| lifeless | stevebaker: ^ | 00:01 |
|---|---|---|
| stevebaker | lifeless: I was installing dib with pip, and a 1.4 bug means that all files loose their execute bit | 00:02 |
| stevebaker | lifeless: and then fedora 20 came out which had heat-cfntools pre-installed, which met our testing needs for a little while | 00:03 |
| *** rpothier has joined #heat | 00:03 | |
| *** radez is now known as radez_g0n3 | 00:13 | |
| *** openstack has joined #heat | 00:19 | |
| *** IlyaE has joined #heat | 00:19 | |
| openstackgerrit | Sean Dague proposed a change to openstack/heat: heat is enabled by default in devstack now https://review.openstack.org/82678 | 00:20 |
| *** RockKuo has joined #heat | 00:22 | |
| *** IlyaE has quit IRC | 00:26 | |
| *** matsuhashi has joined #heat | 00:26 | |
| *** Tross has joined #heat | 00:26 | |
| *** duncanjw has quit IRC | 00:27 | |
| *** jpeeler has quit IRC | 00:30 | |
| *** fandi has quit IRC | 00:34 | |
| *** lindsayk has quit IRC | 00:36 | |
| sdake_ | stevebaker if you want to take a look at this patch set it should be pretty easy to review and result in one less open bug :) https://review.openstack.org/#/c/82593/ | 00:40 |
| *** lipinski has joined #heat | 00:42 | |
| sdake_ | looks like systemd made its way into oslo-incubator - guess we can fix that in juno https://review.openstack.org/#/c/81211/ | 00:43 |
| stevebaker | sdake_: done, thanks | 00:43 |
| sdake_ | thanks stevebaker | 00:43 |
| sdake_ | need anything done besides reviews? I went thrugh the bugs and looks like all are waiting on developer action | 00:43 |
| sdake_ | https://bugs.launchpad.net/bugs/1296859 is unassigned but I'm not sure if i could fix it in a couple days | 00:45 |
| uvirtbot | Launchpad bug 1296859 in heat "stack-list throws an error if any stack returned contains a template error" [Medium,Triaged] | 00:45 |
| *** samstav has quit IRC | 00:45 | |
| sdake_ | i guess we could turn off validation for stack-list as randall suggests | 00:45 |
| *** wchrisj has quit IRC | 00:45 | |
| sdake_ | seems so -- hacky :) | 00:45 |
| stevebaker | we should make sure validation is turned off for delete too | 00:46 |
| sdake_ | well if you like i'll take a look at fixing those two scenarios | 00:46 |
| sdake_ | but I can't guarantee i'll be able to get er done in the next couple days | 00:46 |
| stevebaker | sdake_: a review on this would be handy https://review.openstack.org/#/c/80882/ | 00:47 |
| stevebaker | sdake_: and this https://review.openstack.org/#/c/80868/ | 00:48 |
| openstackgerrit | A change was merged to openstack/heat: Document software config classes https://review.openstack.org/81407 | 00:49 |
| sdake_ | stevebaker I -1 that first change | 00:49 |
| sdake_ | I think it lacks internationalization of the exception | 00:50 |
| sdake_ | other then that looks good | 00:50 |
| sdake_ | the 2nd patch - let me eat first :) | 00:50 |
| stevebaker | sdake_: actually there are zero i18n messages in heatclient, we haven't done that yet | 00:51 |
| *** fandi has joined #heat | 00:51 | |
| sdake_ | oh that was client | 00:52 |
| sdake_ | I didn't notice the different repo :( | 00:52 |
| sdake_ | i'll go ahead and +a then | 00:52 |
| *** wchrisj has joined #heat | 00:56 | |
| sdake_ | stevebaker just to be clear - i3-rc1 is cut as soon as the bug list is zero/ | 01:01 |
| sdake_ | or would we possibly remove some bugs? | 01:01 |
| stevebaker | sdake_: yep, bug list needs to be zero, either by fixin-em or kickin-em | 01:01 |
| sdake_ | not sure how else I can help this week other then try to tackle that validate problem | 01:02 |
| sdake_ | and stay on top of reviews | 01:02 |
| sdake_ | so I guess i'll work on the validate problem and if it doesn't make it we can kick it | 01:02 |
| sdake_ | iirc it was medium | 01:02 |
| stevebaker | sdake_: yep, and if it gets kicked but still lands before rc1 it will still be released in rc1 | 01:04 |
| sdake_ | cool i'll focus on that | 01:04 |
| sdake_ | my other commitments that I thought I had this week are off so I'm at your disposal :) | 01:04 |
| stevebaker | good to know | 01:05 |
| *** RockKuo has quit IRC | 01:09 | |
| openstackgerrit | Dong Liu proposed a change to openstack/heat: Add subnets as a dependency for router https://review.openstack.org/82452 | 01:16 |
| *** rwsu has quit IRC | 01:26 | |
| *** nosnos has joined #heat | 01:29 | |
| *** matsuhashi has quit IRC | 01:30 | |
| *** matsuhashi has joined #heat | 01:30 | |
| *** wchrisj has quit IRC | 01:31 | |
| *** rpothier_ has joined #heat | 01:35 | |
| *** rpothier has quit IRC | 01:39 | |
| *** wchrisj has joined #heat | 01:39 | |
| *** rwsu has joined #heat | 01:42 | |
| *** wchrisj has quit IRC | 01:44 | |
| *** wchrisj has joined #heat | 01:49 | |
| *** wchrisj_ has joined #heat | 01:53 | |
| *** wchrisj has quit IRC | 01:53 | |
| *** IYozhikov_away has quit IRC | 01:56 | |
| *** IgorYozhikov has joined #heat | 01:56 | |
| *** IgorYozhikov is now known as IYozhikov_away | 01:57 | |
| *** RockKuo has joined #heat | 01:57 | |
| *** denis_makogon has quit IRC | 01:58 | |
| *** denis_makogon has joined #heat | 01:58 | |
| *** wchrisj_ has quit IRC | 02:00 | |
| openstackgerrit | A change was merged to openstack/python-heatclient: Do not use the '+' operation with dict_items() https://review.openstack.org/82523 | 02:09 |
| openstackgerrit | A change was merged to openstack/heat: Error and NotFound inherit HeatException class https://review.openstack.org/82593 | 02:10 |
| openstackgerrit | A change was merged to openstack/heat: Properly encode heat.common.exception in rpc https://review.openstack.org/82386 | 02:15 |
| *** duncanjw has joined #heat | 02:28 | |
| openstackgerrit | A change was merged to openstack/heat: heat is enabled by default in devstack now https://review.openstack.org/82678 | 02:29 |
| *** duncanjw has quit IRC | 02:32 | |
| *** mestery has joined #heat | 02:34 | |
| *** liang has joined #heat | 02:35 | |
| *** mestery has quit IRC | 02:46 | |
| *** mestery has joined #heat | 02:46 | |
| *** fandi has quit IRC | 02:50 | |
| *** cmyster_ has quit IRC | 02:54 | |
| *** Tross has left #heat | 02:57 | |
| *** wchrisj has joined #heat | 03:02 | |
| sdake_ | stevebaker all done with https://review.openstack.org/#/c/80868/ | 03:04 |
| sdake_ | actually an easy review - looked daunting originally :) | 03:04 |
| *** wchrisj has quit IRC | 03:06 | |
| *** wchrisj has joined #heat | 03:07 | |
| *** matsuhashi has quit IRC | 03:12 | |
| *** wchrisj has quit IRC | 03:14 | |
| *** wchrisj has joined #heat | 03:15 | |
| *** david-lyle has joined #heat | 03:17 | |
| *** wchrisj_ has joined #heat | 03:19 | |
| *** wchrisj has quit IRC | 03:20 | |
| *** wchrisj has joined #heat | 03:24 | |
| *** wchrisj_ has quit IRC | 03:25 | |
| *** gokrokve has joined #heat | 03:27 | |
| *** wchrisj_ has joined #heat | 03:28 | |
| *** wchrisj has quit IRC | 03:29 | |
| *** nosnos has quit IRC | 03:29 | |
| *** matsuhashi has joined #heat | 03:30 | |
| *** wchrisj has joined #heat | 03:32 | |
| *** wchrisj_ has quit IRC | 03:32 | |
| *** chandankumar_ has joined #heat | 03:33 | |
| *** matsuhashi has quit IRC | 03:36 | |
| *** wchrisj_ has joined #heat | 03:39 | |
| *** wchrisj has quit IRC | 03:40 | |
| *** chandankumar_ has quit IRC | 03:40 | |
| *** mspreitz has joined #heat | 03:40 | |
| *** wchrisj has joined #heat | 03:43 | |
| *** wchrisj_ has quit IRC | 03:43 | |
| *** wchrisj has quit IRC | 03:47 | |
| *** harlowja_ is now known as harlowja_away | 03:49 | |
| *** gokrokve has quit IRC | 03:50 | |
| *** gokrokve has joined #heat | 03:52 | |
| *** pablosan_ has joined #heat | 03:53 | |
| *** pablosan has quit IRC | 03:54 | |
| *** chandankumar_ has joined #heat | 03:55 | |
| *** gokrokve has quit IRC | 03:57 | |
| *** killer_prince is now known as lazy_prince | 04:06 | |
| *** nati_ueno has quit IRC | 04:13 | |
| *** Linz has quit IRC | 04:16 | |
| *** Linz has joined #heat | 04:16 | |
| *** cfriesen has quit IRC | 04:17 | |
| *** nkhare has joined #heat | 04:19 | |
| *** matsuhashi has joined #heat | 04:23 | |
| *** chandankumar_ has quit IRC | 04:23 | |
| *** nosnos has joined #heat | 04:25 | |
| openstackgerrit | A change was merged to openstack/heat: Reimplement DHCPAgent as net's property https://review.openstack.org/81726 | 04:28 |
| *** duncanjw has joined #heat | 04:29 | |
| *** duncanjw has quit IRC | 04:33 | |
| *** nati_ueno has joined #heat | 04:40 | |
| *** pasquier-s has quit IRC | 04:41 | |
| *** nkhare_ has joined #heat | 04:42 | |
| *** lipinski has quit IRC | 04:44 | |
| *** ramishra_ has joined #heat | 04:45 | |
| *** Linz has quit IRC | 04:46 | |
| *** Tross has joined #heat | 04:47 | |
| *** pasquier-s has joined #heat | 04:54 | |
| *** nkhare has quit IRC | 04:54 | |
| *** nkhare_ has quit IRC | 04:54 | |
| *** nkhare has joined #heat | 04:55 | |
| openstackgerrit | Andrew Plunk proposed a change to openstack/heat: Chef solo resource https://review.openstack.org/55208 | 05:08 |
| *** matsuhashi has quit IRC | 05:14 | |
| *** pasquier-s has quit IRC | 05:16 | |
| *** matsuhashi has joined #heat | 05:18 | |
| *** skraynev_afk is now known as skraynev | 05:19 | |
| skraynev | Morning | 05:20 |
| *** gokrokve has joined #heat | 05:21 | |
| *** gokrokve has quit IRC | 05:25 | |
| *** IlyaE has joined #heat | 05:31 | |
| openstackgerrit | Mitsuru Kanabuchi proposed a change to openstack/heat: Change router's prop name agent_id to l3_agent_id https://review.openstack.org/82716 | 05:34 |
| *** nati_ueno has quit IRC | 05:47 | |
| *** chandan_kumar has quit IRC | 05:48 | |
| *** e0ne has joined #heat | 05:49 | |
| *** e0ne has quit IRC | 05:49 | |
| *** nati_ueno has joined #heat | 05:51 | |
| *** cmyster has joined #heat | 05:54 | |
| cmyster | morning | 05:55 |
| *** ramishra_ has quit IRC | 06:06 | |
| skraynev | zaneb: Are you still here? | 06:07 |
| openstackgerrit | Mitsuru Kanabuchi proposed a change to openstack/heat: Change router's prop name agent_id to l3_agent_id https://review.openstack.org/82716 | 06:09 |
| *** cmyster_ has joined #heat | 06:11 | |
| openstackgerrit | Jenkins proposed a change to openstack/heat: Imported Translations from Transifex https://review.openstack.org/72566 | 06:12 |
| *** cmyster has quit IRC | 06:14 | |
| *** mestery has quit IRC | 06:18 | |
| *** mspreitz has quit IRC | 06:26 | |
| *** duncanjw has joined #heat | 06:29 | |
| *** duncanjw has quit IRC | 06:36 | |
| *** ramishra has joined #heat | 06:51 | |
| *** ramishra_ has joined #heat | 06:53 | |
| *** ramishra has quit IRC | 06:54 | |
| *** ramishra_ has quit IRC | 06:55 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/heat: Add docker network_gateway attribute https://review.openstack.org/82511 | 06:57 |
| *** lsmola_ has quit IRC | 07:02 | |
| *** lsmola_ has joined #heat | 07:03 | |
| chmouel | pas-ha: the python docker client? yes that would be the location of the library | 07:03 |
| *** ramishra has joined #heat | 07:13 | |
| *** IlyaE has quit IRC | 07:14 | |
| chmouel | morning all | 07:18 |
| openstackgerrit | huangtianhua proposed a change to openstack/python-heatclient: Deprecate preview --create-timeout in favor of --timeout https://review.openstack.org/82727 | 07:21 |
| *** gokrokve has joined #heat | 07:21 | |
| *** gokrokve has quit IRC | 07:26 | |
| *** saju_m has joined #heat | 07:30 | |
| cmyster_ | morning chmouel | 07:32 |
| *** nati_ueno has quit IRC | 07:37 | |
| therve | 'morning | 07:37 |
| cmyster_ | morn | 07:41 |
| *** tspatzier has joined #heat | 07:45 | |
| *** pasquier-s has joined #heat | 07:48 | |
| *** ifarkas has joined #heat | 07:48 | |
| *** tomek_adamczewsk has joined #heat | 07:48 | |
| *** FL1SK has quit IRC | 07:51 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/heat: Rename docker plugin to dockerheat https://review.openstack.org/82730 | 07:53 |
| *** jstrachan has joined #heat | 07:59 | |
