| *** blomquisg has quit IRC | 00:07 | |
| *** gokrokve has joined #heat | 00:12 | |
| *** gokrokve has quit IRC | 00:17 | |
| *** blomquisg has joined #heat | 00:21 | |
| *** matsuhashi has joined #heat | 00:23 | |
| *** e0ne_ has quit IRC | 00:31 | |
| *** matsuhashi has quit IRC | 00:33 | |
| *** matsuhashi has joined #heat | 00:34 | |
| *** matsuhashi has quit IRC | 00:38 | |
| *** rpothier_ has quit IRC | 00:42 | |
| *** rpothier__ has joined #heat | 00:42 | |
| *** fandi has quit IRC | 00:59 | |
| *** asalkeld has joined #heat | 01:01 | |
| *** matsuhashi has joined #heat | 01:04 | |
| *** nosnos has joined #heat | 01:05 | |
| *** gokrokve has joined #heat | 01:13 | |
| *** gokrokve has quit IRC | 01:17 | |
| *** faramir has joined #heat | 01:23 | |
| *** faramir has quit IRC | 01:27 | |
| *** fandi has joined #heat | 01:27 | |
| *** rpothier__ has quit IRC | 01:34 | |
| *** gokrokve has joined #heat | 01:40 | |
| *** e0ne has joined #heat | 01:42 | |
| *** gokrokve has quit IRC | 01:45 | |
| *** e0ne has quit IRC | 01:47 | |
| *** blomquisg has quit IRC | 01:48 | |
| *** ramishra has joined #heat | 01:55 | |
| *** liang has joined #heat | 01:56 | |
| *** ramishra has joined #heat | 01:56 | |
| *** blomquisg has joined #heat | 02:02 | |
| *** agileclipse has joined #heat | 02:12 | |
| *** fandi has quit IRC | 02:15 | |
| *** matsuhashi has quit IRC | 02:18 | |
| *** erkules_ has joined #heat | 02:19 | |
| *** erkules has quit IRC | 02:21 | |
| *** sgordon has joined #heat | 02:24 | |
| *** sgordon has quit IRC | 02:24 | |
| *** sgordon has joined #heat | 02:24 | |
| *** matsuhashi has joined #heat | 02:26 | |
| *** gokrokve has joined #heat | 02:28 | |
| *** gokrokve_ has joined #heat | 02:30 | |
| *** gokrokve has quit IRC | 02:33 | |
| openstackgerrit | Jun Jie Nan proposed a change to openstack/python-heatclient: Add support for software config resources https://review.openstack.org/58885 | 02:34 |
|---|---|---|
| *** fandi has joined #heat | 02:35 | |
| *** gokrokve_ has quit IRC | 02:35 | |
| *** hndak has joined #heat | 02:49 | |
| *** hndak has quit IRC | 02:51 | |
| *** kfox1111 has joined #heat | 03:11 | |
| *** erkules_ is now known as erkules | 03:12 | |
| *** fandi has quit IRC | 03:13 | |
| *** kfox1111 has quit IRC | 03:27 | |
| *** gokrokve has joined #heat | 03:28 | |
| *** fandi has joined #heat | 03:34 | |
| *** matsuhashi has quit IRC | 03:48 | |
| *** nkhare has joined #heat | 04:01 | |
| *** matsuhashi has joined #heat | 04:32 | |
| *** Qiming has joined #heat | 04:39 | |
| *** gokrokve has quit IRC | 04:41 | |
| *** gokrokve has joined #heat | 04:41 | |
| openstackgerrit | A change was merged to openstack/heat: Rename Openstack to OpenStack https://review.openstack.org/73174 | 04:43 |
| *** sgordon has quit IRC | 04:44 | |
| *** gokrokve_ has joined #heat | 04:44 | |
| *** edmund has quit IRC | 04:45 | |
| *** gokrokve has quit IRC | 04:46 | |
| *** ramishra has quit IRC | 04:53 | |
| *** rpothier has joined #heat | 04:55 | |
| *** agileclipse has quit IRC | 04:58 | |
| *** chandan_kumar has joined #heat | 05:02 | |
| *** ramishra has joined #heat | 05:04 | |
| *** david-lyle has joined #heat | 05:09 | |
| *** agileclipse has joined #heat | 05:15 | |
| *** gokrokve_ has quit IRC | 05:19 | |
| openstackgerrit | A change was merged to openstack/heat: Don't try to stop a stack on a dead engine https://review.openstack.org/72739 | 05:24 |
| *** gokrokve has joined #heat | 05:27 | |
| *** killer_prince has quit IRC | 05:35 | |
| *** tspatzier has joined #heat | 05:39 | |
| *** killer_prince has joined #heat | 05:39 | |
| openstackgerrit | Jun Jie Nan proposed a change to openstack/python-heatclient: Replace try...except with assertRaises https://review.openstack.org/73916 | 05:50 |
| *** tspatzier has quit IRC | 05:53 | |
| openstackgerrit | A change was merged to openstack/heat: Updates template_validate call to validate parameter_groups. https://review.openstack.org/61601 | 06:02 |
| *** ramishra has quit IRC | 06:06 | |
| openstackgerrit | Jenkins proposed a change to openstack/heat: Imported Translations from Transifex https://review.openstack.org/72566 | 06:08 |
| *** cmyster has joined #heat | 06:09 | |
| *** cmyster has joined #heat | 06:09 | |
| *** saju_m has joined #heat | 06:15 | |
| *** topol has quit IRC | 06:17 | |
| skraynev | morning ;) | 06:23 |
| *** akuznetsov has quit IRC | 06:26 | |
| skraynev | zaneb, are you around? | 06:27 |
| *** akuznetsov has joined #heat | 06:29 | |
| *** zpatten has quit IRC | 06:35 | |
| cmyster | morning | 06:39 |
| *** nosnos has quit IRC | 06:42 | |
| *** tspatzier has joined #heat | 06:43 | |
| openstackgerrit | Yongli He proposed a change to openstack/heat: Use six.moves cStringIO instead of cStringIO https://review.openstack.org/73940 | 06:51 |
| *** nosnos has joined #heat | 06:52 | |
| *** cmyster has quit IRC | 06:55 | |
| openstackgerrit | Lee Li proposed a change to openstack/heat: Remove dependent module py3kcompat https://review.openstack.org/73765 | 06:57 |
| *** cmyster has joined #heat | 07:09 | |
| *** cmyster has joined #heat | 07:09 | |
| *** zpatten has joined #heat | 07:13 | |
| openstackgerrit | Yongli He proposed a change to openstack/heat: Use six.moves cStringIO instead of cStringIO https://review.openstack.org/73940 | 07:14 |
| *** asalkeld has quit IRC | 07:17 | |
| *** TonyBurn_ has joined #heat | 07:28 | |
| openstackgerrit | Ton Ngo proposed a change to openstack/heat: Add Heat engine support for troubleshooting failed stack https://review.openstack.org/73952 | 07:29 |
| *** ifarkas has joined #heat | 07:30 | |
| *** TonyBurn_ has quit IRC | 07:31 | |
| *** TonyBurn_ has joined #heat | 07:31 | |
| *** rmiki has joined #heat | 07:32 | |
| *** zpatten has quit IRC | 07:40 | |
| *** jprovazn has joined #heat | 07:47 | |
| *** akuznetsov has quit IRC | 07:48 | |
| *** akuznetsov has joined #heat | 07:51 | |
| *** bvandenh has joined #heat | 08:03 | |
| *** asalkeld has joined #heat | 08:07 | |
| therve | Good morning! | 08:17 |
| skraynev | hi therve ;) | 08:17 |
| openstackgerrit | Ton Ngo proposed a change to openstack/heat: Add API suppport for troubleshooting failed stack https://review.openstack.org/71367 | 08:18 |
| *** zpatten has joined #heat | 08:19 | |
| *** pasquier-s has joined #heat | 08:22 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Implement glance image constraint https://review.openstack.org/72919 | 08:23 |
| *** pshchelo has joined #heat | 08:26 | |
| *** pshchelo has quit IRC | 08:27 | |
| *** pshchelo has joined #heat | 08:27 | |
| pshchelo | morning :) | 08:27 |
| cmyster | morning | 08:30 |
| *** saju_m has quit IRC | 08:31 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/heat: Don't disable SELinux in cloud-init's boothook.sh https://review.openstack.org/73669 | 08:41 |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Implement glance image constraint https://review.openstack.org/72919 | 08:42 |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Enable usage of custom constraint in parameters https://review.openstack.org/71295 | 08:42 |
| *** gokrokve has quit IRC | 08:45 | |
| *** gokrokve has joined #heat | 08:46 | |
| *** gokrokve has quit IRC | 08:50 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Remove empty unit test directory https://review.openstack.org/73972 | 08:51 |
| *** ppetit has joined #heat | 08:52 | |
| *** stannie has joined #heat | 08:55 | |
| *** saju_m has joined #heat | 08:58 | |
| *** jistr has joined #heat | 09:01 | |
| *** e0ne has joined #heat | 09:02 | |
| shardy | morning all | 09:04 |
| skraynev | shardy: hello | 09:06 |
| *** chandan_kumar has quit IRC | 09:10 | |
| skraynev | shardy: what do you think about https://review.openstack.org/#/c/72454/ ? I was confused that static method was used instead valid. Unfortunately I have not caught stevebaker for asking his about it. | 09:10 |
| *** e0ne has quit IRC | 09:11 | |
| shardy | skraynev: I'm not sure I understand your concern, can you explain, or add comments to the review please? | 09:13 |
| *** e0ne has joined #heat | 09:13 | |
| *** derekh has joined #heat | 09:14 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/heat: Add neutron network constraint https://review.openstack.org/73976 | 09:14 |
| *** alexheneveld has joined #heat | 09:15 | |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: Revert config mocking to cfg.CONF.set_override https://review.openstack.org/73977 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: Modify stack_user_domain config option to take domain ID https://review.openstack.org/73978 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: migrate User/AccessKey resources to StackUser base class https://review.openstack.org/72763 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: StackUser add _delete_keypair function https://review.openstack.org/72762 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: engine: allow stack_user_project users to retrieve stack https://review.openstack.org/71300 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: Add test for StackUser._create_keypair https://review.openstack.org/72761 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: StackUser add suspend/resume support https://review.openstack.org/71930 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: migrate StackUser base class to stack domain users https://review.openstack.org/71210 | 09:15 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: heat_keystoneclient add delete_stack_domain_user_keypair https://review.openstack.org/71929 | 09:15 |
| *** gokrokve has joined #heat | 09:16 | |
| *** gokrokve_ has joined #heat | 09:18 | |
| *** chandan_kumar has joined #heat | 09:20 | |
