| *** swygue has quit IRC | 00:04 | |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add call to preview stack with template and params https://review.openstack.org/65576 | 00:14 |
|---|---|---|
| *** sjmc has quit IRC | 00:15 | |
| *** rongze has joined #heat | 00:23 | |
| *** alexheneveld has quit IRC | 00:25 | |
| *** alexheneveld has joined #heat | 00:28 | |
| *** slong has joined #heat | 00:30 | |
| *** slong_ has quit IRC | 00:31 | |
| *** IlyaE has joined #heat | 00:46 | |
| *** rongze has quit IRC | 00:52 | |
| *** tango has quit IRC | 00:59 | |
| *** rwsu has quit IRC | 01:00 | |
| openstackgerrit | Randall Burt proposed a change to openstack/heat: Refactor CLB to work with groups https://review.openstack.org/65586 | 01:05 |
| *** yogesh has joined #heat | 01:06 | |
| *** randallburt has quit IRC | 01:13 | |
| *** pvaneck has quit IRC | 01:13 | |
| *** rongze has joined #heat | 01:23 | |
| *** rongze has quit IRC | 01:28 | |
| *** nosnos has joined #heat | 01:29 | |
| *** yogesh has quit IRC | 01:34 | |
| *** yogesh has joined #heat | 01:35 | |
| *** kebray_ has quit IRC | 01:37 | |
| *** yogesh has quit IRC | 01:39 | |
| *** slong has quit IRC | 01:42 | |
| *** larsks is now known as larsks|Zzz | 01:45 | |
| *** sballe has quit IRC | 01:47 | |
| *** sballe has joined #heat | 01:47 | |
| *** sballe has quit IRC | 01:55 | |
| *** rongze has joined #heat | 01:59 | |
| *** IlyaE has quit IRC | 02:12 | |
| *** plomakin_ has joined #heat | 02:14 | |
| *** TonyBurn has joined #heat | 02:14 | |
| *** sgran_ has joined #heat | 02:14 | |
| *** grncdr has quit IRC | 02:15 | |
| *** sgran has quit IRC | 02:15 | |
| *** plomakin has quit IRC | 02:15 | |
| *** TonyBurn_ has quit IRC | 02:15 | |
| *** erkules_ has joined #heat | 02:16 | |
| *** grncdr has joined #heat | 02:18 | |
| *** erkules has quit IRC | 02:19 | |
| *** grncdr has left #heat | 02:29 | |
| *** alexheneveld has quit IRC | 02:34 | |
| *** yogesh has joined #heat | 02:45 | |
| *** nati_ueno has quit IRC | 02:50 | |
| *** yogesh has quit IRC | 02:50 | |
| *** alienyyg has joined #heat | 02:52 | |
| *** andersonvom has quit IRC | 02:52 | |
| *** alienyyg has quit IRC | 02:56 | |
| *** alienyyg has joined #heat | 02:59 | |
| *** blamar has quit IRC | 03:05 | |
| *** alienyyg has quit IRC | 03:05 | |
| *** spzala has quit IRC | 03:05 | |
| *** liang has joined #heat | 03:06 | |
| *** alienyyg has joined #heat | 03:08 | |
| *** kebray_ has joined #heat | 03:11 | |
| *** andersonvom has joined #heat | 03:12 | |
| *** andersonvom has quit IRC | 03:12 | |
| *** andersonvom1 has joined #heat | 03:12 | |
| *** blamar has joined #heat | 03:14 | |
| asalkeld_ | stevebaker, you busy with the get_file bp? | 03:27 |
| stevebaker | asalkeld_: yep, well I'm actually busy practicing for tomorrow's webinar | 03:28 |
| asalkeld_ | O what webinar is that? | 03:28 |
| asalkeld_ | ptl? | 03:28 |
| stevebaker | state of the project, this is the last one | 03:28 |
| asalkeld_ | ok, power on then ;) | 03:29 |
| *** alienyyg has quit IRC | 03:29 | |
| *** gokrokve has joined #heat | 03:29 | |
| openstackgerrit | A change was merged to openstack/heat-templates: Fix MultiNode_DevStack template https://review.openstack.org/64833 | 03:33 |
| *** alienyyg has joined #heat | 03:33 | |
| *** kebray_ has quit IRC | 03:38 | |
| *** alienyyg has quit IRC | 03:39 | |
| *** kebray_ has joined #heat | 03:40 | |
| *** kebray_ has quit IRC | 03:55 | |
| stevebaker | asalkeld_: this one needs one more review https://review.openstack.org/#/c/64900/ | 04:02 |
| asalkeld_ | looking | 04:02 |
| *** kebray_ has joined #heat | 04:03 | |
| *** kebray_ has quit IRC | 04:04 | |
| asalkeld_ | zip(*[(( | 04:05 |
| asalkeld_ | ga | 04:05 |
| *** harlowja is now known as harlowja_away | 04:07 | |
| asalkeld_ | stevebaker, i didn't review that earlier cos' it hurt my eyes/brain | 04:11 |
| asalkeld_ | still does | 04:11 |
| andersonvom1 | asalkeld_: I feel ya! ;) | 04:12 |
| asalkeld_ | don't like code like that | 04:12 |
| *** andersonvom1 is now known as andersonvom | 04:12 | |
| asalkeld_ | clever code is not clever IMHO | 04:12 |
| andersonvom | hey, since you guys are reviewing stuff... I've been having problems getting jenkins to accept these two patches because of tempest tests. do you have any idea on how to proceed? | 04:13 |
| andersonvom | https://review.openstack.org/#/c/58641/ | 04:13 |
| andersonvom | https://review.openstack.org/#/c/63041/ | 04:14 |
| *** harlowja_away is now known as harlowja | 04:14 | |
| andersonvom | asalkeld_: agreed. clever code to me means it's concise and anyone can immediately understand | 04:14 |
| stevebaker | andersonvom: can you run those tempest tests locally? | 04:17 |
| asalkeld_ | andersonvom, is says: Please rebase your change and upload a new patchset. | 04:17 |
| *** cody-somerville has quit IRC | 04:17 | |
| *** kebray_ has joined #heat | 04:19 | |
| *** arbylee has joined #heat | 04:22 | |
| *** nkhare has joined #heat | 04:25 | |
| andersonvom | asalkeld_: which patch? | 04:28 |
| andersonvom | stevebaker: I have not. I have actually been trying to figure out how to do that. :$ | 04:29 |
| stevebaker | andersonvom: if you use devstack its pretty easy | 04:29 |
| andersonvom | I do | 04:29 |
| *** kebray_ has quit IRC | 04:30 | |
| stevebaker | step 1: enable tempest in devstack | 04:30 |
| stevebaker | there is no step 2 | 04:30 |
| asalkeld_ | well you have to run tempest | 04:30 |
| andersonvom | huh | 04:30 |
| andersonvom | interesting | 04:30 |
| stevebaker | but seriously, once the devstack is running, cd tempest; testr run test_non_empty_stack | 04:31 |
| *** cody-somerville has joined #heat | 04:31 | |
| *** kebray_ has joined #heat | 04:32 | |
| andersonvom | cool. just chatting to the misses here and will try that in a short while. =D | 04:33 |
| andersonvom | thanks. will keep you posted | 04:33 |
| *** kebray_ has quit IRC | 04:38 | |
| *** nosnos_ has joined #heat | 04:47 | |
| *** rongze has quit IRC | 04:47 | |
| *** rushiagr has joined #heat | 04:49 | |
| *** cody-somerville_ has joined #heat | 04:49 | |
| *** sdake_ has joined #heat | 04:49 | |
| *** asalkeld__ has joined #heat | 04:50 | |
| *** cody-somerville has quit IRC | 04:55 | |
| *** liang has quit IRC | 04:55 | |
| *** nosnos has quit IRC | 04:55 | |
| *** asalkeld_ has quit IRC | 04:55 | |
| *** sdake has quit IRC | 04:55 | |
| *** liang has joined #heat | 04:57 | |
| *** arbylee has quit IRC | 05:10 | |
| *** rongze has joined #heat | 05:18 | |
| *** kebray_ has joined #heat | 05:19 | |
| *** rongze has quit IRC | 05:22 | |
| *** rongze has joined #heat | 05:23 | |
| andersonvom | so, once I enable tempest, do I run it with "testr run —parallel tempest"? | 05:26 |
| andersonvom | oh, just saw stevebaker's msg | 05:26 |
| *** chandankumar has joined #heat | 05:27 | |
| *** rongze has quit IRC | 05:27 | |
| *** skraynev has quit IRC | 05:30 | |
| *** skraynev has joined #heat | 05:33 | |
| skraynev | Morning All) It was good holidays))) Congratulations to all a Happy 2014 :) | 05:35 |
| skraynev | therve, did you ask me? | 05:36 |
| *** gokrokve has quit IRC | 05:38 | |
| *** chandankumar_ has joined #heat | 05:39 | |
| *** nati_ueno has joined #heat | 05:43 | |
| *** kebray_ has quit IRC | 05:44 | |
| openstackgerrit | Jenkins proposed a change to openstack/heat: Imported Translations from Transifex https://review.openstack.org/64504 | 06:07 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Check resources are in expected states https://review.openstack.org/58641 | 06:14 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add way to group exceptions in DependencyTaskGroup https://review.openstack.org/62479 | 06:14 |
| *** rongze has joined #heat | 06:23 | |
| *** rongze has quit IRC | 06:24 | |
| *** rongze_ has joined #heat | 06:24 | |
| *** chandankumar has quit IRC | 06:25 | |
| *** harlowja is now known as harlowja_away | 06:25 | |
| *** gokrokve has joined #heat | 06:49 | |
| *** sgran_ is now known as sgran | 06:53 | |
| *** bvandenh has joined #heat | 06:54 | |
| *** gokrokve has quit IRC | 06:54 | |
| *** andersonvom has quit IRC | 06:56 | |
| *** sergmelikyan has joined #heat | 07:09 | |
| *** SergeyLukjanov has joined #heat | 07:18 | |
| openstackgerrit | Serg Melikyan proposed a change to openstack/heat: Implements Nova FloatingIP resources https://review.openstack.org/62836 | 07:20 |
| *** rushiagr is now known as rushiagr_away | 07:26 | |
| *** ifarkas has joined #heat | 07:39 | |
| *** tspatzier has joined #heat | 07:46 | |
| *** SergeyLukjanov has quit IRC | 07:47 | |
| *** SergeyLukjanov has joined #heat | 07:48 | |
| *** sergmelikyan has quit IRC | 07:49 | |
| *** sergmelikyan has joined #heat | 07:54 | |
| *** alienyyg has joined #heat | 07:54 | |
| *** andersonvom has joined #heat | 07:57 | |
| *** erkules_ is now known as erkules | 07:57 | |
| *** jprovazn has joined #heat | 07:58 | |
| *** renlt has joined #heat | 08:00 | |
| *** renlt has quit IRC | 08:00 | |
| *** rushiagr_away is now known as rushiagr | 08:00 | |
| *** renlt has joined #heat | 08:01 | |