| therve | chmouel, I don't understand, we already use the docker module and it works? | 08:00 |
| *** FL1SK has joined #heat | 08:03 | |
| cmyster_ | I am getting this: http://paste.openstack.org/show/74220/ I remember something about it, but can't find if there is already an open bug here | 08:04 |
| *** che-arne has quit IRC | 08:04 | |
| cmyster_ | a UTF8 issue in the DB | 08:04 |
| *** e0ne has joined #heat | 08:10 | |
| *** jprovazn has joined #heat | 08:10 | |
| *** TonyBurn has joined #heat | 08:11 | |
| *** fandi has joined #heat | 08:15 | |
| cmyster_ | fixed with alter table migrate_version CHARACTER SET utf8 COLLATE utf8_unicode_ci; | 08:19 |
| cmyster_ | for some reason it was latin_swedish for me | 08:20 |
| *** pasquier-s has quit IRC | 08:22 | |
| *** pasquier-s has joined #heat | 08:25 | |
| *** duncanjw has joined #heat | 08:30 | |
| shardy | morning all | 08:31 |
| cmyster_ | morning | 08:31 |
| shardy | cmyster_: I had the same issue, it's due to the mysql default AFAIK and/or something not being explicitly specified in the schema | 08:32 |
| cmyster_ | https://bugs.launchpad.net/heat/+bug/1276880 | 08:32 |
| uvirtbot | Launchpad bug 1276880 in heat "heat-db-setup should set default character set to utf-8" [Medium,Fix released] | 08:32 |
| shardy | cmyster_: I fixed it by manually alterning the collation via the mysql cli | 08:32 |
| cmyster_ | worked around it by updating the table manualy | 08:32 |
| cmyster_ | ya | 08:32 |
| shardy | ++ ;) | 08:32 |
| openstackgerrit | Thomas Herve proposed a change to openstack/python-heatclient: Resolve files in resource types https://review.openstack.org/80528 | 08:34 |
| *** duncanjw has quit IRC | 08:36 | |
| *** alexheneveld has joined #heat | 08:42 | |
| *** giulivo has joined #heat | 08:46 | |
| *** matsuhashi has quit IRC | 08:49 | |
| *** jistr has joined #heat | 08:49 | |
| *** saju_m has quit IRC | 08:50 | |
| *** saju_m has joined #heat | 08:52 | |
| *** jistr is now known as jistr|training | 08:54 | |
| pas-ha | morning | 08:55 |
| *** matsuhashi has joined #heat | 08:56 | |
| *** DaveJ__ has joined #heat | 09:01 | |
| *** gokrokve has joined #heat | 09:05 | |
| *** gokrokve has quit IRC | 09:05 | |
| *** gokrokve has joined #heat | 09:05 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/heat: Don't create cloud-init user unless specified https://review.openstack.org/79678 | 09:10 |
| *** chandan_kumar has joined #heat | 09:13 | |
| openstackgerrit | A change was merged to openstack/heat-templates: Add template for separated node/broker OpenShift https://review.openstack.org/81632 | 09:13 |
| *** matsuhashi has quit IRC | 09:14 | |
| *** derekh has joined #heat | 09:15 | |
| *** matsuhashi has joined #heat | 09:18 | |
| *** cmyster has joined #heat | 09:19 | |
| *** cmyster_ has quit IRC | 09:22 | |
| *** lipinski has joined #heat | 09:24 | |
| *** alexheneveld has quit IRC | 09:35 | |
| *** saju_m has quit IRC | 09:38 | |
| *** nosnos has quit IRC | 09:39 | |
| *** duncanjw has joined #heat | 09:39 | |
| *** che-arne has joined #heat | 09:41 | |
| *** duncanjw has quit IRC | 09:42 | |
| *** cmyster has quit IRC | 09:45 | |
| *** bvandenh has quit IRC | 09:47 | |
| openstackgerrit | huangtianhua proposed a change to openstack/heat: Remove the property validate for stack-adopt https://review.openstack.org/82759 | 09:56 |
| *** saju_m has joined #heat | 09:57 | |
| *** matsuhashi has quit IRC | 10:08 | |
| *** matsuhashi has joined #heat | 10:10 | |
| *** bvandenh has joined #heat | 10:10 | |
| *** e0ne_ has joined #heat | 10:16 | |
| *** e0ne has quit IRC | 10:19 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/python-heatclient: Resolve files in resource types https://review.openstack.org/80528 | 10:21 |
| *** faramir1 has joined #heat | 10:39 | |
| *** liang has quit IRC | 10:54 | |
| openstackgerrit | Roman Podoliaka proposed a change to openstack/heat: Fix listing of stacks with non-ascii parameters https://review.openstack.org/82545 | 10:54 |
| *** saju_m has quit IRC | 11:07 | |
| *** saju_m has joined #heat | 11:08 | |
| *** saju_m has quit IRC | 11:09 | |
| *** ramishra has quit IRC | 11:11 | |
| openstackgerrit | A change was merged to openstack/heat: Store stack domain credentials for deployments https://review.openstack.org/80868 | 11:12 |
| *** saju_m has joined #heat | 11:15 | |
| *** ramishra has joined #heat | 11:19 | |
| *** alexheneveld has joined #heat | 11:23 | |
| *** gokrokve has quit IRC | 11:23 | |
| *** RockKuo has quit IRC | 11:32 | |
| *** e0ne_ has quit IRC | 11:33 | |
| *** alexheneveld has quit IRC | 11:33 | |
| *** e0ne has joined #heat | 11:33 | |
| *** gokrokve has joined #heat | 11:34 | |
| *** gokrokve_ has joined #heat | 11:38 | |
| *** saju_m has quit IRC | 11:38 | |
| *** e0ne has quit IRC | 11:38 | |
| *** aweiteka has quit IRC | 11:38 | |
| *** saju_m has joined #heat | 11:38 | |
| *** gokrokve has quit IRC | 11:39 | |
| *** jistr|training has quit IRC | 11:40 | |
| *** jistr has joined #heat | 11:41 | |
| *** jistr is now known as jistr|training | 11:41 | |
| *** gokrokve_ has quit IRC | 11:42 | |
| *** ifarkas has quit IRC | 11:43 | |
| *** rpothier_ has quit IRC | 11:46 | |
| *** fandi has quit IRC | 11:47 | |
| *** ifarkas_ has joined #heat | 11:48 | |
| *** ifarkas_ has quit IRC | 11:50 | |
| *** ifarkas has joined #heat | 11:51 | |
| *** skraynev is now known as skraynev_afk | 12:01 | |
| *** harlowja_away has quit IRC | 12:03 | |
| *** aweiteka has joined #heat | 12:06 | |
| *** gokrokve has joined #heat | 12:10 | |
| *** e0ne has joined #heat | 12:12 | |
| *** e0ne has quit IRC | 12:13 | |
| *** duncanjw has joined #heat | 12:14 | |
| *** e0ne has joined #heat | 12:14 | |
| *** TonyBurn has quit IRC | 12:16 | |
| *** TonyBurn has joined #heat | 12:16 | |
| *** matsuhashi has quit IRC | 12:16 | |
| *** e0ne has quit IRC | 12:18 | |
| *** gokrokve has quit IRC | 12:19 | |
| *** gokrokve has joined #heat | 12:22 | |
| *** matsuhashi has joined #heat | 12:23 | |
| *** networkn8 has joined #heat | 12:23 | |
| *** david-lyle has quit IRC | 12:23 | |
| *** nkhare has quit IRC | 12:26 | |
| *** gokrokve has quit IRC | 12:28 | |
| *** gokrokve has joined #heat | 12:32 | |
| *** andersonvom has joined #heat | 12:33 | |
| *** bgorski has joined #heat | 12:34 | |
| *** sgordon has joined #heat | 12:38 | |
| *** alexheneveld has joined #heat | 12:40 | |
| *** saju_m has quit IRC | 12:41 | |
| *** gokrokve has quit IRC | 12:43 | |
| *** gokrokve has joined #heat | 12:43 | |
| *** mkollaro has joined #heat | 12:44 | |
| *** gokrokve has quit IRC | 12:47 | |
| *** asalkeld has quit IRC | 12:48 | |
| *** dims_ has quit IRC | 12:52 | |
| *** achampion has quit IRC | 12:52 | |
| *** gokrokve has joined #heat | 12:53 | |
| *** insanidade has joined #heat | 12:55 | |
| *** saju_m has joined #heat | 12:55 | |
| insanidade | sdake_: are you there ? quick question, if you don't mind. | 12:55 |
| *** saju_m has quit IRC | 12:59 | |
| *** mwheckmann has joined #heat | 12:59 | |
| *** saju_m has joined #heat | 13:00 | |
| *** rpothier has joined #heat | 13:01 | |
| *** dims_ has joined #heat | 13:02 | |
| *** mkollaro has quit IRC | 13:03 | |
| *** mkollaro1 has joined #heat | 13:03 | |
| *** mkollaro1 is now known as mkollaro | 13:03 | |
| *** matsuhashi has quit IRC | 13:03 | |
| *** saju_m has quit IRC | 13:04 | |
| *** saju_m has joined #heat | 13:05 | |
| aignatov | hi heaters, I'm trying to create stack with the following template http://paste.openstack.org/show/74245/ | 13:06 |
| aignatov | but always I get StackValidationFailed: Property error : eba: image "1" does not validate glance.image | 13:07 |
| aignatov | what does 'glance.image' mean? is it special type of some prop? | 13:07 |
| *** bgorski has quit IRC | 13:07 | |
| *** saju_m has quit IRC | 13:09 | |
| therve | aignatov, That means you don't have an image named "1" | 13:10 |
| *** saju_m has joined #heat | 13:10 | |
| *** asalkeld has joined #heat | 13:11 | |
| aignatov | therve: if i pass existed image id I get the same | 13:12 |
| aignatov | existing | 13:12 |
| therve | I don't know then | 13:13 |
| therve | The message should be improved | 13:13 |
| therve | If you have access to heat logs you may have some more info | 13:14 |
| *** spzala has joined #heat | 13:16 | |
| *** TonyBurn has quit IRC | 13:17 | |
| *** blomquisg has joined #heat | 13:18 | |
| *** blomquisg has quit IRC | 13:18 | |
| *** bgorski has joined #heat | 13:19 | |
| *** blomquisg has joined #heat | 13:19 | |
| *** arbylee has joined #heat | 13:20 | |
| openstackstatus | NOTICE: the gerrit event stream is currently hung, blocking all testing. troubleshooting is in progress (next update at 14:00 utc) | 13:20 |
| *** ChanServ changes topic to "the gerrit event stream is currently hung, blocking all testing. troubleshooting is in progress (next update at 14:00 utc)" | 13:20 | |
| *** wchrisj has joined #heat | 13:20 | |
| *** pablosan_ is now known as pablosan | 13:20 | |
| *** lazy_prince has quit IRC | 13:22 | |
| *** lazy_prince has joined #heat | 13:24 | |
| lipinski | Heat allows for retrieving templates. Is there any way to retrieve environment used for stack -creation | 13:25 |
| *** jcru has joined #heat | 13:26 | |
| *** wchrisj has quit IRC | 13:26 | |
| *** mestery has joined #heat | 13:28 | |
| shardy | lipinski: At the moment, no | 13:28 |
| *** mestery has quit IRC | 13:28 | |
| *** pafuent has joined #heat | 13:28 | |
| *** ChanServ changes topic to "support @ https://ask.openstack.org | developer wiki @ https://wiki.openstack.org/wiki/Heat | development @ https://launchpad.net/heat | logged @ http://eavesdrop.openstack.org/irclogs/%23heat/" | 13:29 | |
| openstackstatus | NOTICE: the issue with gerrit cleared on its own before any corrective action was taken | 13:29 |
| *** mestery has joined #heat | 13:29 | |
| *** ramishra has quit IRC | 13:30 | |
| *** lazy_prince is now known as killer_prince | 13:35 | |
| *** wchrisj has joined #heat | 13:35 | |
| shardy | therve: quick question re 80528: | 13:40 |
| shardy | what's the use-case for the recurse_if logic? | 13:40 |
| therve | shardy, resource groups and autoscaling groups | 13:40 |
| *** nati_ueno has joined #heat | 13:40 | |
| shardy | therve: do you have an example of what the type: syntax would look like? I didn't see a test covering it? | 13:41 |
| shardy | therve: the reason for my question is last week I was discussing an idea with some Tuskar folks, and they'd like to be able to actually embed a nested template via type (the actual body via a reference to another resource) | 13:42 |
| therve | shardy, http://paste.openstack.org/show/74246/ for example. Maybe it's missing a test indeed | 13:42 |
| therve | shardy, get_file may be the way for that | 13:43 |
| *** fandi has joined #heat | 13:44 | |
| *** cfriesen has joined #heat | 13:44 | |
| *** duncanjw has quit IRC | 13:47 | |
| *** blomquisg has quit IRC | 13:47 | |
| *** samstav has joined #heat | 13:50 | |
| shardy | therve: Ok, thanks, the language in the commit message made it sound a bit similar to what we discussed, which was something like: | 13:50 |
| shardy | http://paste.openstack.org/show/74248/ | 13:50 |
| *** radez_g0n3 is now known as radez | 13:51 | |
| shardy | where you could for example have a provider template which took various parameters, and output a stack template and an image which had been built and uploaded to glance | 13:51 |
| shardy | I guess we'd have to resolve that in the parser anyway not the client | 13:52 |
| *** andersonvom has quit IRC | 13:53 | |
| *** zns has joined #heat | 13:57 | |
| *** wchrisj has quit IRC | 13:57 | |
| *** wchrisj_ has joined #heat | 13:59 | |
| therve | shardy, That part would need to be done on the server side, I think | 13:59 |
| *** blomquisg has joined #heat | 13:59 | |
| therve | Oh you just said that | 13:59 |
