dprince was opposed to requiring that, but it seems like it's going to be required for validations to work anyway | 13:20 |
| shadower | mandre: wasn't dprince against that? ^ | 13:20 |
| mandre | shardy: afaik, you can only copy one key with cloud init, and we already copy the current user's key | 13:20 |
| shardy | mandre: No, it's possible to copy another key IIRC but you need to write a script to do it | 13:20 |
| * shardy finds example | 13:20 | |
| dprince | shardy: I don't like setting up a camp on the undercloud. If we start having single directories for keys on disk now it will cause problems for HA underclouds... if and when we have them. Furthermore it just seems like having to do that is a bad practice | 13:21 |
| *** jprovazn has quit IRC | 13:21 | |
| *** jprovazn has joined #tripleo | 13:22 | |
| mandre | shardy: then the limitation was somewhere else, heat templates maybe? | 13:22 |
| shardy | dprince: I agree - but shadower is talking about doing it anyway when validations are enabled, so I don't really get the difference between creating a new key for validations vs documenting that the private key must exist if you enable that feature | 13:22 |
| shardy | mandre: what limitation? | 13:22 |
| *** sshnaidm is now known as sshnaidm|afk | 13:23 | |
| mandre | shardy: providing more than one key to the provisioned node | 13:23 |
| shardy | mandre: let me try, as far as I can remember it is possible | 13:24 |
| *** yamahata has quit IRC | 13:25 | |
| shadower | shardy: but as far as I understand it, this doesn't change the original question, which is: how do we figure out whether the keys should be uploaded | 13:27 |
| openstackgerrit | Merged openstack/instack-undercloud: Disable unsupported drivers and fix DRAC dependency https://review.openstack.org/367237 | 13:27 |
| shardy | shadower: you make a deploy time decision, either based on the undercloud configuration, or by asking the user (e.g expecting them to opt-in by including some configuration) | 13:27 |
| *** sshnaidm|afk has quit IRC | 13:27 | |
| *** tzumainn has joined #tripleo | 13:28 | |
| *** liverpooler has quit IRC | 13:28 | |
| shardy | shadower: note you wouldn't necessarily have to shell out to hiera from mistral, it could be enough to have the undercloud install write a file to ~/.tripleo/environments | 13:29 |
| shardy | which would then get automatically picked up every deploy | 13:29 |
| shardy | ~/.tripleo/environments/validation_enabled.yaml | 13:29 |
| shardy | I guess that won't work for the UI tho | 13:30 |
| shadower | yeah | 13:30 |
| *** liverpooler has joined #tripleo | 13:30 | |
| *** pgadiya has joined #tripleo | 13:32 | |
| openstackgerrit | Julie Pichon proposed openstack/tripleo-common: Clearer error when the Mistral env already exists https://review.openstack.org/367379 | 13:33 |
| *** pgadiya has quit IRC | 13:33 | |
| *** pkovar1 has quit IRC | 13:35 | |
| mandre | shardy: since I got your attention... I have a totally unrelated question :) | 13:35 |
| * shardy hides ;) | 13:35 | |
| mandre | shardy: my overcloud deploy fails with "Could not find data item memcached_node_ips in any Hiera data file" | 13:35 |
| mandre | shardy: I know you've been around this part of the code recently | 13:35 |
| *** redhatkj has quit IRC | 13:37 | |
| *** jayg|g0n3 is now known as jayg | 13:39 | |
| shardy | mandre: sounds like you're running a puppet-tripleo with https://github.com/openstack/puppet-tripleo/commit/932a9f0409395e1f7e8d83efdd6418d55b040370 | 13:39 |
| *** mbound_ has joined #tripleo | 13:39 | |
| mandre | shardy: you've renamed memcache_node_ips to memcached_node_ips in https://review.openstack.org/#/c/353582/ | 13:39 |
| shardy | but without the corresponding tripleo-heat-templates change | 13:39 |
| EmilienM | what happens if I deploy ceph and ommit to declare --ceph-storage-scale ? | 13:39 |
| *** bvandenh_ has quit IRC | 13:39 | |
| EmilienM | omit even | 13:39 |
| shardy | mandre: Yeah, either your templates or puppet-tripleo are out of date | 13:40 |
| shardy | I made corresponding changes in both places | 13:40 |
| mandre | shardy: ok, I though I was running master for both, I'll double check | 13:40 |
| shardy | EmilienM: I don't think it will work unless you put the OSD service onto another node | 13:40 |
| mandre | shardy: thanks for the help | 13:41 |
| EmilienM | shardy: I put the OSD on another node | 13:41 |
| shardy | EmilienM: gfidente has tested that and it does work, but you need to have an environment which moves the OSD service into a *Services list for a role that is getting deployed | 13:41 |
| EmilienM | shardy: I'm investigating https://bugs.launchpad.net/tripleo/+bug/1621467 | 13:41 |
| openstack | Launchpad bug 1621467 in tripleo "all-in-one overcloud with Glance + RBD: image upload hangs forever" [High,Triaged] - Assigned to Emilien Macchi (emilienm) | 13:41 |
| *** mbound has quit IRC | 13:41 | |
| shardy | EmilienM: ack, that should work AFAIK | 13:41 |
| EmilienM | shardy: can you review https://review.openstack.org/#/c/366810/6/test-environments/scenario001-multinode.yaml ? | 13:41 |
| *** weshay is now known as weshay_pto | 13:41 | |
| *** mburned is now known as mburned_out | 13:43 | |
| mandre | shardy: in master t-h-t, I can only find memcached_node_ips_v6 in hieradata but no sign of memcached_node_ips | 13:44 |
| mandre | shardy: am I looking at the right place? | 13:44 |
| shardy | mandre: Yeah, we auto-generate those hiera keys now based on the service_name in puppet/services/foo.yaml | 13:44 |
| mandre | shardy: oh.. I see | 13:44 |
| *** mburned_out is now known as mburned | 13:45 | |
| shardy | mandre: https://github.com/openstack/tripleo-heat-templates/blob/master/network/ports/net_ip_list_map.yaml#L65 | 13:46 |
| *** pkovar has joined #tripleo | 13:46 | |
| shardy | that's where it happens | 13:47 |
| shardy | all services which are enabled and have an entry in the ServiceNetMap get $service_node_ips added via allNodesConfig | 13:47 |
| EmilienM | shardy: I'm not sure ceph has been tested all in one before | 13:49 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fix ansible-lint errors in all playbooks and roles https://review.openstack.org/366837 | 13:49 |
| mandre | shardy: good to know :) | 13:49 |
| shardy | EmilienM: https://review.openstack.org/#/c/338088/ | 13:50 |
| shardy | gfidente definitely tested it, and at one time he said it worked | 13:50 |
| shardy | well, that's hyperconverged not all-in-one | 13:51 |
| shardy | so perhaps there are issues specific to co-location with the controller services | 13:51 |
| EmilienM | shardy: i'll look thanks | 13:51 |
| shardy | EmilienM: when we land https://review.openstack.org/#/c/365763/ the issues re steps there will be resolved | 13:52 |
| EmilienM | shardy: nice | 13:53 |
| EmilienM | shardy: is it safe to land https://review.openstack.org/#/c/364748/ even if OVB is red? | 13:53 |
| shardy | bnemec: Hey, any idea if we can expect ovb jobs to start going green again today? | 13:53 |
| shardy | I'm not sure we can wait much longer with RC1 looming, we might have to rely on the multinode job | 13:54 |
| EmilienM | shardy: maybe in my case I need the "ComputeServices: merge" ? | 13:54 |
| EmilienM | shardy: FYI scenario jobs are now voting. | 13:54 |
| EmilienM | so we have 4 multinode jobs (3 running only when required) | 13:55 |
| EmilienM | and 1 for undercloud, so 5 | 13:55 |
| *** pkovar has quit IRC | 13:55 | |
| shardy | EmilienM: Yeah we've not yet adopted the new heat environment merging, so you need to copy the existing *Services list then add the ceph services | 13:55 |
| *** andrey-mp has joined #tripleo | 13:55 | |
| *** fultonj has quit IRC | 13:55 | |
| *** liverpooler has quit IRC | 13:55 | |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Re-add undercloud.yaml https://review.openstack.org/352037 | 13:55 |
| EmilienM | shardy: can you look https://review.openstack.org/#/c/366810/6/test-environments/scenario001-multinode.yaml please? Did I miss something? | 13:56 |
| EmilienM | shardy: do I need to merge something or? | 13:56 |
| shardy | EmilienM: I think you have a duplicate parameter_defaults key | 13:57 |
| shardy | not sure that's related tho as the last one should win when the yaml is parsed | 13:57 |
| andrey-mp | Hi! is it possible to understand node's role in NodeExtraConfigPost? | 13:57 |
| EmilienM | shardy: ah right | 13:58 |
| *** fultonj has joined #tripleo | 13:58 | |
| EmilienM | shardy: it's a typo, but I don't think it's related to my ceph issue | 13:58 |
| shardy | EmilienM: otherwise that looks OK to me | 13:58 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario001: deploy Ceph https://review.openstack.org/366810 | 13:59 |
| jrist | shadower: did you see this? https://bugs.launchpad.net/tripleo/+bug/1620329 | 14:00 |
| openstack | Launchpad bug 1620329 in tripleo "The network-environment validator crashes if the cidrs are missing" [Undecided,New] | 14:00 |
| shardy | andrey-mp: Not directly, can you look at the hostname to indirectly derive it? | 14:01 |
| shadower | jrist: I haven't yet. Adding it to my queue | 14:01 |
| jrist | thx shadower | 14:01 |
| shardy | andrey-mp: if you're using puppet, we could perhaps wire in a hiera key which tells you the role | 14:02 |
| *** rajinir has joined #tripleo | 14:02 | |
| andrey-mp | shardy: is it predefined key? | 14:02 |
