| *** tochi has quit IRC | 00:10 | |
| *** Marga__ has joined #heat | 00:15 | |
| *** Raj2 has quit IRC | 00:16 | |
| *** Marga_ has quit IRC | 00:17 | |
| *** e0ne has quit IRC | 00:18 | |
| *** tochi has joined #heat | 00:22 | |
| *** Drago has joined #heat | 00:26 | |
| *** Drago has quit IRC | 00:36 | |
| *** achanda has joined #heat | 00:46 | |
| *** achanda has quit IRC | 00:59 | |
| *** kse has quit IRC | 01:01 | |
| *** jaime49 has quit IRC | 01:03 | |
| *** gokrokve has joined #heat | 01:05 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack cli stack actions https://review.openstack.org/247804 | 01:06 |
|---|---|---|
| *** Marga__ has quit IRC | 01:06 | |
| *** Marga_ has joined #heat | 01:07 | |
| *** EricGonc_ has joined #heat | 01:13 | |
| *** rpothier has quit IRC | 01:14 | |
| *** gokrokve has quit IRC | 01:17 | |
| *** EricGonczer_ has quit IRC | 01:17 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add stack hook poll and clear to openstack client https://review.openstack.org/252656 | 01:19 |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack client resource type list and show https://review.openstack.org/254330 | 01:22 |
| *** pt_15 has quit IRC | 01:25 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: OSC plugin for stack snapshot create https://review.openstack.org/242938 | 01:31 |
| *** shakamunyi has quit IRC | 01:40 | |
| *** bana_k has quit IRC | 01:46 | |
| *** achanda has joined #heat | 02:00 | |
| *** sridhar_ram1 has quit IRC | 02:03 | |
| *** sid14726 has joined #heat | 02:11 | |
| *** Raj1 has joined #heat | 02:15 | |
| *** hdd has quit IRC | 02:19 | |
| *** alex_xu has quit IRC | 02:28 | |
| *** alex_xu has joined #heat | 02:30 | |
| *** Raj1 has quit IRC | 02:38 | |
| *** dims_ has quit IRC | 02:45 | |
| *** alex_xu has quit IRC | 02:49 | |
| *** alex_xu has joined #heat | 02:52 | |
| *** links has joined #heat | 03:00 | |
| *** hdd has joined #heat | 03:05 | |
| *** Marga_ has quit IRC | 03:05 | |
| *** EricGonc_ has quit IRC | 03:07 | |
| *** logan-- has joined #heat | 03:10 | |
| *** logan-- has quit IRC | 03:14 | |
| *** logan- has quit IRC | 03:22 | |
| *** logan- has joined #heat | 03:23 | |
| *** dims has joined #heat | 03:27 | |
| *** sridhar_ram has joined #heat | 03:33 | |
| *** dims has quit IRC | 03:37 | |
| *** gokrokve has joined #heat | 03:37 | |
| *** gokrokve_ has joined #heat | 03:43 | |
| *** gokrokve has quit IRC | 03:45 | |
| *** gokrokve has joined #heat | 03:48 | |
| *** gokrokve_ has quit IRC | 03:48 | |
| *** gokrokve has quit IRC | 03:53 | |
| *** gokrokve has joined #heat | 03:53 | |
| *** KanagarajM has joined #heat | 04:24 | |
| *** mragupat has joined #heat | 05:00 | |
| *** gokrokve_ has joined #heat | 05:05 | |
| *** mragupat has quit IRC | 05:06 | |
| *** gokrokve has quit IRC | 05:06 | |
| *** mragupat has joined #heat | 05:06 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat: Resolve name/id using translation rules https://review.openstack.org/265137 | 05:06 |
| *** shangxdy has joined #heat | 05:10 | |
| *** bana_k has joined #heat | 05:21 | |
| openstackgerrit | Merged openstack/python-heatclient: OpenStackClient plugin for stack output list https://review.openstack.org/267191 | 05:24 |
| openstackgerrit | Merged openstack/python-heatclient: OpenstackClient plugin for template show https://review.openstack.org/267701 | 05:26 |
| *** huangtianhua has quit IRC | 05:39 | |
| *** huangtianhua has joined #heat | 05:40 | |
| *** gokrokve_ has quit IRC | 05:42 | |
| *** gokrokve has joined #heat | 05:45 | |
| *** gokrokve has quit IRC | 05:48 | |
| *** gokrokve_ has joined #heat | 05:48 | |