| *** andersonvom has joined #heat | 13:59 | |
| *** nati_ueno has quit IRC | 13:59 | |
| *** achampion has joined #heat | 14:00 | |
| *** ramishra has joined #heat | 14:00 | |
| *** ramishra has quit IRC | 14:02 | |
| *** ramishra has joined #heat | 14:04 | |
| *** pas-ha has quit IRC | 14:06 | |
| *** duncanjw has joined #heat | 14:08 | |
| *** vijendar has joined #heat | 14:09 | |
| *** faramir1 has quit IRC | 14:18 | |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Fail if non-existent security group referenced https://review.openstack.org/81558 | 14:21 |
| *** alexheneveld_ has joined #heat | 14:21 | |
| *** alexheneveld has quit IRC | 14:22 | |
| *** alexheneveld_ is now known as alexheneveld | 14:22 | |
| therve | zaneb, Should you use HeatException? | 14:22 |
| zaneb | therve: check the comments for patch set 2 ;) | 14:23 |
| aignatov | therve: hi, i conducted log for the issue I faced above | 14:23 |
| aignatov | http://paste.openstack.org/show/74251/ | 14:23 |
| *** mestery has quit IRC | 14:24 | |
| therve | zaneb, What's the benefit of not using it though? | 14:25 |
| zaneb | it's just a replacement for a client exception... those don't inherit from HeatException either... what is the point? | 14:26 |
| *** mkollaro has quit IRC | 14:26 | |
| * zaneb was happy when it was just "raise Exception(...)" | 14:27 | |
| *** alexpilotti has quit IRC | 14:28 | |
| *** david-lyle has joined #heat | 14:31 | |
| therve | zaneb, It provides the msg_fmt thing | 14:31 |
| *** mestery has joined #heat | 14:33 | |
| *** jmckind has joined #heat | 14:35 | |
| jasond | why does heat validate resources that weren't changed during update? | 14:36 |
| aignatov | therve: could you please try my example above? I think it's easy to reproduce | 14:36 |
| shardy | jasond: we have to validate the whole updated stack because unchanged resources may reference those which have changed | 14:39 |
| therve | aignatov, Well, it doesn't work for me because I don't have an image with such an id | 14:41 |
| *** mkollaro has joined #heat | 14:41 | |
| therve | aignatov, What's the output of glance image-list? | 14:41 |
| *** cmyster has joined #heat | 14:43 | |
| *** cmyster has joined #heat | 14:43 | |
| cmyster | back (network outage) | 14:44 |
| aignatov | therve: http://paste.openstack.org/show/74252/ | 14:44 |
| aignatov | the last one in the list | 14:44 |
| therve | aignatov, What about nova image-list? | 14:45 |
| aignatov | therve: http://paste.openstack.org/show/74253/ | 14:46 |
| jasond | shardy: i see. i guess you can't validate the state of something that changes after create then https://github.com/openstack/heat/blob/master/heat/engine/resources/nova_keypair.py#L130-L137 | 14:46 |
| therve | aignatov, Well I don't know. | 14:47 |
| aignatov | so, it seems there is some bug in heat, because before adding glance constraints it was worked well | 14:48 |
| therve | It works for me, I start instances all the time | 14:49 |
| therve | The glance constraint does mostly the same thing as before | 14:49 |
| therve | Except with a crappy error message | 14:50 |
| aignatov | ok, I'll try to check again my environment | 14:50 |
| shardy | jasond: that code actually looks wrong to me - I think that validation should probably be a constratint in the schema for the name property, not a validation check | 14:50 |
| shardy | constraint even | 14:50 |
| *** wwallnrr__ has joined #heat | 14:52 | |
| shardy | jasond: In general, I expect resource validate() methods to implement more static checking, like dependent or mutually exclusive properties | 14:52 |
| therve | aignatov, You can try this patch to get more info: http://paste.openstack.org/show/74254/ | 14:52 |
| *** Qiming has joined #heat | 14:53 | |
| jasond | shardy: makes sense. i don't think it's being used that way everywhere | 14:53 |
| *** rcleere has joined #heat | 14:54 | |
| jasond | shardy: i'll see if a parameter constraint will work. thanks | 14:54 |
| aignatov | therve: thx, I'll try | 14:56 |
| *** arbylee has quit IRC | 14:58 | |
| *** bgorski has quit IRC | 14:59 | |
| *** bgorski has joined #heat | 14:59 | |
| *** kgriffs_afk is now known as kgriffs | 14:59 | |
| *** IlyaE has joined #heat | 15:00 | |
| * shardy just did his first ever stack-abandon && stack-adopt | 15:00 | |
| *** arbylee has joined #heat | 15:00 | |
| shardy | kinda interesting/scary | 15:00 |
| aignatov | therve: hehe, I've got the following http://paste.openstack.org/show/74255/ | 15:00 |
| aignatov | seems issue is clear for me now | 15:00 |
| aignatov | but definitely it should be raised before validation :) | 15:01 |
| therve | Hum what | 15:01 |
| *** alexheneveld has quit IRC | 15:01 | |
| therve | I guess it created the keystone client at that point? That's pretty bad | 15:02 |
| aignatov | where could I find it? in logs? | 15:03 |
| shardy | well heat.conf is misconfigured, stack_user_domain has been modified from the default without setting up an admin | 15:04 |
| shardy | aignatov: how did you install heat? | 15:04 |
| aignatov | locally, into my laptop, my heat looks into the devstack env outside of my machine | 15:06 |
| pscheie | Is there a heat command that shows all the instances belonging to a stack? | 15:06 |
| *** sgordon has quit IRC | 15:07 | |
| pscheie | resource-list shows all the resources, but, for example, doesn't show that we have 10 instances of a certain resource (spun via instancegroup) | 15:08 |
| shardy | aignatov: sounds like either the heat.conf has been manually updated or devstack is not up to date | 15:08 |
| pscheie | I can do 'nova list |grep <stackname>' but that seems...crude. | 15:09 |
| therve | pscheie, heat resource-show ? | 15:09 |
| aignatov | shardy: yes, my heat.conf contains only stack_user_domain property | 15:10 |
| shardy | aignatov: in particular perhaps you're missing https://review.openstack.org/#/c/76036/2/lib/heat and https://review.openstack.org/#/c/75424/ | 15:10 |
| *** spzala has quit IRC | 15:10 | |
| zaneb | shardy: what was that summit session you went to about Neutron reimplementing orchestration? | 15:11 |
| zaneb | shardy: I'm looking here but can't spot it: https://wiki.openstack.org/wiki/Summit/Icehouse/Etherpads#Neutron | 15:11 |
| pscheie | therve, that gives me info about an individual resource, but not of the whole stack. | 15:13 |
| *** saju_m has quit IRC | 15:13 | |
| *** spzala has joined #heat | 15:13 | |
| Qiming | heat resource-list --recursive <stack_name> | 15:13 |
| shardy | zaneb: https://etherpad.openstack.org/p/icehouse-neutron-service-insertion-chaining I think | 15:13 |
| therve | pscheie, I guess I meant resource-list | 15:13 |
| zaneb | shardy: ah, thanks | 15:13 |
| shardy | zaneb: they wanted to have a template defined combination of resources, which they would orchestrate IIRC | 15:13 |
| zaneb | it's becoming clear to me why they want that | 15:14 |
| zaneb | the API is so broken that it's illegible to external orchestration | 15:14 |
| *** packet has joined #heat | 15:14 | |
| shardy | zaneb: haha, well we've known that for a while ;) | 15:15 |
| zaneb | so instead of fixing it, they want to reimplement orchestration inside the API | 15:15 |
| pscheie | therve, yes, that sort of works--except we're starting to use InstanceGroups to create multiple instances, and resource-list only shows one resource, that of the InstanceGroup. | 15:15 |
| *** wchrisj_ has quit IRC | 15:15 | |
| zaneb | I always new it was bad, but I'm starting to get a concrete sense of the particular ways in which it is broken | 15:15 |
| zaneb | knew* | 15:16 |
| pscheie | Qiming, what version of heat provides the --recursive option? | 15:17 |
| *** spzala has quit IRC | 15:17 | |
| Qiming | pscheie: no version, it is a feature request, :) | 15:18 |
| pscheie | Qiming, ah, yes, that would be useful. | 15:18 |
| Qiming | not only CLI output is confusing, horizon is suffering from the same problem | 15:19 |
| *** cmyster has quit IRC | 15:19 | |
| *** e0ne has joined #heat | 15:19 | |
| Qiming | when you create a xxxGroup resource, the members can not be shown, so the output is not correct | 15:20 |
| shardy | pscheie: You can use the nested stack id available via resource-show <stack> <instancegroupresource> to resource-list the nested stack containing the resources | 15:20 |
| shardy | Qiming: The output is correct, just not what you're expecting - the resources in xxGroup are created in a nested stack, so they are not in the top level stack containingthe xxGroup resource | 15:21 |
| *** cmyster has joined #heat | 15:21 | |
| *** cmyster has joined #heat | 15:21 | |
| shardy | arguably some convenient way to traverse nested resources from the client would be nice | 15:21 |
| shardy | atm you have to do it in several steps | 15:21 |
| *** rbuilta has joined #heat | 15:22 | |
| Qiming | if you are referencing some other resources in the LaunchConfiguration, in the horizon topology view, those resources are linked to the group resource, because the individual members cannot be shown | 15:22 |
| shardy | http://hardysteven.blogspot.co.uk/2013_08_01_archive.html | 15:22 |
| pscheie | yeah, I'm doing it in several steps currently. | 15:23 |
| shardy | that provides a simple worked example of how to get the resource-list for the nested stack | 15:23 |
| pscheie | was hoping there was something built-in for it. | 15:23 |
| zaneb | shardy: looks like you'll have more work to do in Atlanta: http://summit.openstack.org/cfp/details/19 ;) | 15:23 |
| *** alexheneveld has joined #heat | 15:23 | |
| aignatov | shardy, therve: yes, it's working now, I've just removed stack_user_domain definition from my heat.conf | 15:24 |
| shardy | zaneb: sigh | 15:24 |
| *** mestery has quit IRC | 15:24 | |
| shardy | aignatov: good to hear, at some point you may want to unstack/stack with a recent devstack to configure that correctly | 15:25 |
| Qiming | shardy: I said that it is incorrect, because heat really need a way to express 'contains' relationship, besides 'references' relationship among resources | 15:25 |
| *** pas-ha has joined #heat | 15:25 | |
| aignatov | shardy, thx, domain specific things is new for me, when should I use it? | 15:25 |
| shardy | Qiming: And I agreed the client interfaces can be improved :) | 15:25 |
| aignatov | is there any docs about it? | 15:26 |
| shardy | aignatov: It's turned on by default in devstack now, just looks like you were unlucky and somehow ended up with a broken config | 15:26 |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Don't catch all exceptions in image constraint https://review.openstack.org/82846 | 15:26 |
| openstackgerrit | Jun Jie Nan proposed a change to openstack/python-heatclient: get_file: encode non utf-8 encoding files via base64 https://review.openstack.org/76443 | 15:27 |
| openstackgerrit | Jun Jie Nan proposed a change to openstack/python-heatclient: get_file: do not read same url once again https://review.openstack.org/80882 | 15:27 |
| shardy | aignatov: There are links to the BP and key patches here: https://bugs.launchpad.net/puppet-heat/+bug/1288880 | 15:27 |
| uvirtbot | Launchpad bug 1288880 in puppet-heat "Updates required for stack domain users" [Undecided,New] | 15:27 |
| *** alexheneveld_ has joined #heat | 15:28 | |
| shardy | aignatov: I've not done docs for deployers yet, but I'm planning to write something | 15:28 |
| *** alexheneveld has quit IRC | 15:30 | |
| shardy | aignatov: note heat auth_token needs this https://review.openstack.org/#/c/80341/ to avoid the issues discussed in bug #1291097 | 15:30 |
| uvirtbot | Launchpad bug 1291097 in heat "deployment native signalling failing" [High,Fix committed] https://launchpad.net/bugs/1291097 | 15:30 |
| shardy | I need to figure out how to get some response from devstack-core on that one :\ | 15:31 |
| *** alexheneveld has joined #heat | 15:31 | |
