| *** jmlowe_ has quit IRC | 00:00 | |
| *** jmlowe has joined #heat | 00:00 | |
| *** namnh has joined #heat | 01:03 | |
| *** kiennt26 has joined #heat | 01:08 | |
| *** yangyapeng has joined #heat | 01:10 | |
| *** kiennt26 has quit IRC | 01:21 | |
| *** higgins` has quit IRC | 01:45 | |
| *** higgins has joined #heat | 01:48 | |
| *** zhenguo has joined #heat | 02:14 | |
| *** pbourke_ has quit IRC | 02:22 | |
| *** yangyapeng has quit IRC | 02:22 | |
| *** yangyapeng has joined #heat | 02:23 | |
| *** pbourke_ has joined #heat | 02:24 | |
| *** kiennt26 has joined #heat | 02:51 | |
| *** yangyape_ has joined #heat | 02:55 | |
| *** yangyapeng has quit IRC | 02:55 | |
| *** links has joined #heat | 03:34 | |
| *** yangyapeng has joined #heat | 03:41 | |
| *** yangyape_ has quit IRC | 03:41 | |
| *** ramishra has joined #heat | 04:08 | |
| *** nicolasbock has quit IRC | 04:38 | |
| *** skramaja has joined #heat | 05:45 | |
| *** skramaja has quit IRC | 06:09 | |
| *** skramaja_ has joined #heat | 06:09 | |
| *** skramaja_ has quit IRC | 06:13 | |
| *** skramaja_ has joined #heat | 06:14 | |
| *** armaan has quit IRC | 06:31 | |
| *** armaan has joined #heat | 06:31 | |
| *** ivveh has quit IRC | 06:39 | |
| *** ivveh has joined #heat | 06:40 | |
| *** ivveh has quit IRC | 06:51 | |
| *** openstackgerrit has joined #heat | 07:03 | |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/heat-dashboard master: Use assertRegex instead of assertRegexpMatches https://review.openstack.org/530689 | 07:03 |
|---|---|---|
| *** jtomasek has joined #heat | 08:10 | |
| *** jtomasek has quit IRC | 08:12 | |
| *** jtomasek has joined #heat | 08:12 | |
| *** armaan has quit IRC | 08:20 | |
| *** armaan has joined #heat | 08:26 | |
| *** mdnadeem has joined #heat | 08:51 | |
| *** gmoro has joined #heat | 09:04 | |
| *** yassine has joined #heat | 09:06 | |
| *** yassine is now known as Guest10726 | 09:06 | |
| *** SmearedBeard has joined #heat | 09:10 | |
| *** gmoro has quit IRC | 09:14 | |
| *** gmoro has joined #heat | 09:16 | |
| *** shardy has joined #heat | 09:32 | |
| *** dbecker has joined #heat | 09:34 | |
| *** gfidente has joined #heat | 09:52 | |
| *** gfidente has joined #heat | 09:52 | |
| *** derekh has joined #heat | 09:55 | |
| *** armaan has quit IRC | 09:56 | |
| *** pbourke_ has quit IRC | 10:02 | |
| *** pbourke has joined #heat | 10:10 | |
| *** namnh has quit IRC | 10:14 | |
| *** huangtianhua_ has joined #heat | 10:20 | |
| *** kiennt26 has quit IRC | 10:21 | |
| *** huangtianhua has quit IRC | 10:23 | |
| *** liusheng has quit IRC | 10:23 | |
| *** liusheng has joined #heat | 10:23 | |
| *** ricolin has joined #heat | 10:31 | |
| *** ramishra has quit IRC | 10:35 | |
| *** armaan has joined #heat | 10:38 | |
| *** armaan_ has joined #heat | 10:40 | |
| *** armaan has quit IRC | 10:40 | |
| *** ramishra has joined #heat | 10:48 | |
| *** maestropandy has joined #heat | 10:55 | |
| *** maestropandy has left #heat | 10:55 | |
| *** Guest10726 has quit IRC | 11:00 | |
| *** hjensas has quit IRC | 11:09 | |
| *** hjensas has joined #heat | 11:09 | |
| *** hjensas has quit IRC | 11:09 | |
| *** hjensas has joined #heat | 11:09 | |
| *** blinky_ghost has joined #heat | 11:11 | |
| blinky_ghost | shardy: Hi, can you give me a help here? | 11:12 |
| shardy | blinky_ghost: possibly, what's the problem? | 11:12 |