| *** Marga_ has joined #heat | 05:50 | |
| *** bana_k has quit IRC | 05:51 | |
| *** Marga_ has quit IRC | 05:55 | |
| *** mohankumar_ has joined #heat | 06:00 | |
| *** gokrokve_ has quit IRC | 06:06 | |
| *** mohankumar_ has quit IRC | 06:06 | |
| openstackgerrit | Thomas Herve proposed openstack/heat: Handle missing plugin in is_service_available https://review.openstack.org/278081 | 06:12 |
| *** rakesh_hs has joined #heat | 06:15 | |
| *** mohankumar_ has joined #heat | 06:19 | |
| therve | ramishra, Hi! | 06:26 |
| therve | You should paste some tracebacks in bugs when you can, tempest logs are gone way too fast for bug tracking purposes | 06:27 |
| ramishra | therve: hi | 06:28 |
| ramishra | therve: ok | 06:28 |
| therve | Thanks for opening that bug, I just opened it myself | 06:28 |
| therve | ramishra, So CI people were not happy with that tempest change | 06:29 |
| ramishra | therve: yeah, I saw that | 06:29 |
| ramishra | therve: reg https://review.openstack.org/#/c/278081/ | 06:30 |
| therve | Yep? | 06:30 |
| *** spotz_zzz is now known as spotz | 06:31 | |
| ramishra | therve: what I meant is we should handle the client exception from is_service_available when listing resource types https://github.com/openstack/heat/blob/master/heat/engine/service.py#L1330-L1333 | 06:31 |
| therve | Hum, ok | 06:31 |
| therve | ramishra, Why do you think so? | 06:32 |
| *** shangxdy has quit IRC | 06:32 | |
| ramishra | therve: I saw the error and I think that's where it's coming from. We should not include the resource type if the there is any issue with the client. I'm looking at the log from your discussion yesterday. | 06:37 |
| therve | Hum | 06:38 |
| therve | ramishra, Right, but that won't prevent the resource from being used later on. | 06:38 |
| therve | At any rate, that may be reasonable, but I think it deserves a different patch | 06:39 |
| openstackgerrit | Thomas Herve proposed openstack/heat: Remove db refresh in Stack object get https://review.openstack.org/278225 | 06:39 |
| ramishra | therve: agreed, I can submit a patch later | 06:40 |
| therve | Thanks | 06:40 |
| ramishra | therve: I've commented on the tempest bug you created. | 06:40 |
| therve | Yeah I saw. Still not clear what the proper fix ought to be. Is it just about creating the role correctly? | 06:41 |
| therve | Because I thought it did that, but apparently not? | 06:41 |
| ramishra | yes, the role stack_owner_role/_member_ should be pre-created | 06:42 |
| ramishra | I think openstack-puppet CI is doing that | 06:42 |
| *** bana_k has joined #heat | 06:42 | |
| ramishra | whereas the tempest gate job was leverging the v2 functionality of it being implicitly created on first use. | 06:43 |
| ramishra | that's what my understanding is | 06:44 |
| *** Marga_ has joined #heat | 06:50 | |
| *** Marga_ has quit IRC | 06:50 | |
| *** Marga_ has joined #heat | 06:51 | |
| *** mohankumar_ has quit IRC | 06:51 | |
| *** tochi has quit IRC | 06:52 | |
| *** unlaudable has joined #heat | 06:57 | |
| *** steveg_afk has quit IRC | 06:58 | |
| *** mohankumar_ has joined #heat | 06:58 | |
| *** mohankumar_ has quit IRC | 06:59 | |
| openstackgerrit | Sirushti Murugesan proposed openstack/heat: Convergence: Load resources from DB for API calls https://review.openstack.org/278231 | 07:05 |
| *** gokrokve has joined #heat | 07:06 | |
| *** hdd has quit IRC | 07:08 | |
| skraynev | morning all | 07:09 |
| *** jprovazn has joined #heat | 07:12 | |
| *** spotz is now known as spotz_zzz | 07:13 | |
| *** gokrokve has quit IRC | 07:14 | |
| *** steveg_afk has joined #heat | 07:26 | |
| *** mragupat has quit IRC | 07:27 | |