| openstackgerrit | Feng Ju proposed a change to openstack/heat: Fix heat-keystone-setup error when try to create heat user https://review.openstack.org/65238 | 08:02 |
| *** andersonvom has quit IRC | 08:03 | |
| *** rushiagr2 has joined #heat | 08:09 | |
| *** rushiagr has quit IRC | 08:13 | |
| shardy | morning | 08:21 |
| alienyyg | hi shardy | 08:21 |
| *** jistr has joined #heat | 08:24 | |
| *** SergeyLukjanov has quit IRC | 08:44 | |
| *** nati_ueno has quit IRC | 08:47 | |
| *** aignatov has joined #heat | 08:49 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/heat-templates: Add swift hot example. https://review.openstack.org/65323 | 08:54 |
| chmouel | it could be nice to have yaml validation on heat-templates checks for heat-templates repo | 08:54 |
| *** tomek_adamczewsk has joined #heat | 09:00 | |
| shardy | chmouel: Yeah I've thought the same thing | 09:02 |
| *** tomek_adamczews1 has joined #heat | 09:02 | |
| *** _ruhe is now known as ruhe | 09:04 | |
| *** tomek_adamczewsk has quit IRC | 09:05 | |
| *** che-arne has quit IRC | 09:07 | |
| *** rushiagr2 is now known as rushiagr | 09:08 | |
| *** tomek_adamczews1 has quit IRC | 09:09 | |
| therve | skraynev, Yeah let me remember :) | 09:09 |
| *** tomek_adamczewsk has joined #heat | 09:09 | |
| skraynev | therve: no problem, I will wait your memories))) | 09:10 |
| therve | Oh yeah | 09:10 |
| therve | skraynev, https://bugs.launchpad.net/heat/+bug/1256836 | 09:10 |
| therve | i don't think there is anything to do here, do you think otherwise? | 09:11 |
| chmouel | shardy: ok filled a bp for that https://blueprints.launchpad.net/heat-templates/+spec/templates-validations-jenkins-job | 09:11 |
| *** tomek_adamczewsk has quit IRC | 09:12 | |
| *** tomek_adamczewsk has joined #heat | 09:12 | |
| therve | chmouel, We could call heat template-validate on them | 09:12 |
| chmouel | therve: cool | 09:13 |
| skraynev | therve, yeap, I agree with you. It's completed thing. and we have solution. (unfortunately it looks as "hack") | 09:13 |
| therve | skraynev, OK I'll close it then | 09:14 |
| *** alexheneveld has joined #heat | 09:15 | |
| *** giulivo has joined #heat | 09:15 | |
| skraynev | therve, currently, I offer close it and wait changes in neutron (many vip to lbaas) | 09:15 |
| skraynev | and then make it more clear) | 09:16 |
| *** tomek_adamczewsk has quit IRC | 09:18 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/heat-templates: Add swift hot example. https://review.openstack.org/65323 | 09:19 |
| *** tomek_adamczewsk has joined #heat | 09:19 | |
| *** MaximB has joined #heat | 09:23 | |
| openstackgerrit | A change was merged to openstack/heat: Avoid error on double-delete of nested stack https://review.openstack.org/62752 | 09:24 |
| openstackgerrit | A change was merged to openstack/heat: Fix schema format in template generation unit tests https://review.openstack.org/64899 | 09:25 |
| shardy | woohoo, gate seems to be working again :) | 09:25 |
| *** SergeyLukjanov has joined #heat | 09:25 | |
| skraynev | aha))) I hope, that it's not exception. | 09:26 |
| MaximB | shardy: how would openstack/heat work if I install them on/as a VM on esxi? | 09:27 |
| *** tomek_adamczewsk has quit IRC | 09:28 | |
| openstackgerrit | A change was merged to openstack/heat: Validate number of instance metadata entries https://review.openstack.org/61864 | 09:29 |
| openstackgerrit | A change was merged to openstack/heat: Improve tools/uninstall-heat https://review.openstack.org/64694 | 09:29 |
| chmouel | therve: about heat validate-templates, i think that come back to the discussion we had at the office in the other day if the client or the server do the validation | 09:29 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Added API reference document https://review.openstack.org/63665 | 09:29 |
| skraynev | shardy, You was right;) take off my hat to you | 09:30 |
| therve | chmouel, Yeah. There was a problem that you define resources on the server side, so we wouldn't be able to validate that on the client | 09:30 |
| chmouel | therve: trying to add validate-templates in heat-templates would have to be done on a running v | 09:30 |
| therve | Sure that sounds like an easy tempest task | 09:30 |
| chmouel | i am not sure how that plugs all together in jenkins | 09:31 |
| chmouel | you upload a review at first to heat-templates, jenkins pick it up and launch a devstack | 09:32 |
| *** sergmelikyan has quit IRC | 09:32 | |
| *** giulivo has quit IRC | 09:32 | |
| *** giulivo has joined #heat | 09:33 | |
| chmouel | but I am not sure how tempest takes that get validated by tempest | 09:33 |
| *** rushiagr is now known as rushiagr_away | 09:36 | |
| shardy | MaximB: It will work, but ideally the nova-compute part of openstack needs to be on bare metal, or VM's running on VM's can be very slow (unless you can use nested virt) | 09:37 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Remove vim header https://review.openstack.org/64326 | 09:37 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Enable deleting multiple stacks with single call https://review.openstack.org/60664 | 09:37 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Remove ununsed httplib2 requirement https://review.openstack.org/65251 | 09:37 |
| shardy | chmouel: I think it would just be a case of wiring heat-templates into the devstack gate, but we could just have a "unit" test which does a json/yaml syntax check first | 09:38 |
| *** renlt has quit IRC | 09:38 | |
| MaximB | shardy: but how do you install openstack on esxi? you install it on a vm machine inside esx or just any machine or? | 09:38 |
| shardy | the sytax check should be simple, and would be better than what we have now :) | 09:38 |
| shardy | MaximB: Yeah, for test/evaluation purposes you could install it all inside one VM | 09:39 |
| shardy | that is what many folks do for development, but as I said, it is slow for anything other than trivial testing | 09:40 |
| openstackgerrit | huangtianhua proposed a change to openstack/heat: Fixes template not using the JSON or YAML format https://review.openstack.org/65644 | 09:40 |
| shardy | personally I prefer running it all natively on my laptop | 09:40 |
| chmouel | shardy: yeah i was thinking about the unit test at first | 09:40 |
| *** alienyyg has quit IRC | 09:41 | |
| MaximB | shardy: so I can install it on my laptop and connect to the esxi? | 09:41 |
| *** alienyyg has joined #heat | 09:41 | |
| shardy | MaximB: Probably, there is a vmware driver for nova-compute, but I've never tried it | 09:44 |
| shardy | http://docs.openstack.org/trunk/config-reference/content/vmware.html | 09:44 |
| *** rushiagr_away is now known as rushiagr | 09:47 | |
| *** derekh has joined #heat | 09:52 | |
| *** julienvey has quit IRC | 09:53 | |
| *** bgorski has joined #heat | 09:54 | |
| *** tspatzier has quit IRC | 09:56 | |
| *** tspatzier has joined #heat | 09:58 | |
| *** nati_ueno has joined #heat | 09:58 | |
| *** andersonvom has joined #heat | 10:01 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/heat-templates: Make pep8-cfn-json2yaml PEP8 free https://review.openstack.org/65646 | 10:01 |
| *** plomakin_ has quit IRC | 10:01 | |
| *** plomakin has joined #heat | 10:01 | |
| *** nati_ueno has quit IRC | 10:02 | |
| *** sergmelikyan has joined #heat | 10:05 | |
| *** andersonvom has quit IRC | 10:05 | |
| *** tomek_adamczewsk has joined #heat | 10:08 | |
| *** asalkeld__ is now known as asalkeld | 10:14 | |
| *** rushiagr2 has joined #heat | 10:14 | |
| *** liang has quit IRC | 10:15 | |
| *** rushiagr has quit IRC | 10:18 | |
| sergmelikyan | shardy, I would like to implement https://blueprints.launchpad.net/heat/+spec/discover-catalog-resources, if there is no objections. This seems like valuable feature. | 10:27 |
| shardy | sergmelikyan: Ok, I can assign it to you | 10:28 |
| shardy | sergmelikyan: done! | 10:29 |
| sergmelikyan | shardy, Thx! | 10:30 |
| *** SergeyLukjanov has quit IRC | 10:30 | |
| *** aignatov has quit IRC | 10:37 | |
| *** jprovazn has quit IRC | 10:39 | |
| *** ruhe is now known as _ruhe | 10:42 | |
| *** jpeeler has quit IRC | 10:44 | |
| *** nosnos_ has quit IRC | 10:46 | |
| *** aignatov has joined #heat | 11:05 | |
| *** _ruhe is now known as ruhe | 11:06 | |
| *** tspatzier_ has joined #heat | 11:06 | |
| *** tspatzier has quit IRC | 11:10 | |
| *** nkhare has quit IRC | 11:11 | |
| *** jprovazn has joined #heat | 11:11 | |
| *** ruhe is now known as _ruhe | 11:18 | |
| *** _ruhe is now known as ruhe | 11:23 | |
| *** asalkeld has quit IRC | 11:24 | |
| *** asalkeld has joined #heat | 11:27 | |
| *** che-arne has joined #heat | 11:37 | |
| *** jamieh has joined #heat | 11:44 | |
| *** rongze_ has quit IRC | 11:46 | |
| alienyyg | hi guys,thank you for your help, now the autoscaling works for me, and after tha I wanna to learn more about this project, I mean be a devoloper, not only user, so can you guys give me some advice how can I get into this? | 11:46 |
| MaximB | shardy: is there a minimum specs for a vm install of openstack? | 11:46 |
| MaximB | also the wiki says http://docs.openstack.org/havana/install-guide/install/apt/content/ ubuntu 12.04, would it work for a new version like 13.04 or 13.10? | 11:49 |