| *** gokrokve has quit IRC | 09:21 | |
| *** gokrokve_ has quit IRC | 09:23 | |
| *** tomek_adamczewsk has joined #heat | 09:24 | |
| therve | shardy, FWIW https://github.com/therve/heat/tree/as-spike | 09:29 |
| *** zenoway has joined #heat | 09:33 | |
| skraynev | shardy: done | 09:35 |
| shardy | therve: Are you planning to post that in time for FPF? | 09:47 |
| shardy | therve: Ideally, I think we need to remove the dependency on ec2tokens signed requests for the native resources | 09:49 |
| shardy | use your native signal API, and make python-heatclient support sending signals given a username/password from the SignalResponder via instance metadata | 09:49 |
| therve | shardy, I do. And yes, I guess that means changing SignalResponder? | 09:49 |
| shardy | therve: The changes to SignalResponder are there, from the instance-users work | 09:50 |
| shardy | the thing subclassing just needs to set a random password | 09:50 |
| therve | Hum.. | 09:51 |
| therve | shardy, "There" you mean in master? | 09:51 |
| shardy | therve: Sorry, no in the review queue: | 09:51 |
| shardy | https://review.openstack.org/#/c/71210/8/heat/engine/stack_user.py | 09:51 |
| shardy | therve: I need to sync up with stevebaker, as I believe he was looking at using that for the software-config work | 09:53 |
| *** Qiming has quit IRC | 09:53 | |
| shardy | but the last time we spoke I believe the agreed approach was for signalling to happen via python-heatclient, using username/password to request a token from keystone | 09:53 |
| shardy | the only disadvantage is instances need to be able to see keystone | 09:54 |
| *** matsuhashi has quit IRC | 09:54 | |
| shardy | but overall it seems like a reasonable approach which removes the ec2 requirement | 09:54 |
| shardy | and avoid inventing our own auth scheme | 09:54 |
| shardy | *avoids* | 09:54 |
| * therve nods | 09:55 | |
| therve | I can work on the client change then | 09:55 |
| *** samkottler has quit IRC | 09:55 | |
| *** rmiki has quit IRC | 09:56 | |
| therve | shardy, You need something to get a URL? | 09:57 |
| *** lazy_prince has joined #heat | 09:57 | |
| therve | Although I guess it's simply the ARN of the resource | 09:57 |
| *** killer_prince has quit IRC | 09:58 | |
| *** lazy_prince is now known as killer_prince | 09:58 | |
| shardy | therve: Yup probably, we just need to provide a resource identifier for the client | 09:58 |
| shardy | and the client needs to be able to parse a username/password optionally from the nova openstack meta_data.json | 09:58 |
| shardy | then we can pass in the credentials when creating the instance | 09:58 |
| shardy | I'm going to look at adding a property to OS::Nova::Server which allows you do do that | 09:59 |
| shardy | I guess we may also want a property which allows you to pass in a username/password shared between a group of instances, e.g from the ScalingPolicy | 09:59 |
| shardy | and the scaling policy needs to expose the credentials as attributes | 09:59 |
| shardy | sigh. This dev cycle seems to have gone very quickly :( | 10:00 |
| therve | You don't say | 10:02 |
| *** matsuhashi has joined #heat | 10:02 | |
| *** aignatov is now known as aignatov_ | 10:02 | |
| *** liang has quit IRC | 10:03 | |
| *** samkottler has joined #heat | 10:03 | |
| *** agileclipse has quit IRC | 10:05 | |
| *** fandi has quit IRC | 10:05 | |
| shardy | therve: http://paste.openstack.org/show/66414/ | 10:07 |
| shardy | Just to clarify - that's what I was thinking, and we could do something similar for native wait condition signalling | 10:08 |
| therve | Yeah | 10:08 |
| shardy | just use your native signal API and some common client code for both | 10:08 |
| therve | Some people are going to complain about metadata security | 10:08 |
| shardy | well how is userdata any more secure? | 10:08 |
| *** akuznetsov has quit IRC | 10:08 | |
| *** akuznetsov has joined #heat | 10:09 | |
| therve | Oh it's not | 10:10 |
| therve | Wait you used metadata | 10:10 |
| therve | I don't remember what it does :) | 10:10 |
| shardy | http://docs.openstack.org/grizzly/openstack-compute/admin/content//metadata-service.html | 10:11 |
| therve | Oh right | 10:13 |
| shardy | I was planning to work on a native waitcondition patch today, so so if you can do the client part we might be able to get something working :) | 10:14 |
| *** matsuhashi has quit IRC | 10:17 | |
| *** matsuhashi has joined #heat | 10:17 | |
| *** gokrokve has joined #heat | 10:18 | |
| *** matsuhashi has quit IRC | 10:22 | |
| *** gokrokve has quit IRC | 10:23 | |
| *** mkollaro has joined #heat | 10:28 | |
| *** e0ne_ has joined #heat | 10:29 | |
| openstackgerrit | A change was merged to openstack/heat: Use integer when appropriate in autoscaling schema https://review.openstack.org/72944 | 10:29 |
| openstackgerrit | Thomas Herve proposed a change to openstack/python-heatclient: Support the native signal API https://review.openstack.org/73996 | 10:31 |
| *** e0ne has quit IRC | 10:33 | |
| *** stannie has left #heat | 10:35 | |
| *** aignatov_ is now known as aignatov | 10:37 | |
| *** ramishra has joined #heat | 10:38 | |
| *** denis_makogon has joined #heat | 10:43 | |
| *** jamieh has joined #heat | 10:44 | |
| *** matsuhashi has joined #heat | 10:47 | |
| *** ramishra has quit IRC | 11:08 | |
| *** ramishra has joined #heat | 11:09 | |
| *** gokrokve has joined #heat | 11:17 | |
| shardy | skraynev: 72454 update, you are right IMO | 11:22 |
| *** gokrokve has quit IRC | 11:22 | |
| skraynev | shardy: thanks ;) | 11:23 |
| openstackgerrit | A change was merged to openstack/heat: Enable usage of custom constraint in parameters https://review.openstack.org/71295 | 11:29 |
| *** nkhare has quit IRC | 11:32 | |
| openstackgerrit | A change was merged to openstack/heat: Revert config mocking to cfg.CONF.set_override https://review.openstack.org/73977 | 11:33 |
| *** e0ne has joined #heat | 11:37 | |
| *** matsuhashi has quit IRC | 11:37 | |
| *** aignatov is now known as aignatov_ | 11:40 | |
| *** e0ne_ has quit IRC | 11:41 | |
| *** blomquisg has quit IRC | 11:43 | |
| *** aignatov_ is now known as aignatov | 11:50 | |
| *** aignatov is now known as aignatov_ | 11:51 | |
| *** zenoway has quit IRC | 11:53 | |
| *** aignatov_ is now known as aignatov | 11:57 | |
| *** blomquisg has joined #heat | 11:57 | |
| *** killer_prince has quit IRC | 12:01 | |
| *** bvandenh has quit IRC | 12:01 | |
| *** zhiyan_ is now known as zhiyan | 12:04 | |
| *** cmyster has quit IRC | 12:07 | |
| *** alexpilotti has quit IRC | 12:13 | |
| *** rpothier has quit IRC | 12:13 | |
| *** cmyster has joined #heat | 12:14 | |
| *** cmyster has joined #heat | 12:14 | |
| *** topol has joined #heat | 12:16 | |
| *** gokrokve has joined #heat | 12:17 | |
| *** e0ne has quit IRC | 12:18 | |
| *** e0ne_ has joined #heat | 12:18 | |
| openstackgerrit | A change was merged to openstack/heat: Use six.moves cStringIO instead of cStringIO https://review.openstack.org/73940 | 12:20 |
| *** gokrokve has quit IRC | 12:22 | |
| ramishra | shardy: Hi Steve | 12:23 |
| *** vvechkanov1 has joined #heat | 12:25 | |
| openstackgerrit | Thomas Herve proposed a change to openstack/python-heatclient: Support the native signal API https://review.openstack.org/73996 | 12:26 |
| *** e0ne_ has quit IRC | 12:26 | |
| *** e0ne has joined #heat | 12:26 | |
| *** tspatzier has quit IRC | 12:26 | |
| *** e0ne_ has joined #heat | 12:28 | |
| *** e0ne has quit IRC | 12:28 | |
| *** blomquisg has quit IRC | 12:31 | |
| shardy | ramishra: Hi! | 12:41 |
| *** e0ne_ has quit IRC | 12:41 | |
| *** e0ne has joined #heat | 12:42 | |
| ramishra | shardy: regarding the code review...I have put some comments .let me know if they make sense.. else, I have the changed patch ready and will upload it... | 12:42 |
| *** blomquisg has joined #heat | 12:46 | |
| *** mkollaro has quit IRC | 12:52 | |
| *** mkollaro1 has joined #heat | 12:52 | |
| *** zhiyan is now known as zhiyan_ | 12:54 | |
| *** pafuent has joined #heat | 12:57 | |
| *** nosnos has quit IRC | 13:08 | |
| *** zhiyan_ is now known as zhiyan | 13:15 | |
| *** mkollaro has joined #heat | 13:17 | |
| *** gokrokve has joined #heat | 13:17 | |
| *** gokrokve_ has joined #heat | 13:19 | |
| *** bvandenh has joined #heat | 13:19 | |
| *** mkollaro1 has quit IRC | 13:21 | |
| *** gokrokve has quit IRC | 13:22 | |
| *** alexpilotti has joined #heat | 13:22 | |
| *** gokrokve_ has quit IRC | 13:24 | |
| *** tspatzier has joined #heat | 13:24 | |
| therve | It's gate-sitting day. Again :/ | 13:38 |
| *** cmyster has quit IRC | 13:39 | |
| *** jdob has joined #heat | 13:41 | |
| *** aignatov is now known as aignatov_ | 13:43 | |
| openstackgerrit | Rabi Mishra proposed a change to openstack/heat: Add ability to create provider networks (vlan,flat). https://review.openstack.org/72454 | 13:47 |
| *** funzo has joined #heat | 13:52 | |
| *** aweiteka has joined #heat | 13:57 | |
| *** funzo has quit IRC | 13:58 | |
| *** agileclipse has joined #heat | 13:58 | |
| *** agileclipse has quit IRC | 13:59 | |
| *** ramishra_ has joined #heat | 14:01 | |
| *** sgordon has joined #heat | 14:01 | |
| *** sgordon has quit IRC | 14:01 | |
| *** sgordon has joined #heat | 14:01 | |
| *** topol has quit IRC | 14:01 | |
| *** ramishra has quit IRC | 14:02 | |
| radix | did something change about {get_attr} recently? | 14:03 |
| *** aignatov_ is now known as aignatov | 14:03 | |
| *** mkollaro has quit IRC | 14:05 | |
| *** ramishra_ has quit IRC | 14:06 | |
| therve | radix, Relatively | 14:06 |
| shardy | radix: Yes, most significantly https://review.openstack.org/#/c/67562/ | 14:06 |
| therve | I think Zane is trying to revert it | 14:06 |
| radix | hm ok, yeah | 14:06 |