| shardy | andrey-mp: we don't write the hiera key now, I'm saying we could possibly add it | 14:03 |
| shardy | andrey-mp: for now the easiest way is to look for e.g "controller" in the hostname | 14:03 |
| *** pkovar has joined #tripleo | 14:03 | |
| shardy | unless you're overriding the node names to non-role-related values | 14:03 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario002: deploy Swift https://review.openstack.org/367401 | 14:03 |
| *** mat128 has left #tripleo | 14:05 | |
| andrey-mp | shardy: ok. got it. i try to setup my software on the overcloud and as I understood I can use only all-node-post-config (there are no controller/compute post configs). so i need to know role of machine... | 14:05 |
| andrey-mp | shardy: when i can set hostnames for ironic nodes? | 14:05 |
| andrey-mp | when -> where | 14:05 |
| shardy | andrey-mp: There are two ways, either set the *HostnameFormat parameters to something other than the default, or pass in an explicit HostnameMap which overrides the default names: | 14:07 |
| shardy | http://docs.openstack.org/developer/tripleo-docs/advanced_deployment/node_placement.html#custom-hostnames | 14:07 |
| ansiwen | chem, mwhahaha: can you tell me what's going wrong here? http://logs.openstack.org/80/364580/5/check/gate-puppet-tempest-puppet-unit-3.6-centos-7/26b0a0e/console.html | 14:08 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud.yaml#L319 | 14:08 |
| andrey-mp | shardy: ok, thanks | 14:08 |
| shardy | andrey-mp: ^^ | 14:08 |
| *** noslzzp has quit IRC | 14:08 | |
| *** noslzzp has joined #tripleo | 14:09 | |
| trown | derekh: could you point me to the IP of the current mirror server? | 14:09 |
| jrist | shadower: is there a hard drive space validator? | 14:09 |
| *** _coolsvap_ is now known as coolsvap | 14:09 | |
| mwhahaha | ansiwen: it's erroring because thats not what the code actually does | 14:09 |
| shadower | jrist: https://github.com/openstack/tripleo-validations/blob/master/validations/undercloud-disk-space.yaml | 14:10 |
| shadower | i.e. yep | 14:10 |
| jrist | nice | 14:10 |
| jrist | thanks :) | 14:10 |
| jrist | thought so | 14:11 |
| shadower | -) | 14:11 |
| derekh | trown: sorry we had to bring up a new one yesterday http://66.187.229.139/ | 14:11 |
| derekh | trown: what was the old IP http://66.187.229.18 ? let me see if I can get that back | 14:11 |
| ansiwen | mwhahaha: ok, I just saw the static aws config is added, that test succeeds... but it fails with the "creates ec2 credentials" test | 14:11 |
| jrist | shadower: how about overcloud? | 14:11 |
| mwhahaha | ansiwen: so the unit tests check the catalog, and your provider is writing out to the file directly. You'd need to write a unit test for your provider if you want to ensure the value is written out correctly | 14:11 |
| mwhahaha | ansiwen: I'll comment on the unit test as to what needs to get fixed but if you want to validate it's going to write out to the files correctly you need a unit test for your provider | 14:12 |
| trown | derekh: oh now worries, I think I may have had even a different one | 14:12 |
| trown | derekh: https://github.com/openstack/tripleo-quickstart/blob/master/config/release/master-tripleo.yml#L3 | 14:12 |
| ansiwen | mwhahaha: I'm afraid I don't understand >_< | 14:13 |
| shadower | jrist: not yet. It's a little more complex -- we want to do it by talking to ironic (post-deployment would be too late) plus different roles would require different disk sizes | 14:13 |
| trown | derekh: but I am returning to the idea of booting an overcloud-full.qcow2 as an undercloud | 14:13 |
| mwhahaha | ansiwen: ok i'll dig up some examples too | 14:13 |
| ansiwen | mwhahaha: the other tests also check for configs in the tempest.conf, why doesn't it work in my case? | 14:13 |
| mandre | shardy: I'm still puzzled about that missing memcached_node_ips hiera data, I have a recent checkout of t-h-t that I'm deploying with on the undercloud, and I'm using master puppet-tripleo that I deploy to the overcloud with the mechanism you described in https://hardysteven.blogspot.fr/2016/08/tripleo-deploy-artifacts-and-puppet.html | 14:13 |
| thrash | jaosorior: so, it's definitely just related to ssl... (Which you've probably already determined) | 14:14 |
| bnemec | shardy: I'm not sure. It depends on how long it takes to work out these networking issues. AFAICT that's the major blocker right now. | 14:14 |
| mwhahaha | ansiwen: if your provider actually called tempest_config you could do what you're doing but it doesn't | 14:14 |
| shardy | mandre: do you have the latest tripleo-common for your upload-puppet-modules script? | 14:14 |
| thrash | jaosorior: i replaced the endpoint with non-ssl and it works. | 14:14 |
| shardy | mandre: https://review.openstack.org/#/c/365783/ | 14:15 |
| derekh | trown: that IP was on RH2 and not being updated, hopefully we can leave it up until you can make use of the overcloud-full image | 14:15 |
| mandre | shardy: nope, that's probably it! I knew I was missing a step | 14:15 |
| mwhahaha | ansiwen: the unit tests check the output of the puppet catalog and your provider is not manipulating the catalog but rather writing out directly to the file | 14:15 |
| ansiwen | mwhahaha: well, I just copied it from the setters... could I call tempest_config instead of writing to the file directly? | 14:15 |
| shardy | mandre: Yeah, we broke the modules upload when switching to mistral for plan storage, that patch works around it | 14:15 |
| mwhahaha | ansiwen: you can but you still need a unit test for the provider itself | 14:16 |
| d0ugal | dprince: Hey, if you get a moment I could do with input on https://bugs.launchpad.net/tripleo/+bug/1621462 and https://bugs.launchpad.net/tripleo/+bug/1620932 | 14:16 |
| openstack | Launchpad bug 1621462 in tripleo "Provide a way to update a plan after the templates are changed in swift" [Critical,Confirmed] - Assigned to Dougal Matthews (d0ugal) | 14:16 |
| trown | derekh: k, if it goes away it is not such a big deal, devmode is in pretty rough shape at the moment anyways | 14:16 |
| openstack | Launchpad bug 1620932 in tripleo "openstack overcloud deploy will use the old plan and wont update on a second deploy" [Critical,In progress] - Assigned to Dougal Matthews (d0ugal) | 14:16 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 14:16 |
| dprince | d0ugal: looking | 14:16 |
| derekh | trown: ok | 14:16 |
| ansiwen | mwhahaha: also I was not sure which user/project I can use. are there predefined one? I just chose random names for now | 14:17 |
| d0ugal | dprince: Thanks, both have a bit of discussion in them (from my ideas) but I'm not sure which is the best choice. | 14:17 |
| EmilienM | note for everyone, I see gate-tripleo-ci-centos-7-scenario001-multinode failing 'sometimes', when creating Gnocchi metrics. The job is now voting. If you see too many scenario001 failure, please let me know we'll disable voting on this one | 14:17 |
| mwhahaha | ansiwen: in the test it can just be random ones. or do you mean in the manifest? | 14:17 |
| trown | derekh: http://66.187.229.139/builds/current-tripleo/instack.qcow2 is copied or actually produced? | 14:17 |
| jaosorior | thrash: yep | 14:18 |
| jaosorior | thrash: that's the case | 14:18 |
| jaosorior | thrash: aaand we can't break ssl ;) | 14:18 |
| derekh | trown: it was copied, let me find out from when | 14:19 |
| thrash | jaosorior: lol | 14:19 |
| thrash | jaosorior: gonna try something on the client side right quick. | 14:19 |
| ansiwen | mwhahaha: in the test instantiation of the tempest class... so the users will be crated implicitly? because there is the openstack request with the --user and --project parameters... I thought, maybe I have to create the user and project first. | 14:19 |
| *** Guest_84848 has joined #tripleo | 14:19 | |
| trown | derekh: ah ok, cool, just wanted to make sure we didnt change to producing it again | 14:19 |
| jaosorior | thrash: I even went to the zaqar channel and asked... but nobody answered :( | 14:19 |
| jaosorior | thrash: we really need someone that actually knows websocket internals | 14:20 |
| trown | derekh: dont worry about when it was copied, not a big deal | 14:20 |
| mwhahaha | ansiwen: the unit test won't actually execute it | 14:20 |
| Guest_84848 | allah is doing | 14:20 |
| dprince | d0ugal: I like your patch for re-deploying (i.e. re-creating a plan) | 14:20 |
| Guest_84848 | sun is not doing allah is doing | 14:20 |
| dprince | d0ugal: https://review.openstack.org/#/c/366541/4 +2 from me | 14:20 |
| Guest_84848 | moon is not doing allah is doing | 14:20 |
| ansiwen | mwhahaha: ok, and how can then a reasonable value be returned? | 14:20 |
| Guest_84848 | stars are not doing allah is doing | 14:20 |
| Guest_84848 | planets are not doing allah is doing | 14:21 |
| Guest_84848 | galaxies are not doing allah is doing | 14:21 |
| mwhahaha | ansiwen: I commented on the review | 14:21 |
| Guest_84848 | oceans are not doing allah is doing | 14:21 |
| d0ugal | dprince: cool, i'm fairly happy with that one. I am less sure about updating the mistral-env to match the new plan files (the other bug) | 14:21 |
| Guest_84848 | mountains are not doing allah is doing | 14:21 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Update plan files when re-deploying https://review.openstack.org/366541 | 14:21 |