| blinky_ghost | I'm using a resourcegroup nested stack, however I need to use predictable ips for my servers in the same fashion as tripleo. That is working fine, however when I try to "heal" the stack, marking a resource group nova node as unhealthy it messes all my nodes because of indexes. This doesn't happen when I don't use predictable ips. Is there a better way to specify which node I want to remove from the stack without messing it up? | 11:17 |
| *** armaan_ has quit IRC | 11:19 | |
| *** huangtianhua_ has quit IRC | 11:23 | |
| *** huangtianhua_ has joined #heat | 11:24 | |
| *** SmearedBeard has quit IRC | 11:31 | |
| *** mdnadeem has quit IRC | 11:40 | |
| *** serlex has joined #heat | 11:46 | |
| *** SmearedBeard has joined #heat | 11:47 | |
| *** nicolasbock has joined #heat | 11:58 | |
| *** mdnadeem has joined #heat | 12:04 | |
| *** links has quit IRC | 12:13 | |
| *** gfidente has quit IRC | 12:30 | |
| *** SmearedBeard has quit IRC | 12:31 | |
| *** Smeared_Beard has joined #heat | 12:34 | |
| *** maestropandy has joined #heat | 12:47 | |
| *** maestropandy has quit IRC | 12:51 | |
| *** maestropandy has joined #heat | 12:52 | |
| *** maestropandy has left #heat | 12:52 | |
| *** jcoufal has joined #heat | 12:59 | |
| *** skramaja_ has quit IRC | 13:22 | |
| *** gfidente has joined #heat | 13:23 | |
| *** gfidente has joined #heat | 13:23 | |
| *** ygl has joined #heat | 13:23 | |
| *** ramishra has quit IRC | 13:26 | |
| *** ygl_ has joined #heat | 13:27 | |
| *** yassine has joined #heat | 13:28 | |
| *** yassine has quit IRC | 13:28 | |
| *** ygl has quit IRC | 13:29 | |
| *** yassine has joined #heat | 13:29 | |
| *** yassine is now known as Guest74108 | 13:29 | |
| ygl_ | hi all | 13:34 |
| ygl_ | i need help on nested templates | 13:34 |
| ygl_ | i have nested templates n1 and n2 which are normal OS::Nova:server resources. I am using them both in a parent template p1. How can I capture the value of the mac address of a network port of n1 to use it in the post configuration of n2 ? | 13:35 |
| *** serlex has quit IRC | 13:38 | |
| *** ygl_ has quit IRC | 13:40 | |
| *** ygl has joined #heat | 13:41 | |
| ygl | i need some help | 13:42 |
| ygl | i have nested templates n1 and n2 which are normal OS::Nova:server resources. I am using them both in a parent template p1. How can I capture the value of the mac address of a network port of n1 to use it in the post configuration of n2 ? | 13:42 |
| ygl | is anyone there ? | 13:46 |
| *** mchlumsky has joined #heat | 14:03 | |
| *** zaneb has joined #heat | 14:05 | |
| *** mchlumsky has quit IRC | 14:13 | |
| *** ygl has quit IRC | 14:15 | |
| *** mchlumsky has joined #heat | 14:16 | |
| shardy | If ygl appears again, I created http://paste.openstack.org/show/636146/ to demonstrate how to pass an output between nested stacks | 14:17 |
| *** serlex has joined #heat | 14:19 | |
| *** mchlumsky has quit IRC | 14:26 | |
| *** mchlumsky has joined #heat | 14:28 | |
| *** gmoro has quit IRC | 14:31 | |
| blinky_ghost | shardy: you saw my question? | 14:37 |
| shardy | blinky_ghost: Sorry I missed it - the way we do this in TripleO is use the removal_policies interface to resource group | 14:38 |
| shardy | blinky_ghost: that means you can permanently blacklist a node without changing the indexes (the input data for predictable IPs will contain some unused list items though) | 14:39 |
| blinky_ghost | shardy: do you have an example how to do that? | 14:39 |
| shardy | https://docs.openstack.org/heat/latest/template_guide/openstack.html#OS::Heat::ResourceGroup-prop-removal_policies | 14:40 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud.j2.yaml#L194 | 14:40 |