| *** Marga_ has quit IRC | 07:33 | |
| therve | KanagarajM, Hey, around? To chat about https://review.openstack.org/278081 | 07:34 |
| openstackgerrit | Merged openstack/python-heatclient: Adds ignore_errors for template-validate command https://review.openstack.org/257211 | 07:42 |
| *** openstackgerrit has quit IRC | 07:47 | |
| *** openstackgerrit has joined #heat | 07:48 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat: Handle client exceptions in list_resource_types https://review.openstack.org/278239 | 07:48 |
| *** jtomasek_ has joined #heat | 07:51 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat: Remove server uuid property specific checks https://review.openstack.org/275098 | 07:51 |
| openstackgerrit | Rabi Mishra proposed openstack/heat: Cleanup neutron finder functions https://review.openstack.org/275246 | 07:51 |
| *** e0ne has joined #heat | 07:58 | |
| *** pt_15 has joined #heat | 08:03 | |
| *** fgimenez has joined #heat | 08:08 | |
| *** fgimenez has joined #heat | 08:08 | |
| *** achanda has quit IRC | 08:13 | |
| openstackgerrit | Thomas Herve proposed openstack/heat: Add post hooks https://review.openstack.org/252522 | 08:14 |
| *** palexster has joined #heat | 08:14 | |
| *** palexster has quit IRC | 08:14 | |
| *** d0ugal has joined #heat | 08:17 | |
| *** d0ugal has quit IRC | 08:17 | |
| *** d0ugal has joined #heat | 08:17 | |
| *** jistr|doc has joined #heat | 08:22 | |
| *** Guest66498 has joined #heat | 08:33 | |
| KanagarajM | therve, hi | 08:37 |
| *** achanda has joined #heat | 08:37 | |
| KanagarajM | therve, i was in lunch break, missed to see your message | 08:37 |
| *** e0ne has quit IRC | 08:39 | |
| *** gokrokve has joined #heat | 08:40 | |
| therve | KanagarajM, replied in the review | 08:41 |
| KanagarajM | therve, ok | 08:41 |
| *** Drago has joined #heat | 08:43 | |
| *** Drago has quit IRC | 08:44 | |
| *** Drago has joined #heat | 08:44 | |
| *** gokrokve has quit IRC | 08:44 | |
| *** jaosorior has joined #heat | 08:48 | |
| *** Drago has quit IRC | 09:00 | |
| *** jcoufal has joined #heat | 09:03 | |
| *** zeih_ has joined #heat | 09:09 | |
| *** shardy has joined #heat | 09:11 | |
| therve | KanagarajM, Can we chat here now? :) | 09:29 |
| *** tochi has joined #heat | 09:30 | |
| *** bana_k has quit IRC | 09:32 | |
| *** Marga_ has joined #heat | 09:33 | |
| *** Marga_ has quit IRC | 09:38 | |
| *** gokrokve has joined #heat | 09:41 | |
| *** jistr|doc has quit IRC | 09:46 | |
| *** gokrokve has quit IRC | 09:48 | |
| *** derekh has joined #heat | 09:50 | |
| *** yassine has joined #heat | 09:51 | |
| openstackgerrit | Merged openstack/heat: Resolve name/id using translation rules https://review.openstack.org/265137 | 10:00 |
| *** tochi has quit IRC | 10:00 | |
| *** achanda has quit IRC | 10:06 | |
| *** getvasanth has joined #heat | 10:06 | |
| *** e0ne has joined #heat | 10:07 | |
| *** achanda has joined #heat | 10:08 | |
| KanagarajM | therve, hi, sure | 10:09 |
| *** kmARC has quit IRC | 10:09 | |
| therve | KanagarajM, So I don't really understand what your issue is | 10:09 |
| therve | Or what you're suggesting | 10:09 |
| KanagarajM | ok, let me explain here. | 10:10 |
| KanagarajM | existing is_service_available() checks for whether service got listed in keystone catalog and if service is not listed then ResourceTypeUnavailable() exception is raised. | 10:11 |
| KanagarajM | this exception currently says: Service for resoruce type xxx is not available, reason: keystone catalog does not list the service ..... | 10:12 |