| Guest_84848 | trees are not doing allah is doing | 14:22 |
| Guest_84848 | mom is not doing allah is doing | 14:22 |
| Guest_84848 | dad is not doing allah is doing | 14:22 |
| Guest_84848 | boss is not doing allah is doing | 14:23 |
| shadower | any ops around here? | 14:23 |
| Guest_84848 | job is not doing allah is doing | 14:23 |
| *** sshnaidm|afk has joined #tripleo | 14:23 | |
| *** sshnaidm|afk is now known as sshnaidm | 14:23 | |
| *** zoliXXL is now known as zoli|brb | 14:23 | |
| dtantsur | is it Friday already? | 14:23 |
| derekh | trown: actually I got no idea where that came from, sshnaidm ? probably copied it from the old mirror server | 14:23 |
| d0ugal | lol | 14:23 |
| Guest_84848 | dollar is not doing allah is doing | 14:23 |
| dtantsur | nova is not doing tripleo is doing | 14:23 |
| dtantsur | cinder is not doing tripleo is doing | 14:23 |
| trown | hahaha | 14:23 |
| dtantsur | meh, I like it, I will respond this way to all questions | 14:23 |
| shardy | \/ignore Guest_84848 | 14:24 |
| sshnaidm | derekh, what are talking about? | 14:24 |
| trown | sshnaidm: http://66.187.229.139/builds/current-tripleo/instack.qcow2 | 14:24 |
| dtantsur | shardy, thanks for reminder | 14:24 |
| dprince | d0ugal: commented here as well https://bugs.launchpad.net/tripleo/+bug/1621462 | 14:24 |
| openstack | Launchpad bug 1621462 in tripleo "Provide a way to update a plan after the templates are changed in swift" [Critical,Confirmed] - Assigned to Dougal Matthews (d0ugal) | 14:24 |
| *** bnemec has quit IRC | 14:24 | |
| derekh | sshnaidm: the instack.qcow2 file on the new mirror server did you put it there? | 14:24 |
| d0ugal | dprince: Thanks. | 14:24 |
| sshnaidm | trown, yeah, I've copied it from rh2 | 14:24 |
| shardy | Guest_84848 is not doing, ignore is doing ;) | 14:24 |
| * dprince hopes this is somewhat helpful | 14:24 | |
| sshnaidm | derekh, ^ | 14:24 |
| dtantsur | lol | 14:24 |
| derekh | sshnaidm: cool, +1 | 14:25 |
| sshnaidm | trown, derekh isn't it good? | 14:25 |
| shadower | shardy: ooh, I didn't know about /ignore. Thanks | 14:25 |
| derekh | sshnaidm: yup, I was just confused of how it got there, all is good ;-) | 14:25 |
| sshnaidm | derekh, ok :) | 14:26 |
| jpeeler | d0ugal: were you planning on implementing https://bugs.launchpad.net/tripleo/+bug/1621097 ? if not, i'll take it | 14:26 |
| openstack | Launchpad bug 1621097 in tripleo "Password generation should be moved from tripleoclient to mistral workflows" [High,Confirmed] | 14:26 |
| sshnaidm | derekh, the previous was erased by cleanup script, I've also updated the promote.sh | 14:26 |
| thrash | jaosorior: fyi... wsdump == websocket client debug tool | 14:26 |
| derekh | sshnaidm: ok | 14:26 |
| jaosorior | thrash: nice! | 14:26 |
| sshnaidm | derekh, https://review.openstack.org/#/c/367244/ | 14:26 |
| derekh | trown: this was the origional build of the file BTW june 30th http://8.43.87.241/builds/current-tripleo-20160630/ | 14:27 |
| d0ugal | jpeeler: I was, but I would be very happy for somebody else to do it :) | 14:27 |
| d0ugal | jpeeler: because I already feel fairly swamped | 14:27 |
| jpeeler | ok i'll take it | 14:27 |
| thrash | jaosorior: I think I have a lead... look at the output of wsdump -vv wss://192.0.2.2:14000 | 14:27 |
| *** bnemec has joined #tripleo | 14:27 | |
| d0ugal | jpeeler: oh, sorry - rbrady might be looking at that one | 14:27 |
| * d0ugal had the wrong tab open | 14:28 | |
| thrash | jaosorior: the initial request is to http://192.0.2.2:14000 | 14:28 |
| slagle | rbrady: in DeployStackAction, i thought self.container would be the name of the plan that was used? | 14:28 |
| d0ugal | jpeeler, rbrady - but happy for either or both of you to look at it. | 14:28 |
| d0ugal | jpeeler, rbrady - I am also looking for help with https://review.openstack.org/365625 | 14:28 |
| jaosorior | thrash: I guess that's not gonna work, it should be from wss no? | 14:29 |
| jaosorior | thrash: or at least https | 14:29 |
| derekh | sshnaidm: +2 | 14:29 |
| thrash | jaosorior: https I think is the key... | 14:29 |
| jaosorior | thrash: do you have a machine I can also see what you're doing at? | 14:29 |
| thrash | jaosorior: sure. | 14:29 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: DONT MERGE: WIP: Nothing here https://review.openstack.org/365324 | 14:29 |
| *** Guest_84848 has quit IRC | 14:32 | |
| jpeeler | d0ugal: with just passing CI or what (365625) | 14:32 |
| rbrady | slagle: correct. | 14:34 |
| d0ugal | jpeeler: it's known to be broken, it was merged and reverted. I then put it up again to work on it but have not gotten there yet. | 14:34 |
| *** rodrigods has quit IRC | 14:35 | |
| *** rodrigods has joined #tripleo | 14:35 | |
| jpeeler | d0ugal: ok i see, i'll take a look | 14:35 |
| d0ugal | jpeeler: rbrady probably knows more about that one, he wrote the original patch. | 14:35 |
| d0ugal | jpeeler: This is the bug for it: https://bugs.launchpad.net/tripleo/+bug/1621099 | 14:36 |
| openstack | Launchpad bug 1621099 in tripleo "Deploy parameters ashould be set in the mistral workflow" [High,Confirmed] | 14:36 |
| d0ugal | jpeeler: (not much detail, I can add some if it isn't clear) | 14:36 |
| *** anshul has quit IRC | 14:37 | |
| *** scarab_ has joined #tripleo | 14:38 | |
| *** scarab_ has left #tripleo | 14:38 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-common: Run copy_ssh_keys after overcloud finishes https://review.openstack.org/367422 | 14:39 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: [WIP] Set Deployment Parameters https://review.openstack.org/365625 | 14:39 |
| d0ugal | jpeeler: ^ just linked it to the bug and added a link in the bug to where this is in the CLI now. | 14:41 |
| jpeeler | thanks! | 14:41 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-common: Run copy_ssh_keys after overcloud finishes https://review.openstack.org/367422 | 14:43 |
| derekh | bnemec: sshnaidm latest theory ::: random things are failing because 8.8.8.8 is either trottling us or packets are being lost | 14:43 |
| derekh | bnemec: sshnaidm I've seem random dns errors on various ci jobs, on the overcloud controller and in the te-broker the point being they can happen anyways and may be the reason we havn't been able to nail down whats failing | 14:43 |
| derekh | > timeout 60 tcpdump -B 4096 -i em1 -nn dst host 8.8.8.8 | 14:44 |
| derekh | 9920 packets captured | 14:44 |
| derekh | bnemec: sshnaidm thats 165 dns requests per second too 8.8.8.8 over the last minute... | 14:44 |
| bnemec | derekh: Yeah, I can see where they might not like that. | 14:45 |
| sshnaidm | derekh, yeah, I saw the same errors, I was sure we have some internal dns there | 14:45 |
| *** limao has joined #tripleo | 14:45 | |
| bnemec | derekh: Although I don't understand the temporary dns failure messages though. The overcloud endpoint is in /etc/hosts on te-broker. :-/ | 14:45 |
| sshnaidm | bnemec, I think it's problem of resolving repositories and pip server | 14:46 |
| sshnaidm | at least those errors I saw | 14:46 |
| derekh | bnemec: I put it there this morning, I also put it on the overcloud controller yesterday(I think it was yesterday) | 14:47 |
| bnemec | derekh: Ah, that probably explains it. | 14:47 |
| *** limao has quit IRC | 14:47 | |
| * derekh doesn't know if there is another dns server we can use at the moment | 14:48 | |
| sshnaidm | derekh, maybe to add 8.8.4.4 | 14:48 |
| bnemec | derekh: Where are those 165 dns requests coming from? Do we have any idea? | 14:48 |
| derekh | bnemec: most of them look to be cloud instances, the tcpdump shows their FIP, I'll run it for another minute an log them | 14:49 |
| derekh | sshnaidm: 8.8.4.4 might be a good backup, a local option would be great though, there might be one already if we look | 14:50 |
| jaosorior | ayoung: duuuuuude, keystone works :D | 14:50 |
| jaosorior | ayoung: and so does nova | 14:50 |
| jaosorior | ayoung: seems that TLS everywhere in HAProxy broke glance though | 14:50 |
| ayoung | jaosorior, duuuuuude no it doesn't! | 14:50 |
| ayoung | Ah | 14:51 |
| bnemec | derekh: Ah, so maybe the problem is that previously we weren't allowing DNS traffic out of the instances? | 14:51 |
| bnemec | And now that we are it's pissing off the googles. | 14:51 |
| derekh | bnemec: we weren't allowing ...? | 14:51 |
| jaosorior | ayoung: I meant haproxy terminating all endpoints with SSL | 14:51 |
| ayoung | jaosorior, I know. I am just a bitter man/ | 14:52 |
| derekh | "If queries from a specific source IP address exceed the maximum QPS, or exceed the average bandwidth or amplification limit consistently (the occasional large response will pass), we return (small) error responses or no response at all." | 14:52 |
| jaosorior | now I just need to merge a shit ton a patches :D | 14:52 |
| ayoung | jaosorior, yesterday I got an overcloud deploy where all of the nodes were IPA enrolled. We are getting closer | 14:52 |
| jaosorior | ayoung: nice! | 14:52 |
| derekh | bnemec: sshnaidm from google ^ | 14:52 |
| jaosorior | ayoung: well, TLS is IPA powered ;) | 14:52 |
| bnemec | derekh: I thought the testenvs didn't have access to get out to external DNS, which was why the ssh connections in postci took so long. | 14:52 |
| ayoung | is that a Metric shit ton or imperial? | 14:52 |