| shardy | blinky_ghost: ^^ | 14:40 |
| shardy | blinky_ghost: basically something like removal_policies: [{'resource_list': ['0']}] | 14:40 |
| shardy | blinky_ghost: however note you will have to maintain that list on update, e.g update the list instead of overwriting it | 14:41 |
| *** hongbin has joined #heat | 14:41 | |
| shardy | we have a workflow in TripleO which does that, but heat doesn't provide a way to do it by default AFAIK | 14:41 |
| blinky_ghost | shardy: so you're saying that I have to scale out instead of replacing the node? Can I reuse the ip of the blacklisted node? | 14:43 |
| shardy | blinky_ghost: if you want to replace the node, you can use the resource mark unhealthy interface | 14:43 |
| shardy | blinky_ghost: then heat will replace the resource group member you marked unhealthy on the next stack update | 14:44 |
| shardy | Sorry I don't have a good example of that, but I've done it with TripleO deployments and it works fine | 14:44 |
| shardy | blinky_ghost: it'll be something like stack resource mark unhealthy <resource group nested stack ID> "0" | 14:45 |
| shardy | then do a stack update (or in the case of TripleO run the deploy command again | 14:45 |
| blinky_ghost | shardy: ok, so I have to mark the interface unhealthy, meaning the neutron port instead of nova server. Then I have to scale out, add a node to the counter? | 14:46 |
| shardy | blinky_ghost: No you don't need to change the count | 14:46 |
| shardy | just mark whatever you want to replace unhealthy then update | 14:46 |
| shardy | heat treats the unhealthy resource as failed so it gets replaced | 14:47 |
| blinky_ghost | shardy: but I tested the unhealthy on the nova server and it messes the indexes because I use fixed ips. Are you saying I need to mark unhealthy the neutron port? | 14:47 |
| shardy | blinky_ghost: you need to mark the ResoureGroup member unhealthy, whatever that is | 14:48 |
| shardy | blinky_ghost: in the case of tripleo, it's a nested stack containing a server and some ports | 14:48 |
| shardy | they all get replaced on the next update | 14:48 |
| shardy | marking just the server (or port associated) probably won't work IME | 14:49 |
| blinky_ghost | shardy: In this example http://paste.openstack.org/show/636188/ I have a resource group with 3 servers and I want to replace node 0. Which resources should I mark unhealthy? | 14:59 |
| openstackgerrit | Merged openstack/heat master: Refactor unit tests for server network https://review.openstack.org/529608 | 14:59 |
| shardy | blinky_ghost: is each server, and the ports etc it depends on, encapsulated in a nested stack you scale out via ResourceGroup? | 15:01 |
| shardy | blinky_ghost: the way I'd do it is do openstack stack resource list <nested stack ID> "0" | 15:01 |
| shardy | blinky_ghost: where the ID is e.g 0b2333d2-2d58-4b9a-a497-b3373f00116b for rabbitmqcluster | 15:01 |
| shardy | blinky_ghost: if you have coupling based on the index over multiple resource groups (which is a bad idea) then you may have to replicate the mark unhealthy in more than one place | 15:02 |
| blinky_ghost | shardy: It's seems to be working openstack stack resource mark unhealthy f23d7cfc-1653-43e1-bac2-ab5533dfaf9d rabbitmqserver. I marked unhealthy the OS::Nova::Server and it's seems to be working :) | 15:03 |
| blinky_ghost | thanks | 15:03 |
| blinky_ghost | shardy: usually I would mark all the resource unhealthy for that index (without fixed ips), but since I want to reuse the ips, I just mark the Nova Server. Does it make sense? | 15:05 |