| KanagarajM | but now, we are trying to add one more logic that, if required client is missing, then consider that as service not available. | 10:12 |
| therve | I don't think thats true | 10:13 |
| therve | It will say 'Service endpoint not in service catalog.' | 10:13 |
| therve | KanagarajM, Is that your issue? That the message will be wrong? | 10:14 |
| KanagarajM | here, i was suggesting that, raise the ResourceTypeUnavalible() with different reason saying client is not loaded and change this exception message to 'Resrource type xxx is not availeble and reason: as appropirate based on either keystone catalog miss or client miss | 10:14 |
| therve | So is_service_available doesn't raise that exception | 10:15 |
| therve | In master right now, it will raise a ResourceTypeUnavalaible error, but the message will be 'NoneType has no attribute service_types' | 10:15 |
| therve | I'd argue my patch is an improvement | 10:15 |
| KanagarajM | your patch is surely arresting the issue, but would like to suggest to raise exception with proper reasoning. | 10:17 |
| therve | OK, that's fair, let me cook up something | 10:18 |
| KanagarajM | therve, sure. :) | 10:20 |
| *** jaosorior has quit IRC | 10:21 | |
| *** atuvenie has joined #heat | 10:22 | |
| openstackgerrit | Thomas Herve proposed openstack/heat: Handle missing plugin in is_service_available https://review.openstack.org/278081 | 10:23 |
| therve | KanagarajM ^^^ when you have a few mins | 10:24 |
| *** shardy has quit IRC | 10:25 | |
| *** shardy has joined #heat | 10:25 | |
| KanagarajM | therve, changes are fine for this patch, but we may need to correct the ResourceTypeUnavailable exception message to 'Resource type xxx is not available' | 10:35 |
| therve | KanagarajM, From what? | 10:39 |
| KanagarajM | therve, current message in that exception | 10:39 |
| KanagarajM | "Service %(service_name)s is not available for resource type %(resource_type)s" | 10:40 |
| therve | I don't know, it's not totally unreasonable | 10:41 |
| therve | With the reason being explanatory | 10:41 |
| *** achanda has quit IRC | 10:45 | |
| *** dims has joined #heat | 10:48 | |
| *** Marga_ has joined #heat | 10:52 | |
| openstackgerrit | Thomas Herve proposed openstack/heat: Handle missing plugin in is_service_available https://review.openstack.org/278081 | 10:55 |
| therve | ramishra, So I ended up stealing your patch | 10:55 |
| therve | Sigh | 10:55 |
| therve | ramishra, Thanks for the test :) | 10:55 |
| *** Marga_ has quit IRC | 10:56 | |
| ramishra | therve: np:) Is catching ClientNotAvailable going be enough? I think there can be other exceptions like auth exceptions. no? | 10:58 |
| therve | ramishra, Yeah I added more | 10:58 |
| therve | Like ConnectionFailed and stuff | 10:58 |
| therve | I really stole your patch :) | 10:58 |
| therve | The unit test was failing otherwise | 10:58 |
| *** fgimenez has quit IRC | 11:02 | |
| *** fgimenez has joined #heat | 11:03 | |
| *** Marga_ has joined #heat | 11:31 | |
| *** atuvenie has quit IRC | 11:36 | |
| *** achanda has joined #heat | 11:45 | |
| *** Marga_ has quit IRC | 11:46 | |
| *** Marga_ has joined #heat | 11:46 | |
| *** achanda has quit IRC | 11:50 | |
| *** xek has quit IRC | 11:53 | |
| *** sid14726 has quit IRC | 11:54 | |
| *** permalac has joined #heat | 11:55 | |
| *** links has quit IRC | 11:56 | |
| *** Guest66498 has quit IRC | 12:03 | |
| *** EricGonczer_ has joined #heat | 12:31 | |
| *** fgimenez has quit IRC | 12:31 | |
| *** fgimenez has joined #heat | 12:32 | |
| *** EricGonczer_ has quit IRC | 12:36 | |
| *** links has joined #heat | 12:36 | |
| *** Guest66498 has joined #heat | 12:38 | |
| *** krotscheck_dcm is now known as krotscheck | 12:38 | |