| slagle | derekh: bnemec : there is named running on the bastion. i think kosh had mentioned earlier we could use that if we needed | 14:53 |
| jaosorior | ayoung: all hail metric | 14:53 |
| ayoung | (2,205 lb). | 14:53 |
| sshnaidm | derekh, bnemec we used before 10.1.8.10, what is it was? | 14:54 |
| derekh | bnemec: back in the old system? at one stage they were configured with a DNS server that didn't exist so we had to wait for a timeout | 14:54 |
| derekh | [centos@jumphost ~]$ host -v -t A git.openstack.org 10.1.8.10 | 14:54 |
| derekh | Received 138 bytes from 10.1.8.10#53 in 0 ms | 14:54 |
| derekh | she runs like a hot snot, lets do it | 14:55 |
| *** andrey-mp has left #tripleo | 14:55 | |
| bnemec | 0 ms! | 14:55 |
| derekh | this one is infinity times slower | 14:55 |
| derekh | Received 138 bytes from 10.1.8.10#53 in 1 ms | 14:55 |
| shardy | d0ugal: when an error happens running an action in the plan create workflow, I just get "Exception creating plan:" | 14:55 |
| sshnaidm | derekh, I've got 3 :) | 14:56 |
| shardy | d0ugal: there's a backtrace and exception in the mistral logs from the action, what do I need to raise to get that error surfaced via the client? | 14:56 |
| *** zoli|brb is now known as zoli | 14:56 | |
| sshnaidm | derekh, bnemec so 10.1.8.1 and 8.8.8.8, 8.8.4.4 as backups? | 14:56 |
| *** zoli is now known as zoliXXL | 14:56 | |
| *** rhallisey has joined #tripleo | 14:56 | |
| *** dprince has quit IRC | 14:57 | |
| sshnaidm | bnemec, derekh like here: https://github.com/openstack-infra/tripleo-ci/blob/master/toci_instack.sh#L122 | 14:57 |
| derekh | sshnaidm: hmm, we're doing something that here scripts/common_functions.sh: echo -e "nameserver 10.1.8.10\nnameserver 8.8.8.8" | sudo dd of=$MOUNTDIR/etc/resolv.conf | 14:57 |
| *** dprince has joined #tripleo | 14:57 | |
| derekh | oo, that image update | 14:58 |
| *** oshvartz has quit IRC | 14:58 | |
| sshnaidm | derekh, yeah, it's not on undercloud afaiu | 14:58 |
| d0ugal | shardy: hmm, https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/workflows/plan_management.py#L68 | 14:59 |
| d0ugal | shardy: so there is no message in the payload - it might be that we need to catch the exception in the action and then return that? I'd need to take a look | 14:59 |
| d0ugal | shardy: I can do shortly, just about to go into a meeting | 14:59 |
| derekh | sshnaidm: bnemec slagle we could change the network on the overcloud (the rh1 overcloud) | 14:59 |
| *** jraju has quit IRC | 14:59 | |
| shardy | d0ugal: ack, thanks, not urgent | 15:00 |
| *** mcornea has quit IRC | 15:00 | |
| bnemec | derekh: To what? | 15:01 |
| derekh | bnemec: this one http://paste.openstack.org/show/569241/ | 15:01 |
| slagle | derekh: you mean set the nameserver in the neutron subnet? | 15:01 |
| derekh | bnemec: Change 8.8.8.8 -> 10.1.8.10 | 15:01 |
| derekh | slagle: yup | 15:01 |
| slagle | yea | 15:01 |
| bnemec | derekh: Oh, gotcha. Yeah, we should do that. | 15:01 |
| slagle | 10.1.8.10 is indeed the bastion ip, which we should probably be using | 15:01 |
| derekh | that should get the majority of the traffic | 15:02 |
| derekh | and we were using it in the old instack job before OVB | 15:03 |
| *** akshai_ has quit IRC | 15:04 | |
| derekh | bnemec: sshnaidm slagle changed the subnet patch incomming | 15:06 |
| bnemec | WTF? I just got prompted for a password during the undercloud install. :-/ | 15:09 |
| *** ebarrera has quit IRC | 15:09 | |
| d0ugal | :/ | 15:09 |
| d0ugal | bnemec: Does your user have sudo? | 15:10 |
| bnemec | And it accepted "password". | 15:10 |
| d0ugal | lol | 15:10 |
| bnemec | d0ugal: Yes, it's a base centos 7 cloud image. | 15:10 |
| bnemec | The prompt is coming from osc. | 15:10 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 15:10 |
| dtantsur | bnemec, I'd say OSC somewhat regressed in terms of authentication | 15:12 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use a DNS server local to rh1 https://review.openstack.org/367444 | 15:12 |
| dtantsur | bnemec, e.g. I could not make auth with token work with 3.2 | 15:12 |
| EmilienM | ok scenario001 is super unstable | 15:13 |
| EmilienM | I'm disabling voting on it to not block our CI | 15:13 |
| EmilienM | gnocchi is failing | 15:13 |
| EmilienM | we also see it in Puppet CI | 15:13 |
| *** akshai has joined #tripleo | 15:14 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Fix some Typos https://review.openstack.org/367448 | 15:14 |
| EmilienM | that patch will disable voting for scenario001 until we stabilize it: https://review.openstack.org/367453 | 15:15 |
| EmilienM | please run recheck only when this patch is merged | 15:15 |
| shardy | d0ugal: Don't worry re the error thing, I think I've figured it out | 15:15 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-common: Run copy_ssh_keys after overcloud finishes https://review.openstack.org/367422 | 15:15 |
| trown | EmilienM: I wonder if we shouldnt also disable gnocchi by default for all of tripleo | 15:15 |
| d0ugal | shardy: ack | 15:16 |
| EmilienM | trown: what other error have you seen? | 15:16 |
| trown | EmilienM: I cant reproduce locally, but on slower hardware deploy takes more than 90 minutes, and gnocchi CPU usage seems responsible | 15:16 |
| EmilienM | :( | 15:17 |
| EmilienM | pradk: ^ fyi | 15:17 |
| sshnaidm | bnemec, the same for me for a last week | 15:17 |
| pradk | EmilienM, this should help ..https://review.openstack.org/#/c/367436/ | 15:18 |
| sshnaidm | bnemec, seems like it's os client config tries to authenticate, it's something from last week changes | 15:18 |
| pradk | EmilienM, atleast the issue we saw | 15:18 |
| sshnaidm | bnemec, it accepts anything you type, even empty.. | 15:18 |
| bnemec | sshnaidm: Ugh. I suppose we don't see it in CI because there's no terminal so it doesn't bother prompting. | 15:18 |
| pradk | trown, do you have any numbers to support that? | 15:18 |
| *** mbound_ has quit IRC | 15:18 | |
| sshnaidm | bnemec, right, it detects the shell | 15:18 |
| pradk | trown, in what cases, is it apache or the services | 15:19 |
| sshnaidm | bnemec, I'm thinking how to workaround this, it breaks local scripts | 15:19 |
| *** akshai has quit IRC | 15:19 | |
| pradk | trown, jus saying gnocchi cpu usage doesnt help at all | 15:19 |
| bnemec | sshnaidm: Yeah, this is kind of a serious problem. Do we have a bug open for it? | 15:19 |
| *** akshai has joined #tripleo | 15:19 | |
| pradk | trown, how many measure is it processing etc | 15:20 |
| trown | pradk: k, like I said I can not reproduce it locally | 15:20 |
| sshnaidm | bnemec, no, didn't open yet | 15:20 |
| sshnaidm | bnemec, I'm not sure which project to submit it to | 15:21 |
| trown | this job failed to deploy in 90 minutes though https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-master-delorean-minimal_pacemaker-114/ and we see "High CPU load detected" on the overcloud https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-master-delorean-minimal_pacemaker-114/overcloud-controller-0/var/log/messages.gz | 15:21 |
| bnemec | sshnaidm: I would open it against both tripleo and openstackclient. | 15:21 |
| trown | pradk: ^ | 15:21 |
| sshnaidm | bnemec, ok, will do | 15:22 |
| openstackgerrit | Brent Eagles proposed openstack/tripleo-heat-templates: Add support for configuring the OVS firewall driver https://review.openstack.org/357556 | 15:26 |
| sshnaidm | bnemec, https://bugs.launchpad.net/python-openstackclient/+bug/1621524 | 15:26 |
| openstack | Launchpad bug 1621524 in tripleo "Openstack undercloud install asks for a password" [Undecided,New] | 15:26 |
| *** tremble has quit IRC | 15:26 | |
| pradk | trown, looking at the logs 0 metrics have been processed.. so its really not doing anything | 15:27 |
| beagles | EmilienM, shardy: if you have a moment - https://review.openstack.org/#/c/357556/... | 15:27 |
| pradk | trown, i'll set up a local env with minimal_pacemaker and see whats taking up cpu | 15:27 |
| pradk | trown, i dont think i saw it before, may be some recent changes upstream are causing it to peg | 15:28 |
| pradk | i'll verify now | 15:28 |
| beagles | shardy: you had commented on the bug previously - I guess I had uploaded the patch before the bug or something as the patch wasn't linked. I added it manually. | 15:28 |
| trown | pradk: seems maybe swift related... as the highest spikes in https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-master-delorean-minimal_pacemaker-114/overcloud-controller-0/var/log/messages.gz are around proxy-server logs | 15:28 |
| pradk | hmm could be | 15:29 |
| *** dtantsur is now known as dtantsur|afk | 15:29 | |
| trown | Sep 8 02:18:55 localhost crmd[14954]: notice: High CPU load detected: 17.520000 on is where things get really bogged down | 15:29 |
| EmilienM | beagles: looking | 15:30 |
| *** dprince has quit IRC | 15:30 | |
| trown | pradk: and it is a bunch of metricsd swift calls in there | 15:30 |
| EmilienM | beagles: lgtm | 15:30 |
| beagles | EmilienM: thx | 15:30 |
| *** dprince has joined #tripleo | 15:31 | |
| pradk | trown, hmm ok, could you add that info in a bug and open against gnocchi in launchpad .. worth tracking, may be we're hitting swift proxy too often | 15:31 |
| *** akshai has quit IRC | 15:32 | |
| pradk | there is literally nothing to process, weird | 15:33 |
| trown | pradk: mind if I just add on to https://bugs.launchpad.net/gnocchi/+bug/1621164 | 15:33 |