| shardy | That won't affect you unless you're using the new software-config resources | 15:31 |
| aignatov | shardy: Thanks a lot for the info, I'm just trying to figure out what configuration needed to fix savanna(sahara) heat provisioning engine | 15:32 |
| *** alexheneveld_ has quit IRC | 15:32 | |
| aignatov | because in the mid-icehouse there were no needs in domain specific things | 15:33 |
| shardy | aignatov: Yeah, it's new stuff which recently landed under the instance-users BP | 15:33 |
| shardy | aignatov: If you do nothing then heat falls back to the old behavior with log warnings | 15:34 |
| shardy | aignatov: the advantage of the new approach is you no longer need to be admin to create heat stacks containing certain resources | 15:34 |
| *** wchrisj has joined #heat | 15:35 | |
| *** wchrisj has quit IRC | 15:35 | |
| aignatov | shardy: any special resource types? | 15:36 |
| aignatov | or is it something like delegating admin privileges? | 15:36 |
| *** wchrisj has joined #heat | 15:36 | |
| shardy | aignatov: Anything which creates a keystone user, so WaitConditionHandle, ScalingPolicy, User/AccessKey and the new *SoftwareDeployment resources | 15:37 |
| shardy | aignatov: it's for the credentials we deploy inside instances e.g for signalling waitcondition events | 15:37 |
| shardy | aignatov: the users associated with those credentials are now created in a separate heat-specific domain, completely isolated from the stack owner's project | 15:37 |
| shardy | which is more secure, and removes the admin requirement, because you configure a domain-admin user which heat uses to manage the create/delete of those users | 15:38 |
| *** sgordon has joined #heat | 15:39 | |
| aignatov | shardy: makes sense, thanks for the explanation, I'll try to play with it for more comprehension | 15:42 |
| *** Qiming has quit IRC | 15:42 | |
| *** alexpilotti has joined #heat | 15:44 | |
| sdake_ | morning | 15:45 |
| *** alexpilotti has quit IRC | 15:46 | |
| aignatov | shardy, one simple question, Sahara has one use case when it creates trust for long living clusters to be able remove it after some time. So using heat, do I need special configuration in heat to work with trusts? | 15:47 |
| shardy | aignatov: Not for you to use trusts, but if you want heat to use trusts (instead of storing a password in the DB), e.g for deferred operations like autoscaling, you may want to specify deferred_auth_method=trusts | 15:48 |
| shardy | that also became the default in devstack recently and we're moving towards making it the heat default for everyone: | 15:49 |
| shardy | https://review.openstack.org/#/c/80002/ | 15:50 |
| openstackgerrit | Rabi Mishra proposed a change to openstack/heat: Provides ability to reference existing neutron resources https://review.openstack.org/82853 | 15:50 |
| aignatov | shardy: thanks | 15:54 |
| *** jergerber has joined #heat | 15:54 | |
| *** Linz has joined #heat | 15:57 | |
| *** Linz has quit IRC | 15:57 | |
| *** packet has quit IRC | 15:57 | |
| *** Linz has joined #heat | 15:57 | |
| *** killer_prince is now known as lazy_prince | 15:57 | |
| sdake_ | shadower nice work on the sellinux issue | 15:58 |
| sdake_ | shadower++ | 15:58 |
| *** alexheneveld_ has joined #heat | 15:59 | |
| *** packet has joined #heat | 15:59 | |
| *** mestery has joined #heat | 15:59 | |
| *** alexheneveld has quit IRC | 16:00 | |
| *** alexheneveld_ is now known as alexheneveld | 16:00 | |
| *** fandi has quit IRC | 16:06 | |
| *** fandi has joined #heat | 16:09 | |
| *** mestery has quit IRC | 16:09 | |
| *** mestery has joined #heat | 16:10 | |
| *** skraynev_afk is now known as skraynev | 16:14 | |
| skraynev | zaneb: are you here? | 16:14 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Updated from global requirements https://review.openstack.org/76698 | 16:15 |
| zaneb | skraynev: I am :) | 16:15 |
| jasond | shardy: apparently even parameters that haven't been change get validated during update, so a property constraint isn't going to work :/ | 16:15 |
| skraynev | zaneb: couple questions. | 16:16 |
| skraynev | zaneb: first about https://review.openstack.org/#/c/81558/ | 16:16 |
| skraynev | zaneb: could you add assert for checking status_reason of stack | 16:16 |
| *** nati_ueno has joined #heat | 16:17 | |
| zaneb | I guess that would make sense | 16:17 |
| shardy | jasond: not sure I follow, so the issue is you want to update to create a keypair which already exists with the same name? | 16:17 |
| skraynev | zaneb: I suppose, it give additional confirmation of cause failing | 16:17 |
| skraynev | zaneb: ok | 16:18 |
| jasond | shardy: if you update a stack without changing the nova keypair, nova keypair validation will fail https://bugs.launchpad.net/heat/+bug/1296747 | 16:18 |
| uvirtbot | Launchpad bug 1296747 in heat "Nova keypair cannot be updated" [High,Triaged] | 16:18 |
| skraynev | zaneb: and other question about validation of outputs. | 16:18 |
| shardy | jasond: maybe that logic should move so we only check in handle_create? | 16:18 |
| skraynev | zaneb: in comment you offered call validate(self.args). May be you meant validate(self, self.args) or something else? | 16:19 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add OS:Barbican:Order resource https://review.openstack.org/81906 | 16:19 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add OS:Barbican:Secret resource https://review.openstack.org/79355 | 16:19 |
| jasond | shardy: wouldn't that be the same as removing validation completely? the stack would still be created and then fail | 16:20 |
| jasond | that's what i'm leaning toward | 16:20 |
| zaneb | skraynev: do you have the link handy? | 16:20 |
| skraynev | yeap... sec | 16:21 |
| skraynev | zaneb: https://review.openstack.org/#/c/82486/1/heat/engine/function.py | 16:21 |
| zaneb | ta | 16:21 |
| zaneb | skraynev: nope, my comment is correct :) | 16:22 |
| *** ifarkas has quit IRC | 16:22 | |
| therve | jasond, I remember commenting on https://bugs.launchpad.net/heat/+bug/1278532 and not getting a response | 16:22 |
| uvirtbot | Launchpad bug 1278532 in heat "The KeyPair resource needs validation" [Undecided,Fix released] | 16:22 |
| *** sdague has joined #heat | 16:22 | |
| therve | Considering the issue maybe we should revert that change? | 16:22 |
| shardy | jasond: what about comparing the resource_id with the id from the get_keypair instead? | 16:22 |
| sdague | so reiterating a heat question from -dev in wrapping my brain around things enough to help build out some tests | 16:23 |
| jasond | therve: your comment was vague so i didn't comment. i should've asked for clarification | 16:23 |
| sdague | heat question, I shouldn't need anything more than cloud-init in an instance to get user_data executed in an OS::Nova::Server, should I? | 16:23 |
| therve | jasond, Well, yes. Please do :) | 16:23 |
| skraynev | zaneb: Well, it will be class method or you suggest call it in validate method of Function class? | 16:23 |
| shardy | jasond: and just pass instead of raising if the id's match? | 16:23 |
| lipinski | Mentioned this yesterday and just got around to trying again. - Nested provider templates. | 16:23 |
| lipinski | These seem to work only if I set all the resources in a registry under /etc/heat/environment.d | 16:24 |
| shardy | sdague: No, if you specify user_data_format=RAW, otherwise you also need heat-cfntools | 16:24 |
| lipinski | If I set a resource registry in a passed environment file, nested ones are not found. | 16:24 |
| zaneb | skraynev: it's not a method, it's a function. you already defined it right there in that patch | 16:24 |
| sdague | shardy: ok, I've got user_data_format=RAW | 16:24 |
| therve | lipinski, That'd be a bug I suppose | 16:24 |
| lipinski | e.g., template -> ResourceA (providerA) -> ResourceB (providerB) = ResourceB not found | 16:24 |
| shardy | sdague: Ok, it should just get passed to cloud-init then | 16:24 |
| lipinski | therve: ok - will submit. thanks | 16:24 |
| shardy | sdague: otherwise you can use e.g the Fedora cloud image (F20+) which contains heat-cfntools | 16:25 |
| sdague | oh, it looks like it mostly did work, but blew up earlier than I expected | 16:25 |
| sdague | shardy: yeh, I was mostly trying to figure out what's testable with just cloud-init on some of these, as that should make us be able to do more things in the gate | 16:26 |
| sdague | as we could even do quite a bit with cirros | 16:26 |
| lipinski | ok, one more, plese confirm as a bug (and not user error) before I submit: When I use a file URL in resource_registry, there is some corruption of the url | 16:27 |
| skraynev | zaneb: Ou you talk about ancillary function. | 16:27 |
| lipinski | resource_registry: "My::Res::Node": "file:///root/My-Res-Node.hot.yaml" | 16:27 |
| sdague | anyway, at least I'm slightly sanity checked now, I can sort out the specific fail | 16:27 |
| sdague | thanks | 16:28 |
| lipinski | error: Could not fetch file:///root/file:///root/My-Res-Node.hot.yaml from the environment | 16:28 |
| shardy | sdague: Sounds good - you may also be interested in the new OS::Heat::MultipartMime and OS::Heat::CloudConfig resources then | 16:28 |
| lipinski | It's automatically adding on 'file:///$PWD' | 16:28 |
| sdague | shardy: probably, but I need to walk before I can run | 16:28 |
| skraynev | zaneb: but it looks like recursion , because if we call func.validate, it will call validate() and that function.validate and etc. | 16:29 |
| jasond | shardy: good idea. i'll see if that will work. thanks | 16:29 |
| shardy | sdague: btw, I'm going to be focussing on tempest tests over the next couple of weeks, please shout if there are any concerns re where I'm putting stuff in the tree ;) | 16:29 |
| zaneb | skraynev: why is recursion bad? | 16:30 |
| *** varora has left #heat | 16:30 | |
| *** nati_ueno has quit IRC | 16:30 | |
| shardy | sdague: I'm not 100% clear where API testing ends and scenario testing begins, but hopefully we can add some better coverage to both | 16:30 |
| skraynev | zaneb: no, recursion is cool thing, and sometimes it gives head explosion. But I understand my mistake, I thought that it will be infinite recursion | 16:31 |
| zaneb | skraynev: say you have a tree like this: http://paste.openstack.org/show/74265/ | 16:32 |
| sdague | shardy: will do | 16:32 |
| zaneb | you need to recurse to make sure you validate both functions | 16:32 |
| zaneb | you can't stop at the first one | 16:32 |
| *** alexheneveld has quit IRC | 16:33 | |
| zaneb | and you can't call the validate method on its args, because args is a list, not a Function | 16:33 |
| zaneb | so you call validate and it continues traversing the tree | 16:33 |
| skraynev | zaneb: gotcha. | 16:33 |
| zaneb | cool :) | 16:34 |
| skraynev | zaneb: I forgot, that in validate we will call on args of function not on this function. | 16:34 |
| zaneb | I can see why that would have been confusing :) | 16:35 |
| *** pvaneck has joined #heat | 16:35 | |
| skraynev | zaneb: I guess it is my fatigue today ;) never mind :) | 16:38 |
| jasond | shardy: bah, resource_id isn't set at that point. i'll just revert the commit that added validation | 16:40 |
| *** sgordon has quit IRC | 16:40 | |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Revert "Add validation to KeyPair resource" https://review.openstack.org/82862 | 16:41 |
| sdague | shardy: yeh, especially with the fact that a lot of heat's API is really the HOT dsl, it's an interesting question. Honestly, my take right now is to determine it by backend image. If we can pull it off quickly with cirros, it's probably game for API | 16:43 |
| sdague | if it's going to be longer and require a real image, put it in scenario | 16:44 |
| lipinski | Is there a way to perform a count in a template? Say, get the size of a list? | 16:44 |