| shardy | blinky_ghost: yes that makes sense, but IME it's best to scale out the server and all the things it depends on, as it's more likely you'll encounter issues re dependencies if you try to scale ports and servers via independent resource groups | 15:34 |
| shardy | blinky_ghost: that said, if it works for you, then all good :) | 15:35 |
| *** blomquisg_gone is now known as blomquisg | 15:38 | |
| *** tiantian has joined #heat | 15:45 | |
| *** Guest74108 has quit IRC | 15:45 | |
| *** bnemec-xmas is now known as bnemec | 15:46 | |
| *** huangtianhua_ has quit IRC | 15:47 | |
| *** mchlumsky has quit IRC | 15:48 | |
| *** mchlumsky has joined #heat | 15:49 | |
| blinky_ghost | shardy: In this case all I want is to replace the server and reuse the ips, for example if the compute node dies. Scale out is a different thing, I would change the counter and use new ips. | 15:52 |
| *** ricolin has quit IRC | 15:53 | |
| *** vishwanathj has quit IRC | 16:12 | |
| *** armaan has joined #heat | 17:00 | |
| *** mwynne has joined #heat | 17:19 | |
| *** d0ugal has quit IRC | 17:19 | |
| *** d0ugal has joined #heat | 17:19 | |
| *** fragatina has joined #heat | 17:29 | |
| *** fragatin_ has joined #heat | 17:31 | |
| *** fragatina has quit IRC | 17:34 | |
| *** gkadam_ has joined #heat | 17:36 | |
| *** armaan has quit IRC | 17:41 | |
| *** mdnadeem has quit IRC | 17:42 | |
| *** armaan has joined #heat | 17:43 | |
| *** armaan has quit IRC | 17:43 | |
| *** armaan has joined #heat | 17:45 | |
| *** armaan has quit IRC | 17:51 | |
| *** armaan has joined #heat | 17:51 | |
| *** catintheroof has joined #heat | 17:51 | |
| *** armaan has quit IRC | 17:55 | |
| *** derekh has quit IRC | 17:59 | |
| *** armaan has joined #heat | 17:59 | |
| *** armaan has quit IRC | 18:07 | |
| *** harlowja has joined #heat | 18:12 | |
| *** fragatin_ has quit IRC | 18:31 | |
| *** serlex has quit IRC | 18:37 | |
| *** blinky_ghost has quit IRC | 18:56 | |
| *** chenying has quit IRC | 18:58 | |
| *** fishbone_ has joined #heat | 19:08 | |
| *** gfidente has quit IRC | 19:15 | |
| *** dbecker has quit IRC | 19:25 | |
| *** ayoung has joined #heat | 19:42 | |
| *** flwang1 has quit IRC | 19:53 | |
| *** d0ugal has quit IRC | 20:00 | |
| *** shardy has quit IRC | 20:04 | |
| *** dbecker has joined #heat | 20:13 | |
| *** d0ugal has joined #heat | 20:18 | |
| *** dbecker has quit IRC | 20:24 | |
| *** fragatina has joined #heat | 20:24 | |
| *** flwang1 has joined #heat | 20:26 | |
| *** catintheroof has quit IRC | 20:46 | |
| *** catintheroof has joined #heat | 20:46 | |
| *** catintheroof has quit IRC | 20:51 | |
| *** fragatina has quit IRC | 21:00 | |
| *** Smeared_Beard has quit IRC | 21:05 | |
| *** jose-phillips has joined #heat | 21:23 | |
| *** fragatina has joined #heat | 21:31 | |
| *** fragatina has quit IRC | 21:32 | |
| *** fragatina has joined #heat | 21:32 | |
| *** jcoufal has quit IRC | 21:42 | |
| *** rcernin has joined #heat | 21:45 | |
| *** hongbin has quit IRC | 21:45 | |
| *** fragatina has quit IRC | 22:44 | |
| *** catintheroof has joined #heat | 22:46 | |
| openstackgerrit | Nicolas Helgeson proposed openstack/heat master: Add project tags to heat orchestration https://review.openstack.org/507962 | 22:56 |
| *** blomquisg is now known as blomquisg_gone | 23:05 | |
| *** gkadam has joined #heat | 23:16 | |
| *** gkadam_ has quit IRC | 23:16 | |
| *** jtomasek has quit IRC | 23:17 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!