| radix | I guess that's not related to my problem (or rather, the problem of person I'm trying to help) | 14:07 |
| shardy | therve: I think there's a fix for bug #1280526 which doesn't need reverting? | 14:07 |
| radix | time to actually try to reproduce it | 14:07 |
| therve | shardy, Well no revert, but change the behavior back mostly | 14:07 |
| therve | That's what I understood at leasst | 14:08 |
| *** TonyBurn_ has quit IRC | 14:09 | |
| shardy | therve: Hmm, well that is not at all clear from the commit message, or lack of reverted docs.. :\ | 14:09 |
| therve | I mentioned the latter in the review, yeah | 14:10 |
| *** Linz has joined #heat | 14:12 | |
| *** lpmulligan has joined #heat | 14:12 | |
| *** funzo has joined #heat | 14:15 | |
| *** ramishra has joined #heat | 14:15 | |
| *** saju_m has quit IRC | 14:15 | |
| *** zhiyan is now known as zhiyan_ | 14:16 | |
| *** gokrokve has joined #heat | 14:17 | |
| *** zhiyan_ is now known as zhiyan | 14:18 | |
| *** jdob has quit IRC | 14:19 | |
| *** Linz has quit IRC | 14:19 | |
| *** jdob has joined #heat | 14:19 | |
| *** Linz has joined #heat | 14:20 | |
| *** tomek_adamczewsk has quit IRC | 14:20 | |
| *** gokrokve has quit IRC | 14:22 | |
| *** tomek_adamczewsk has joined #heat | 14:22 | |
| *** jcru has joined #heat | 14:23 | |
| *** sabeen has joined #heat | 14:24 | |
| *** Linz has quit IRC | 14:24 | |
| *** jpeeler has joined #heat | 14:24 | |
| *** jpeeler has quit IRC | 14:24 | |
| *** jpeeler has joined #heat | 14:24 | |
| *** sabeen has left #heat | 14:24 | |
| *** radez_g0n3 is now known as radez | 14:30 | |
| SpamapS | Hrm | 14:31 |
| SpamapS | looks like bug 1276517 was deeper than the commit that fixed it | 14:31 |
| *** ramishra has quit IRC | 14:31 | |
| shardy | Treating 500's as acceptable responses is just so wrong IMO :( | 14:32 |
| shardy | paper over the cracks and carry on \o/ :( | 14:32 |
| zaneb | skraynev: o/ | 14:34 |
| skraynev | zaneb: hi) | 14:34 |
| SpamapS | shardy: 500 is not a permanent error | 14:35 |
| SpamapS | shardy: and right now, without retry, 500 means tear down the entire cloud. | 14:35 |
| skraynev | zaneb: I have addressed my question to shardy yet ) | 14:35 |
| zaneb | ok :) | 14:35 |
| shardy | SpamapS: well IME more often than not it means nova broken or misconfigured | 14:35 |
| skraynev | zaneb: it was about using static methods on review https://review.openstack.org/#/c/72454/ | 14:35 |
| shardy | so delaying reporting that for the duration of a stack timeout seems really counter productive | 14:36 |
| zaneb | skraynev: fwiw I am now in the US, so shardy is probably closer to your time zone :) | 14:36 |
| SpamapS | right now it means that nova has a timeout... we have about a 1% error rate on nova-api... we're digging it out.. but.. just try again works fine. | 14:36 |
| shardy | and also, if we don't fail on 500's (which is *always* the wrong response), how will the bugs get fixed? | 14:36 |
| skraynev | zaneb: Yeap) I had such guess) | 14:36 |
| SpamapS | shardy: I'd say error rates should be measured in logs, not failed stacks. :-P | 14:36 |
| *** jdob_ has joined #heat | 14:37 | |
| *** jdob has quit IRC | 14:37 | |
| skraynev | zaneb: I was wondering why do you work all night in Europe ) | 14:37 |
| *** TonyBurn_ has joined #heat | 14:37 | |
| *** blamar has joined #heat | 14:38 | |
| SpamapS | shardy: oh actually.. interesting enough, since last night I turned on debugging .. I think I may have found the bug in nova baremetal that is causing the error rate | 14:38 |
| SpamapS | 2014-02-17 11:52:43,410.410 23065 TRACE nova.api.openstack File "/opt/stack/venvs/nova/local/lib/python2.7/site-packag | 14:38 |
| SpamapS | es/nova/virt/baremetal/db/sqlalchemy/api.py", line 131, in bm_node_get | 14:38 |
| SpamapS | 2014-02-17 11:52:43,410.410 23065 TRACE nova.api.openstack bm_node_id = int(bm_node_id) | 14:38 |
| SpamapS | 2014-02-17 11:52:43,410.410 23065 TRACE nova.api.openstack ValueError: invalid literal for int() with base 10: 'ed0363be | 14:38 |
| *** jdob_ is now known as jdob | 14:38 | |
| SpamapS | -e673-4baf-821c-d2c98699b200' | 14:38 |
| zaneb | skraynev: lol. I moved from Munich in December | 14:38 |
| skraynev | zaneb: Anyway your opinion about it question is welcome for me | 14:38 |
| skraynev | zaneb: I don't remember when exactly this idea visited my head | 14:39 |
| *** blamar has quit IRC | 14:40 | |
| SpamapS | shardy: so.. still.. 500 is "the servers problem" not "the heat request's problem" .. so I respectfully disagree that it is wrong for Heat to continue retrying when it gets 500's | 14:40 |
| *** blamar has joined #heat | 14:40 | |
| shardy | SpamapS: 503 means we should retry, 500 means the server is broken, IMO | 14:40 |
| SpamapS | the box could be low on RAM too.. or out of disk space.. or partitioned from rabbitmq.. | 14:41 |
| shardy | SpamapS: so we add to the problem by polling for all stack, by default for an *hour*? | 14:41 |
| SpamapS | shardy: I think that 503 is a nuance that most blanket error handlers aren't quite ready to fully support. | 14:42 |
| SpamapS | shardy: Yes... but its not our fault. :) | 14:42 |
| SpamapS | shardy: I'd be _way_ more inclined to say yeah let's be pedantic about server side errors if we could actually retry once the provider says "ok we fixed it" | 14:43 |
| shardy | SpamapS: It's like fixing race conditions by putting sleeps in your code IMO, but.. <shrugs> | 14:43 |
| shardy | SpamapS: That's exactly what the retry from failed states stuff is supposed to do isn't it? | 14:44 |
| SpamapS | but, as zaneb has pointed out, that is a feature which requires a fair amount of re-plumbing for updates. :-/ | 14:44 |
| SpamapS | yes retry from failed states is important. I'm kind of shocked nobody has freaked out about using Heat without it actually. | 14:44 |
| zaneb | skraynev: NeutronResource.validate_properties is also a staticmethod, so it makes sense to me to override it with another staticmethod | 14:44 |
| SpamapS | shardy: it is non-trivial unfortunately. I was hoping it wouldn't be, but you have to start storing the updated snippets for each resource as the update progresses so that a retried update doesn't retry everything that already worked. | 14:45 |
| shardy | SpamapS: Hmm, yeah that sounds messy :( | 14:46 |
| SpamapS | because right now.. I have a _perfectly good working openstack cloud_ ... | 14:46 |
| SpamapS | but its heat stack is broken | 14:46 |
| SpamapS | so I cannot update it | 14:46 |
| SpamapS | How that isn't a critical feature.. I'm not sure. :-/ | 14:47 |
| shardy | SpamapS: Yeah, well we first talked about this nearly a year ago IIRC | 14:47 |
| skraynev | zaneb: Yeap. It's really possible, but looks different from other resource implementation at all. | 14:47 |
| shardy | https://bugs.launchpad.net/heat/+bug/1160052 | 14:47 |
| SpamapS | shardy: indeed, and it hasn't gotten fixed because it is non-trivial. | 14:47 |
| SpamapS | I'm poking at abandon/fix/adopt right now as a stop-gap until that feature can land, because I suspect that people will freak out if we do major surgery on updates right before i3 | 14:48 |
| zaneb | skraynev: no other resource has ever overridden that, so anything you do will look different to any other resource implementation | 14:49 |
| shardy | zaneb: But multiple existing neutron resources extend validate() for the same reason | 14:52 |
| *** killer_prince has joined #heat | 14:52 | |
| *** vijendar has joined #heat | 14:52 | |
| *** tims has quit IRC | 14:53 | |
| zaneb | shardy: so if we're saying that we shouldn't override it at all, then I would probably agree. I haven't looked into it | 14:53 |
| *** andersonvom has joined #heat | 14:53 | |
| shardy | zaneb: I think either way is probably OK, but it's desirable to keep the resources fairly consistent | 14:54 |
| zaneb | if we're arguing about whether it should be static, then that's just not a very interesting argument | 14:54 |
| *** arbylee has joined #heat | 14:54 | |
| SpamapS | hm, abandon/adopt haven't landed in heatclient yet? | 14:54 |
| therve | Nope | 14:57 |
| shardy | abandon has | 14:57 |
| therve | The actions are broken server side anyway | 14:57 |
| therve | Please review! | 14:57 |
| *** achampion has quit IRC | 14:57 | |
| SpamapS | <sigh> | 14:57 |
| *** david-lyle has quit IRC | 14:58 | |
| skraynev | shardy, zaneb: have we any guide or rules for writing new resource? | 14:58 |
| SpamapS | do we have any actual users? | 14:58 |
| SpamapS | update users I should say | 14:58 |
| SpamapS | sorry | 14:58 |
| SpamapS | I suspect no. | 14:58 |
| *** vvechkanov1 has quit IRC | 14:58 | |
| zaneb | SpamapS: I suspect you may be right | 14:58 |
| zaneb | skraynev: there is a wiki page, and I think there was some added to the developer docs, or at least there was a review up for it | 14:59 |
| SpamapS | zaneb: ok, so perhaps I should attempt surgery.. | 15:00 |
| *** Michalik has quit IRC | 15:00 | |
| SpamapS | zaneb: Other than storing the updated snippets so we can assemble a partially updated template for retry... what do you see as a blocker for update retries? | 15:01 |
| *** skraynev is now known as skraynev_afk | 15:02 | |
| *** skraynev_afk is now known as skraynev_afk_goi | 15:03 | |
| *** skraynev_afk_goi is now known as skraynev_afk | 15:03 | |
| zaneb | SpamapS: that's the main one that I know of. because the stack is in an inconsistent state after a failed update, we can't safely perform more operations on it | 15:03 |
| zaneb | (including delete, btw, but we still allow that) | 15:03 |
| zaneb | since deleting can't make things worse | 15:04 |
| zaneb | so if we can store a consistent template, we would be ok | 15:04 |