| shardy | sdague: Ok, sounds good, I'm starting with fairly minimal exercising of API calls we don't currently hit (in API tests) and plan to move on to more complex "real" tests in scenarios | 16:45 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add OS:Barbican:Order resource https://review.openstack.org/81906 | 16:46 |
| pas-ha | anyone having problems with latest devstack? | 16:49 |
| *** jpeeler has joined #heat | 16:52 | |
| *** jpeeler has quit IRC | 16:52 | |
| *** jpeeler has joined #heat | 16:52 | |
| *** tomek_adamczewsk has quit IRC | 16:58 | |
| *** jistr|training has quit IRC | 16:58 | |
| *** IYozhikov_away is now known as IgorYozhikov | 16:59 | |
| *** nati_ueno has joined #heat | 17:02 | |
| *** ifarkas has joined #heat | 17:03 | |
| radix | good morning | 17:03 |
| *** harlowja has joined #heat | 17:03 | |
| *** cmyster has quit IRC | 17:05 | |
| *** mkollaro has quit IRC | 17:07 | |
| *** tspatzier has quit IRC | 17:08 | |
| *** kgriffs is now known as kgriffs_afk | 17:10 | |
| *** mkollaro has joined #heat | 17:10 | |
| *** ramishra has quit IRC | 17:11 | |
| *** saurabhs has joined #heat | 17:12 | |
| *** ramishra has joined #heat | 17:13 | |
| *** alexheneveld_ has joined #heat | 17:15 | |
| lipinski | any way to set per-instance data on a resourcegroup? | 17:16 |
| openstackgerrit | A change was merged to openstack/heat: Don't create cloud-init user unless specified https://review.openstack.org/79678 | 17:17 |
| *** saurabhs1 has joined #heat | 17:17 | |
| radix | lipinski: can you be more specific? | 17:18 |
| lipinski | I create a ResourceGroup with say 4 instances. Can I give them each a unique name? | 17:19 |
| *** sgordon has joined #heat | 17:19 | |
| lipinski | e.g., node-1, node-2, node-3, etc. | 17:19 |
| *** saurabhs has quit IRC | 17:19 | |
| *** ifarkas_ has joined #heat | 17:20 | |
| *** zns has quit IRC | 17:21 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Include exception in custom constraints errors https://review.openstack.org/82881 | 17:27 |
| lipinski | radix: you know if that's possible? | 17:27 |
| radix | ahem, sorry | 17:27 |
| radix | lipinski: no, names are generated for you | 17:28 |
| radix | lipinski: oh, wait, do you mean resource names or server names? | 17:29 |
| lipinski | server names | 17:29 |
| radix | hmm | 17:29 |
| lipinski | (well, any parameters for that matter). | 17:29 |
| radix | I don't _think_ that's possible either | 17:29 |
| lipinski | It would be nice if I could pass a list of data for any resource property and that list would be cycled through. | 17:29 |
| *** rpothier_ has joined #heat | 17:29 | |
| radix | lipinski: can you explain your use case? | 17:30 |
| lipinski | e.g., I could have a ResourceGroup of 4 instances. For a avail_zone, I could pass a [ 'zone1', 'zone2' ] value. When the group was instantiated, instance 0 would be assigned zone1; instance 1 zone2; instance2 zone1; instance 3 zone2 | 17:30 |
| lipinski | I have multiple use cases where I want to have a group of resources that have same properties (for the most part) - e.g., flavor, image, volumes, etc. But, I just want to pass some data that may differ for each instance. | 17:31 |
| *** skraynev is now known as skraynev_afk | 17:32 | |
| *** jprovazn has quit IRC | 17:32 | |
| *** rpothier has quit IRC | 17:33 | |
| therve | lipinski, You can create 2 groups I suppose for now | 17:34 |
| lipinski | therve: yeah - by the time I address the handful of different permutations - I end up having to individually specify each server - negating the usefulness of ResourceGroup at all. | 17:34 |
| *** zns has joined #heat | 17:35 | |
| lipinski | avail_zone was just an example. Plus, I have a requirement for each VM to be created with a unique name - ResourceGroup can't seem to do that either. | 17:35 |
| lipinski | Right now, on the OS Dashboard all VMs are spun up with same name. | 17:35 |
| therve | lipinski, servers are created with a unique name already though | 17:36 |
| *** e0ne has quit IRC | 17:36 | |
| lipinski | therve: I specify a name property to OS::Nova::Server. | 17:37 |
| lipinski | I have a (now conflicting) requirement about having a user-defined prefix to the server names. | 17:37 |
| therve | Well yeah don't :) | 17:37 |
| lipinski | I wish. | 17:37 |
| therve | You can use the stack name maybe | 17:37 |
| lipinski | customers have long wish lists :) | 17:37 |
| lipinski | and short timeframes | 17:37 |
| lipinski | I'm trying to get to a static template that can be used and support multiple configuration options simply by changing the paraemeters/environment. It's not working as I'd hope and now will have to have software that generates a custom template for every deployment - at that point, templates start to lose their value pretty quickly as I could just invoke the orchestration myself... | 17:38 |
| * therve shrugs | 17:38 | |
| lipinski | I know. I do appreciate the help, though... | 17:39 |
| lipinski | does any of this sound like something I should create a blueprint for? Would that be possibly useful for others? | 17:39 |
| radix | it's unclear what features you would want | 17:39 |
| lipinski | some way to specify per-instance parameters via a ResourceGroup. probably easiest though the use of lists | 17:40 |
| therve | That may be doable | 17:41 |
| lipinski | I would think it wouldn't be too hard. I'm no python developer, or I would prototype something myself. I may try to anyways to learn. I see the majority of the code in heat/engine/resources/resource_group.py | 17:42 |
| *** mspreitz has joined #heat | 17:44 | |
| *** jstrachan has quit IRC | 17:47 | |
| *** mestery_ has joined #heat | 17:48 | |
| *** mestery has quit IRC | 17:48 | |
| *** mspreitz_ has joined #heat | 17:49 | |
| *** mspreitz_ has left #heat | 17:49 | |
| *** mspreitz has left #heat | 17:50 | |
| *** mestery_ is now known as mestery | 17:50 | |
| *** derekh has quit IRC | 17:55 | |
| *** imtiaz has joined #heat | 17:55 | |
| *** Darnoth has joined #heat | 17:56 | |
| *** imtiaz has quit IRC | 17:57 | |
| *** mspreitz has joined #heat | 17:59 | |
| mspreitz | SpamapS: are you there? | 18:00 |
| *** yogesh_ has joined #heat | 18:00 | |
| mspreitz | zaneb: are you there? | 18:00 |
| *** kgriffs_afk is now known as kgriffs | 18:02 | |
| therve | lipinski, Ideally you would create a blueprint describing what you'd like, with use cases and example templates | 18:06 |
| therve | Otherwise a mail on the list | 18:06 |
| lipinski | therve: ok - will do. Thanks | 18:06 |
| *** packet has quit IRC | 18:07 | |
| SpamapS | mspreitz: here for a bit.. whassup? | 18:08 |
| mspreitz | SpamapS: just wanted to confirm some thinking about how to fix that problem where Heat is stuck if an update fails partway through. | 18:09 |
| mspreitz | First, is there an official bug for this? | 18:09 |
| sdake | zaneb looks like the murano cats get it to me | 18:10 |
| sdake | zaneb whats your take? | 18:10 |
| SpamapS | mspreitz: Yes... | 18:10 |
| SpamapS | mspreitz: https://bugs.launchpad.net/heat/+bug/1160052 | 18:10 |
| uvirtbot | Launchpad bug 1160052 in heat "Need a way to retry failed operations" [High,In progress] | 18:10 |
| mspreitz | thanks | 18:11 |
| SpamapS | mspreitz: and there is a blueprint attached https://blueprints.launchpad.net/heat/+spec/update-failure-recovery | 18:11 |
| *** packet has joined #heat | 18:11 | |
| SpamapS | mspreitz: I have recently confirmed that you can manually recover with abandon/adopt | 18:12 |
| mspreitz | I am having trouble finding the link from bug to blueprint | 18:12 |
| mspreitz | Is there one? | 18:13 |
| mspreitz | I am guessing not. Moving along.. | 18:14 |
| SpamapS | mspreitz: there is | 18:14 |
| SpamapS | mspreitz: it is on the right side, after all the subscribers | 18:14 |
| mspreitz | Ah, thanks! | 18:14 |
| SpamapS | mspreitz: and on the blueprint there is a link back to the bug | 18:14 |
| mspreitz | Yes, the link in the other direction is nicely obvious | 18:15 |
| mspreitz | SpamapS: In our earlier discussion, and in that blueprint, I do not see the word "workflow". This gratifies me, I think Heat orchestration is not and should not be as simple as creating and running a workflow. | 18:16 |
| SpamapS | sdake: hey.. I'm a bit concerned about https://review.openstack.org/#/c/79678/ ... | 18:16 |
| SpamapS | shardy: ^^ | 18:16 |
| mspreitz | I just wanted to double-check for agreement | 18:16 |
| sdake | SpamapS which part | 18:16 |
| SpamapS | that has changed a fairly fundamental thing about how OS::Nova::Server works.. a change I 100% support.. but I wonder if it will be too much surprise. | 18:16 |
| sdake | SpamapS I guess there are a couple ways to mitigate that | 18:17 |
| sdake | we could communicate the change (mailing lists, blog posts) | 18:17 |
| therve | What's that rule about empty lines between imports? | 18:17 |
| mspreitz | sdake: yes, socialize the change | 18:17 |
| sdake | If this is the way forward, we have to accept that at some point our users are going to have to deal with it | 18:18 |
| sdake | might as well be sooner rather then later | 18:18 |
| sdake | mspreitz yup but the change has already been made | 18:18 |
| *** blomquisg has quit IRC | 18:18 | |
| SpamapS | sdake: we should to that anyway. But IMO we have to have a cycle where we deprecate the behavior. Which is why I was so surprised at your respons on the users list taht this was "almost merged. | 18:18 |
| sdake | it can also be documented in the release notes | 18:18 |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Don't catch all exceptions in image constraint https://review.openstack.org/82846 | 18:19 |
| sdake | SpamapS I see, you mean wait 6 months to fix the bug while warning users its coming? | 18:19 |
| sdake | I guess my general take is a 6 month warning seems a bit much | 18:21 |
| SpamapS | sdake_: the fix involves changing a default behavior. | 18:21 |
| sdake | if we take that approach for every change, it quickly becomes untenable | 18:21 |
| mspreitz | SpamapS: sorry to double-team you, but regarding broken updates, I wanted to confirm that others agree that making and running a workflow is not the answer | 18:21 |
| mspreitz | others here can chime in too | 18:21 |
| sdake | SpamapS I guess there was a communication breakdown, I did read the review comments in detail before +a the review | 18:22 |
| mspreitz | I mean, not the best answer | 18:22 |
| sdake | SpamapS suggest adding an item for the wednesday meeting to discuss a plan to either mitigate or revert | 18:22 |
| SpamapS | sdake: we can make the new behavior optional. That makes it a config decision, not a 6 month release upgrade decision. | 18:23 |
| shardy | SpamapS: can you clarify what is backwards incompatible? | 18:23 |
| mspreitz | sdake: or make it a per-instance option? | 18:23 |
| SpamapS | mspreitz: ACK, window_size=1 | 18:23 |
| shardy | SpamapS: we still have the ADMIN_USER property, marked as support.DEPRECATED | 18:23 |
| SpamapS | shardy: users who expect to ssh as admin_user will be broken now. | 18:24 |
| therve | SpamapS, is the behavior really changing? | 18:24 |
| therve | If the conf value stayed the same? | 18:24 |
| SpamapS | shardy: right, we should have it marked as deprecated, but still defaulting to ec2-user .. wait.. we do.. doh | 18:24 |
| shardy | therve: only if they haven't specified a user in heat.conf | 18:24 |
| therve | admin_user is not even in Havana AFAICT | 18:25 |
| SpamapS | I misread the diff.. so the default stays ec2-user .. and OS::Nova::Server will still, by default, use it? | 18:25 |
| SpamapS | therve: it is. I put it there. | 18:25 |
| *** shakamunyi has joined #heat | 18:25 | |
| *** shakamunyi has quit IRC | 18:25 | |
| therve | hum... | 18:25 |
| *** shakamunyi has joined #heat | 18:26 | |
| *** shakamunyi has quit IRC | 18:26 | |
| shardy | SpamapS: Yes, it's still defaulted to ec2-user, but you get the new behavior if you set it to an empty string | 18:26 |