| shardy | MaximB: There are details of hardware requirements in the openstack operations guide | 11:51 |
| shardy | MaximB: for a minimal all-in-one test environment, I'd suggest 8G ram minimum | 11:52 |
| *** cody-somerville_ has quit IRC | 11:52 | |
| shardy | Re Ubuntu, I expect so but I think the docs only cover the LTS version | 11:53 |
| shardy | I run only on Fedora and RHEL ;) | 11:53 |
| sergmelikyan | alienyyg, as usually :) Send changes on review to review.openstack.org, See bugs and BP on launchpad.net/Heat | 11:54 |
| shardy | alienyyg: good news, start by looking at https://wiki.openstack.org/wiki/Heat/DevelopmentProcess | 11:55 |
| *** ruhe is now known as ruhe_away | 12:05 | |
| *** cody-somerville_ has joined #heat | 12:06 | |
| *** ruhe_away is now known as ruhe | 12:08 | |
| *** cody-somerville_ is now known as cody-somerville | 12:10 | |
| *** dims_ is now known as dims | 12:11 | |
| MaximB | shardy: about this guide: http://docs.openstack.org/havana/install-guide/install/apt/content/basics-networking.html ,do I need to create 2 machines? if so, what specs each machine should have? | 12:15 |
| *** rongze has joined #heat | 12:21 | |
| *** rongze_ has joined #heat | 12:22 | |
| *** derekh has quit IRC | 12:23 | |
| *** ruhe is now known as _ruhe | 12:23 | |
| *** radez_g0n3 is now known as radez | 12:25 | |
| *** rongze has quit IRC | 12:26 | |
| *** rpothier has quit IRC | 12:32 | |
| *** renlt has joined #heat | 12:33 | |
| *** renlt has quit IRC | 12:34 | |
| *** renlt has joined #heat | 12:35 | |
| *** rushiagr2 has quit IRC | 12:36 | |
| sdake_ | morning | 12:39 |
| *** _ruhe is now known as ruhe | 12:50 | |
| alienyyg | shardy: too many there, so it means the developer is review the code ,fix bugs, or raise blueprint and then make it become reality, so know how to do it now | 12:51 |
| *** derekh has joined #heat | 12:57 | |
| *** alienyyg has quit IRC | 13:07 | |
| *** rongze_ has quit IRC | 13:07 | |
| *** sgordon has joined #heat | 13:09 | |
| chmouel | anyone knows if heat bulitins ssl setup is supposed to work? | 13:12 |
| *** rongze has joined #heat | 13:17 | |
| *** larsks|Zzz is now known as larsks | 13:20 | |
| openstackgerrit | Thomas Spatzier proposed a change to openstack/heat: Refactor Parameters Schema based on common Schema https://review.openstack.org/65688 | 13:20 |
| *** alexpilotti has joined #heat | 13:25 | |
| *** radez is now known as radez_g0n3 | 13:31 | |
| *** rushiagr2 has joined #heat | 13:37 | |
| *** rpothier has joined #heat | 13:39 | |
| *** renlt has quit IRC | 13:41 | |
| *** rushiagr2 is now known as rushiagr | 13:51 | |
| *** sballe has joined #heat | 13:52 | |
| *** jdob has joined #heat | 14:01 | |
| *** andersonvom has joined #heat | 14:01 | |
| *** jpeeler has joined #heat | 14:05 | |
| *** jpeeler has quit IRC | 14:05 | |
| *** jpeeler has joined #heat | 14:05 | |
| *** andersonvom has quit IRC | 14:06 | |
| *** andersonvom has joined #heat | 14:08 | |
| *** andersonvom has quit IRC | 14:18 | |
| *** jcru has joined #heat | 14:22 | |
| *** SergeyLukjanov has joined #heat | 14:23 | |
| therve | tspatzier_, Looking at your patch, it seems that changing things into lower-case is backward incompatible | 14:24 |
| therve | In particular as the ouput of template validation | 14:25 |
| tspatzier_ | hi therve | 14:25 |
| therve | Sorry, hi :) | 14:26 |
| tspatzier_ | The change should be only internally in the code used by the engine. The templates can still have uppercase, and when read in it is transformed to lowercase | 14:26 |
| therve | tspatzier_, The output changes, though | 14:26 |
| *** tspatzier_ is now known as tspatzier | 14:27 | |
| therve | See http://paste.openstack.org/show/60878/ as a before/after | 14:27 |
| tspatzier | What is the output used for? I thought it was "just" objects used inside the engine and in testcases. So nothing really critical to the user. | 14:28 |
| therve | No the output is returned by an API | 14:28 |
| therve | It is used for example by horizon | 14:28 |
| therve | There are already some bad stuff in it, like constraints are normalized between HOT and CFN | 14:29 |
| therve | But I'm not sure we should break compatibility here. I guess lower-case isn't so bad, but still | 14:29 |
| tspatzier | Ok, got it. | 14:30 |
| *** pbacterio has joined #heat | 14:31 | |
| *** pbacterio has left #heat | 14:31 | |
| tspatzier | This is really a mess actually. | 14:31 |
| therve | Yeah I know :/ | 14:32 |
| tspatzier | So I think internally we should really just go with one option. And then maybe have some to_external function? | 14:33 |
| tspatzier | Currently, everything is kept as it comes from the template, right? | 14:34 |
| therve | Well HOT is translated | 14:34 |
| therve | At least some keys are | 14:35 |
| tspatzier | That also something I am changing in that patch. No translation of params any more, but plain read as schema objects. | 14:37 |
| therve | Well you're translation CFN presumably? | 14:38 |
| *** andersonvom has joined #heat | 14:38 | |
| tspatzier | previously we translated parts into CFN | 14:38 |
| tspatzier | therve: So for validate, is it always input == output? | 14:42 |
| *** vijendar has joined #heat | 14:42 | |
| therve | tspatzier, No, it's output == CFN (mostly) | 14:42 |
| *** arbylee has joined #heat | 14:46 | |
| tspatzier | Hm, so maybe so to_external code would be best. Internally it would really be nice to just deal with the objects and not distinguish between CFN and HOT. | 14:47 |
| tspatzier | And then some code to externalize the schema objects. I could also revert the constants to uppercase and deal with upper/lower in comparisions inside the code, but that would leave the additional attributes (hidden, required) in your sample. | 14:49 |
| therve | I think additional fields are okay, you can ignore them | 14:51 |
| *** arbylee1 has joined #heat | 14:51 | |
| therve | I also think it may be reasonable to change the output for HOT, especially if we stard handling constraints properly | 14:51 |
| therve | But it seems that CFN template should still get out as CFN syntax | 14:52 |
| tspatzier | Sure | 14:52 |
| tspatzier | Is template-validate the only place where this hits? | 14:52 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Unscoped List Stacks https://review.openstack.org/63041 | 14:53 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add filter and pagination to stack_get_all https://review.openstack.org/63039 | 14:53 |
| therve | I would say so | 14:53 |
| *** arbylee has quit IRC | 14:54 | |
| *** MaximB has quit IRC | 14:55 | |
| tspatzier | Ok. I guess I will add a comment to the review about what we talked. Also what the options would be. Then think about it again and change it. ... and wait for other comments :-) | 14:56 |
| therve | Thansk :) | 14:57 |
| *** alienyyg has joined #heat | 15:01 | |
| *** alienyyg has quit IRC | 15:03 | |
| *** swygue has joined #heat | 15:06 | |
| *** john-n-seattle1 has joined #heat | 15:07 | |
| *** john-n-seattle1 has left #heat | 15:09 | |
| *** john-n-seattle1 has joined #heat | 15:10 | |
| *** zaneb has quit IRC | 15:12 | |
| *** alienyyg has joined #heat | 15:13 | |
| *** alienyyg has left #heat | 15:14 | |
| sdake_ | I figured out how to mark all read in thunderbird!! | 15:15 |
| *** andersonvom has quit IRC | 15:15 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/python-heatclient: Use python requests lib for all our http thingies https://review.openstack.org/65704 | 15:15 |
| *** topol has joined #heat | 15:19 | |
| *** andersonvom has joined #heat | 15:22 | |
| *** rwsu has joined #heat | 15:23 | |
| *** dims has quit IRC | 15:23 | |
| *** tspatzier has quit IRC | 15:24 | |
| *** dims has joined #heat | 15:24 | |
| *** IlyaE has joined #heat | 15:27 | |
| *** rongze has quit IRC | 15:31 | |
| *** granny has joined #heat | 15:31 | |
| *** jdob has quit IRC | 15:35 | |
| *** jdob has joined #heat | 15:36 | |
| *** zaneb has joined #heat | 15:38 | |
| *** randallburt has joined #heat | 15:41 | |
| *** randallburt has quit IRC | 15:41 | |
| *** randallburt has joined #heat | 15:42 | |
| *** sjmc has joined #heat | 15:47 | |
| *** rcleere has joined #heat | 15:48 | |
| *** granny has quit IRC | 15:48 | |
| zaneb | therve: see https://review.openstack.org/#/c/65325/1/heat/engine/service.py for an example of how we should handle it | 15:51 |
| *** jmckind has joined #heat | 15:52 | |
| *** chandankumar_ has quit IRC | 15:54 | |
| *** tims has joined #heat | 15:56 | |
| *** tims has quit IRC | 15:56 | |
| *** tims has joined #heat | 15:56 | |
| *** jhurlbert has quit IRC | 15:57 | |
| *** funzo has quit IRC | 15:59 | |
| *** funzo has joined #heat | 16:01 | |
| openstackgerrit | DennyZhang proposed a change to openstack/heat: Remove vim header https://review.openstack.org/64691 | 16:09 |
| *** spzala has joined #heat | 16:09 | |
| *** jdag is now known as jdaggett | 16:10 | |
| *** ifarkas has quit IRC | 16:11 | |
| *** alexpilotti has quit IRC | 16:20 | |
| shadower | I've been reading the docs and it occurred to me that OS::Heat::RandomString should be using a cryptographically secure PRNG instead of Mersenne twister | 16:21 |