| *** nkhare has joined #heat | 15:05 | |
| shardy | Would converge be an alternative solution - where the only allowable update from a failed state is a converge where we inspect the underlying resource and align the resource with the template? | 15:05 |
| zaneb | SpamapS: note though that rollback exists and should work | 15:05 |
| zaneb | shardy: it's possible to end up with resources that aren't even in the template | 15:06 |
| SpamapS | zaneb: rolling back an updated database schema isn't really something we want to orchestrate | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Add Docker resources to docs https://review.openstack.org/68753 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Refactor resource loading functions https://review.openstack.org/71989 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Restructure contrib/ directories https://review.openstack.org/73417 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Fix resource mapping for Docker https://review.openstack.org/68749 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Fix resource mapping for Rackspace https://review.openstack.org/69374 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Change Docker resource mapping name https://review.openstack.org/68750 | 15:06 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Generate docs for contrib plugins https://review.openstack.org/68751 | 15:06 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/heat: Update contrib docstrings to match guidelines https://review.openstack.org/73070 | 15:06 |
| shardy | zaneb: humm, yeah I see | 15:06 |
| zaneb | SpamapS: why not? migrations are supposed to work in both directions :p | 15:07 |
| SpamapS | with the graceful actions thing we'd have to tell the server "you're about to be deleted, but for a rollback, so please run the downgrade part of db_sync" | 15:07 |
| *** sabeen has joined #heat | 15:07 | |
| SpamapS | zaneb: except the previous image doesn't have the new migration on it | 15:07 |
| *** chandan_kumar has quit IRC | 15:07 | |
| SpamapS | so you're manually running a db_sync somewhere .. and hopefully that works | 15:07 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add API support for stack preview https://review.openstack.org/68406 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add StackResource specific preview behavior https://review.openstack.org/68407 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add preview for LoadBalancer https://review.openstack.org/68412 | 15:08 |
| zaneb | SpamapS: if a previous image exists and hasn't been deleted yet, it will just be swapped back in | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add engine preview stack with Template and Params https://review.openstack.org/65576 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add preview for NestedStack https://review.openstack.org/68408 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add preview for ResourceGroup https://review.openstack.org/68409 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add preview for InstanceGroup https://review.openstack.org/68410 | 15:08 |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Add preview for TemplateResource https://review.openstack.org/68411 | 15:08 |
| *** rustlebee is now known as russellb | 15:08 | |
| SpamapS | zaneb: yes, swapped back in, and then the code comes up, and "wtf is this string column that used to be an int?" | 15:08 |
| *** rwsu has joined #heat | 15:09 | |
| zaneb | ok, I thought you were talking about the DB instance. | 15:09 |
| * zaneb is just confused | 15:09 | |
| SpamapS | No I'm saying the schema migrations are the devil. | 15:10 |
| therve | shardy, There is a problem with the new signal API | 15:10 |
| SpamapS | We've been operating under an assumption that if we ever have to do that, we slay a goat first. | 15:10 |
| therve | shardy, The self._load_user_creds call in the engine seems to break it somehow | 15:10 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/heat: Update api docstrings to match guidelines https://review.openstack.org/73515 | 15:11 |
| *** bada has quit IRC | 15:13 | |
| andersonvom | whois radix | 15:13 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/heat: Update common docstrings to match guidelines https://review.openstack.org/73519 | 15:13 |
| shardy | therve: doh! | 15:14 |
| therve | shardy, I guess you see why? :) | 15:15 |
| radix | andersonvom: hello :) | 15:15 |
| *** bada has joined #heat | 15:15 | |
| andersonvom | radix: whoops, sorry about that. missed a slash ;) | 15:15 |
| *** daneyon has joined #heat | 15:17 | |
| shardy | therve: https://review.openstack.org/#/c/71300/ will fix it for stack domain users | 15:17 |
| shardy | (I think?) | 15:17 |
| *** bada_ has joined #heat | 15:18 | |
| shardy | therve: oh, hmm, maybe I don't see why :) | 15:18 |
| therve | shardy, Cool, because I don't either | 15:18 |
| *** gokrokve has joined #heat | 15:19 | |
| *** daneyon has quit IRC | 15:19 | |
| shardy | therve: I'm about to start testing a native waitcondition resource, with your client patch, so I guess I'll find out soon enough | 15:19 |
| *** daneyon has joined #heat | 15:20 | |
| *** bada has quit IRC | 15:20 | |
| therve | shardy, I get non-obvious failure. When I try to signal, the server.get() call fails in neutron and nova returns a 500 | 15:20 |
| *** nkhare has quit IRC | 15:20 | |
| therve | (using HARestarter) | 15:20 |
| *** giulivo has joined #heat | 15:22 | |
| *** Michalik has joined #heat | 15:23 | |
| *** gokrokve has quit IRC | 15:23 | |
| *** achampion has joined #heat | 15:24 | |
| *** coolsvap has joined #heat | 15:26 | |
| *** ramishra has joined #heat | 15:32 | |
| *** jergerber has joined #heat | 15:32 | |
| *** beekneemech is now known as bnemec | 15:34 | |
| *** tims has joined #heat | 15:35 | |
| *** david-lyle has joined #heat | 15:35 | |
| *** ramishra has quit IRC | 15:37 | |
| *** e0ne has quit IRC | 15:38 | |
| zaneb | so if testcase finds a test file that it can't load, it now just skips over it silently and reports that everything passed? | 15:39 |
| zaneb | testr, that is | 15:39 |
| therve | zaneb, Because of import error? | 15:40 |
| zaneb | no, just syntax error | 15:40 |
| therve | Right | 15:40 |
| therve | That's probably true yes | 15:41 |
| zaneb | you have to watch the number of test cases actually run like a hawk, just to know if your "pass" means anything or not | 15:41 |
| zaneb | at least pep8 will find it but man, that's horrible | 15:42 |
| *** topol has joined #heat | 15:43 | |
| *** ramishra has joined #heat | 15:44 | |
| *** edmund has joined #heat | 15:45 | |
| openstackgerrit | Ton Ngo proposed a change to openstack/python-heatclient: Add CLI suppport for troubleshooting failed stack https://review.openstack.org/71438 | 15:45 |
| *** e0ne has joined #heat | 15:47 | |
| * IgorYozhikov is now away: went away... | 15:50 | |
| *** IgorYozhikov is now known as IYozhikov_away | 15:50 | |
| *** aignatov is now known as aignatov_ | 15:54 | |
| *** ramishra has quit IRC | 15:55 | |
| *** jprovazn has quit IRC | 15:56 | |
| *** jergerber has quit IRC | 15:57 | |
| *** e0ne_ has joined #heat | 15:59 | |
| *** e0ne has quit IRC | 16:02 | |
| *** randallburt has joined #heat | 16:03 | |
| *** jergerber has joined #heat | 16:05 | |
| *** rcleere has joined #heat | 16:05 | |
| radix | is there some failure condition that would lead to a stack's outputs just showing the unprocessed template chunk specifying the value of the output? | 16:07 |
| radix | e.g. the stack's output includes {"output_value": {"get_attr": ["my-resource", "my-attribute"]}, ...} | 16:08 |
| therve | radix, If your template is wrong? | 16:09 |
| shardy | radix: You forgot to use str_replace? | 16:10 |
| andersonvom | shardy: good morning/afternoon! | 16:14 |
| radix | hrmph | 16:14 |
| shardy | andersonvom: hi | 16:15 |
| *** jamieh has quit IRC | 16:15 | |
| radix | unfortunately this is happening in an internal rackspace QA template for testing rackspace resources, so it won't be reproducible outside of the rax cloud of course :P | 16:15 |
| radix | anyway I am pretty sure there is just something wrong with the template, I just have no idea what | 16:15 |
| therve | Can you share it? | 16:15 |
| radix | probably, lemme double-check for s3kr1ts | 16:16 |
| radix | never mind, randallburt just figured it out, the resource name was wrong :P | 16:16 |
| therve | :) | 16:17 |
| radix | that's a bit strange, perhaps get_attr should fail if the first argument isn't actually a resource that can be found | 16:17 |
| therve | radix, https://review.openstack.org/#/c/70358/ fwiw | 16:17 |
| *** jamieh has joined #heat | 16:17 | |
| therve | Maybe it would help | 16:17 |
| radix | hmm | 16:18 |
| randallburt | therve: seems like we're waiting too late to validate that. Once the stack is created, its too late IMO. | 16:18 |
| *** pablosan has joined #heat | 16:19 | |
| andersonvom | shardy: so, we're trying to get the keystone v2 shim to work. would you happen to have any suggestions on how to go about that? any caveats or anything? | 16:19 |
| *** gokrokve has joined #heat | 16:19 | |
| shardy | andersonvom: Well basically you can take a lot of the code I just ripped out in the keystone-v3-only BP ;) | 16:20 |
| shardy | and all the stack_domain* functions will need to be mapped to some other non-domain container (heat specific project?) | 16:20 |
| shardy | Or you just use the non-admin client for everything and live with bug #1089261 until you get v3 | 16:21 |
| shardy | i.e all the stack_domain functions just create users in the stack owning users tenant, as they do in Havana | 16:22 |