| openstack | Launchpad bug 1621164 in tripleo "gnocchi statsd consumes all overcloud resources when configured with swift backend" [Critical,In progress] - Assigned to Emilien Macchi (emilienm) | 15:33 |
| pradk | trown, yea thats fine too | 15:33 |
| *** akshai has joined #tripleo | 15:33 | |
| pradk | ideally i would want to use ceph as the backend, but i was told we dont deploy ceph by default | 15:34 |
| *** aufi has quit IRC | 15:34 | |
| pradk | too bad, as ceph for gnocchi really just works | 15:34 |
| pradk | swift should too, but aside from what you're noticing | 15:35 |
| EmilienM | pradk: i'm working on scenario001 to add ceph backend, but I have issue with Glance | 15:35 |
| pradk | ok | 15:36 |
| EmilienM | https://bugs.launchpad.net/tripleo/+bug/1621467 for the record | 15:36 |
| openstack | Launchpad bug 1621467 in tripleo "all-in-one overcloud with Glance + RBD: image upload hangs forever" [High,Triaged] - Assigned to Emilien Macchi (emilienm) | 15:36 |
| *** dprince has quit IRC | 15:37 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Fix promote script and add logs https://review.openstack.org/367244 | 15:38 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fix ansible-lint errors in all playbooks and roles https://review.openstack.org/366837 | 15:38 |
| shardy | rbrady: Hey, I've been testing the jinja2 stuff more, found a few issues, would you be OK with me squashing https://review.openstack.org/#/c/362465/ with https://review.openstack.org/#/c/366877/ combined with my latest fixes? | 15:38 |
| ayoung | EmilienM, do you need help with getting to the root of the issue with the failing gate test here? http://logs.openstack.org/94/366394/4/check/gate-puppet-openstack-integration-3-scenario001-tempest-centos-7/cd0bd1e/console.html | 15:38 |
| ayoung | Er...check test | 15:39 |
| shardy | rbrady: just thinking one patch will hopefully be easier to get through CI, as I think this is really nearly ready now | 15:39 |
| ayoung | seems like the same one is consistently failing | 15:39 |
| shardy | works pretty well locally | 15:39 |
| EmilienM | ayoung: no, it's not related to keystone | 15:39 |
| EmilienM | ayoung: it's transient issue | 15:39 |
| ayoung | EmilienM, a permanent transient? I had one of those in a house I rented once... | 15:39 |
| rbrady | shardy: I have no problem with the squash | 15:40 |
| shardy | rbrady: ack, thanks | 15:40 |
| ayoung | EmilienM, I'm not even sure what the problem is there. Just that the same error has been reported for the past couple days, so it does not sound transient | 15:40 |
| EmilienM | ayoung: gnocchi is breaking us | 15:41 |
| rbrady | I built a new undercloud this morning and following the upstream docs I get "ERROR: Timed out waiting for a reply to message" during a deployment. Has anyone else seen this result recently? | 15:43 |
| shardy | rbrady: how many cpus does the undercloud VM have? | 15:43 |
| shardy | sounds like either rabbit or one of the backend services died, or they are running too slowly | 15:44 |
| rbrady | shardy: nproc says 2 | 15:44 |
| shardy | rbrady: ack, should be OK - you definitely see that error if you try to get away with 1 CPU | 15:45 |
| beagles | shardy, dprince: just filed an MTU related bug https://bugs.launchpad.net/tripleo/+bug/1621533 | 15:46 |
| openstack | Launchpad bug 1621533 in tripleo "Some network environments need MTU adjustment" [High,Confirmed] - Assigned to Brent Eagles (beagles) | 15:46 |
| beagles | shardy, dprince: I didn't target to rc 1 "off-hand", but I'd like to get your thoughts on the matter | 15:47 |
| beagles | EmilienM: you too ^ and anyone else of course | 15:47 |
| shardy | beagles: looks like something we have to fix before the final release, so I added it to RC1 | 15:48 |
| shardy | beagles: next week, we'll cut RC1 with whatever has landed, then we'll need to scrub the remainder of bugs to decide what is actually a blocker for newton, and I'll create an RC2 milestone | 15:49 |
| beagles | shardy: cool, its now officially my top priority :) | 15:49 |
| beagles | shardy: right on | 15:49 |
| shardy | beagles: ack, thanks! :) | 15:49 |
| *** leanderthal is now known as leanderthal|afk | 15:51 | |
| *** nyechiel_ has quit IRC | 15:54 | |
| *** shivrao has joined #tripleo | 15:58 | |
| sshnaidm | bnemec, derekh seems like the issue with slow network still exists: http://logs.openstack.org/96/366896/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/e12ed6b/console.html#_2016-09-08_15_04_06_491767 | 15:58 |
| sshnaidm | bnemec, derekh and it's not related to afs, it's just slow for everything | 15:58 |
| *** akshai has quit IRC | 15:58 | |
| *** oshvartz has joined #tripleo | 15:59 | |
| *** akshai has joined #tripleo | 15:59 | |
| sshnaidm | but it's not for all jobs, maybe it could be problem with specific compute nodes | 15:59 |
| bnemec | sshnaidm: I guess that's not surprising. Neither DNS nor AFS would explain why the cirros image download is timing out, which I've seen in a few jobs. | 16:00 |
| *** saneax is now known as saneax-_-|AFK | 16:00 | |
| derekh | is the cirros image being downloaded via the proxy I wonder? | 16:01 |
| *** shivrao has left #tripleo | 16:04 | |
| bnemec | derekh: It should be. | 16:04 |
| *** shivrao has joined #tripleo | 16:05 | |
| sshnaidm | bnemec, derekh are pip install via proxy as well? | 16:05 |
| bnemec | sshnaidm: I'm not sure whether pip respect http_proxy. | 16:06 |
| bnemec | *respects | 16:06 |
| sshnaidm | because it's also extremely slow, delroean setup takes about a hour | 16:06 |
| sshnaidm | maybe it's proxy that is broken | 16:06 |
| *** liverpooler has joined #tripleo | 16:08 | |
| derekh | sshnaidm: the timeout cirros download | 16:08 |
| derekh | 1473349009.077 1465408 192.168.102.180 TCP_HIT_TIMEDOUT/200 3440566 GET http://download.cirros-cloud.net/0.3.4/cirros-0.3.4-x86_64-kernel - HIER_NONE/- text/plain | 16:08 |
| derekh | TCP_HIT_TIMEDOUT/200 | 16:08 |
| *** shivrao has quit IRC | 16:10 | |
| sshnaidm | derekh, yeah.. | 16:11 |
| sshnaidm | I submitted a patch without a proxy, just for check | 16:12 |
| derekh | sshnaidm: ok, | 16:12 |
| derekh | those timouts are between the undercloud and the proxy arn't they? | 16:12 |
| derekh | it was a HIT so no outside traffic would have happened | 16:13 |
| *** zoliXXL is now known as zoli|gone | 16:13 | |
| *** zoli|gone is now known as zoli_gone-proxy | 16:14 | |
| *** b00tcat has quit IRC | 16:14 | |
| bnemec | derekh: sshnaidm: We're definitely having issues downloading from the cirros servers. I'm looping through the compute nodes and curling it directly and it's taking forever on some of them. | 16:17 |
| bnemec | Not necessarily the same ones every time. :-/ | 16:17 |
| bnemec | Doesn't explain why the proxy isn't isolating us from that though. | 16:17 |
| derekh | bnemec: no it doesn't .... | 16:19 |
| *** mbound has joined #tripleo | 16:19 | |
| sshnaidm | bnemec, I'd suspect neutron.. | 16:19 |
| derekh | bnemec: actually now that I think about it, infra have put a cirros image on the image we use, so we probably don't need to download it | 16:19 |
| derekh | sshnaidm: bnemec: I know its not a fix but the more of this traffic we can avoid the better | 16:20 |
| sshnaidm | derekh, agree | 16:20 |
| bnemec | Yeah, that won't fix the pip failures and extremely slow image builds. | 16:20 |
| *** pabelanger has joined #tripleo | 16:21 | |
| *** tesseract- has quit IRC | 16:21 | |
| derekh | bnemec: I was hopping in a few hours when the dns change has propagated to all the instances that it would help those , could be just a pipe dream though | 16:22 |
| sshnaidm | bnemec, derekh by method of excluding, what could it be else? proxy, MTU, neutron, tcp window, .. ? | 16:23 |
| *** trown is now known as trown|outtypewww | 16:23 | |
| derekh | sshnaidm: I doubt its MTU, nothing would have changed there | 16:24 |
| *** nyechiel_ has joined #tripleo | 16:24 | |
| *** mbound has quit IRC | 16:24 | |
| *** dbecker has joined #tripleo | 16:24 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Terminate Zaqar websocket endpoint in HAProxy https://review.openstack.org/360329 | 16:24 |
| *** hrybacki is now known as hrybacki|food | 16:24 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Terminate Zaqar websocket endpoint in HAProxy https://review.openstack.org/360329 | 16:26 |
| *** bana_k has joined #tripleo | 16:26 | |
| jaosorior | thrash: ^^ | 16:26 |
| *** abregman_ is now known as abregman | 16:27 | |
| openstackgerrit | Merged openstack/tripleo-common: Include environments in capabilities output https://review.openstack.org/355598 | 16:27 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Enable TLS for Zaqar's websocket endpoint https://review.openstack.org/360350 | 16:28 |
| jaosorior | thrash: and that's the one where we should see it working ^^ | 16:28 |
| thrash | jaosorior: excellent | 16:29 |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates: Populate vnc_api_lib.ini on compute nodes with OpenContrail https://review.openstack.org/367497 | 16:29 |
| *** nyechiel_ has quit IRC | 16:30 | |
| *** dprince has joined #tripleo | 16:31 | |
| mandre | shardy: it turns out I was using a custom environment file that needed to be added to the capability map and require overcloud-resource-registry-puppet.yaml | 16:36 |
| mandre | shardy: seems to be going on now | 16:36 |