| shadower | http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Heat::RandomString | 16:21 |
| shadower | happy to look into it myself -- just checking if I'm not missing something obvious | 16:21 |
| therve | zaneb, Hum it's not totally clear what this code does :) | 16:22 |
| zaneb | therve: it retrieves the label from the parameter class by calling label() and adds it to the output to be returned with the key expected by the API | 16:23 |
| zaneb | therve: we need to start doing that for all of the stuff returned from the API, instead of grabbing it from the dictionary the parameter was defined with | 16:24 |
| zaneb | and move the function into heat.engine.api | 16:24 |
| therve | Mokay, it looks a bit awkward though | 16:24 |
| *** IlyaE has quit IRC | 16:25 | |
| *** kebray_ has joined #heat | 16:25 | |
| zaneb | more awkward than the format_*() functions in https://github.com/openstack/heat/blob/master/heat/engine/api.py ? | 16:25 |
| therve | Yeah, because we do "give me the param schema. Oh but just replace that one key" | 16:26 |
| *** alexheneveld has quit IRC | 16:27 | |
| zaneb | it's adding, not replacing | 16:27 |
| zaneb | but yes, it's awkward at the moment because there's only one thing we're accessing that way | 16:27 |
| zaneb | but after tspatzier's patch goes in, we need to access _everything_ that way | 16:27 |
| zaneb | and then it won't be awkward | 16:27 |
| therve | It's not replacing? Why would label not be in the dict? | 16:27 |
| zaneb | we don't make it available in ParamSchema | 16:28 |
| therve | Okay | 16:29 |
| zaneb | actually, you might be right | 16:29 |
| *** tomek_adamczewsk has quit IRC | 16:29 | |
| therve | It seems the answer is "it depends" :) | 16:30 |
| zaneb | https://review.openstack.org/#/c/62418/8/heat/engine/hot.py | 16:30 |
| zaneb | yeah, we do if it's available, but it's not necessarily the one you want | 16:30 |
| therve | Anyway I see your point | 16:30 |
| therve | I really wish we used HOT everywhere internally, and just have translation at the borders | 16:31 |
| zaneb | I really wish we were completely independent of the template format everywhere internally, and just have translation at the borders | 16:31 |
| zaneb | going fishing through the template keys is how we got into this mess | 16:32 |
| therve | Yeah | 16:32 |
| zaneb | going fishing through HOT template keys doesn't solve the problem, it just changes it | 16:32 |
| therve | Right by HOT I mean HOT schema objects | 16:32 |
| zaneb | if we have programmatic APIs for everything, we will be able to plug in a new template formate easily | 16:33 |
| zaneb | or, for that matter, modify HOT easily | 16:33 |
| zaneb | yep, that is essentially the goal | 16:33 |
| zaneb | there will likely be HOT schema objects and cfn schema objects that read different templates, but have the same API | 16:34 |
| * therve nods | 16:35 | |
| therve | zaneb, On that subject, do you think we should change the output of template-validate for HOT templates? | 16:35 |
| therve | It looks like we would have trouble translation constraints, for example | 16:36 |
| *** randallburt1 has joined #heat | 16:36 | |
| zaneb | yeah | 16:36 |
| *** rushiagr2 has joined #heat | 16:36 | |
| zaneb | tbh I think we will have to add a new validate API | 16:36 |
| zaneb | or perhaps a separate get_params API | 16:36 |
| therve | Oh noes new APIs :/ | 16:37 |
| zaneb | and fix some of my original mistakes, and some of the things that have emerged from HOT | 16:37 |
| zaneb | better a new API than to make an old one incompatible with existing clients | 16:38 |
| *** randallburt has quit IRC | 16:38 | |
| therve | Agreed. We still need to make the old compatible, though | 16:38 |
| *** rushiagr has quit IRC | 16:39 | |
| *** randallburt1 has quit IRC | 16:42 | |
| *** che-arne has quit IRC | 16:49 | |
| *** che-arne has joined #heat | 16:50 | |
| *** aignatov has quit IRC | 16:51 | |
| *** pablosan has joined #heat | 16:53 | |
| *** tspatzier has joined #heat | 16:53 | |
| *** derekh has quit IRC | 17:02 | |
| *** rushiagr2 is now known as rushiagr_away | 17:03 | |
| *** gokrokve has joined #heat | 17:03 | |
| *** alexpilotti has joined #heat | 17:09 | |
| *** alexheneveld has joined #heat | 17:10 | |
| *** randallburt has joined #heat | 17:15 | |
| *** randallburt has quit IRC | 17:20 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/python-heatclient: Use python requests lib for all our http thingies https://review.openstack.org/65704 | 17:20 |
| *** m4dcoder has joined #heat | 17:23 | |
| m4dcoder | alexpilotti: you there? | 17:23 |
| alexpilotti | m4dcoder: sure | 17:23 |
| *** m4dcoder_ has joined #heat | 17:25 | |
| m4dcoder_ | alexpilotti: i want to see what you have done so far with the windows instance bp | 17:26 |
| m4dcoder_ | alexpilotti: any room for me to work with you on that bp? | 17:26 |
| alexpilotti | m4dcoder_: you mean since yesterday? :-) | 17:26 |
| *** ruhe is now known as _ruhe | 17:27 | |
| m4dcoder_ | alexpilotti: :) i didn't know how long you've been working on that. | 17:27 |
| alexpilotti | m4dcoder_: sure, I'm going to publish the heat templates for Windows this week | 17:28 |
| alexpilotti | m4dcoder_: we have Heat support already in the Windows cloud init: https://github.com/cloudbase/cloudbase-init | 17:28 |
| *** m4dcoder has quit IRC | 17:28 | |
| alexpilotti | m4dcoder_: so basically that bp is mostly related to documenting the Windows specific stuff | 17:29 |
| alexpilotti | m4dcoder_: if you'd like to help, there's always space! :-) | 17:29 |
| m4dcoder_ | alexpilotti: sure. so if i understand what you're saying, the cloudbase init implementation already works on Windows? there's not much dev work, just documentation? | 17:30 |
| alexpilotti | m4dcoder_: correct. We're finishing up the heat templates this week (MSSQL, Exchane, Sharepoint, etc) | 17:30 |
| alexpilotti | m4dcoder_: so if you'd like to help on testing them it'd be great :-) | 17:31 |
| shardy | alexpilotti: are you going to publish those templates somewhere? | 17:31 |
| *** jistr has quit IRC | 17:31 | |
| alexpilotti | shardy: definitely yes | 17:31 |
| m4dcoder_ | alexpilotti: that's awesome! i would like to test that. can you review the ML? shardy posted some comments. does your implementation address those? | 17:32 |
| shardy | alexpilotti: great! If you like, I'm pretty sure we'd be happy to have them in the heat-templates repo :) | 17:32 |
| *** alexheneveld has quit IRC | 17:32 | |
| *** pvaneck has joined #heat | 17:33 | |
| *** swygue has quit IRC | 17:33 | |
| shardy | m4dcoder_: My ML comments may be obsolete by the sounds of it | 17:33 |
| alexpilotti | shardy: sounds good! We're also planning to put cloudbase-init on stackforge | 17:33 |
| *** swygue has joined #heat | 17:34 | |
| shardy | alexpilotti: Are there any plans for cloudbase-init to merge back into cloud-init? | 17:34 |
| shardy | (I remember it was mentioned at one point) | 17:34 |
| alexpilotti | shardy: we're also looking at that, but there's really too much refactoring involved | 17:34 |
| alexpilotti | shardy: actually cloudbase-init started because it was impossible to simply port cloud-init | 17:35 |
| alexpilotti | shardy: there's also teh licensing issue | 17:35 |
| alexpilotti | shardy: cloudbase-init is Apache2, while cloud-init is GPL | 17:35 |
| shardy | alexpilotti: Ok, the one concern I have is divergence between the two tools, but are you aiming to maintain compatibility with cloud-init interfaces indefinitely? | 17:35 |
| alexpilotti | shardy: if by interfaces you mean metadata services, then yes | 17:36 |
| shardy | alexpilotti: Oh really, I incorrectly assumed cloudbase-init was a fork | 17:36 |
| alexpilotti | shardy: nope, it's a total rewrite | 17:36 |
| shardy | alexpilotti: well the main things Heat currently cares about is stuff like the cloud-config DSL and part-handler functionality | 17:37 |
| alexpilotti | shardy: it was written with multiplatform portability in mind and some guys lately used it for a FreeBSD version | 17:37 |
| alexpilotti | shardy: yep, that's part of the multi part user data plugin | 17:37 |
| shardy | alexpilotti: Ok, sounds good :) | 17:38 |
| alexpilotti | shardy: do you have plans for design changes on the part-handler functionality? | 17:39 |
| shardy | alexpilotti, m4dcoder_: I'm really glad you guys are picking this up, windows support is something many users have asked for and so far we've not had the bandwidth to investigate | 17:39 |
| m4dcoder_ | alexpilotti: when you've published, can you send a shout out on the ML along with some instructions on setup? | 17:40 |
| shardy | alexpilotti: No, but we need to ensure what we have stays working (i.e without hacks to support two different cloud-init implemenations) | 17:40 |
| *** rushiagr_away is now known as rushiagr2 | 17:40 | |
| alexpilotti | m4dcoder_: sure! | 17:41 |
| alexpilotti | shardy: I also hate the idea of two separate tools, so if we can overcome the licensing issues and merge the two tools (possibly as a openstack core or at least stackforge project), that would be great | 17:42 |