| therve | SpamapS, https://github.com/openstack/heat/blob/stable/havana/heat/engine/resources/server.py where? | 18:26 |
| sdake | ya imo that seems ok | 18:26 |
| *** lipinski1 has joined #heat | 18:26 | |
| SpamapS | Ok yeah I read it wrong. Ignore me. | 18:26 |
| sdake | but definately communicating it makes sense | 18:26 |
| SpamapS | but thanks for the reassurance :) | 18:27 |
| *** lipinski has quit IRC | 18:27 | |
| mspreitz | If old config gets old behavior, we are good | 18:27 |
| mspreitz | well, pretty good | 18:27 |
| shardy | mspreitz: AFAICS we do | 18:27 |
| mspreitz | that just moves the problem from our laps to the operators' | 18:27 |
| shardy | https://review.openstack.org/#/c/79678/5/heat/common/config.py | 18:27 |
| mspreitz | they still have the problem of whether to surprise all their users | 18:27 |
| SpamapS | mspreitz: that is ok. Operators can be expected to read release notes and look for deprecation errors in logs. | 18:28 |
| SpamapS | mspreitz: right, the idea is to let the users know _well_ in advance of the change. | 18:28 |
| mspreitz | Yeah, but it's the same kind of problem: at some point an operator has to throw the switch, pissing off any user who has not yet caught up. | 18:29 |
| SpamapS | mspreitz: Yeah.. 6 months may be too short for those users. | 18:30 |
| sdake | we dont have to deprecate in a 6 month window | 18:30 |
| SpamapS | mspreitz: it is measurable though. An operator has direct contact information for _every_ user. | 18:31 |
| sdake | we can keep the deprecation in for 12 or 18 months | 18:31 |
| *** blomquisg has joined #heat | 18:31 | |
| mspreitz | SpamapS: agreed, the operators have better contact with users | 18:31 |
| SpamapS | sdake: I think we might want to keep that one around until K | 18:31 |
| *** alexheneveld has joined #heat | 18:31 | |
| sdake | SpamapS no sweat off my back as long as selinux works ;) | 18:31 |
| SpamapS | mspreitz: excellent point. So, on to workflowing things | 18:32 |
| sdake | we still have things like describe which were deprecated 2 cycles ago | 18:32 |
| mspreitz | SpamapS: yes, please, I have to leave soon | 18:32 |
| SpamapS | mspreitz: me too. Ok so you are correct. THe idea is that Heat would in fact develop its own workflow to get from point a (where it is now) to point b (where the user wants it to go) | 18:33 |
| mspreitz | I have two reasons for not liking a workflow oriented approach. 1: it is not good at quickly reacting to exogenous changes on either side | 18:33 |
| *** alexheneveld_ has quit IRC | 18:34 | |
| mspreitz | 2: it is not good at retry logic that has to back up before it can move forward | 18:34 |
| *** ifarkas_ has quit IRC | 18:35 | |
| mspreitz | SpamapS: What I am saying is that I think the best approach is different from: Heat computes workflow to move from A to B and then hands it off to dumb workflow engine that can only run that workflow | 18:36 |
| mspreitz | Rather I think the best approach is a heat engine that, after making one change, can reconsider A and B and decide what to do next. | 18:37 |
| mspreitz | That naturally is more pwerful. | 18:37 |
| mspreitz | It can accommodate changes in A and B that happened while it was working on one thing | 18:37 |
| SpamapS | mspreitz: yes, you're saying, use convergence, rather than workflow | 18:37 |
| mspreitz | yes | 18:37 |
| SpamapS | we agree | 18:38 |
| mspreitz | I'm late for my next task, I am going away thinking there is agreement here. pls let me know asynchronously if I'm wrong. | 18:38 |
| SpamapS | mspreitz: so I'd like for that to be the way Heat works. I also think that to do that without a lot of spurious errors and retries, we need to get Heat's engine out of the business of tracking state for resources. | 18:39 |
| mspreitz | Yes, let Heat use the tracking that others already do, or get info from the true source. | 18:40 |
| mspreitz | Lot's more to say, but gotta go now. | 18:40 |
| SpamapS | mspreitz: ciao | 18:40 |
| *** mspreitz has left #heat | 18:40 | |
| *** e0ne has joined #heat | 18:41 | |
| therve | SpamapS, that's interesting. Do you think we shouldn't detect when a server goes into error? Or do you have something else in mind? | 18:41 |
| openstackgerrit | Rabi Mishra proposed a change to openstack/heat: Provides ability to reference existing neutron resources https://review.openstack.org/82853 | 18:42 |
| SpamapS | therve: Heat has no business monitoring things no. | 18:44 |
| SpamapS | therve: heat-engine I should say | 18:44 |
| SpamapS | therve: but Heat knowing when the monitoring service has detected a failure, and acting automatically, yes. | 18:44 |
| SpamapS | therve: the point is, Heat shouldn't keep track of the state of the resources. It should query or subscribe to notifications only. | 18:45 |
| shardy | +1 on notifications, I'd like to see all the polling for state stuff die | 18:47 |
| therve | Yes :) | 18:47 |
| therve | Vote for my design summit session! | 18:47 |
| sdake | ya notify would be good | 18:47 |
| sdake | prefer event driven - more scalable | 18:47 |
| shardy | Ooh, have submissions opened now? | 18:47 |
| therve | shardy, About 2 weeks ago | 18:47 |
| shardy | haha | 18:48 |
| therve | :) | 18:48 |
| * shardy missed that somehow | 18:48 | |
| *** kgriffs is now known as kgriffs_afk | 18:50 | |
| sdake | where is the link for the design summit sessions therve? | 18:50 |
| *** shakayumi has joined #heat | 18:50 | |
| *** shakayumi has quit IRC | 18:50 | |
| * sdake needs better filters | 18:50 | |
| therve | sdake: http://summit.openstack.org/ | 18:50 |
| *** shakayumi has joined #heat | 18:51 | |
| sdake | therve thanks | 18:52 |
| radix | zaneb: are you planning on a session for healing/convergence/whatever? | 18:53 |
| *** ifarkas_ has joined #heat | 18:53 | |
| radix | therve: wait, there's voting now? | 18:53 |
| therve | radix, Not really. During meetings I guess. | 18:54 |
| radix | SpamapS: or you? | 18:54 |
| radix | SpamapS: I'm wondering if there should be something where an external source can say "hey I noticed resource <x> is messed up" | 18:55 |
| *** ramishra has quit IRC | 18:55 | |
| *** ramishra has joined #heat | 18:55 | |
| SpamapS | radix: I have been really focused on getting graceful notifications into some kind of state for TripleO's push for HA.. once that is done, I will think about design sessions. | 18:56 |
| therve | radix, Presumably that's whare notifications are? | 18:56 |
| SpamapS | radix: But yes I want that to be the push | 18:56 |
| therve | s/whare/what | 18:56 |
| radix | SpamapS: well, yes, but I'm specifically thinking it needs to be a notification about specific resources | 18:56 |
| radix | er, s/SpamapS/therve/ | 18:56 |
| radix | therve: just telling heat "hey fix yourself" means checking all resources in a stack, which can be too much | 18:57 |
| radix | if we say "I found that resource X is dead" would allow heat to only repair that one resource without having to recheck everything else | 18:57 |
| shardy | radix: but convergence starts with checking all resources anyway | 18:57 |
| radix | shardy: no, not "checking" | 18:58 |
| therve | radix, Who's "I" in "I found"? | 18:58 |
| sdake | radix therve there is no voting for design sessions | 18:58 |
| shardy | radix: well that's the model we've been working towards - stack-check aligns DB status with the world, stack-converge fixes the world so all is good again :) | 18:58 |
| radix | shardy: hm, stack-align sounds unscalable | 18:58 |
| SpamapS | radix: oh .. hmmm | 18:59 |
| *** che-arne has quit IRC | 18:59 | |
| shardy | radix: or are you saying notify some out of band metric such as application status? | 18:59 |
| SpamapS | radix: that is a bit backwards from what I think Heat should do | 18:59 |
| *** derekh has joined #heat | 18:59 | |
| therve | sdake: I meant that some needs to be selected for the summit. It's more free flow indeed | 18:59 |
| *** jistr has joined #heat | 18:59 | |
| SpamapS | radix: I think Heat should subscribe to "server X status" ... the thing that knows about server X doesn't know about "resources" | 18:59 |
| shardy | radix: no more unscalable than create or update - asycnronous notifications/monitoring should happen via ceilometer, not heat | 18:59 |
| *** ramishra has quit IRC | 18:59 | |
| radix | shardy: no, I'm saying e.g. some monitoring system that would watch nova RPC events (or in the future, hopefully marconi queues or whatever that nova publishes) | 19:00 |
| radix | whatever it is, I don't think "check the world" is feasible | 19:00 |
| therve | radix, Yeah those are notifications | 19:00 |
| SpamapS | shardy: so I think we should just eliminate the DB status. | 19:00 |
| shardy | radix: Ah, Ok well that makes sense, I thought you were talking about operator interaction | 19:00 |
| *** bgorski has quit IRC | 19:00 | |
| sdake | I wonder how many session slots we have available | 19:01 |
| shardy | SpamapS: possbly, but it's useful as a cache so we don't have to poll for every user resource-show request | 19:01 |
| SpamapS | we still have to poll. | 19:02 |
| SpamapS | well.. sorry.. we still SHOULD | 19:02 |
| shardy | SpamapS: we'd just use it for caching the latest notification | 19:02 |
| *** jprovazn has joined #heat | 19:02 | |
| SpamapS | that status is a lie.. | 19:02 |
| sdake | if you use the notification system you dont have to poll | 19:02 |
| sdake | however, not all services implement notifications | 19:02 |
| shardy | sdake: ++ | 19:02 |
| sdake | so I think the expectation that we could rely on that as a general means may not be possible to meet | 19:02 |
| SpamapS | right so notifications are an optimization. | 19:02 |
| sdake | yup optimization plain and simple | 19:03 |
| therve | Well we need to do it incrementally anyway | 19:03 |
| shardy | sdake: Yeah I just added a comment to therve's session proposal saying we should start with an analysis of the state of notifications accross projects | 19:03 |
| therve | It's like like we can flip a switch | 19:03 |
| SpamapS | Polling for resource-show seems fine. Polling for resource-list might be a bit much.. but we can give the user ids.. and a --poll to get real statuses. | 19:03 |
| therve | It's *not* like we can flip a switch | 19:03 |
| SpamapS | But the status we show in resource-list ... is a lie. | 19:03 |
| SpamapS | as are many caches. :) | 19:03 |
| shardy | therve: Yeah, just be good to know that the functionality exists in a usable state in many/most of the projects | 19:03 |
| sdake | we may have to bootstrap some of the notification efforts in other projects if we really want to use that as part of our resource lifecycle | 19:04 |
| sdake | which is pretty straightforward, notifications are dead smiple to develop | 19:04 |
| SpamapS | imagine there are no notifications | 19:04 |
| therve | shardy, Oh yeah. I'm sure it can be half broken. It was in nova not long ago | 19:04 |
| *** julienvey has joined #heat | 19:04 | |
| SpamapS | how bad is it, really, to poll statuses of servers when you need them. When do you _actually_ need them? | 19:04 |
| sdake | the big problem I see with notifications is that theyare less reliable then polling given the half-hearted attempt at notifications most projects undertake ;) | 19:05 |
| radix | sorry got distracted | 19:05 |
| *** shakayumi has quit IRC | 19:05 | |
| radix | SpamapS: I want to make sure there's a design that's _possible_, or a way forward, that doesn't involve polling | 19:05 |
| shardy | radix: +1 | 19:06 |
| sdake | radix who is this cat http://summit.openstack.org/cfp/details/126 | 19:06 |
| sdake | I'm not familiar with him | 19:06 |
| SpamapS | radix: why? | 19:06 |
| radix | SpamapS: because it's unscalable? :) | 19:06 |
| SpamapS | what makes it unscalable? | 19:06 |
| radix | SpamapS: imagine a stack with 1000 nodes | 19:06 |