| *** fragatina has joined #tripleo | 16:37 | |
| mandre | shardy: nope, still the same issue :( | 16:38 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fix ansible-lint errors in all playbooks and roles https://review.openstack.org/366837 | 16:39 |
| *** jlinkes_ has quit IRC | 16:39 | |
| bnemec | derekh: sshnaidm: I wonder if we should just restart squid. I'm not having the same speed problems scp'ing directly from the proxy server. | 16:40 |
| derekh | bnemec: worth a try | 16:40 |
| sshnaidm | yeah, will break nothing | 16:41 |
| *** fragatina has quit IRC | 16:41 | |
| bnemec | derekh: Alright, restarting it. | 16:41 |
| bnemec | Done. | 16:42 |
| jaosorior | EmilienM: got time for a review? https://review.openstack.org/#/c/367176/ | 16:42 |
| jaosorior | or actually, if people can take a look at anything from this https://review.openstack.org/#/q/branch:master+topic:bp/tls-via-certmonger+status:open it would be awesome :D | 16:43 |
| jaosorior | REALLY need reviews there :/ | 16:43 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Stop removing the puppet-ceph git repository https://review.openstack.org/367505 | 16:44 |
| ansiwen | mwhahaha: thanks for your review! | 16:44 |
| mwhahaha | ansiwen: no problem | 16:44 |
| bnemec | derekh: sshnaidm: I also pointed the proxy at the bastion for DNS. | 16:44 |
| ansiwen | mwhahaha: so, the contain_tempest_ec2_credentials() finction is something I have to write, or is it created automatically? | 16:44 |
| derekh | bnemec: +1 | 16:44 |
| mwhahaha | ansiwen: it's handled automatically by rspec, it automagically figures it out | 16:45 |
| bnemec | I don't know that it's helping though. I'm still seeing some pretty slow downloads through the proxy. :-/ | 16:45 |
| mwhahaha | contain_<resource name> | 16:45 |
| ansiwen | mwhahaha: ok, and what does it do? how to use it? some docs about that? | 16:46 |
| mwhahaha | ansiwen: http://rspec-puppet.com/matchers/ | 16:47 |
| derekh | bnemec: btw, did you update the dns entry on the proxy befor the squid restart ? | 16:48 |
| *** lucasagomes is now known as lucas-dinner | 16:48 | |
| derekh | bnemec: it doesn't get picked up by squid if it was after, which surprised me one day | 16:48 |
| derekh | at least not straight away | 16:48 |
| * derekh has gotta go | 16:49 | |
| bnemec | derekh: Oh, no I didn't. I'll restart it again. | 16:49 |
| derekh | bnemec: it probbaly wont help the download rate but worth making sure | 16:49 |
| derekh | ttyl | 16:49 |
| bnemec | derekh: Thanks for the help. | 16:49 |
| derekh | bnemec: no prob | 16:49 |
| *** derekh has quit IRC | 16:49 | |
| bnemec | Failed connect to 192.168.103.252:3128; Connection refused | 16:51 |
| bnemec | Ruh roh | 16:51 |
| bnemec | Also, WTF? | 16:51 |
| bnemec | Okay, it's back. Not entirely sure what was wrong, but I configured the hostname in /etc/hosts so it stopped warning about hostname issues. | 16:57 |
| *** pkovar has quit IRC | 16:58 | |
| ansiwen | mwhahaha: ok, I think I slowly get it. so it just checks for the presence of the declarative structure, but no actual code is running... so this is not yet a unit test. correct? | 16:58 |
| mwhahaha | it's a puppet unit test not a unit test of the provider itself | 16:59 |
| ansiwen | mwhahaha: got it | 16:59 |
| ansiwen | mwhahaha: temptest_config_path is really '...' by default? or is this a placeholder? | 17:00 |
| mwhahaha | placeholder | 17:00 |
| mwhahaha | i was lazy and didn't want to go looking :D | 17:00 |
| *** pkovar has joined #tripleo | 17:00 | |
| *** egafford has quit IRC | 17:01 | |
| *** jaosorior has quit IRC | 17:01 | |
| bnemec | sshnaidm: Okay, squid seems like it might be happier now. Maybe the DNS throttling was causing something to block internally. | 17:02 |
| *** abregman has quit IRC | 17:02 | |
| *** ohamada has quit IRC | 17:05 | |
| ansiwen | mwhahaha: I'm using $tempest_conf in init.pp as the other providers did, but for me it's not really obvious what the value of $tempest_conf is. It's only defined in a deeper scope, so this is defined outside of init.pp? | 17:08 |
| ansiwen | mwhahaha: there is a $tempest_config_file variable, but it's not used there... weird | 17:09 |
| mwhahaha | gimme a minute and i'll find it | 17:09 |
| pabelanger | bnemec: sshnaidm: just catching up on back scroll, which nodes are having issues with DNS? Is it ovb? | 17:11 |
| bnemec | sshnaidm: Okay, squid seems like it might be happier now. Maybe the DNS throttling was causing something to block internally. | 17:11 |
| sshnaidm | pabelanger, yes | 17:11 |
| bnemec | pabelanger: No, we think we were getting throttled by Google's DNS. | 17:11 |
| sshnaidm | bnemec, ok, let's see | 17:12 |
| bnemec | Whoops, how did that happen? | 17:12 |
| openstackgerrit | Waldemar Znoinski proposed openstack/diskimage-builder: don't configure 'lo' for dhcp https://review.openstack.org/367527 | 17:12 |
| pabelanger | bnemec: sshnaidm: Right, so we run unbound on nodepool nodes to avoid issues with that. So, trying to figure out where the DNS issues are | 17:12 |
| pabelanger | might be worth adding unbound to OVB too | 17:13 |
| openstackgerrit | Waldemar Znoinski proposed openstack/diskimage-builder: don't configure 'lo' for dhcp https://review.openstack.org/367527 | 17:14 |
| bnemec | pabelanger: It turns out we actually have a local DNS server in the env. Is there a benefit to using unbound instead? | 17:14 |
| *** pkovar has quit IRC | 17:14 | |
| openstackgerrit | Waldemar Znoinski proposed openstack/diskimage-builder: don't configure 'lo' for dhcp https://review.openstack.org/367527 | 17:15 |
| sshnaidm | pabelanger, what is unbound? is it cache dns? | 17:15 |
| pabelanger | bnemec: It would make my life easier is we didn't have a local DNS server :) Trying to clean up all the private infrastructure. In fact, it would be straightfowrad to configure the nodepool image to be a DNS server, then have ovb use that. | 17:16 |
| pabelanger | sshnaidm: yes: https://www.unbound.net/ | 17:17 |
| *** hrybacki|food is now known as hrybacki | 17:17 | |
| sshnaidm | bnemec, btw, zuul installs packages with pip on node without delays at all... | 17:18 |
| pabelanger | let me check something | 17:18 |
| bnemec | sshnaidm: Yeah, I haven't had a slow download since squid got up and running with the local DNS. | 17:18 |
| sshnaidm | bnemec, but now it's again ERROR - Couldn't retrieve env :( | 17:19 |
| *** egafford has joined #tripleo | 17:19 | |
| ansiwen | mwhahaha: for me it looks like the usage of $tempest_conf in all the providers is just wrong... | 17:20 |
| *** jpena is now known as jpena|off | 17:21 | |
| bnemec | sshnaidm: That's probably not dns though. | 17:21 |
| ansiwen | mwhahaha: maybe something like Tempest_config[:path] or so would be correct? because Tempest_config { path => ... } is always defined | 17:22 |
| *** bana_k has quit IRC | 17:22 | |
| mwhahaha | ansiwen: most likely yes, i think it's a parameter that we pass in so in the test it may be a variable | 17:22 |
| bnemec | sshnaidm: Hmm, a lot of gateway timeouts in the te-worker logs., | 17:22 |
| bnemec | heatclient.exc.HTTPException: ERROR: <html><body><h1>504 Gateway Time-out</h1> | 17:22 |
| bnemec | The server didn't respond in time. | 17:22 |
| mwhahaha | ansiwen: if it's elsewhere in the tests you could just steal that check | 17:23 |
| ansiwen | mwhahaha: so you think if I see something which is wrong I should rather ignore it? ;-) | 17:23 |
| mwhahaha | well of course not, but it may not be wrong :D | 17:24 |
| openstackgerrit | Dimitri Savineau proposed openstack/puppet-tripleo: Cinder: Add iSCSI protocol parameter https://review.openstack.org/367542 | 17:24 |
| ansiwen | mwhahaha: ok, I interpreted your "most likely yes" is referring to "wrong" :-) | 17:24 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements https://review.openstack.org/361501 | 17:25 |
| sshnaidm | bnemec, also see "Connection to neutron failed: ('Connection aborted." | 17:25 |
| mwhahaha | ansiwen: :tempest_conf_path => '/var/lib/tempest/etc/tempest.conf' | 17:25 |
| ansiwen | mwhahaha: yes, just saw it in that second | 17:25 |
| bnemec | sshnaidm: Yeah, lots of issues in heat-api. :-( | 17:26 |
| bnemec | sshnaidm: Ugh, many dead stacks again. | 17:27 |
| ansiwen | mwhahaha: thanks a lot... I will push that and will try write a provider unit tests tomorrow | 17:28 |
| pabelanger | bnemec: sshnaidm: so unbound is listening on all interfaces by default, so we should just be able to open the port on the firewall, and have ovb use it | 17:30 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart: Add back socket for ci-scripts - OVB and baremetal https://review.openstack.org/367545 | 17:31 |
| pabelanger | bnemec: sshnaidm: sudo iptables -I openstack-INPUT 1 -p udp -i eth0 --sport 53 -j ACCEPT | 17:32 |
| EmilienM | shardy, slagle, dprince: do you mind landing https://review.openstack.org/#/c/367324/ so we can have scenarios in dashboard? | 17:32 |
| pabelanger | need to test it still | 17:32 |
| sshnaidm | bnemec, started a cleaning script | 17:37 |
| bnemec | sshnaidm: Okay. It does look like the testenvs are cleaning up after themselves to some extent. It may be that a bunch of jobs failed at once while we were messing around with stuff. | 17:38 |
| *** bana_k has joined #tripleo | 17:40 | |