| shardy | alexpilotti: ++ | 17:42 |
| *** swygue has quit IRC | 17:42 | |
| *** nati_ueno has joined #heat | 17:42 | |
| *** randallburt has joined #heat | 17:44 | |
| *** randallburt has quit IRC | 17:44 | |
| *** randallburt has joined #heat | 17:44 | |
| *** gokrokve has quit IRC | 17:49 | |
| *** gokrokve has joined #heat | 18:01 | |
| *** m4dcoder_ has quit IRC | 18:02 | |
| *** gokrokve has quit IRC | 18:03 | |
| openstackgerrit | A change was merged to openstack/heat: Update log message for scale out/in https://review.openstack.org/64606 | 18:03 |
| *** aignatov has joined #heat | 18:05 | |
| *** rushiagr2 is now known as rushiagr_away | 18:06 | |
| *** harlowja_away is now known as harlowja | 18:07 | |
| *** derekdon has joined #heat | 18:08 | |
| *** sdake_ has quit IRC | 18:12 | |
| *** rushiagr_away is now known as rushiagr2 | 18:12 | |
| *** derekdon has quit IRC | 18:13 | |
| *** yogesh has joined #heat | 18:24 | |
| *** tspatzier has quit IRC | 18:39 | |
| *** tango|2 has joined #heat | 18:42 | |
| *** che-arne has quit IRC | 18:45 | |
| *** rushiagr2 has quit IRC | 18:47 | |
| *** bgorski has quit IRC | 18:49 | |
| *** _ruhe is now known as ruhe | 18:52 | |
| stevebaker | shardy: hey, should I mention V2 API as a possible icehouse feature in the Project Update webinar? | 18:57 |
| *** alexheneveld has joined #heat | 18:59 | |
| radix | hmm | 19:03 |
| radix | is there any code that actually uses heatclient in heat? I see that there's a method to return a heat client, but I don't see anything calling it | 19:03 |
| *** randallburt has quit IRC | 19:03 | |
| radix | (maybe in review?) | 19:05 |
| *** alexpilotti has quit IRC | 19:06 | |
| *** alexpilotti has joined #heat | 19:09 | |
| *** wwallnrr__ has joined #heat | 19:09 | |
| stevebaker | radix: https://review.openstack.org/#/c/58879/ https://review.openstack.org/#/c/58880/ | 19:11 |
| radix | ah, of course, the config management stuff | 19:11 |
| *** SergeyLukjanov has quit IRC | 19:13 | |
| *** aignatov has quit IRC | 19:15 | |
| *** jamieh has quit IRC | 19:18 | |
| zaneb | radix: when we go multiregion we'll need it for nested stacks too | 19:18 |
| *** wwallnrr__ has left #heat | 19:18 | |
| radix | zaneb: so, the other day you said something that I didn't quite notice at the time | 19:21 |
| radix | (changing subjects entirely) | 19:21 |
| radix | <zaneb> 1) split out the database, 2) split the implementation out of the resources into a separate python api 3) create native resources using this api 4) wrap a ReST API around the API | 19:21 |
| radix | zaneb: can you explain #3 a little more? | 19:21 |
| zaneb | so currently we have AWS autoscaling resources... | 19:22 |
| *** rcleere has quit IRC | 19:22 | |
| zaneb | there's no point implementing native ones just yet, because the implementation is still buried in the AWS ones | 19:22 |
| zaneb | but when the implementation is encapsulated behind its own API, we can go ahead and start using that | 19:23 |
| zaneb | and add the ReST client/server interface in the middle later | 19:23 |
| radix | zaneb: are you talking about using the python code in-process, or talking to the separate AS API? | 19:23 |
| zaneb | using the python code in-process for 3 | 19:24 |
| radix | the design that was discussed at the summit and laid out in the specs was resources talking to the AS REST API | 19:24 |
| zaneb | separate api for 4 | 19:24 |
| zaneb | (rest api for 4) | 19:24 |
| zaneb | right, and that is absolutely the goal IMO | 19:24 |
| radix | ok, so you're suggesting this as an intermediate step? | 19:24 |
| zaneb | but that doesn't mean that there are no intermediate steps that are useful | 19:25 |
| zaneb | exactly, yeah | 19:25 |
| zaneb | I'm just concerned about schedule risk | 19:25 |
| radix | yeah, me too | 19:25 |
| zaneb | we're nearly at I-2 | 19:25 |
| zaneb | therve I gather is not devoting time to it any more | 19:25 |
| radix | though I'm wary of spending time on internal resources before getting to the API | 19:25 |
| radix | er, "internal resources" was a bad way of putting that | 19:26 |
| zaneb | I understand | 19:26 |
| radix | ok :) | 19:26 |
| zaneb | I am wary of doing a top down implementation where we only discover if everything fits at the last possible moment | 19:26 |
| radix | yeah | 19:26 |
| radix | I guess ideally those intermediate resources should be trivial, if the other parts are amenable | 19:27 |
| zaneb | indeed | 19:27 |
| radix | anyway, I'm basically trashing all of the BPs right now and recreating the chain of dependencies | 19:27 |
| zaneb | btw the intermediate resource thing was something that shardy pointed out to me | 19:27 |
| radix | they were basically backwards from what they needed to be, and I didn't realize it | 19:27 |
| zaneb | which I agree with, but I would still think we should do it in that order even without them | 19:28 |
| radix | yeah | 19:28 |
| radix | okay | 19:28 |
| radix | well, I will create a BP for them | 19:28 |
| radix | zaneb: btw, afaik the thing you put in #1 doesn't actually need to be #1 | 19:28 |
| zaneb | I think #1 & #2 are somewhat linked | 19:29 |
| radix | but maybe I'm mistaken | 19:29 |
| radix | I'm imagining that the first step is "create a library that generates templates" (basically what you said as #2). then the intermediate resources could be implemented in terms of that library. | 19:30 |
| radix | the intermediate resources would just use a nested stack | 19:30 |
| radix | once we get to a separate process / engine, that needs a different database | 19:30 |
| zaneb | by separate database I mean separate table | 19:31 |
| zaneb | not necessarily it's own db | 19:31 |
| zaneb | but don't store any data in the resource | 19:31 |
| radix | hmm | 19:31 |
| zaneb | storing data in the nested stack is fine | 19:32 |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Close SSH connections in Cloud Servers resource https://review.openstack.org/65761 | 19:33 |
| zaneb | but not in the AutoscalingGroup resource, and especially not in the LaunchConfig resource, &c. | 19:33 |
| radix | okay, I think I know what you mean | 19:34 |
| radix | to be clear | 19:34 |
| jdob | hey all, are there any docs on heat templates besides the example templates repo? i've found the client docs but nothing specific to the template format | 19:34 |
| radix | the state that isn't represented by the generated template or nested stack is what we're talking about, and what's left is basically cooldowns and rolling update tracking info (and maybe something else I'm forgetting) | 19:34 |
| larsks | jdob: There http://docs.openstack.org/developer/heat/template_guide/openstack.html which lists all the supported resource types. | 19:35 |
| zaneb | radix: yeah, exactly. we can't have any of that data tied to a resource in the local stack | 19:35 |
| radix | right | 19:35 |
| radix | okay, that sounds reasonable | 19:35 |
| jdob | larsks: ah cool, that looks like a good starting point, thanks | 19:36 |
| zaneb | radix: as far as everything that happens in the nested stack is concerned, that's already pretty well abstracted and should port right across without much difficulty when that becomes a process boundary | 19:36 |
| *** tomek_adamczewsk has joined #heat | 19:37 | |
| radix | huh. where the heck is that state stored in the AWS-based resources? I'm not seeing any resource_data | 19:37 |
| zaneb | a lot of it is just in the properties | 19:38 |
| *** rcleere has joined #heat | 19:38 | |
| radix | hmm, looks like last adjustment time is stored in metadata | 19:38 |
| zaneb | e.g. when we want to know the launch config, we go fishing for it in the local stack and copy all of its properties into a server | 19:39 |
| radix | cool: metadata.keys()[0] :-( | 19:39 |
| zaneb | yikes | 19:39 |
| radix | I wonder what key that is, haha | 19:39 |
| zaneb | yeah, that's the level of stuff we're dealing with ;) | 19:39 |
| radix | ohhh, it's a dynamic key, it's a timestamp | 19:39 |
| radix | okay, I look forward to obsoleting that code ;-) | 19:39 |
| zaneb | so, this is why we need separate DB tables :D | 19:40 |
| *** IlyaE has joined #heat | 19:40 | |
| radix | hehe | 19:42 |
| *** spzala has quit IRC | 19:44 | |
| *** ruhe is now known as _ruhe | 19:44 | |
| *** blamar has quit IRC | 19:54 | |
| *** gokrokve has joined #heat | 19:56 | |
| jdob | is there a way to specify arbitrary metadata on a resource? more generally, if I make two Server resources with slightly different values and I wanted to tell them apart when i view the resources on a stack | 19:59 |
| *** radez_g0n3 is now known as radez | 20:05 | |
| *** SergeyLukjanov has joined #heat | 20:09 | |
| zaneb | jdob: other than the name you mean? | 20:09 |
| zaneb | there is a metadata section on each resource, but note that is contains data that is available inside the instance | 20:10 |
| zaneb | that may or may not be what you want in this case | 20:10 |
| jdob | ya, because my example was too small. i have 10, of two different types, and i need to tell them apart | 20:10 |