| sdake | polling is inherently less scalable | 19:07 |
| radix | that's at least 1000 calls to nova every time we need to check | 19:07 |
| shardy | SpamapS: part of the problem is the cost of all the API requests, e.g creating 100 empty stacks on my machine takes nearly a minute due to all the API overhead (keystone etc) | 19:07 |
| sdake | but it isn't unscalable | 19:07 |
| radix | if we can just *be notified* when a node goes down, we can double check that node's status and recreate it if necessary | 19:07 |
| SpamapS | If I have a cloud.. which can host 1000 nodes, I expect that the status for those nodes will be queried quite often. | 19:07 |
| therve | sdake: Were you in Portland? Ken was there. | 19:07 |
| *** lindsayk has joined #heat | 19:07 | |
| sdake | therve no | 19:08 |
| sdake | is he a rackspace guy? | 19:08 |
| radix | sdake: yes | 19:08 |
| sdake | engineer, manager ? | 19:08 |
| SpamapS | shardy: yes if I tried to host a cloud on a developer machine I would expect everything to serialize and basically suck ;) | 19:08 |
| radix | engineer manager :) | 19:08 |
| sdake | so both then radix ? :) | 19:08 |
| radix | sdake: yes | 19:08 |
| *** zns has quit IRC | 19:09 | |
| radix | SpamapS: rate limiting! | 19:09 |
| SpamapS | is a crime | 19:09 |
| therve | radix, Note that you'll have thousands of API calls when you create the stack anyway | 19:09 |
| radix | it's *really* annoying to try to check the status of 1000 nova nodes | 19:09 |
| SpamapS | radix: batching! | 19:09 |
| therve | So if you redo a check later, it's not more expensive than creating one | 19:09 |
| sdake | this looks interesting | 19:09 |
| radix | therve: I create 1000 node stacks way fewer than they would be checked | 19:09 |
| sdake | http://summit.openstack.org/cfp/details/1 | 19:10 |
| therve | radix, Getting the status of a server is also way cheaper than creating one :) | 19:10 |
| therve | So all in all... | 19:11 |
| radix | we can have a polling mechanism if you want | 19:11 |
| therve | At any rate, I agree with you that we should optimize our number of calls. Or the quality of them. But we need some | 19:12 |
| *** lindsayk1 has joined #heat | 19:12 | |
| *** lindsayk has quit IRC | 19:12 | |
| radix | but I hope that whatever mechanisms we have in heat will allow us to take advantage of notification if/when it becomes available to tell it which resources have died | 19:12 |
| therve | +1 | 19:12 |
| therve | People uses notifications for billing. They must work :) | 19:12 |
| radix | rackspace already has an event feed from its cloud servers that I'd like to take advantage of by writing a bit of glue to heat | 19:13 |
| radix | and avoid having to check all resources in a stack when some failure is detected | 19:13 |
| radix | and I *hope* that nova itself grows support for sending push-y notifications to something like a marconi queue | 19:14 |
| shardy | SpamapS: My point is the API overhead is significant, using lightweight amqp notifications and caching the status locally is likely to perform better, especially if you have lots of customers in a large cloud :) | 19:14 |
| radix | therve: yeah, though on the private RPC :P | 19:14 |
| therve | radix, I think we should use those for now | 19:14 |
| therve | If the interface becomes better, we'll switch | 19:14 |
| radix | well, someone can use them, I don't think we can :) | 19:15 |
| radix | RS heat deployment is outside of the nova datacenter | 19:15 |
| radix | (actually it's running *on* that cloud...) | 19:16 |
| therve | The key is probably to separate the resource creation from the status check | 19:16 |
| therve | Then you could be able to plug a different event feed | 19:16 |
| *** alexheneveld has quit IRC | 19:16 | |
| radix | by creation do you mean re-creation? | 19:16 |
| therve | No I mean creation in general | 19:17 |
| radix | well... what about the deletion of the resource? | 19:17 |
| radix | when it turns bad | 19:17 |
| therve | Yeah same thing | 19:17 |
| radix | it's not really identical to the initial creation | 19:17 |
| therve | It is in the sense that we do an action, and then wait for the result by doing get forever | 19:17 |
| *** openstackgerrit has quit IRC | 19:18 | |
| radix | therve: I think the proposed "stack-converge", if there's an alternative mechanism than the proposed "stack-align" | 19:18 |
| therve | I don't really think about convergence for now | 19:18 |
| *** openstackgerrit has joined #heat | 19:18 | |
| radix | er | 19:18 |
| radix | I meant to say I think stack-converge is good | 19:18 |
| therve | I'm mostly thinking about removing polling from Heat | 19:18 |
| radix | therve: which polling? | 19:18 |
| therve | radix, The one we do after each action | 19:18 |
| radix | ahhh | 19:19 |
| radix | I see | 19:19 |
| therve | eg nova.boot(). while x: server.get() | 19:19 |
| radix | yeah | 19:19 |
| therve | That introduces notifications gently into Heat :) | 19:19 |
| therve | Then maybe convergence and all can benefit from it | 19:19 |
| radix | I guess that'd be good | 19:20 |
| *** kgriffs_afk is now known as kgriffs | 19:21 | |
| *** kfox1111 has joined #heat | 19:21 | |
| kfox1111 | YAY. I just got a leter from legal saying I can contribute to OpenStack. :) | 19:21 |
| radix | kfox1111: hooraaaaaaaaaaaaay | 19:21 |
| kfox1111 | So, should I create blueprints for heat-templates for mongodb and elasticsearch? | 19:22 |
| radix | kfox1111: what, you mean just the example templates? | 19:22 |
| *** denis_makogon has quit IRC | 19:22 | |
| radix | kfox1111: I don't think heat-templates typically has blueprints | 19:22 |
| kfox1111 | so just a review then? | 19:22 |
| *** dmakogon_ has joined #heat | 19:23 | |
| radix | you can just submit them for review in gerrit | 19:23 |
| radix | yeah | 19:23 |
| kfox1111 | k. thanks. :) | 19:23 |
| sdake | kfox1111 HELL FROZE OVER GRATZ! :) | 19:23 |
| kfox1111 | :) | 19:23 |
| sdake | kfox1111 b lueprints for heat-templates aren't really required | 19:23 |
| sdake | the requirements around that repo are pretty informal | 19:23 |
| radix | therve: so I guess there needs to be a way to push a "this resource just transitioned to status X" message to heat | 19:23 |
| sdake | for example, we have a noop check/gate | 19:23 |
| radix | therve: those messages could be subscribed to by things like resource-waiting-for-creation, and the states could also be written to the DB for stack-converge to check later | 19:24 |
| therve | radix, Yep. In an internal API though. | 19:24 |
| radix | therve: well, I think it should be external too :) | 19:24 |
| radix | well... hmm | 19:24 |
| radix | maybe internal is ok regardless | 19:25 |
| radix | so I guess we don't have to argue about that :) | 19:25 |
| *** duncanjw has quit IRC | 19:27 | |
| *** alexheneveld has joined #heat | 19:27 | |
| *** tango has joined #heat | 19:27 | |
| radix | ok, I'm going to grab some lunch | 19:29 |
| radix | bbl | 19:29 |
| *** arbylee has quit IRC | 19:34 | |
| *** arbylee has joined #heat | 19:36 | |
| *** shakayumi has joined #heat | 19:45 | |
| *** shakayumi has quit IRC | 19:45 | |
| *** kfox1111 is now known as Guest59946 | 19:48 | |
| *** kfox1111 has joined #heat | 19:48 | |
| *** kfox11111 has joined #heat | 19:48 | |
| kfox1111 | "Caution: Do not check in changes on your master branch. Doing so will cause merge commits when you pull new upstream changes, and merge commits will not be accepted by Gerrit." | 19:48 |
| kfox1111 | Does that mean you should branch before committing any changes? | 19:48 |
| kfox1111 | does the branch name matter and do you need one per new review, or only per feature? | 19:49 |
| *** IlyaE has quit IRC | 19:49 | |
| *** andersonvom has quit IRC | 19:50 | |
| *** kfox11111 has quit IRC | 19:51 | |
| *** shakayumi has joined #heat | 19:54 | |
| openstackgerrit | Kevin Fox proposed a change to openstack/heat-templates: Add a scalable ElasticSearch cluster set of templates in Hot format. https://review.openstack.org/82903 | 19:55 |
| *** ramishra has joined #heat | 19:56 | |
| *** shakamunyi has joined #heat | 19:56 | |
| *** shakayumi has quit IRC | 19:58 | |
| *** duncanjw has joined #heat | 20:00 | |
| *** rbuilta has quit IRC | 20:02 | |
| *** shakamunyi has quit IRC | 20:03 | |
| *** derekh has quit IRC | 20:05 | |
| *** jistr has quit IRC | 20:06 | |
| *** shakamunyi has joined #heat | 20:06 | |
| *** IlyaE has joined #heat | 20:07 | |
| *** e0ne has quit IRC | 20:15 | |
| *** jprovazn has quit IRC | 20:15 | |
| *** e0ne has joined #heat | 20:16 | |
| *** e0ne has quit IRC | 20:20 | |
| *** BillArnold has joined #heat | 20:22 | |
| *** mkollaro has quit IRC | 20:23 | |
| *** alexheneveld has quit IRC | 20:24 | |
| shardy | kfox1111: You should branch per topic (e.g per bug or per blueprint) | 20:26 |
| shardy | kfox1111: e.g I do git checkout -b bug/123456 or git checkout -b bp/my-new-feature | 20:26 |
| shardy | then commit your changes, rebase as needed using rebase -i origin, and then submit the series via git review | 20:27 |
| shardy | note you should do git review -s before the first commit so the hook which adds the ChangeId is setup | 20:27 |
| therve | stevebaker, I'd like possible to have the constraint fixes in icehouse | 20:28 |
| stevebaker | therve: OK, feel free to add the milestone back, then ping individuals for reviews | 20:29 |
| therve | stevebaker, How does it work? Can gerrit still merge in master? | 20:29 |
| stevebaker | therve: there won't be an rc1 release until bugs go to zero, either by fixing in master or deferring | 20:30 |
| stevebaker | I'm trying to be brutal about deferring because we're quite behind http://old-wiki.openstack.org/rc/ | 20:30 |
| therve | OK no problem. Those bugs are not critical, we can backport them later if deemed useful. | 20:31 |
| *** mkollaro has joined #heat | 20:32 | |
| stevebaker | also I expect there will be an rc2 | 20:34 |
| kfox1111 | shardy: ok, thanks. :) | 20:34 |
| *** vijendar has quit IRC | 20:37 | |
| *** vijendar has joined #heat | 20:37 | |
| *** vijendar has quit IRC | 20:39 | |
| *** insanidade has quit IRC | 20:40 | |
| sdake_ | stevebaker I think that idea of not validating the templates for all operations but update and create will be pretty invasive | 20:46 |
| sdake_ | we don't actually validate, the parser plugin does that work | 20:46 |
| openstackgerrit | Rabi Mishra proposed a change to openstack/heat: Provides ability to reference existing neutron resources https://review.openstack.org/82853 | 20:47 |
| sdake_ | so we need to either hack the plugin or make a new one | 20:47 |
| *** ramishra has quit IRC | 20:47 | |
| *** ramishra has joined #heat | 20:48 | |
| stevebaker | sdake_: another option is reverting that change until we come up with a migration strategy | 20:49 |
| sdake_ | you mean reverting the pluggable parsers? | 20:50 |
| stevebaker | sdake_: revert the specific validation which is causing existing stacks to fail validation (checking template version?) | 20:51 |
| zaneb | what bug are we talking about? | 20:52 |
| *** ramishra has quit IRC | 20:52 | |
| stevebaker | https://bugs.launchpad.net/heat/+bug/1296859 | 20:52 |
| uvirtbot | Launchpad bug 1296859 in heat "stack-list throws an error if any stack returned contains a template error" [Medium,Triaged] | 20:52 |
| stevebaker | ERROR: "2014-03-07" is not a valid heat_template_version. Should be one of: ('2013-05-23',) | 20:52 |
| sdake_ | it isn't just checking tmeplate version | 20:52 |
| sdake_ | but that is easy enough to hack around | 20:53 |
| sdake_ | the entire parser is different | 20:53 |
| zaneb | workaround: don't upload broken templates ;) | 20:53 |
| sdake_ | zaneb the templates are already in the db | 20:53 |
| sdake_ | and they can't be shown/deleted/worked on | 20:53 |
| stevebaker | zaneb: these templates were already in before that validation was added | 20:53 |