| shardy | as that needs admin, I don't really see how you can allow that though, which is the whole reason for all the instance-users stuff | 16:22 |
| pshchelo | hi all, how can I use Neutron floating IP with AWS load balancer? or only EIP is working with it? | 16:22 |
| *** bvandenh has quit IRC | 16:23 | |
| shardy | andersonvom: actually forget the stack specific project comment, that won't work | 16:24 |
| *** gokrokve has quit IRC | 16:24 | |
| *** chandan_kumar has joined #heat | 16:29 | |
| *** chandan_kumar has quit IRC | 16:29 | |
| *** chandan_kumar has joined #heat | 16:30 | |
| *** skraynev_afk is now known as skraynev | 16:31 | |
| andersonvom | shardy: ok. I don't think the bug will affect us that much anyway | 16:31 |
| *** bvandenh has joined #heat | 16:35 | |
| *** jistr has quit IRC | 16:35 | |
| andersonvom | shardy: did anything change in terms of the things we set up in heat from keystone v2 to v3 that we need to watch out for? or just using the previous client and adding a way to switch from v3 to v2 should do it? | 16:35 |
| shardy | andersonvom: the only change really is that we create a project per stack (in the heat domain), and all users associated with stack resources are created in that stack-specific project | 16:36 |
| shardy | andersonvom: if you don't required that isolation, you can just use the v2 calls with the stack-owners's project for everything | 16:37 |
| shardy | andersonvom: as mentioned in the bug, I'd suggest making heat_keystoneclient.KeystoneClient pluggable, then you can just plug in an interface compatible replacement | 16:37 |
| *** alexpilotti has quit IRC | 16:38 | |
| shardy | we can't promise interface stability there, but I don't think it's likely to change much after the instance-users patches land (at least for Icehouse) | 16:38 |
| *** alexpilotti has joined #heat | 16:38 | |
| jpeeler | shardy: do you have a patch that addresses the failure here for havana? http://lists.openstack.org/pipermail/openstack-stable-maint/2014-February/002162.html | 16:40 |
| jpeeler | i'm told that it's because of the new keystone client release | 16:41 |
| shardy | jpeeler: yes, therve fixed it in master on Friday | 16:41 |
| jpeeler | ah i grepped for your name | 16:41 |
| shardy | jpeeler: therve beat me to it ;) | 16:41 |
| therve | I didn't fix any other gate failures happening right now though :/ | 16:41 |
| shardy | jpeeler: https://review.openstack.org/#/c/73544/ | 16:42 |
| *** erkules_ has joined #heat | 16:43 | |
| jpeeler | has it been proposed for stable? (assuming that makes sense) | 16:43 |
| *** erkules has quit IRC | 16:43 | |
| shardy | jpeeler: I don't think it has yet, but yes it probably should be based on those test failures | 16:45 |
| shardy | If you or therve can propose the backport, ping me and I'll review | 16:46 |
| *** e0ne has joined #heat | 16:48 | |
| jpeeler | therve: want to? | 16:48 |
| jpeeler | if not i'll do it | 16:48 |
| therve | jpeeler, Feel free. I'm not used to the backport procedure | 16:49 |
| therve | Also I try to debug gate failures :/ | 16:49 |
| jpeeler | alrighty | 16:49 |
| *** erkules_ has quit IRC | 16:50 | |
| *** e0ne_ has quit IRC | 16:50 | |
| *** erkules_ has joined #heat | 16:51 | |
| *** blomquisg has quit IRC | 16:53 | |
| *** chandan_kumar has quit IRC | 16:55 | |
| *** Linz has joined #heat | 16:55 | |
| *** Linz has quit IRC | 16:55 | |
| *** gokrokve has joined #heat | 16:56 | |
| *** randallburt has quit IRC | 17:02 | |
| *** blomquisg has joined #heat | 17:07 | |
| *** jamieh has quit IRC | 17:10 | |
| *** randallburt has joined #heat | 17:13 | |
| *** TonyBurn_ has quit IRC | 17:14 | |
| *** e0ne has quit IRC | 17:16 | |
| *** erkules_ is now known as erkules | 17:18 | |
| *** akuznetsov has quit IRC | 17:26 | |
| radix | I guess we don't have HOT equivalents for e.g. UpdatePolicy, eh? | 17:34 |
| radix | (no "update_policy") | 17:34 |
| radix | or Metadata -> metadata | 17:35 |
| *** akuznetsov has joined #heat | 17:42 | |
| *** akuznetsov has quit IRC | 17:42 | |
| radix | hm, are nested stacks hidden from stack-list now? | 17:43 |
| *** randallburt has quit IRC | 17:46 | |
| *** akuznetsov has joined #heat | 17:47 | |
| *** zhiyan is now known as zhiyan_ | 17:50 | |
| *** derekh has quit IRC | 17:54 | |
| openstackgerrit | Tim Schnell proposed a change to openstack/heat: Adds Parameter Label to template validate call https://review.openstack.org/65325 | 17:56 |
| *** akuznetsov has quit IRC | 17:57 | |
| *** ppetit has quit IRC | 17:58 | |
| openstackgerrit | A change was merged to openstack/heat: Re-enable lazy translation https://review.openstack.org/69133 | 17:58 |
| *** ppetit has joined #heat | 17:59 | |
| *** randallburt has joined #heat | 17:59 | |
| openstackgerrit | A change was merged to openstack/heat: Handle API limit exception in OS::Trove::Instance https://review.openstack.org/71082 | 17:59 |
| *** tomek_adamczewsk has quit IRC | 17:59 | |
| *** ppetit has quit IRC | 17:59 | |
| *** akuznetsov has joined #heat | 18:00 | |
| *** lindsayk has joined #heat | 18:04 | |
| radix | randallburt: do you have an example of a ResourceGroup that scales StackResources? | 18:05 |
| randallburt | radix: nope. | 18:05 |
| randallburt | and ResourceGroup doesn't scale by itself; you need a stack update to change the number of resources. | 18:06 |
| radix | er, yeah, "scales" was the wrong verb there :) | 18:06 |
| randallburt | radix: that being said, it should "just work" unless you hit nested stack limits | 18:06 |
| openstackgerrit | A change was merged to openstack/heat: Translate constraint description in validation https://review.openstack.org/70372 | 18:07 |
| jasond` | has anybody noticed that there's a huge amount of output from the migration scripts before each failing unit test now? http://dunsmor.com/pastebin/1392660094.txt | 18:08 |
| randallburt | jasond`: does one of us still need to raise their hand for the Heat security team or did that get sorted? | 18:08 |
| jasond` | randallburt: one of us still needs to raise our hand | 18:10 |
| randallburt | jasond`: ok. I don't mind unless you have a preference | 18:10 |
| *** andrew_plunk has joined #heat | 18:11 | |
| jasond` | randallburt: i'll do it. unless you're really into the idea | 18:12 |
| randallburt | jasond`: nah. you go for it and thanks! | 18:12 |
| jasond` | ok. i'll let steve baker know. thanks | 18:13 |
| randallburt | np | 18:13 |
| shardy | therve: when you saw signal errors, did you have the latest policy.json? | 18:13 |
| shardy | I saw an error but it was due to resource:signal being missing from my outdated policy | 18:13 |
| shardy | otherwise heat resource-signal appears to work for me | 18:14 |
| randallburt | shardy (or anyone else who knows) what's the process for cutting a release of python-heatclient? I know its on its own cycle. There are some patches for ssl and other fixes that some of our users are asking about. | 18:14 |
| shardy | randallburt: I haven't actually done it myself, stevebaker is the person to ask as AFAIK he's done most/all of the releases to date | 18:15 |
| randallburt | shardy: yeah, thought so. I'll shoot him an email in case I miss him this afternoon. Thanks! | 18:15 |
| *** pvaneck has joined #heat | 18:25 | |
| *** alexheneveld has quit IRC | 18:27 | |
| *** alexheneveld has joined #heat | 18:29 | |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Move refresh_server() to nova_utils https://review.openstack.org/72008 | 18:37 |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Make server resources use nova_utils.refresh_server() https://review.openstack.org/72009 | 18:37 |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Handle API limit exception in nova_utils.refresh_server https://review.openstack.org/71660 | 18:42 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/python-heatclient: Only call decode() on bytes https://review.openstack.org/74123 | 18:45 |
| *** aignatov_ is now known as aignatov | 18:48 | |
| *** openstackgerrit has quit IRC | 18:53 | |
| *** openstackgerrit has joined #heat | 18:54 | |
| *** andrew_plunk has quit IRC | 18:55 | |
| *** andrew_plunk has joined #heat | 18:55 | |
| *** e0ne has joined #heat | 18:57 | |
| *** giulivo has quit IRC | 18:57 | |
| *** arbylee has quit IRC | 18:58 | |
| *** arbylee has joined #heat | 19:04 | |
| *** arbylee has quit IRC | 19:04 | |
| *** jprovazn has joined #heat | 19:04 | |
| *** arbylee has joined #heat | 19:05 | |
| *** aignatov is now known as aignatov_ | 19:07 | |
| *** gokrokve has quit IRC | 19:07 | |
| *** gokrokve has joined #heat | 19:08 | |
| openstackgerrit | Anderson Mesquita proposed a change to openstack/heat: Move X-Auth-Url logic to auth_url middleware https://review.openstack.org/68800 | 19:08 |
| *** e0ne has quit IRC | 19:09 | |
| *** e0ne has joined #heat | 19:09 | |
| stevebaker | randallburt: randallburt releasing heatclient is essentially just doing a git tag and pushing that to gerrit | 19:11 |
| randallburt | stevebaker: any criteria for doing so? gates and such? votes? | 19:11 |
| *** gokrokve has quit IRC | 19:12 | |
| *** gokrokve has joined #heat | 19:13 | |
| stevebaker | its gated, but what client versions are used for gating seems like a bit of a mess | 19:13 |
| stevebaker | randallburt: I've been waiting for this to land then I was going to do a release. Maybe also some obvious bug fix changes https://review.openstack.org/#/c/58885/ | 19:14 |
| stevebaker | because software config is really hard to try out until it does | 19:14 |
| *** e0ne has quit IRC | 19:14 | |
| jasond` | does anybody have time to review https://review.openstack.org/#/c/58043/ ? | 19:14 |
| randallburt | stevebaker: cool. I'll review that one asap | 19:15 |
| *** gokrokve has quit IRC | 19:17 | |