| therve | ramishra, Regarding that tempest stuff, what should we do in the end? | 12:39 |
| therve | The issue was that devstack was setup using _member_ role even in v3, correct? | 12:39 |
| therve | Should we set the role to '' in devstack instead? | 12:40 |
| therve | That would mean not passing the role in tempest | 12:43 |
| *** vgridnev has joined #heat | 12:44 | |
| ramishra | therve: I don't know what is the issue tempest team has for https://review.openstack.org/#/c/274346/5/tempest/api/orchestration/base.py | 12:47 |
| ramishra | it's conditional create and would work on both devstack and non-devstack cases | 12:48 |
| *** achanda has joined #heat | 12:48 | |
| therve | ramishra, So it's not the "tempest team", it's mtreinish | 12:48 |
| therve | And I guess you should ask him for more details | 12:49 |
| *** shangxdy has joined #heat | 12:49 | |
| ramishra | I've mentioned that in the bug | 12:49 |
| ramishra | he just -2ed a patch and vanished to thin air | 12:49 |
| ramishra | no response to my queries | 12:50 |
| therve | Well he was here yesterday night, so hopefully he'll be around again | 12:50 |
| therve | The build logs are gone, but where the tests indeed failing? | 12:50 |
| therve | ramishra, That said, I still think it sucks that we need to create the _member_ role in v3 | 12:52 |
| therve | It doesn't feel super nice | 12:52 |
| openstackgerrit | Merged openstack/heat: Handle missing plugin in is_service_available https://review.openstack.org/278081 | 12:53 |
| *** achanda has quit IRC | 12:54 | |
| ramishra | They should change it to something else in devstack, we should get https://review.openstack.org/#/c/274612/ merged | 12:54 |
| therve | ramishra, So this change needs your tempest patch to be reverted though, right? | 12:55 |
| ramishra | assuming non-devstack CIs would create stack_owner_role before hand. Having said that they do create _member_ in other projects. https://bugs.launchpad.net/openstack-ansible/+bug/1474916 | 12:55 |
| openstack | Launchpad bug 1474916 in openstack-ansible trunk "Keystone v3 does not automatically create _member_ role" [High,Fix released] - Assigned to Ian Cordasco (icordasc) | 12:56 |
| therve | Well that's fine, non-devstack ci can configured tempest appropriately. They ought to create the role anyway AFAIU | 12:57 |
| *** sid14726 has joined #heat | 12:57 | |
| ramishra | it can work with or without the tempest patch | 12:57 |
| ramishra | I mean for devstack | 12:58 |
| ramishra | No sorry, we 've to revert the tempest patch | 12:58 |
| therve | Ah :) | 12:59 |
| therve | OK that sounds good to me. | 12:59 |
| therve | So let's have https://review.openstack.org/278068 in (though it doesn't pass the gate, ah) | 13:00 |
| ramishra | yeah, I thought adding the roles it the tempest tests is easy. It's test end of the day and addition of the role is conditional. | 13:01 |
| * therve shrugs | 13:01 | |
| *** Guest66498 has quit IRC | 13:02 | |
| therve | It's still somewhat racy, I guess | 13:04 |
| *** sdake has joined #heat | 13:04 | |
| *** jcoufal has quit IRC | 13:07 | |
| *** dprince has joined #heat | 13:09 | |
| *** e0ne has quit IRC | 13:09 | |
| *** vgridnev has quit IRC | 13:09 | |
| *** e0ne has joined #heat | 13:09 | |
| *** pt_15 has quit IRC | 13:15 | |
| *** kmARC has joined #heat | 13:16 | |
| *** Guest66498 has joined #heat | 13:17 | |
| *** vgridnev has joined #heat | 13:19 | |
| *** vgridnev has quit IRC | 13:21 | |
| shardy | Hey all, if anyone has a moment to review https://review.openstack.org/#/c/275602/ I'd appreciate it | 13:22 |