| jdob | thats what i was looking at, the metadata thing. i wasnt sure if the fields were locked into enumerated values or if i could shove my own jdob-type-id sort of thing in there | 20:11 |
| jdob | and have heat return it to me when i look up the stack resources | 20:11 |
| *** alexpilotti has quit IRC | 20:11 | |
| *** SergeyLukjanov has quit IRC | 20:11 | |
| zaneb | yeah, the content of the metadata is entirely up to you | 20:12 |
| jdob | is that available on ever resource, not just Server ? | 20:12 |
| jdob | er, "resource type" | 20:12 |
| zaneb | though in-instance tools like os-collect-config will treat certain bits of data in certain ways | 20:12 |
| zaneb | yes, every resource can have metadata | 20:13 |
| zaneb | actually, OS::Nova::Server is a confusing one, because it has a metadata _property_ as well | 20:13 |
| jdob | oh, thats not the same thing? | 20:14 |
| jdob | (glad you said that :) | 20:14 |
| zaneb | tbh I have no idea what that is | 20:14 |
| *** spzala has joined #heat | 20:14 | |
| radix | Nova servers have arbitrary metadata that can be associated with them | 20:14 |
| jdob | but thats injected into the server? | 20:15 |
| radix | it's completely inert IIRC | 20:15 |
| zaneb | but the metadata you want lives a level up, at the same level as the properties | 20:15 |
| jdob | zaneb: thats still on a resource by resource basis, not at the stack level, right? | 20:15 |
| zaneb | radix: oh, so it's metadata in the Nova API, but not something that is available to the instance? | 20:15 |
| radix | I *think*, yes | 20:15 |
| jdob | all of my 10 instances i described are the same stack | 20:16 |
| *** sdake has joined #heat | 20:16 | |
| *** sdake has quit IRC | 20:16 | |
| *** sdake has joined #heat | 20:16 | |
| zaneb | jdob: yes, so inside each resource you have a 'properties' key, and you can also have a 'metadata' key | 20:16 |
| zaneb | radix: thanks, I had been wondering about that ;) | 20:16 |
| jdob | ah, perfect, thats the sort of thing i was looking for | 20:17 |
| zaneb | jdob: caveat: it only applies to servers | 20:17 |
| jdob | ah, hrm... | 20:18 |
| *** mkollaro has joined #heat | 20:19 | |
| zaneb | jdob: that is to say, the property that radix was talking about is only for servers | 20:19 |
| jdob | oh oh, gotcha | 20:19 |
| zaneb | jdob: the metadata thing is for everything | 20:19 |
| jdob | ya, i'll be querying heat, not nova directly, so i dont care about the one that makes it to that level | 20:19 |
| zaneb | jdob: cool, then the metadata attribute/key/whatever-you-want-to-call-it on each resource will do what you want | 20:21 |
| jdob | excellent, thanks for your help. i still have yet to get heat running on my setup, so it might be a bit before i have more questions, but that definitley helps | 20:21 |
| mkollaro | is there somewhere an example on how to combine multiple HOT files? I want to deploy a lot of various topologies that are somewhat similar and I'd rather not create a different template for each | 20:22 |
| zaneb | mkollaro: o/ :) | 20:22 |
| mkollaro | or maybe some way to reuse parts of other templates | 20:22 |
| zaneb | good question | 20:23 |
| mkollaro | zaneb: \o | 20:23 |
| radix | so, the AS blueprints *weren't* backwards, I was just momentarily confused about what the arrows meant :) | 20:23 |
| radix | but I'm adding more | 20:23 |
| zaneb | mkollaro: we have an AWS::CloudFormation::Stack resource type that creates a new stack | 20:23 |
| zaneb | so you can manage a tree of stacks from a collection of templates like that | 20:24 |
| zaneb | (at some point we will create an OS::Heat::Stack equivalent) | 20:24 |
| mkollaro | I don't understand...what exactly is meant by "stack" here? | 20:24 |
| zaneb | stack = the collection of resources created from a template by Heat | 20:24 |
| zaneb | so stack =~= template | 20:25 |
| mkollaro | oh | 20:25 |
| zaneb | effectively you can compose templates using a resource type that pulls in another template | 20:25 |
| mkollaro | and am I somehow able to reuse the parameters to the template, or do I have to define them anew each time? | 20:26 |
| zaneb | you pass any parameters you want through the Stack resource's 'Parameters' property | 20:27 |
| * zaneb goes to look for the docs | 20:27 | |
| *** alexpilotti has joined #heat | 20:27 | |
| radix | zaneb: can you cast your eyes on the dependency graph here? https://blueprints.launchpad.net/heat/+spec/as-lib-db | 20:29 |
| zaneb | mkollaro: http://docs.openstack.org/developer/heat/template_guide/cfn.html#AWS::CloudFormation::Stack | 20:29 |
| mkollaro | so, if I have one template that creates a single VM (+ floating ip, volume, etc) and I want to have another template that is essentially the same, but creates 3 of them? | 20:29 |
| radix | therve: if you happen to still be around a quick review of that bp tree would be nice :) | 20:29 |
| zaneb | mkollaro: that is not so easy. you can use a ScalingGroup to create a group of servers (with the size determined by a property), but you can't yet do it for an arbitrary template (i.e. with floating ip, volume, &c.). We are working on that though, that's actually what radix is doing at the moment :) | 20:32 |
| mkollaro | oh | 20:32 |
| mkollaro | good luck with that then :D | 20:32 |
| sjmc | mkollaro - i ended up scripting template generation to do this | 20:33 |
| zaneb | that's exactly what I thought when I saw all those blueprints he assigned to himself :D | 20:33 |
| mkollaro | sjmc: templates of templates? :D | 20:34 |
| radix | heh heh | 20:34 |
| sjmc | :) kinda | 20:34 |
| zaneb | radix: but seriously, that looks good ;) | 20:34 |
| radix | zaneb: yay | 20:34 |
| radix | there's probably yet more detail to fill in | 20:34 |
| radix | and it may not be a strict ordering (some patches for as-lib might happen before as-lib-db for example) | 20:35 |
| radix | but it's a, uh, completion order I guess | 20:35 |
| mkollaro | zaneb: thanks for the help, I'll probably be annoying you with a few questions for the next week | 20:35 |
| zaneb | mkollaro: np, please do :) | 20:35 |
| mkollaro | :) | 20:35 |
| zaneb | mkollaro, sjmc: there have been a few folks who asked us to do templating of templates _inside_ heat. Those folks are destined to be eternally disappointed ;) | 20:36 |
| sjmc | zaneb - i know, that's why i didn't ask :) | 20:36 |
| sjmc | but in my case i needed to be able to generate a variable number of servers + floating ips + volumes | 20:37 |
| sjmc | and there doesn't seem to be a way to do it other than copy n pasting | 20:37 |
| zaneb | sjmc: for the record I think doing that _outside_ of Heat is actually a great idea | 20:37 |
| zaneb | sjmc: but good news, we will eventually let you [auto]scale templates, and then you won't have to any more | 20:38 |
| mkollaro | so, before I jump into this, what are currently the biggest problems around heat? | 20:39 |
| *** tomek_adamczewsk has quit IRC | 20:39 | |
| sjmc | :) yeah. i know heat's a work in progress. i may be getting more involved in it soonish, waiting to get more direction from my management | 20:39 |
| *** spzala has quit IRC | 20:41 | |
| *** Gamekiller77 has joined #heat | 20:41 | |
| Gamekiller77 | i know this is listed as a heat dev channel but can i ask some simple support questions for heat here too ? | 20:42 |
| zaneb | Gamekiller77: certainly :) | 20:44 |
| Gamekiller77 | so getting odd auth errors | 20:45 |
| zaneb | paste.openstack.org | 20:46 |
| *** radix__ has quit IRC | 20:46 | |
| Gamekiller77 | will do when i get it to gen again great home sick one day and stuff start to work hehe | 20:47 |
| mkollaro | does heat have some cleanup function? I want to spawn some vms, run something on them and then kill them | 20:48 |
| therve | radix, The first is a bit obscure still | 20:48 |
| Gamekiller77 | got it | 20:48 |
| zaneb | mkollaro: when you delete the stack, it deletes all of the resources again | 20:49 |
| therve | radix, Otherwise it looks good I guess | 20:49 |
| zaneb | mkollaro: but what you're asking about sounds more like a workflow | 20:49 |
| Gamekiller77 | zaneb, http://paste.openstack.org/show/60911/ i just using simple create | 20:50 |
| radix | therve: as-lib-db you mean? | 20:50 |
| therve | radix, Yep | 20:50 |
| radix | therve: yeah, that's kinda why I was thinking some of as-lib might get started before as-lib-db | 20:50 |
| Gamekiller77 | this is the simple text i use http://paste.openstack.org/show/60912/ | 20:50 |
| zaneb | mkollaro: the Mistral project is hoping to implement something like that | 20:50 |
| mkollaro | zaneb: what's that? | 20:50 |
| radix | we could do some basic template generation logic in before we do e.g. cooldown management | 20:50 |
| therve | radix, Sounds like a good idea yeah | 20:50 |
| zaneb | mkollaro: so a workflow could be something like "Create a stack, then wait for this task to run on it, then delete the stack again". Mistral is a project to create an openstack api to do that. (Sometimes I call it 'cron as a service', which really annoys them.) | 20:52 |
| *** denis_makogon has quit IRC | 20:52 | |
| *** dmakogon_ has joined #heat | 20:53 | |
| mkollaro | hehe | 20:53 |