| *** alexheneveld has quit IRC | 19:18 | |
| stevebaker | jasond`: do you have a way of testing all those cloud-init install methods? | 19:18 |
| jasond` | the configuration scripts? | 19:19 |
| stevebaker | yeah | 19:19 |
| jasond` | we test them in our CI infrastructure | 19:19 |
| jasond` | that way, they're always tested against the latest RS Cloud Servers images | 19:21 |
| stevebaker | jasond`: it would be awesome if you could hook your CI into gerrit commits. Its getting quite common for third-parties to do this. Red Hat has a bunch of these jobs in other projects | 19:21 |
| jasond` | stevebaker: yeah. we might end up doing that. i was just looking at https://review.openstack.org/#/c/55987/ this morning | 19:22 |
| jasond` | i'll talk to andrew. he's been setting all of that up | 19:23 |
| jasond` | andrew_plunk, that is | 19:23 |
| *** e0ne has joined #heat | 19:23 | |
| randallburt | stevebaker: https://review.openstack.org/#/c/58885/ lgtm. any reservations on approving? The history looks good and Jun's comment has been addressed. | 19:24 |
| stevebaker | jasond`: and its worth pointing out that we would have had that keystone v2 debate much earlier if your CI notified the breakage on that first commit | 19:24 |
| *** lindsayk has quit IRC | 19:24 | |
| stevebaker | randallburt: I authored the original, so I can't +2. Feel free if you think its done | 19:25 |
| randallburt | stevebaker: I do. | 19:25 |
| *** lindsayk has joined #heat | 19:26 | |
| randallburt | stevebaker: approved. any thing else I should look at for the client prior to releasing? | 19:26 |
| stevebaker | randallburt, jasond`, https://review.openstack.org/#/c/72181/ https://review.openstack.org/#/c/73567/ https://review.openstack.org/#/c/69802/ https://review.openstack.org/#/c/70380/ | 19:28 |
| andersonvom | zaneb: you around? | 19:30 |
| zaneb | andersonvom: yep | 19:30 |
| andersonvom | zaneb: I had to rebase this patch a couple of times after you and shardy approved it, do you have a few secs to take a look at it again? https://review.openstack.org/#/c/71989/ | 19:31 |
| andersonvom | zaneb: no changes though, just non trivial rebases | 19:31 |
| sdake | 146 test failures | 19:32 |
| * sdake ughs | 19:32 | |
| zaneb | andersonvom: approved | 19:33 |
| andersonvom | zaneb: thanks! =) | 19:34 |
| randallburt | stevebaker: any heartburn over https://review.openstack.org/#/c/70380? | 19:37 |
| *** aignatov_ is now known as aignatov | 19:38 | |
| *** yogesh has joined #heat | 19:38 | |
| stevebaker | randallburt: reviewing now, it might need some minor tweaks | 19:38 |
| *** gokrokve has joined #heat | 19:38 | |
| randallburt | stevebaker: k. lgtm, though. I can +2 and leave approve to you then. | 19:39 |
| stevebaker | randallburt: can you hold off on that one? | 19:39 |
| randallburt | stevebaker: sure. I didn't approve | 19:39 |
| stevebaker | I just wonder if it should be possible to specify a URL for the adopt file | 19:40 |
| *** yogesh has quit IRC | 19:40 | |
| *** yogesh has joined #heat | 19:41 | |
| randallburt | stevebaker: could be helpful, though also could be a pretty easy follow-on. | 19:41 |
| *** gokrokve_ has joined #heat | 19:42 | |
| stevebaker | randallburt: we can have frequent heatclient releases up to icehouse release | 19:42 |
| *** aignatov is now known as aignatov_ | 19:42 | |
| randallburt | so far, the use cases have been abandon->save results->change endpoint->adopt | 19:42 |
| stevebaker | randallburt: like we can do another release as soon as its possible to do a full abandon/adopt | 19:43 |
| randallburt | stevebaker: sounds fine to me. The big thing our users are asking about is the ssl fix for Mac anyway, so that's ok by me ;) | 19:43 |
| *** gokrokv__ has joined #heat | 19:43 | |
| stevebaker | yeah, there are already some worthy fixes in there | 19:44 |
| *** gokrokve has quit IRC | 19:44 | |
| randallburt | stevebaker: IMO, lets release with software config and do another later when abandon/adopt is sorted, then. | 19:44 |
| stevebaker | randallburt: yep. and all those python3 and cleanup fixes can wait | 19:45 |
| *** yogesh has quit IRC | 19:45 | |
| randallburt | stevebaker: k. thanks! | 19:45 |
| *** gokrokve_ has quit IRC | 19:47 | |
| *** e0ne has quit IRC | 19:47 | |
| *** e0ne has joined #heat | 19:47 | |
| *** gokrokv__ has quit IRC | 19:48 | |
| *** samkottler has quit IRC | 19:48 | |
| *** samkottler has joined #heat | 19:48 | |
| *** jcru has quit IRC | 19:49 | |
| *** yogesh has joined #heat | 19:50 | |
| skraynev | stevebaker: hey | 19:50 |
| stevebaker | skraynev: ho | 19:50 |
| *** jcru has joined #heat | 19:50 | |
| skraynev | stevebaker: according this review https://review.openstack.org/#/c/72454/ | 19:51 |
| skraynev | you said that it's ok. I only want to know is it good have a resource with validation_properties instead using validate like in other resources? | 19:53 |
| stevebaker | skraynev: Overriding static methods seems odd, but maybe thats just my OldJavaBrain | 19:55 |
| stevebaker | skraynev: I think I agree with shardy, and would prefer overriding validate | 19:55 |
| skraynev | stevebaker: Heh) ok. I have discussed it with him | 19:56 |
| skraynev | and now I am glad to here, that you agree with that | 19:57 |
| *** gokrokve has joined #heat | 19:58 | |
| skraynev | and other question about https://review.openstack.org/#/c/67836/ | 19:58 |
| skraynev | I agree with your last comment, but how about using it in AWS templates? | 19:59 |
| skraynev | AFAIK, only hot support get_attr | 20:00 |
| stevebaker | skraynev: non-hot templates can use Fn::Select | 20:01 |
| *** lindsayk has quit IRC | 20:02 | |
| stevebaker | skraynev: but since it is not an AWS resource there will be no reason not to use HOT | 20:02 |
| *** lindsayk has joined #heat | 20:02 | |
| skraynev | but in case with first_address it will be combination of some functions | 20:02 |
| stevebaker | skraynev: yes, it sucks. Use HOT ;) | 20:02 |
| skraynev | stevebaker: lol ! | 20:03 |
| *** lindsayk has quit IRC | 20:03 | |
| skraynev | stevebaker: I will recommend it all our guys using your quote :) | 20:03 |
| *** lindsayk has joined #heat | 20:04 | |
| zaneb | stevebaker: which hotel are you going to book for the summit? | 20:06 |
| skraynev | stevebaker: ok. I will stay it abandon. Also what are planing do with first_address attribute of server resource? | 20:07 |
| sdake | zaneb mind helping me a bit with mox - http://paste.fedoraproject.org/77980/13926675 is the test script - http://paste.fedoraproject.org/77982/26675631 is the results | 20:07 |
| sdake | AttributeError: 'module' object has no attribute 'call'^M | 20:07 |
| stevebaker | zaneb: maybe the DoubleTree, but I was wondering if it was out of favour now | 20:07 |
| sdake | zaneb I booked option B = free breakfast + free wifi + looks pretty good overall | 20:08 |
| stevebaker | skraynev: maybe deprecated, when that is possible | 20:08 |
| zaneb | stevebaker: reading the reviews it sounds questionable. apparently the fire alarm goes off every morning at 1am for 30 minutes(!) | 20:08 |
| skraynev | stevebaker: Hm before icehouse release? or after? | 20:08 |
| stevebaker | zaneb: btw thanks for volunteering to be the sober driver for all the parties | 20:08 |
| zaneb | sdake: do you have a link where it mentions free breakfast? I'm leaning towards that one | 20:08 |
| zaneb | stevebaker: oh, you wanted a *sober* driver | 20:09 |
| sdake | zaneb ts24 says | 20:09 |
| zaneb | awkward... | 20:09 |
| sdake | "free breakfast" | 20:09 |
| sdake | taht is as much as I know | 20:09 |
| sdake | marriot generally has a decent free breakfast | 20:09 |
| zaneb | ah, I never look at ts24 because it is crap | 20:09 |
| sdake | ya I don't like ts24 either | 20:10 |
| sdake | stevebaker I think we would be better off with walking +cabs :) | 20:10 |
| randallburt | Atlanta isn't terribly walk-able. | 20:10 |
| zaneb | sdake: they usually have buses put on for all the parties | 20:11 |
| randallburt | and traffic is a wonder to behold. | 20:11 |
| sdake | so zaneb, about that paste above, could you take al ook :) | 20:11 |
| * sdake dislikes mox | 20:12 | |
| zaneb | sigh, reading the TS24 "Quickstart Guide" and it fails to answer my first question, WHERE IS THE WEBSITE? | 20:12 |
| sdake | actually I think I saw that on the marriot site zaneb | 20:13 |
| zaneb | oh, wait... (b) is not the one I am looking at | 20:13 |
| zaneb | looking at (a) | 20:14 |
| sdake | a was the westin, b was the fairfield marriot | 20:14 |
| zaneb | (b) is 50% further away than _all_ of the other hotels, and in a different direction | 20:14 |
| zaneb | (and sounds like it might get a bit stabby) | 20:15 |
| sdake | men are generally safe in most parts of the US | 20:15 |
| zaneb | (d) is the Westin, but yes, (b) is the fairfield | 20:15 |
| shardy | lol | 20:15 |
| sdake | oh maybe the order changed :) | 20:15 |
| zaneb | agree on the safety thing, but I may have an entourage to think about | 20:16 |
| *** skraynev is now known as skraynev_afk | 20:19 | |
| larsks | zaneb: The Westin and the Fairfield are approx the same distance from the venue. | 20:19 |
| sdake | .4km takes 10 minutes to walk | 20:19 |
| zaneb | as the crow flies, but I'd be more worried about the direction, tbh. it sucks when everybody at the conference heads off to their hotel and you head off in the opposite direction and miss out on all of those conversations | 20:21 |
| sdake | the alternative is to book everyone at the westin and send 2 less people | 20:22 |
| zaneb | ok, I just booked at the Hampton. Discovered they have free breakfast too | 20:25 |
| *** tspatzier has quit IRC | 20:27 | |
| *** daneyon has quit IRC | 20:27 | |