| zaneb | yeah, it's a bummer | 20:54 |
| zaneb | we could add a DB migration to fix 'em? | 20:54 |
| sdake_ | it makes heat unworkable - the templates cant even be deleted | 20:54 |
| sdake_ | rather the stacks | 20:54 |
| zaneb | like we did with all of the recent HOT changes | 20:54 |
| stevebaker | I thought stack-list avoided parsing each stack anyway, for performance | 20:54 |
| sdake_ | stevebaker it calls stack.load, which calls template.lod which calls the pluggable parser parsing code | 20:55 |
| therve | stevebaker, It doesn't gather outpus | 20:55 |
| therve | outputs | 20:55 |
| sdake_ | one option is to make pluggable parsers in the mainline which are less strict | 20:56 |
| sdake_ | and only call those for cases not in update/create | 20:56 |
| sdake_ | this is what i'm working on now | 20:56 |
| sdake_ | but it won't be tidy | 20:56 |
| zaneb | sdake_: https://review.openstack.org/#/c/73580/7/heat/db/sqlalchemy/migrate_repo/versions/041_migrate_hot_template_resources.py <- something like this | 20:57 |
| *** IlyaE has quit IRC | 21:01 | |
| sdake_ | zaneb you think doing that for just the version field would cover all the cases of mixed hot+cfn use since you already have that migration? | 21:01 |
| *** radez is now known as radez_g0n3 | 21:01 | |
| sdake_ | zaneb it looks like that migration doesn't handle parameters or intrinsics | 21:02 |
| sdake_ | I don't understand the details of the parser enough to know if that matters or not | 21:03 |
| zaneb | afaik you'd just have to go through and wherever you see hot_template_version change it to 2013-05-23, wherever you see HeatTemplateFormatVersion change it to 2012-12-12 and wherever you see AWSTemplateFormatVersion change is to 2012-09-09 | 21:04 |
| sdake_ | cool | 21:04 |
| sdake_ | well that is nice and tidy | 21:04 |
| zaneb | for that bug, anyway | 21:05 |
| sdake_ | i'm wondering if there are other cases outside of the heta version fiel dthough | 21:05 |
| zaneb | we haven't yet removed the old intrinsic functions from HOT | 21:05 |
| zaneb | although tspatzier has suggested it, which means we'd need another migration | 21:05 |
| sdake_ | ok so intrinsics aren't a problem | 21:05 |
| sdake_ | what about parameters? | 21:06 |
| zaneb | yet ;) | 21:06 |
| zaneb | parameters were the first thing implemented in HOT, so I don't think they've changed | 21:06 |
| sdake_ | could folks mix hot and cfn parameter syntax? | 21:06 |
| sdake_ | prior to your rework of the pluggable parsers? | 21:06 |
| zaneb | the pluggable parser stuff didn't really touch that, but tspatzier's work during Icehouse to replace the parameters parsing definitely did | 21:08 |
| *** pafuent has quit IRC | 21:08 | |
| zaneb | but I'm not sure if it allowed a mix beforehand or not | 21:08 |
| zaneb | you might need to check with him | 21:08 |
| sdake_ | i'll test it out on havana | 21:08 |
| sdake_ | so sounds like we may have two bugs | 21:08 |
| sdake_ | and then all the possible failure scenarios are fixed | 21:08 |
| *** shakayumi has joined #heat | 21:08 | |
| zaneb | actually, I think it did allow a mix, because I remember fixing one of the OpenShift templates that had a mix | 21:09 |
| *** blomquisg has quit IRC | 21:09 | |
| *** shakayumi has quit IRC | 21:09 | |
| sdake_ | so assuming I add migrations for those two things, are there more things we need to address? | 21:09 |
| *** alexheneveld has joined #heat | 21:10 | |
| *** shakamunyi has quit IRC | 21:11 | |
| sdake_ | are we allowed to change existing migration patches? One scenario with version is if we come out with a new hot version, and I simply replace every version with 2013-05-23, I dont want to break new versions | 21:12 |
| *** aweiteka has quit IRC | 21:12 | |
| sdake_ | so theoretical migration patch would need to whitelist the new versions as they come into the source base | 21:13 |
| sdake_ | zaneb eg assume that the new versions are good if they are in the template for future migrations | 21:14 |
| zaneb | "are we allowed to change existing migration patches" <- no. | 21:14 |
| sdake_ | so when that migration runs in the future, it would change to the old version of hot if there was a new template version | 21:15 |
| *** ifarkas has quit IRC | 21:15 | |
| zaneb | although... adding to a whitelist might not be the worst thing | 21:15 |
| sdake_ | ya seems like a reasonable exception to me | 21:16 |
| *** ifarkas_ has quit IRC | 21:16 | |
| zaneb | randallburt was talking at one point about having a script that ran on demand, rather than a DB migration | 21:16 |
| zaneb | jasond: any idea what the outcome of that ^ was? | 21:16 |
| sdake_ | stevebaker given that migrations can't be backported, i think we will need to fix both of these before release | 21:19 |
| sdake_ | zaneb since this is our last chance to fix thee, can you think of other scenarios besides version and parameters which should be migrated? | 21:19 |
| zaneb | sdake_: not that I can think of, but ping tspatzier with an email. He has done most of the work in this area | 21:21 |
| sdake_ | zaneb will do thanks | 21:21 |
| *** blomquisg has joined #heat | 21:23 | |
| *** lindsayk1 has quit IRC | 21:25 | |
| sdake_ | stevebaker I'll have migration patches ready for review tomorrow | 21:26 |
| sdake_ | nearing my end of day here and not sure i can get er done today | 21:26 |
| *** asalkeld has quit IRC | 21:28 | |
| stevebaker | sdake_: ok, thanks. I guess migrating to an arbitrary version is better than nothing | 21:29 |
| *** achampion has quit IRC | 21:29 | |
| *** julienvey has quit IRC | 21:30 | |
| zaneb | sdake: rather than whitelisting versions, you could just modify all versions that look like a date and are before the scheduled Icehouse release | 21:31 |
| zaneb | on the theory that if anyone genuinely creates a new version, it will be after then | 21:31 |
| *** gokrokve has quit IRC | 21:32 | |
| *** andersonvom has joined #heat | 21:32 | |
| radix | I wonder if it's worth having a cross-team session about notifications | 21:37 |
| *** rpothier_ has quit IRC | 21:40 | |
| *** andersonvom has quit IRC | 21:40 | |
| sdake_ | stevebaker we will need a general exception to be able to change migrations in the future to add to the whitelist of versions *not* to migrate | 21:40 |
| sdake_ | the migration in quesiton i mean | 21:40 |
| sdake_ | radix i htink that would be great! | 21:40 |
| zaneb | radix: I think randallburt proposed one in the Heat track already | 21:41 |
| zaneb | if he hadn't already, I was going to ;) | 21:41 |
| *** IlyaE has joined #heat | 21:43 | |
| *** alexpilotti has joined #heat | 21:44 | |
| openstackgerrit | A change was merged to openstack/heat: Add subnets as a dependency for router https://review.openstack.org/82452 | 21:44 |
| sdake_ | nova sort of standardized em | 21:45 |
| sdake_ | but not in a real consstent way with other projects | 21:45 |
| sdake_ | zaneb some people have put the current date in the template version | 21:45 |
| sdake_ | in fact the bug does that :( | 21:45 |
| sdake_ | zaneb actually that is a really good idea | 21:45 |
| sdake_ | zaneb sorry didn't see your original msg inbetween | 21:45 |
| radix | just filed http://summit.openstack.org/cfp/details/144 (convergence) | 21:45 |
| radix | zaneb: therve proposed one about heat | 21:45 |
| radix | but I think he's just talking about using existing nova RPC notifications | 21:46 |
| *** lekha has quit IRC | 21:46 | |
| zaneb | radix: I was referring to http://summit.openstack.org/cfp/details/86 | 21:46 |
| sdake_ | bitcoin miners still making 100 a day | 21:46 |
| * sdake_ yays | 21:46 | |
| radix | zaneb: oh | 21:47 |
| *** blomquisg has quit IRC | 21:47 | |
| *** russellb has quit IRC | 21:47 | |
| radix | I think we are talking about different things :) | 21:47 |
| zaneb | perhaps | 21:47 |
| zaneb | ;) | 21:47 |
| *** zigo has quit IRC | 21:47 | |
| *** wendar has quit IRC | 21:47 | |
| radix | I didn't really give much context, that's my fault | 21:47 |
| zaneb | radix: what did *you* mean by 'notifications'? | 21:47 |
| *** lekha has joined #heat | 21:48 | |
| *** lindsayk has joined #heat | 21:48 | |
| radix | I'm talking about stuff like nova sending notifications about lifecycle events for its resources to other services | 21:48 |
| radix | over something a bit more accessible (dare I say "public") than the existing RPC thing | 21:48 |
| *** russellb has joined #heat | 21:48 | |
| radix | e.g. marconi | 21:48 |
| sdake_ | zaneb i parsed radix's suggetstion as getting the various projects in the room together to standardize the notifications across project | 21:48 |
| radix | granted marconi isn't even a part of openstack so it may be early times | 21:48 |
| radix | Too Soon | 21:48 |
| sdake_ | eg - make a notification specification | 21:49 |
| radix | right | 21:49 |
| *** wendar has joined #heat | 21:49 | |
| *** zigo has joined #heat | 21:49 | |
| *** lindsayk has quit IRC | 21:57 | |
| *** jmckind has quit IRC | 22:00 | |
| *** blomquisg has joined #heat | 22:00 | |
| *** lindsayk has joined #heat | 22:01 | |
| *** mwheckmann has quit IRC | 22:02 | |
| openstackgerrit | A change was merged to openstack/python-heatclient: get_file: do not read same url once again https://review.openstack.org/80882 | 22:04 |
| openstackgerrit | A change was merged to openstack/python-heatclient: get_file: encode non utf-8 encoding files via base64 https://review.openstack.org/76443 | 22:04 |
| openstackgerrit | A change was merged to openstack/heat: Revert "Add validation to KeyPair resource" https://review.openstack.org/82862 | 22:04 |
| *** sgordon has quit IRC | 22:04 | |
| *** lindsayk has quit IRC | 22:04 | |
| sdake_ | anyone know when you write migrations if you typically write test cases for them? | 22:06 |
| *** BillArnold has quit IRC | 22:06 | |
| stevebaker | sdake_: we have been doing that recently, yes. but its usually quite superficial; does the column exist... | 22:07 |
| stevebaker | Any reviews for this? https://review.openstack.org/#/c/82394/ | 22:07 |
| *** lindsayk has joined #heat | 22:08 | |
| sdake_ | stevebaker looks good | 22:10 |
| *** jcru has quit IRC | 22:15 | |
| openstackgerrit | Steve Baker proposed a change to openstack/heat: Make OS::Nova::Server networks property updatable https://review.openstack.org/74299 | 22:18 |
| *** dims_ has quit IRC | 22:18 | |
| *** yogesh_ has quit IRC | 22:18 | |
| *** blomquisg has quit IRC | 22:18 | |
| stevebaker | rebased the test ^ feel free to review | 22:18 |
| *** zns has joined #heat | 22:19 | |
| stevebaker | reviews for this please? https://review.openstack.org/#/c/81558/ | 22:19 |
| *** IlyaE has quit IRC | 22:24 | |
| *** zns has quit IRC | 22:25 | |
| *** zns has joined #heat | 22:25 | |
| *** arbylee has quit IRC | 22:27 | |
| *** varora has joined #heat | 22:28 | |
| *** blomquisg has joined #heat | 22:31 | |
| *** dims_ has joined #heat | 22:34 | |
| *** lindsayk has quit IRC | 22:38 | |
| *** lindsayk has joined #heat | 22:39 | |
| *** arbylee has joined #heat | 22:39 | |
| *** arbylee has quit IRC | 22:43 | |
| *** jergerber has quit IRC | 22:44 | |
| *** rcleere has quit IRC | 22:46 | |
| *** achampion has joined #heat | 22:46 | |
| *** jpeeler has quit IRC | 22:54 | |
| openstackgerrit | A change was merged to openstack/heat: Don't catch all exceptions in image constraint https://review.openstack.org/82846 | 23:02 |
| *** david-lyle has quit IRC | 23:08 | |
| *** faramir1 has joined #heat | 23:14 | |
| *** rpothier_ has joined #heat | 23:25 | |
| *** giulivo has quit IRC | 23:28 | |
| *** mkollaro has quit IRC | 23:30 | |
| *** mkollaro has joined #heat | 23:31 | |
| *** tango has quit IRC | 23:47 | |
| *** blomquisg has quit IRC | 23:47 | |
| *** faramir1 has quit IRC | 23:48 | |
| *** cfriesen has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!