| mkollaro | zaneb: I was thinking about running ansible on it and then deleting it | 20:53 |
| zaneb | Heat can takes care of tearing things down cleanly | 20:54 |
| zaneb | but you need an external script to decide when you task has finished and you want to go ahead and delete | 20:54 |
| zaneb | the latter is the part that Mistral will eventually give you an API for | 20:54 |
| mkollaro | zaneb: in what kind of state is that project? | 20:56 |
| zaneb | very early days, I think | 20:56 |
| sdake | they have just got started | 20:56 |
| sdake | fastest openstack projects have been 1 year to relative maturity | 20:57 |
| zaneb | yeah, they just announced that they were going to start it back in October | 20:57 |
| sdake | they are on month 1 or 2 | 20:57 |
| zaneb | I haven't been following along since then to see when they actually did start | 20:57 |
| mkollaro | this is the problem in openstack...stuff I want will maybe exist a few months later :) | 20:58 |
| sdake | I don't think I would have really considered heat relatively mature and reliable until it hit 18 months of dev | 20:58 |
| sdake | mkollaro ya it is a really big space and lots of folks want lots of things | 20:59 |
| sdake | and there are "only" 1k technical contributors to OpenStack (I think 1k is right) | 20:59 |
| sjmc | on the subject of the future, is the direction regarding software setup to use cloudinit to bootstrap chef/puppet/ansible/something? or is there likely to be something more formal | 21:00 |
| sjmc | i know there's been to-ing and fro-ing on that | 21:00 |
| Gamekiller77 | zaneb, did you take a look at my paste ? | 21:00 |
| zaneb | sdake: sounds right | 21:00 |
| *** jamieh has joined #heat | 21:01 | |
| zaneb | Gamekiller77: no, thanks for the reminder! | 21:01 |
| Gamekiller77 | not a problem | 21:01 |
| arbylee1 | hey all, we're looking to update the heatclient to support the limit/marker pagination params on stacks. It looks like there's some pagination code in StackManager from 2012 (before we implemented pagination) and it doesn't look like anyone is calling the method with the paging kwargs currently | 21:01 |
| *** randallburt has joined #heat | 21:02 | |
| arbylee1 | wondering if anyone has context on that part of the code and if it's safe to go ahead and remove/overhaul to support the existing paging params | 21:02 |
| sdake | arbylee1 cool submit patch through workflow - sounds handy | 21:02 |
| zaneb | Gamekiller77: so the problem there is that's not a whole template | 21:02 |
| sdake | I think overhaul is ok - it was added by a contributor who did pagination for a bunch of problems | 21:02 |
| sdake | problems/projects :) | 21:02 |
| sdake | and maybe it could use a more modern view | 21:03 |
| Gamekiller77 | ok | 21:03 |
| Gamekiller77 | good | 21:03 |
| Gamekiller77 | i need to find a simple quick template to spin up a simple server | 21:03 |
| arbylee1 | sdake: thanks, we'll go ahead and change it then | 21:03 |
| Gamekiller77 | to make sure it all works so i can sign it off | 21:03 |
| sdake | sjmc we are currently going to stick with cloudinit :) | 21:04 |
| sdake | sjmc I think the current thinking is to expose more of cloudinit to the user | 21:04 |
| zaneb | Gamekiller77: http://paste.openstack.org/show/60913/ would be the equivalent template to your snippet | 21:04 |
| Gamekiller77 | thanks | 21:04 |
| sdake | sjmc there is alos the concept of adding specific resources for specific config management tools (eg chef/puppet/ansible) | 21:04 |
| sjmc | ok, sdake. any thoughts around pushed software config (ansible-style)? | 21:06 |
| sdake | sjmc early on we started with that approach actually (ssh pushing to the guests) | 21:06 |
| sdake | sjmc however our current method is more scalable because it relies on the vm to do the orhcestration | 21:07 |
| sjmc | yeah, that makes sense | 21:07 |
| sdake | with the previous ssh-push based method we would have had to scale that service out to multiple processors | 21:07 |
| sjmc | right | 21:07 |
| sdake | instead with the current mode we just load the data in the text field, and nova et al is responsible for the scaling :) | 21:07 |
| Gamekiller77 | you guys need to provide just a little more description to the errors that pop up on the dashboard | 21:07 |
| sdake | error reporting in heat is pretty bad | 21:08 |
| sdake | not going to lie :) | 21:08 |
| sjmc | sdake - i've been booting images that don't always have packages i'd expect to be there, which has sometimes meant installing provisioning clients has been tricky | 21:08 |
| sdake | sjmc which provisioning clients? | 21:08 |
| sjmc | ansible, puppet or whatever | 21:09 |
| zaneb | Gamekiller77, sdake: some are better than others. The real problem is a lack of sanity checks. That error should have complained about the properties section being missing altogether, not about being empty | 21:09 |
| sjmc | but if that's definitely going to be the direction i'll work around those issues rather than hoping for my ssh-pushed config | 21:09 |
| Gamekiller77 | yah i just got an error the key was not spelled right | 21:09 |
| zaneb | more errors = better errors | 21:10 |
| sdake | sjmc I am not a big fan of the server approach to config management (eg have a server drive the activity like puppet/chef/ansible do) | 21:10 |
| Gamekiller77 | zaneb, i need to add the network to that template you gave me | 21:10 |
| sdake | but I realize it is a reality of what people want out of heat | 21:10 |
| sjmc | sdake - yeah, i definitely see for heat it's best not to let that scaling monster into the room | 21:10 |
| Gamekiller77 | what line should i add | 21:10 |
| jasond | zaneb: what did you mean "I'd also suggest that each group of EngineService, EngineListener and ThreadGroupManager share a unique engine_id"? | 21:10 |
| sdake | sjmc and I believe we intend to provide proper hooks so folks investment in those toolchains doesn't waste | 21:10 |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Don't query for resource during stack creation https://review.openstack.org/63914 | 21:11 |
| sdake | well, that is why we didn't do it in heat initially | 21:11 |
| sdake | we have enough scaling problems to deal with | 21:11 |
| sjmc | :) yeah | 21:11 |
| sdake | actually why we removed it initially :) | 21:11 |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Allow an in-progress stack to be deleted https://review.openstack.org/63002 | 21:11 |
| zaneb | Gamekiller77: by add a network, you mean add a Neutron network resource? Or just connect the server to an existing network? | 21:11 |
| sjmc | will heat run ok with multiple engines and apis now? i know there was some work recently on that | 21:12 |
| sdake | sjmc I believe jasond did that work on master and last report I got was its all done :) | 21:12 |
| sjmc | woo | 21:12 |
| zaneb | jasond: I meant that e.g. EngineService will generate an engine_id when it is created | 21:12 |
| sdake | sjmc but you will ahve to use icehouse to get it (we aren't backporting) | 21:12 |
| Gamekiller77 | so got stack create error that it needed a network to attach to | 21:13 |
| jasond | sjmc, sdake: yes, it's done. there's a bugfix in for delete during an in-progress action. here is the doc https://review.openstack.org/#/c/63963/ | 21:13 |
| sjmc | sdake - understood. it'll be a while till we get it properly set up anyway i think, we're still experimenting | 21:13 |
| sjmc | thanks jasond | 21:14 |
| zaneb | Gamekiller77: docs for OS::Nova::Server are here: http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Nova::Server | 21:14 |
| Gamekiller77 | zaneb, the code you gave me worked fine just i got this error Create_Failed: Resource create failed: BadRequest: Multiple possible networks found, use a Network ID to be more specific. (HTTP 400) ( | 21:15 |
| Gamekiller77 | perfect i needed that | 21:15 |
| Gamekiller77 | thanks for all the support | 21:15 |
| zaneb | Gamekiller77: np :) | 21:17 |
| Gamekiller77 | OMG i been looking for this guide for a while | 21:17 |
| Gamekiller77 | thanks you so much | 21:17 |
| Gamekiller77 | yah i keep failing at network ID | 21:17 |
| sdake | jasond nice docs | 21:17 |
| jasond | sdake: thanks | 21:17 |
| sdake | jasond have you tested heat at much scale yet? | 21:18 |
| jasond | sdake: not yet | 21:19 |
| sdake | curious how the db lock impacts scaling limits | 21:19 |
| zaneb | sdake: as long as collisions are unusual it should be pretty minimal | 21:20 |
| zaneb | and collisions should be unusual | 21:20 |
| *** gokrokve has quit IRC | 21:21 | |
| *** gokrokve has joined #heat | 21:22 | |
| Gamekiller77 | is there a tool to validate a template for HOT yet ? | 21:22 |
| Gamekiller77 | i saw someone say there was for the yaml | 21:22 |
| jasond | zaneb: how is using stack_lock.engine_id related to singleton classes? | 21:23 |
| zaneb | Gamekiller77: there is a validate_template API in heat | 21:23 |
| Gamekiller77 | ok i see what i need to add to the template just not sure what the formate is i figure this out i need to add a new type to the my template | 21:24 |
| Gamekiller77 | can you have more then one type in a template ? | 21:25 |