| openstackgerrit | Julie Pichon proposed openstack/python-tripleoclient: Provide more information when 'node provide' fails https://review.openstack.org/367553 | 17:46 |
| openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates: Adds http proxy support for subscription-manager on RHEL overcloud nodes https://review.openstack.org/367554 | 17:47 |
| *** jayg is now known as jayg|g0n3 | 17:47 | |
| *** jayg|g0n3 is now known as jayg | 17:48 | |
| openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates: Adds http proxy support for subscription-manager on RHEL overcloud nodes https://review.openstack.org/367554 | 17:49 |
| openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/367554 | 17:49 |
| *** jpich has quit IRC | 17:50 | |
| *** akshai has quit IRC | 17:54 | |
| *** dbecker has quit IRC | 17:56 | |
| bnemec | pabelanger: Would changing the nodepool rate stop it from trying to spin up 15 envs at once? | 17:58 |
| bnemec | rh1 cannot handle that many heat stacks creating at once. | 17:58 |
| dprince | EmilienM: commented, adding scenarios will make it pretty wide | 17:59 |
| EmilienM | dprince: what do you propose? | 17:59 |
| dprince | EmilienM: so I have to propose the idea here :) | 18:00 |
| EmilienM | dprince: maybe drop gate-tripleo-ci- in the display | 18:00 |
| EmilienM | and keep useful infos | 18:00 |
| dprince | EmilienM: The page isn't going to scale horizontally as is | 18:00 |
| dprince | EmilienM: we might need to consider javascript tabs or something? | 18:00 |
| *** fragatina has joined #tripleo | 18:01 | |
| EmilienM | dprince: right | 18:01 |
| dprince | EmilienM: in the short term I'm not opposed to adding the scenarios though | 18:01 |
| dprince | EmilienM: just commenting... | 18:01 |
| EmilienM | yeah your comment is valid | 18:01 |
| pabelanger | bnemec: how many environments could it handle? | 18:02 |
| *** akshai has joined #tripleo | 18:02 | |
| bnemec | pabelanger: I'd prefer to limit it to 4 or 5 at once. Although after giving it more thought I think it won't help. It's not the nodepool nodes that are the problem, it's the zuul jobs coming into the te-broker. | 18:03 |
| pabelanger | bnemec: we can try, but our rate limits for nodepool are around API access, not launching nodes | 18:03 |
| pabelanger | bnemec: it is pretty rough on clouds | 18:03 |
| *** ifarkas is now known as ifarkas_afk | 18:04 | |
| pabelanger | bnemec: Is using heat for ovb a hard requirement or could that change? | 18:04 |
| pabelanger | I assume it is the ironic aspect you need | 18:05 |
| bnemec | pabelanger: It's a hard requirement for the moment. It would be a pretty huge change to provision envs a different way. | 18:05 |
| openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/367554 | 18:05 |
| chem | matbu: he, do you have a ci exemple of running the upgrade ? | 18:05 |
| EmilienM | chem: https://review.openstack.org/#/c/364859/47 | 18:06 |
| chem | EmilienM: thanks | 18:06 |
| EmilienM | I can't figure why matbu dupplicated my patch but it comes from https://review.openstack.org/#/c/351330/ | 18:06 |
| EmilienM | chem: and also https://review.openstack.org/#/c/346995/ | 18:07 |
| matbu | chem: yep this one | 18:07 |
| matbu | chem: actually failling | 18:07 |
| *** abregman has joined #tripleo | 18:07 | |
| matbu | EmilienM: i dupplicated it for debuging, and to avoid to spam a bunch of 10 peoples every CI triggered | 18:07 |
| matbu | :) | 18:08 |
| matbu | chem: currently M to N is broken, i'll try to figure out the issue tomorow | 18:08 |
| EmilienM | matbu: why not continuing in my patch? | 18:09 |
| EmilienM | I don't understand the "spam" issue | 18:09 |
| chem | matbu: EmilienM FYI: I don't know upstream (that's what I would like to check) but ceph upgrade downstream is going from ~0.9 to 2.0 ! yeah \o/. And it's non working: https://bugzilla.redhat.com/show_bug.cgi?id=1374076 | 18:09 |
| openstack | bugzilla.redhat.com bug 1374076 in openstack-tripleo-heat-templates "OSP9/10 Ceph osd node upgrade fails." [Unspecified,New] - Assigned to jstransk | 18:09 |
| matbu | EmilienM: i can no pb with that. | 18:09 |
| matbu | chem: upstream doesn't upgrade ceph for now | 18:10 |
| pabelanger | bnemec: We won't be able to rate limit jobs from nodepool or zuul today. I assume heat doesn't offer that functionality today? | 18:10 |
| EmilienM | matbu: I would avoid duplicated efforts | 18:10 |
| matbu | chem: rdo does an HA with 1 compute and tripleo-ci 2 nodes | 18:10 |
| EmilienM | please take over my patch, put your code and abandon the "WIP -- clean full upgrade review" thing | 18:10 |
| matbu | EmilienM: yep no worries, i'll put my change on the first review tomorow | 18:11 |
| EmilienM | thanks! | 18:11 |
| bnemec | pabelanger: I don't see anything, but I'll see if we can come up with something. | 18:11 |
| chem | matbu: hum, oki, that's why I would like to check more freely in a local test. But I would like to stay as close as possible as what is tested upstream | 18:11 |
| matbu | EmilienM: for the UC upgrade review we've got two +2 , so i think we are close | 18:11 |
| pabelanger | bnemec: how many minutes does it take to bring ovb online using heat? | 18:13 |
| chem | matbu: EmilienM I don't get it, why we don't use oooq upgrade role ? | 18:13 |
| EmilienM | matbu: I'm going to +A it | 18:13 |
| EmilienM | chem: 1) I'm not aware about oooq upgrade role 2) oooq is not tripleo-ci | 18:13 |
| chem | EmilienM: oki, just asking :) | 18:14 |
| EmilienM | chem: what is it? | 18:14 |
| pabelanger | bnemec: almost need to keep about 6 ovb nodes online at all times, vs creating them once a job starts. | 18:14 |
| *** akshai has quit IRC | 18:14 | |
| matbu | EmilienM: i did an ansible role for upgrade a rdo deployment | 18:14 |
| matbu | well a tripleo deployment :) | 18:14 |
| chem | EmilienM: https://github.com/redhat-openstack/ansible-role-tripleo-overcloud-upgrade | 18:14 |
| chem | matbu: EmilienM But I found the interface of ^ confusing to say the least | 18:15 |
| matbu | chem: the goal w/ tripleo-ci would be to gate the review with the upgrade jobs | 18:15 |
| EmilienM | matbu: approving it with a comment, please look | 18:15 |
| matbu | EmilienM: k | 18:16 |
| chem | (for an non ansible expert) | 18:16 |
| *** aufi has joined #tripleo | 18:16 | |
| matbu | chem: which will probably avoid regression (like we hit currently :)) | 18:16 |
| bnemec | pabelanger: Looks like it's taking about 4 minutes for successful stacks right now. | 18:16 |
| EmilienM | matbu: I'm sending an update on ML | 18:16 |
| matbu | EmilienM: /me hopes to land the OC asap | 18:17 |
| chem | matbu: yeap, but then we have two public project testing the upgrade in maybe different ways ... but I'm not here for the rant. I will use the current upstream ci as show in the Emilien patch and use that. | 18:17 |
| bnemec | I would like to have a pool of ovb envs ready to go at all times, but that's tricky to do because they're created dynamically with the appropriate size for the test job. | 18:17 |
| chem | matbu: EmilienM thanks for the clarification | 18:17 |
| matbu | chem: yes sure, i don't like also having 2 differents projects | 18:18 |
| matbu | chem: too much duplicated efforts :/ | 18:18 |
| EmilienM | chem: yes, this is a big problem (tools) | 18:18 |
| pabelanger | bnemec: Ya, I think for that we need nodepool. Since that functionality is out of box today. We'd need whiteboard how to do the heat integration however | 18:18 |
| chem | EmilienM: for the record, the learning curve for the current script is lest steep that for the oooq role :) | 18:19 |
| chem | anyway, going now, see ya | 18:19 |
| *** chem is now known as chem|off | 18:19 | |
| matbu | EmilienM: +1 idk who have the keys for resolving that :) | 18:20 |
| matbu | chem|off: good night | 18:20 |
| openstackgerrit | Ana Krivokapic proposed openstack/tripleo-ui: Add parameter setting to node assignment https://review.openstack.org/367562 | 18:21 |
| *** egafford1 has joined #tripleo | 18:21 | |
| *** athomas has quit IRC | 18:21 | |
| *** egafford has quit IRC | 18:23 | |
| *** bfournie has quit IRC | 18:23 | |
| *** skramaja has quit IRC | 18:24 | |
| *** karthiks has quit IRC | 18:24 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: Implement non-ovb overcloud update job - Newton -> Newton https://review.openstack.org/351330 | 18:26 |
| *** akshai has joined #tripleo | 18:35 | |
| *** dmsimard is now known as dmsimard|afk | 18:45 | |
| *** egafford1 is now known as egafford | 18:50 | |
| EmilienM | sshnaidm, bnemec: (same question as yesterday), I missed context today, what is the status of ovb jobs now? | 18:50 |
| EmilienM | I'm trying to figure if we should land some patches without them or not | 18:50 |
| *** mhenkel has quit IRC | 18:50 | |
| *** aufi has quit IRC | 18:50 | |
| *** fzdarsky has quit IRC | 18:51 | |
| bnemec | EmilienM: There are jobs running ATM, and we think we've got the proxy issues that were causing timeouts fixed. | 18:52 |
| bnemec | EmilienM: But until we get results back from some of those jobs I don't know that we can call it "fixed". | 18:52 |
| openstackgerrit | Dimitri Savineau proposed openstack/tripleo-heat-templates: Added support for pass-through iSER configuration https://review.openstack.org/324781 | 18:53 |
| *** david-lyle has quit IRC | 18:53 | |
| *** david-lyle has joined #tripleo | 18:53 | |
| EmilienM | bnemec: ok, i'll monitor it | 18:53 |