| shardy | simple adjustment to str_replace validation, which we need for a TripleO patch | 13:23 |
| *** palexster has joined #heat | 13:23 | |
| *** e0ne has quit IRC | 13:24 | |
| *** EricGonczer_ has joined #heat | 13:25 | |
| *** e0ne has joined #heat | 13:25 | |
| *** sdake_ has joined #heat | 13:25 | |
| *** vgridnev has joined #heat | 13:27 | |
| *** sdake has quit IRC | 13:29 | |
| *** EricGonczer_ has quit IRC | 13:29 | |
| *** Marga_ has quit IRC | 13:29 | |
| *** hdd has joined #heat | 13:33 | |
| *** ahelkhou has joined #heat | 13:37 | |
| *** xek has joined #heat | 13:38 | |
| *** KanagarajM has quit IRC | 13:39 | |
| *** permalac has quit IRC | 13:39 | |
| *** dyasny has quit IRC | 13:41 | |
| *** dyasny has joined #heat | 13:42 | |
| *** sdake_ has quit IRC | 13:48 | |
| *** gokrokve has joined #heat | 13:49 | |
| *** rakesh_hs has quit IRC | 13:50 | |
| *** gokrokve has quit IRC | 13:53 | |
| *** shangxdy has quit IRC | 14:04 | |
| *** jonesbr has joined #heat | 14:05 | |
| *** dyasny has quit IRC | 14:06 | |
| *** dyasny has joined #heat | 14:06 | |
| *** Delair has joined #heat | 14:06 | |
| *** fgimenez has quit IRC | 14:08 | |
| *** fgimenez has joined #heat | 14:10 | |
| *** neelashah has joined #heat | 14:17 | |
| *** sabeen has joined #heat | 14:19 | |
| *** sabeen3 has joined #heat | 14:22 | |
| *** permalac has joined #heat | 14:22 | |
| Delair | @all Hi can i define an optional parameter in HOT template?.. for example if I configure a parameter to define a network_gateway_IP and want to leave the field blank when deploying the template | 14:22 |
| ryansb | Delair: you can set a default | 14:22 |
| ryansb | e.g: mynetwork: | 14:23 |
| ryansb | type: string | 14:23 |
| Delair | @ryansb.. Yes I can but i dont want to :) | 14:23 |
| ryansb | default: '' | 14:23 |
| ryansb | that way you could leave it out, and you don't need to pass in the param if you don't want to | 14:23 |
| *** sabeen has quit IRC | 14:24 | |
| Delair | ohh sorry you mean just define default with empty colons? | 14:24 |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OSC plugin for stack resource show and list https://review.openstack.org/244927 | 14:24 |
| ryansb | yeah, you'd just define the default as an empty string | 14:24 |
| Delair | @ryansb ohh sorry you mean just define default with empty colons? | 14:24 |
| Delair | @ryan ok cool let me test.. Thanks a lot | 14:24 |
| ryansb | so that way the template will work without you supplying a value, but you can always override. | 14:24 |
| *** sid14726 has quit IRC | 14:25 | |
| Delair | @ryansb working .. Thanks a lot | 14:26 |
| ryansb | no problem :) | 14:26 |
| *** links has quit IRC | 14:27 | |
| openstackgerrit | Steven Hardy proposed openstack/heat: Add user_data_update_policy property to OS::Nova::Server https://review.openstack.org/274149 | 14:27 |
| openstackgerrit | Steven Hardy proposed openstack/heat: Make Server user_data update policy configurable https://review.openstack.org/274101 | 14:28 |
| openstackgerrit | Merged openstack/heat: Allow str_replace to reference parameters/attributes https://review.openstack.org/275602 | 14:28 |
| *** hdd has quit IRC | 14:29 | |
| *** crose has joined #heat | 14:31 | |
| krotscheck | I've got a middleware configuration update up for review, are there any cores that can help me puzzle out the appropriate headers? https://review.openstack.org/#/c/265391/ | 14:31 |
| *** prazumovsky has joined #heat | 14:32 | |
| *** jcoufal has joined #heat | 14:33 | |
| dmsimard | therve, ramishra: I can change the commit message but I otherwise don't understand ramishra's comment | 14:34 |
| *** Delair has quit IRC | 14:34 | |