| *** e0ne has quit IRC | 20:28 | |
| *** e0ne has joined #heat | 20:29 | |
| *** akuznetsov has quit IRC | 20:29 | |
| *** lindsayk has quit IRC | 20:30 | |
| *** e0ne has quit IRC | 20:33 | |
| shardy | Anyone have any ideas why a routes mapper would not pass the body to the controller method? | 20:40 |
| shardy | can't figure out where it's getting lost... | 20:40 |
| *** bada_ has quit IRC | 20:41 | |
| zaneb | shardy: decorator? | 20:41 |
| shardy | zaneb: Yeah, that's what I thought but it doesn't look like the body kwarg is even getting passed into the decorator | 20:42 |
| * shardy looks again | 20:42 | |
| zaneb | I got nothin | 20:42 |
| *** ifarkas has quit IRC | 20:44 | |
| stevebaker | zaneb, shardy, a quick review please? https://review.openstack.org/#/c/72181/ | 20:44 |
| *** lindsayk has joined #heat | 20:44 | |
| shardy | stevebaker: done | 20:45 |
| stevebaker | ta | 20:45 |
| shardy | stevebaker: Did I see you're planning to extend OS::Nova:Server with StackUser? | 20:47 |
| stevebaker | shardy: yes, but currently user will only be created for software-config paths | 20:47 |
| stevebaker | its working locally | 20:47 |
| shardy | stevebaker: Ok, cool, I'm working on a OS::Heat::WaitSignal resource, which counts signals (and stores data if I can get it working..) from the native signal API | 20:48 |
| *** aignatov_ is now known as aignatov | 20:48 | |
| shardy | If we can get credentials into the instance, it could be useful with heatclient resource-signal | 20:48 |
| stevebaker | shardy: I'm just deploying the keypair to the os-collect-config config currently | 20:49 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Add support for software config resources https://review.openstack.org/58885 | 20:52 |
| stevebaker | \o/ | 20:54 |
| stevebaker | of course I have another commit not yet pushed which is needed before a release | 20:55 |
| zaneb | sdake: you have this patch, right? https://github.com/openstack/heat/commit/26fe7a961f644e1c93a92c8a73b25d649137f7b9 | 21:00 |
| sdake | zaneb no but I am using older keystone client atm | 21:01 |
| sdake | I ended up using somethign liek this: | 21:01 |
| shardy | stevebaker: Ok, I guess the de-ec2-keypairification may have to wait until Juno, but I'm trying to move things in the direction where we can just use native auth for all the native resources | 21:01 |
| sdake | + rpcclient = self.m.CreateMockAnything() | 21:02 |
| sdake | + rpcclient.call(dummy_req.context, self.topic, | 21:02 |
| sdake | instead of the stub out | 21:02 |
| sdake | seems to work | 21:02 |
| sdake | atleast the test cases do :) | 21:02 |
| stevebaker | shardy: I'm sure fixing that can be framed as a bug ;) | 21:02 |
| shardy | \o/ ;) | 21:03 |
| *** nati_ueno has joined #heat | 21:03 | |
| *** asalkeld has quit IRC | 21:04 | |
| *** nati_ueno has quit IRC | 21:06 | |
| *** nati_ueno has joined #heat | 21:06 | |
| zaneb | sdake: self.m.StubOutWithMock(messaging, 'call') seems to be the problem. There is no oslo.messaging.call() function | 21:07 |
| andrew_plunk | hey SpamapS are you around? | 21:08 |
| sdake | zaneb yes I know its in rpc, I tried that as well with no luck | 21:08 |
| sdake | oslo.messaging.rpc | 21:08 |
| zaneb | sdake: it's actually in a class now, isn't it? | 21:09 |
| zaneb | oslo.messaging.rpc.RPCClient | 21:09 |
| *** pafuent has left #heat | 21:09 | |
| stevebaker | shardy, sdake, zaneb, could you review this? its a serious bug https://review.openstack.org/#/c/70289/ | 21:09 |
| sdake | ugh | 21:09 |
| zaneb | so you should create an RPCClient, stub out it's call() method, and use that for RPC | 21:10 |
| sdake | stevebaker enjoy | 21:12 |
| sdake | zaneb I already have an rpcclient available in EngineClient called _client | 21:12 |
| sdake | RPCClient that is | 21:13 |
| sdake | how do I stub out it's call method there? | 21:13 |
| zaneb | sdake: then you want self.m.StubOutWithMock(_client, 'call') | 21:13 |
| sdake | zaneb cool I'll give that a go | 21:13 |
| sdake | thanks | 21:13 |
| zaneb | np | 21:13 |
| sdake | and _client.call? | 21:13 |
| stevebaker | sdake: do you mean to change the help text? The default value will avoid the issue for current catalog sizes | 21:14 |
| openstackgerrit | Andrew Plunk proposed a change to openstack/heat: Oslo config option changed by test causing problem https://review.openstack.org/74157 | 21:14 |
| sdake | stevebaker ok, the help seems to imply it doesn't actually set a reasonable default | 21:14 |
| zaneb | sdake: the help is copied from Glance | 21:15 |
| stevebaker | sdake: to me it says that in the future you'll need an even bigger value since the service catalog size continues to grow | 21:15 |
| sdake | ok then I'll alter the -1 | 21:15 |
| sdake | although the help text needs love imo :) just because it was copied from glance doesn't mean they were doing it right :) | 21:16 |
| openstackgerrit | Steve Baker proposed a change to openstack/python-heatclient: REST method to fetch deployments metadata for a server https://review.openstack.org/74159 | 21:18 |
| sdake | zaneb how does that test_api-cfn_v1.py know to load EngineClient and construct it? | 21:21 |
| openstackgerrit | Jason Dunsmore proposed a change to openstack/heat: Remove spurious tracebacks from tests, part 2 https://review.openstack.org/74161 | 21:21 |
| sdake | i am getting: ./heat/tests/test_api_cfn_v1.py:992:1: F821 undefined name '_client'^M | 21:21 |
| sdake | self.m.StubOutWithMock(_client, 'call')^M | 21:21 |
| zaneb | sdake: you tell me, you said you had a _client object ;) | 21:22 |
| sdake | EngineClient creates it on init | 21:23 |
| zaneb | do you have a reference to the EngineClient in your unit test? | 21:23 |
| sdake | no | 21:23 |
| zaneb | that's what you need to get then | 21:23 |
| sdake | how do the tests operate as is without such reference? | 21:23 |
| sdake | eg, how are they getting a reference to "call" | 21:24 |
| openstackgerrit | A change was merged to openstack/heat-cfntools: Log curl stderr on non-zero exit status https://review.openstack.org/69245 | 21:25 |
| stevebaker | zaneb, sdake, could either of you approve? https://review.openstack.org/#/c/70289/ | 21:26 |
| sdake | stevebaker I already +2, someone else needs to approve probably zaneb or shardy | 21:26 |
| sdake | nm i gort it | 21:27 |
| sdake | zaneb can you do something like this: self.m.StubOutWithMock(rpc_client._client, 'call') | 21:27 |
| zaneb | yes | 21:27 |
| sdake | then on the "call" , I do rpc_client._client.call? | 21:28 |
| zaneb | if rpc_client._client is the RPCClient object that is going to get used | 21:28 |
| zaneb | yep | 21:28 |
| sdake | cool I'll try serach and replace for the 30th time today | 21:28 |
| *** alexheneveld has joined #heat | 21:28 | |
| * zaneb wishes all his unit test woes could be fixed by search and replace | 21:28 | |
| *** daneyon has joined #heat | 21:28 | |
| jpeeler | ok has anybody run the unit tests in havana recently? i'm unsure how master works, but stable doesn't | 21:29 |
| *** derekh has joined #heat | 21:31 | |
| *** yogesh has quit IRC | 21:35 | |
| *** alexheneveld has quit IRC | 21:36 | |
| *** yogesh has joined #heat | 21:36 | |
| openstackgerrit | A change was merged to openstack/python-heatclient: Add optional args arg back to do_list https://review.openstack.org/72181 | 21:37 |
| *** e0ne has joined #heat | 21:39 | |
| *** yogesh has quit IRC | 21:40 | |
| *** dims has joined #heat | 21:41 | |
| *** coolsvap has quit IRC | 21:44 | |
| *** e0ne has quit IRC | 21:44 | |
| *** asalkeld has joined #heat | 21:47 | |
| *** tomek_adamczewsk has joined #heat | 21:47 | |
| *** alexheneveld has joined #heat | 21:48 | |
| *** nati_ueno has quit IRC | 21:49 | |
| stevebaker | jpeeler: tests passed for me with tox. I've not tried it with packaged dependencies | 21:49 |
| *** nati_ueno has joined #heat | 21:49 | |
| jpeeler | stevebaker: this is the same thing i'm experiencing, https://bugs.launchpad.net/subunit/+bug/1278539 | 21:51 |
| stevebaker | jpeeler: testr list-tests fails for me too | 21:53 |
| stevebaker | my lazy fingers are happy "heat list" works again | 21:54 |
| jpeeler | well i can't run the unit tests either | 21:54 |
| *** jdob has quit IRC | 21:54 | |
| jpeeler | a backport has turned into an all day affair, which is just sad | 21:55 |
| *** jprovazn has quit IRC | 21:55 | |
| stevebaker | jpeeler: maybe lifeless can help with subunit issues | 21:55 |
| lifeless | hi | 21:56 |
| lifeless | stevebaker: did you see my thread with Mike from IBM about this on the list ? | 21:56 |
| *** tomek_adamczewsk has quit IRC | 21:56 | |
| *** tomek_adamczewsk has joined #heat | 21:57 | |
| stevebaker | jpeeler: http://lists.openstack.org/pipermail/openstack-dev/2014-February/027246.html | 21:57 |
| randallburt | stevebaker: how happy will your fingers be when its removed? It is deprecated after all ;) | 21:59 |
| stevebaker | randallburt: they would be happy strangling your neck D) | 22:00 |
| randallburt | lol | 22:00 |
| randallburt | remind me where you're staying again in ATL? Need to know what hotel *not* to pick :D | 22:01 |
| stevebaker | actually I've been playing with using ipython notebook to run heat commands, so I may never have to type it again | 22:01 |
| randallburt | stevebaker: good plan | 22:01 |
| stevebaker | ipython notebook is the shiz | 22:01 |
| *** spzala has joined #heat | 22:02 | |
| *** nati_ueno has quit IRC | 22:04 | |
| shardy | jpeeler: I don't have that issue, but I can see the backport I asked you do do isn't a straightforward backport, sorry about that | 22:05 |