| EmilienM | bnemec: if by EOD we don't have it "fixed", I suggest we find a plan B | 18:53 |
| EmilienM | RC1 is really close, next week and we need to land patches | 18:54 |
| *** mbound has joined #tripleo | 19:00 | |
| openstackgerrit | Honza Pokorny proposed openstack/tripleo-ui: Update version to match current release https://review.openstack.org/367587 | 19:09 |
| *** dprince has quit IRC | 19:16 | |
| *** rlandy is now known as rlandy|brb | 19:16 | |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 19:17 |
| EmilienM | matbu: I thought you would abandon it? ^ I'm confused | 19:17 |
| *** florianf has quit IRC | 19:22 | |
| *** coolsvap has quit IRC | 19:22 | |
| *** hrybacki is now known as hrybacki|brb | 19:22 | |
| *** egafford has quit IRC | 19:22 | |
| *** kjw3 has joined #tripleo | 19:30 | |
| *** mlupton has joined #tripleo | 19:32 | |
| *** abregman has quit IRC | 19:39 | |
| *** hrybacki|brb is now known as hrybacki | 19:40 | |
| *** sshnaidm is now known as sshnaidm|afk | 19:41 | |
| *** egafford has joined #tripleo | 19:43 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Implement undercloud upgrade job - Mitaka -> Newton https://review.openstack.org/346995 | 19:43 |
| *** mhenkel has joined #tripleo | 19:44 | |
| *** yamahata has joined #tripleo | 19:49 | |
| *** rlandy|brb is now known as rlandy | 19:52 | |
| *** dmsimard|afk is now known as dmsimard | 19:54 | |
| *** mcornea has joined #tripleo | 19:58 | |
| *** jeckersb is now known as jeckersb_gone | 20:02 | |
| *** jprovazn has quit IRC | 20:03 | |
| *** mcornea has quit IRC | 20:07 | |
| slagle | i'm seeing some ovb green | 20:17 |
| EmilienM | alleluia | 20:18 |
| EmilienM | https://review.openstack.org/#/c/367554/ | 20:18 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Convert SwiftDevicesAndProxyConfig to composable format https://review.openstack.org/364748 | 20:19 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move role deployment steps into puppet/post.yaml https://review.openstack.org/365763 | 20:19 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Remove *ExtraConfig parameters from overcloud.yaml https://review.openstack.org/365792 | 20:19 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Convert overcloud.yaml to support jinja2 templating https://review.openstack.org/315679 | 20:20 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move AllNodesDeployments into jinja template loop https://review.openstack.org/337267 | 20:20 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move AllNodesValidationDeployments into jinja template loop https://review.openstack.org/337587 | 20:20 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move per-role NetIpListMap's into jinja template loop https://review.openstack.org/364749 | 20:20 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move role ResourceGroups inside the jinja2 loop https://review.openstack.org/365793 | 20:20 |
| *** yamahata has quit IRC | 20:20 | |
| *** paramite has quit IRC | 20:22 | |
| bnemec | Yeah, things are looking up at the moment. | 20:23 |
| *** jayg is now known as jayg|g0n3 | 20:24 | |
| EmilienM | bnemec, slagle: can I have your vote on https://review.openstack.org/#/c/351330/ please? | 20:27 |
| EmilienM | I want to re-enable update jobs | 20:27 |
| *** Goneri has quit IRC | 20:31 | |
| slagle | you'd make a good politician | 20:35 |
| slagle | so polite asking for votes | 20:35 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-heat-templates: [wip] deploy opstools node as part of the overcloud https://review.openstack.org/367618 | 20:36 |
| bnemec | slagle: Yeah, but you know he'd just be a puppet. ;-) | 20:40 |
| slagle | that is not worthy of a response | 20:40 |
| *** bfournie has joined #tripleo | 20:43 | |
| *** kjw3 has quit IRC | 20:44 | |
| * thrash groans | 20:44 | |
| *** flepied has quit IRC | 20:45 | |
| *** absubram has joined #tripleo | 20:48 | |
| *** shardy has quit IRC | 20:49 | |
| ayoung | can we make gate-puppet-openstack-integration-3-scenario001-tempest-centos-7 nonvoting? | 20:49 |
| EmilienM | ayoung: why that?? | 20:50 |
| EmilienM | ayoung: right channel is puppet-openstack btw | 20:50 |
| ayoung | it keeps failing consistantly | 20:50 |
| EmilienM | yes and we are fixing it | 20:50 |
| ayoung | Ah, cool | 20:50 |
| EmilienM | ayoung: https://review.openstack.org/#/c/367551/ | 20:50 |
| ayoung | EmilienM, its all Tripleo to me... | 20:50 |
| EmilienM | no | 20:50 |
| EmilienM | it's puppet CI. | 20:50 |
| EmilienM | https://github.com/openstack/puppet-openstack-integration | 20:50 |
| ayoung | HA! | 20:50 |
| ayoung | Openstack is Keystone and Everything else. | 20:51 |
| openstackgerrit | Christopher Brown proposed openstack/tripleo-heat-templates: Add proxy parameter for rhel registration https://review.openstack.org/367628 | 20:52 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-common: Wire in jinja templating for custom roles https://review.openstack.org/362465 | 20:53 |
| *** flepied has joined #tripleo | 20:58 | |
| openstackgerrit | Jeff Peeler proposed openstack/tripleo-common: Set Deployment Parameters https://review.openstack.org/365625 | 20:59 |
| *** lblanchard has quit IRC | 21:00 | |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP -- clean full upgrade review https://review.openstack.org/364859 | 21:01 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Add fluentd client service https://review.openstack.org/353506 | 21:04 |
| EmilienM | larsks: rebasing it, if CI pass i'll +A | 21:04 |
| *** jcoufal has quit IRC | 21:06 | |
| EmilienM | bnemec, slagle: wdyt about 1) moving undercloud upgrade job from experimental to check queue (non voting) for instack-undercloud and 2) moving overcloud update job from experimental to check queue non voting for puppet-tripleo and THT ? | 21:06 |
| EmilienM | for 2), maybe tripleoclient too | 21:06 |
| *** mburned is now known as mburned_out | 21:08 | |
| bnemec | EmilienM: wfm | 21:08 |
| openstackgerrit | Jeff Peeler proposed openstack/tripleo-common: Set Deployment Parameters https://review.openstack.org/365625 | 21:09 |
| EmilienM | ok | 21:09 |
| slagle | EmilienM: sounds fine to me | 21:10 |
| EmilienM | ok i'm working on zuul layout | 21:10 |
| *** cdearborn has quit IRC | 21:15 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario002: deploy Swift https://review.openstack.org/367401 | 21:25 |
| *** akrivoka has quit IRC | 21:35 | |
| *** mlupton has quit IRC | 21:45 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Implement non-ovb overcloud update job - Newton -> Newton https://review.openstack.org/351330 | 21:55 |
| *** mlupton has joined #tripleo | 22:00 | |
| *** mbound has quit IRC | 22:08 | |
| *** dtantsur|afk has quit IRC | 22:12 | |
| *** dtantsur has joined #tripleo | 22:14 | |
| *** egafford has quit IRC | 22:15 | |
| *** absubram has quit IRC | 22:15 | |
| *** nyechiel_ has joined #tripleo | 22:17 | |
| *** nyechiel_ has quit IRC | 22:26 | |
| *** dtrainor has quit IRC | 22:32 | |
| *** nyechiel_ has joined #tripleo | 22:33 | |
| *** mhenkel has quit IRC | 22:39 | |
| *** akshai has quit IRC | 22:42 | |
| *** wfoster has quit IRC | 22:46 | |
| *** sai has quit IRC | 22:49 | |
| *** fultonj has quit IRC | 22:50 | |
| *** rhallisey has quit IRC | 22:51 | |
| *** nyechiel_ has quit IRC | 22:51 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Enable SSL for undercloud-only job https://review.openstack.org/359131 | 22:52 |
| *** sai has joined #tripleo | 22:54 | |
| *** nyechiel_ has joined #tripleo | 22:54 | |
| openstackgerrit | Emilien Macchi proposed openstack/python-tripleoclient: Get template contents from plan, not local path https://review.openstack.org/365735 | 22:55 |
| *** wfoster has joined #tripleo | 22:56 | |
| *** rlandy has quit IRC | 23:01 | |
| *** mlupton has quit IRC | 23:06 | |
| *** mlupton has joined #tripleo | 23:07 | |
| *** pradk has quit IRC | 23:08 | |
| *** sai has quit IRC | 23:11 | |
| *** rook has quit IRC | 23:11 | |
| *** mlupton has quit IRC | 23:12 | |
| *** lucas-dinner has quit IRC | 23:12 | |
| *** wfoster has quit IRC | 23:12 | |
| *** saneax-_-|AFK is now known as saneax | 23:14 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenarios: set Debug to True https://review.openstack.org/366896 | 23:21 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: scenario001: deploy Ceph https://review.openstack.org/366810 | 23:21 |
| *** ipsecguy has joined #tripleo | 23:22 | |
| *** rook has joined #tripleo | 23:26 | |
| *** wfoster has joined #tripleo | 23:26 | |
| *** sai has joined #tripleo | 23:26 | |
| *** rook is now known as Guest90983 | 23:26 | |
| *** lucasagomes has joined #tripleo | 23:26 | |
| openstackgerrit | Merged openstack/tripleo-common: Test baremetal: Correctly stop the mocks https://review.openstack.org/365958 | 23:27 |
| *** egafford has joined #tripleo | 23:35 | |
| *** shivrao has joined #tripleo | 23:40 | |
| *** shivrao has quit IRC | 23:44 | |
| *** Ryjedo_ has joined #tripleo | 23:54 | |
| *** Ryjedo has quit IRC | 23:55 | |
| *** Ryjedo_ is now known as Ryjedo | 23:55 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Work around deletion of _member_ role assignments on upgrade https://review.openstack.org/307352 | 23:55 |
| openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/367554 | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!