| zaneb | jasond: right now you need a global thread list, because there's only one engine_id per process (so if you encapsulated it in a class, it would be a Singleton). If there were one engine_id per EngineService, then you can store all the thread group list locally | 21:25 |
| *** gokrokve has quit IRC | 21:26 | |
| zaneb | Gamekiller77: maybe https://github.com/openstack/heat-templates/blob/master/hot/servers_in_existing_neutron_net.yaml is a good example to look at | 21:28 |
| Gamekiller77 | yup already made it there | 21:29 |
| Gamekiller77 | hehe | 21:29 |
| jasond | there's only one EngineService and one EngineListener (listening on different AMQP queues). they both need access to the thread group list | 21:29 |
| Gamekiller77 | just read that site thanks man i get up to speed fast | 21:29 |
| jasond | *read/write access, that is | 21:30 |
| zaneb | the EngineService creates the listener, so it's no problem to pass the thread group list to it | 21:31 |
| zaneb | s/listener/EngineListener/ | 21:31 |
| zaneb | "there's only one EngineService and one EngineListener" <- so, they're Singletons at the moment | 21:32 |
| zaneb | because they're using some global state | 21:33 |
| zaneb | but there's no reason they couldn't encapsulate that global state | 21:33 |
| randallburt | zaneb: true but is there ever a reason to have more than one of each per engine? | 21:33 |
| zaneb | randallburt: no | 21:34 |
| zaneb | but encapsulating global state is desirable in itself | 21:34 |
| randallburt | zaneb: so not a big deal atm, IMO. I do like the idea of encapsulating the thread group management, however. that could certainly be a follow on. | 21:34 |
| *** nkhare has joined #heat | 21:35 | |
| zaneb | well, we have to design this new bit (thread group management) _now_. So why not define it the way we want at the start - it isn't really any more work | 21:36 |
| randallburt | zaneb: understood, I just don't think one (ThreadGroupManager) is predicated on changing the object hierarchy in the current patch. | 21:37 |
| Gamekiller77 | thanks you zaneb i am up and running. Will there be more logic in the template down the line to pull the local project ID for networking and images and keys already in the project ? | 21:38 |
| *** sgordon has quit IRC | 21:38 | |
| jasond | randallburt: i agree, i'd rather put in a follow up patch. it's been difficult getting people to look at this review. would rather get the current patch in now while people are thinking about it | 21:39 |
| zaneb | randallburt: I haven't looked at the latest patchset yet | 21:39 |
| jasond | zaneb: although i'm still not sure how changing stack_lock.engine_id would help | 21:40 |
| randallburt | zaneb: tbh me either, but its on my todo today. | 21:40 |
| zaneb | the object hierarchy in patch set 6 made my eyes bleed | 21:40 |
| zaneb | (only saying that 'cos it's not jasond's fault ;) | 21:40 |
| randallburt | zaneb, jasond: so much for being acommodating ;) TBH, I didn't have an issue with the original organization, but I was on vacation at the time. At least that's the story I'm sticking to. | 21:42 |
| jasond | zaneb: :) can i quote you on saying that i've invented a radical new paradigm of OO design? | 21:42 |
| randallburt | jasond: should def include that in your performance review. | 21:42 |
| zaneb | jasond: you may :) | 21:43 |
| *** blomquisg has quit IRC | 21:43 | |
| *** topol has quit IRC | 21:43 | |
| zaneb | jasond: http://paste.openstack.org/show/60917/ | 21:45 |
| jasond | zaneb: that makes sense, thanks | 21:47 |
| zaneb | reading the review again, it's clear that I didn't explain it very well | 21:48 |
| jasond | zaneb: it's no problem, thanks for the example | 21:49 |
| zaneb | cool. that snippet was a more succinct explanation of what I was trying to suggest ;) | 21:51 |
| *** jdob has quit IRC | 21:52 | |
| zaneb | jasond: btw ThreadGroupManager could be a defaultdict with an overridded __delitem__() method | 21:53 |
| zaneb | perhaps not unidentical to https://github.com/openstack/heat/blob/master/heat/engine/dependencies.py#L94 | 21:53 |
| jasond | cool, that helps | 21:55 |
| *** nati_ueno has quit IRC | 21:56 | |
| *** gfidente has joined #heat | 21:57 | |
| stevebaker | sjmc: I think when ansible software config support is added it will be pull-based. This is what ansible recommends for large scale anyway, but instead of calling ansible from cron it is called by the software config polling mechanism | 21:58 |
| *** nati_ueno has joined #heat | 21:59 | |
| sjmc | ok, thanks stevebaker. that was the impression i'd got from following amiling list discussions | 21:59 |
| sjmc | do you see heat being involved getting those clients available, or do you think that'll remain part of a template's cloudinit ? | 21:59 |
| sjmc | i can't use waitconditions at the moment so i think we're stuck with pushing configs for the near term (otherwise we never know when things are finished) | 22:02 |
| *** nati_ueno has quit IRC | 22:03 | |
| *** nati_ueno has joined #heat | 22:04 | |
| *** dims has quit IRC | 22:12 | |
| *** gfidente has quit IRC | 22:12 | |
| *** gokrokve has joined #heat | 22:14 | |
| *** mestery has quit IRC | 22:16 | |
| stevebaker | sjmc: the recommended method would be to create a custom image containing os-collect-config, os-refresh-config, os-apply-config, your configuration tool, and the heat config hook script | 22:17 |
| *** alexpilotti has quit IRC | 22:17 | |
| stevebaker | sjmc: but if you can't or won't use custom images you could install all that with cloud-init on boot, which can be specified in the template | 22:18 |
| stevebaker | sjmc: why can't you use waitconditions? | 22:18 |
| *** dims has joined #heat | 22:18 | |
| *** mestery has joined #heat | 22:19 | |
| *** rpothier has quit IRC | 22:21 | |
| *** jprovazn has quit IRC | 22:24 | |
| *** jamieh has quit IRC | 22:24 | |
| *** asalkeld has quit IRC | 22:25 | |
| lifeless | stevebaker: don't they still require you to be admin? | 22:26 |
| sjmc | stevebaker, lifeless: yeah, that | 22:26 |
| sjmc | sorry, on phone | 22:26 |
| *** radix_ has joined #heat | 22:27 | |
| *** asalkeld has joined #heat | 22:27 | |
| *** asalkeld has left #heat | 22:29 | |
| *** asalkeld has joined #heat | 22:29 | |
| stevebaker | sjmc: ah, ok. I just need to check which reason | 22:33 |
| *** denis_makogon has joined #heat | 22:33 | |
| *** randallburt has quit IRC | 22:34 | |
| denis_makogon | goodday 2 all, i've got few questions to all, does heat able to perform resource updating through stack update (example - resize instance/volume, add security group, resize volume) ? | 22:34 |
| *** kebray_ has quit IRC | 22:38 | |
| zaneb | denis_makogon: yes, but support for each one has to be implemented independently. So e.g. we support resize for instances but not for volumes at the moment | 22:40 |
| *** jmckind has quit IRC | 22:40 | |
| denis_makogon | zaneb, what about security groups ? | 22:40 |
| zaneb | denis_makogon: on which resource? | 22:41 |
| denis_makogon | zaneb, perhaps Neutron | 22:41 |
| zaneb | OS::Neutron::Port | 22:42 |
| zaneb | that should work | 22:42 |
| denis_makogon | zaneb, you said that heat doesn't support update for volumes, right ? | 22:43 |
| *** jcru has quit IRC | 22:43 | |
| zaneb | volumes will be replaced if you modify them | 22:43 |
| zaneb | which is probably not what you want | 22:44 |
| denis_makogon | yes | 22:44 |
| asalkeld | denis_makogon: look here: http://docs.openstack.org/developer/heat/template_guide/openstack.html | 22:44 |
| asalkeld | and look for "Can be updated without replacement." | 22:44 |
| *** rwsu has quit IRC | 22:45 | |
| *** Linz has quit IRC | 22:48 | |
| *** kebray_ has joined #heat | 22:50 | |
| *** radez is now known as radez_g0n3 | 22:51 | |
| *** alexheneveld has quit IRC | 22:52 | |
| *** nati_ueno has quit IRC | 22:54 | |
| *** randallburt has joined #heat | 22:55 | |
| *** alexheneveld has joined #heat | 22:55 | |
| *** randallburt has quit IRC | 22:55 | |
| *** randallburt has joined #heat | 22:55 | |
| *** nati_ueno has joined #heat | 22:56 | |
| *** nkhare has quit IRC | 22:56 | |
| *** nati_ueno has quit IRC | 22:57 | |
| *** nati_ueno has joined #heat | 22:57 | |
| *** yogesh has quit IRC | 23:06 | |
| *** Linz has joined #heat | 23:06 | |
| *** yogesh has joined #heat | 23:06 | |
| *** Linz has quit IRC | 23:07 | |
| *** vijendar has quit IRC | 23:07 | |
| *** Linz has joined #heat | 23:07 | |
| *** denis_makogon has quit IRC | 23:07 | |
| *** mkollaro has quit IRC | 23:08 | |
| *** yogesh has quit IRC | 23:11 | |
| *** mkollaro has joined #heat | 23:15 | |
| *** rcleere has quit IRC | 23:19 | |
| *** pablosan has quit IRC | 23:20 | |
| *** pablosan has joined #heat | 23:23 | |
| *** slong has joined #heat | 23:28 | |
| *** alexpilotti has joined #heat | 23:35 | |
| *** sballe has quit IRC | 23:36 | |
| *** sballe has joined #heat | 23:36 | |
| *** tims has quit IRC | 23:39 | |
| *** Gamekiller77 has quit IRC | 23:46 | |
| *** Linz has quit IRC | 23:46 | |
| *** Linz has joined #heat | 23:49 | |
| *** sjmc has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!