| shardy | jpeeler: I'm taking a look at it now | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Refactor re-resolving of templates https://review.openstack.org/73814 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Re-resolve functions during stack update https://review.openstack.org/73815 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Prevent user introspection of Python objects https://review.openstack.org/73812 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Add a base class for pluggable functions https://review.openstack.org/73813 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Replace function resolution with lazy evaluation https://review.openstack.org/73822 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: unit tests: Fix types in Fn::ResourceFacade tests https://review.openstack.org/73820 | 22:05 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: unit tests: Refactor to accomodate lazy evaluation https://review.openstack.org/73821 | 22:06 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: unit tests: Always resolve static functions https://review.openstack.org/73818 | 22:06 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Fix test_association_eip unit test https://review.openstack.org/73819 | 22:06 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Calculate dependencies based on unresolved template https://review.openstack.org/73816 | 22:06 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Evaluate lazy functions in autoscaling launch config https://review.openstack.org/73817 | 22:06 |
| jpeeler | shardy: thanks, i think i could have finished it if i could have debugged locally | 22:07 |
| stevebaker | jpeeler: does tox work for you? having a clean venv might help | 22:08 |
| jpeeler | nothing works, and this was on a fresh install | 22:08 |
| jpeeler | well post devstack... | 22:08 |
| stevebaker | jpeeler: using tox should isolate that | 22:09 |
| *** aignatov is now known as aignatov_ | 22:10 | |
| lifeless | stevebaker: jpeeler: that bug is a dupe, and fixed. | 22:11 |
| sdake | zaneb this seems to not be working: @mock.patch.object(rpc_client._client, 'call') | 22:12 |
| sdake | any suggestions? | 22:12 |
| zaneb | sdake: I've never used mock, so I don't know | 22:12 |
| sdake | hrm, some tests are written using mox, and others using mock it appears | 22:13 |
| *** tomek_adamczewsk has quit IRC | 22:14 | |
| jpeeler | using tox i get 244 failures, so the tests run, but this can't be right | 22:15 |
| shardy | sdake: Are you setting the return value of the patched object? | 22:15 |
| shardy | sdake: test_api_openstack_v1.py has quite a few examples of mocking rpc responses | 22:16 |
| sdake | shardy it can't find the object at all, I think because it is constructed at __init_- time | 22:16 |
| sdake | shardy yes ,I am doing oslo.messaging | 22:16 |
| sdake | all that shit needs to be rewritten | 22:16 |
| sdake | which I am doing atm | 22:16 |
| shardy | sdake: Hmm, and yes the tests are a mess | 22:16 |
| sdake | no what I mean is all the tests need to be rewritten to support oslo.messaging | 22:16 |
| sdake | or reworked atleast | 22:17 |
| sdake | in EngineClient, i created an _client RPCCLient object | 22:17 |
| sdake | and use that throughout heat now | 22:17 |
| *** tomek_adamczewsk has joined #heat | 22:17 | |
| shardy | sdake: Oh, okay, well the tests are still a mess ;D | 22:17 |
| sdake | and now I need a way to mock the .call operation in _client | 22:17 |
| sdake | got Mock part to work, but the patch decorators appear to not work properly | 22:18 |
| openstackgerrit | A change was merged to openstack/heat: Raise the default max header to accommodate large tokens https://review.openstack.org/70289 | 22:18 |
| sdake | every patch decorator I am using fails because it can't find the _client object (which is created on init EngineClient object | 22:18 |
| zaneb | sdake: are you using mock or mox? I'm confused. | 22:19 |
| sdake | zaneb, I am using the existing tests | 22:19 |
| sdake | the existing tests use BOTH mox and mock apparently | 22:19 |
| zaneb | which use mox iirc | 22:19 |
| zaneb | orly? | 22:19 |
| *** topol has quit IRC | 22:20 | |
| sdake | srs, check out test_cloudwatch.py | 22:20 |
| sdake | it inherits HeatTestCase (mox) | 22:20 |
| sdake | and uses mock | 22:20 |
| shardy | AFAIK "the plan" was to move towards mock openstack wide, and we've seen a few patches converting tests, and new stuff using mock | 22:21 |
| shardy | but then I geuss folks have realized what an unimaginably huge amount of work it is and found other things to do ;) | 22:21 |
| shardy | as I've not seen (m)any porting to mock patches lately | 22:21 |
| *** tomek_adamczewsk has quit IRC | 22:22 | |
| stevebaker | for a greenfields test I use mock, but use mox when extending/copypasting existing tests | 22:22 |
| shardy | jpeeler: https://review.openstack.org/#/c/74129/2 | 22:23 |
| sdake | shardy so how do I beat mock into submission for this particular case? | 22:23 |
| sdake | I've got like 40 test cases failing in the same way | 22:23 |
| sdake | undo the patch decorators and muse Mock directory? | 22:23 |
| sdake | directly? | 22:23 |
| shardy | sdake: I'm sorry but I don't really understand your issue clearly enough to comment | 22:25 |
| zaneb | sdake: use "with mock.patch.object(rpc_client._client, 'call'):" instead of @mock.patch.object | 22:25 |
| * shardy is getting tired also ;) | 22:25 | |
| zaneb | it can be a decorator _or_ a context manager | 22:25 |
| sdake | interesting | 22:26 |
| zaneb | http://www.voidspace.org.uk/python/mock/patch.html#mock.patch | 22:26 |
| zaneb | or, rather, http://www.voidspace.org.uk/python/mock/patch.html#patch-object | 22:27 |
| sdake | zaneb can that be done with a function? | 22:28 |
| sdake | eg with @mock: | 22:28 |
| sdake | def abc | 22:28 |
| zaneb | no | 22:28 |
| sdake | that is the situation I got :( | 22:28 |
| jpeeler | shardy: thanks, i'm going to go ahead and put that patch in our package. the differences are more than i was expecting, so again thanks | 22:29 |
| zaneb | right, but it's a 2-line fix to each function plus some indenting changes | 22:29 |
| sdake | zaneb you mean put at top of def and reindenet the whole schebang | 22:29 |
| shardy | jpeeler: np, sorry it turned out to be a hassle | 22:29 |
| zaneb | sdake: yeah, exactly | 22:30 |
| zaneb | @mock.patch.object(messaging, 'call') | 22:30 |
| zaneb | def test_list_rmt_aterr(self, mock_call): | 22:30 |
| zaneb | becomes | 22:30 |
| sdake | ya got it I'll give it a go | 22:31 |
| zaneb | def test_list_rmt_aterr(self): | 22:32 |
| zaneb | with mock.patch.object(self.engine._rpcclient, 'call') as mock_call: | 22:32 |
| zaneb | and everything gets indented another level | 22:32 |
| openstackgerrit | A change was merged to openstack/python-heatclient: Fixes environment file using correct YAML format https://review.openstack.org/69802 | 22:34 |
| shardy | ha, found my routes issue - if the body doesn't match the content type it silently discards the data from the request routed to the controller, even though it still exists in the request object | 22:40 |
| shardy | sigh | 22:40 |
| *** vijendar has quit IRC | 22:40 | |
| *** yogesh has joined #heat | 22:46 | |
| *** derekh has quit IRC | 22:49 | |
| *** aweiteka has quit IRC | 22:50 | |
| *** yogesh has quit IRC | 22:51 | |
| stevebaker | zaneb, shardy, sdake, any takers for reviewing this? https://review.openstack.org/#/c/72591/ | 22:56 |
| zaneb | not today | 22:56 |
| openstackgerrit | Steven Dake proposed a change to openstack/heat: Use olso.messaging https://review.openstack.org/72798 | 22:58 |
| sdake | i'm taking a break for awhile been rolling since 5am - bbiaf | 22:58 |
| randallburt | radix: you about? | 23:02 |
| radix | randallburt: yo :) just started a VC meeting | 23:03 |
| randallburt | radix: k. quick question when you have a second. | 23:03 |
| radix | shoot | 23:04 |
| randallburt | radix: for the rackspace autoscale, does the launch config need image and flavor ID's or can I use names like the other resources? | 23:08 |
| radix | randallburt: umm, whatever nova accepts for those parameters | 23:11 |
| *** lindsayk has quit IRC | 23:11 | |
| radix | randallburt: it literally just passes them through to nova's server create request | 23:11 |
| radix | I think they're IDs | 23:11 |
| randallburt | radix: yeah, finally found the spot in the code. Looks like it just uses the ID's. | 23:11 |
| *** lindsayk has joined #heat | 23:13 | |
| *** alexheneveld has quit IRC | 23:16 | |
| *** asalkeld has quit IRC | 23:20 | |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Make Keystone client pluggable https://review.openstack.org/74190 | 23:20 |
| openstackgerrit | Richard Lee proposed a change to openstack/heat: Add Keystone V2 plugin https://review.openstack.org/74191 | 23:20 |
| *** alexpilotti has quit IRC | 23:30 | |
| *** asalkeld has joined #heat | 23:31 | |
| * zaneb shakes fist at Python 2.6 | 23:35 | |
| *** rcleere has quit IRC | 23:36 | |
| *** jergerber has quit IRC | 23:37 | |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: Replace function resolution with lazy evaluation https://review.openstack.org/73822 | 23:39 |
| openstackgerrit | Zane Bitter proposed a change to openstack/heat: unit tests: Refactor to accomodate lazy evaluation https://review.openstack.org/73821 | 23:39 |
| *** sabeen has quit IRC | 23:39 | |
| *** pablosan has quit IRC | 23:39 | |
| *** sgordon has quit IRC | 23:39 | |
| *** tims has quit IRC | 23:42 | |
| *** pablosan has joined #heat | 23:43 | |
| *** daneyon has quit IRC | 23:48 | |
| *** jcru has quit IRC | 23:52 | |
| *** alexheneveld has joined #heat | 23:53 | |
| *** jcru has joined #heat | 23:53 | |
| *** jcru has quit IRC | 23:57 | |
| *** andrew_plunk has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!