| *** jreeves has quit IRC | 14:34 | |
| *** rpothier has joined #heat | 14:35 | |
| *** Delair has joined #heat | 14:36 | |
| therve | dmsimard, His point is that the tests will fail with devstack v3 changes. We need another change in devstack later on to make this work, we discussed it here | 14:37 |
| Delair | @ryansb.. Sorry it is still not working.. throws an error message "Bad router request: Subnet for router interface must have a gateway IP" | 14:37 |
| therve | So I don't think you have anything else to do except press tempest guys to review it :) | 14:37 |
| dmsimard | therve: okay, it just wasn't clear if there was anything expected of us | 14:37 |
| dmsimard | I'll submit the edited commit. | 14:37 |
| ryansb | Delair: that may be because you're using the parameter somewhere that has stricter requirements | 14:37 |
| therve | Cools thanks | 14:38 |
| Delair | @ryansb.. It is working if i define an ip or remove that parameter option competely then it automatically take .1 as a gateway | 14:38 |
| ryansb | e.g. some parameters must be valid IP addresses | 14:38 |
| *** sdake has joined #heat | 14:38 | |
| ryansb | which sounds like the case for you | 14:38 |
| *** swygue has quit IRC | 14:38 | |
| ryansb | is there a sane default you can have for gateway? Like 192.168.1.1 or something? | 14:39 |
| Delair | @ryansb but i assume that if i leave the field blank then it auto takes .1 as a gateway | 14:39 |
| Delair | @ryansb yes i am using it in a subnet resource | 14:39 |
| dmsimard | therve: done, thanks | 14:39 |
| dmsimard | mtreinish: I think https://review.openstack.org/#/c/278068/ will be good to go now | 14:39 |
| Delair | @ryansb .. so i have a parameter defined as "net_gateway" and associates that in "net_subnet" as "gateway_ip: get_param: net_gateway" | 14:40 |
| Delair | @ryansb and when deploying this template i want to leave this field blank so that it automatically takes .1 as a gateway | 14:41 |
| dmsimard | therve: do you think there's something other than https://review.openstack.org/#/c/278081/ that we need or is that sorted out ? (saw the discussion around that in the bug) | 14:41 |
| ryansb | if you look at the docs http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Neutron::Subnet it looks like it should take an empty string | 14:42 |
| ryansb | what heat version are you on? | 14:42 |
| Delair | @ryansb .. 2015-10-15 | 14:42 |
| therve | dmsimard, I think we're okay on your side | 14:42 |
| dmsimard | therve: Okay I'll give CI a shot once that revert merges thanks | 14:43 |
| Delair | @ryan.. it doc doesnt mention about leaving a field blank | 14:45 |
| ryansb | yeah it does, under the gateway_ip parameter docs "The gateway IP address. Set to any of [ null | ~ | “” ] to create/update a subnet without a gateway | 14:46 |
| ryansb | " | 14:46 |
| *** gokrokve has joined #heat | 14:50 | |
| Delair | @ryan yes but i do want to define gateway but dont want to manually input the IP in the field when deploying template | 14:50 |
| *** jpeeler has joined #heat | 14:50 | |
| Delair | @ryan the reason is that i am conf a template with many parameters but want to use some of then as optional for example if i want gateway ip to .20 then i input that in the gateway_ip field otherwise leave it blank and it auto takes .1 | 14:51 |
| *** achanda has joined #heat | 14:52 | |
| *** gokrokve has quit IRC | 14:56 | |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: Separate get_live_state on several methods https://review.openstack.org/276729 | 14:57 |
| *** achanda has quit IRC | 14:57 | |
| *** gokrokve has joined #heat | 14:58 | |
| *** getvasanth has quit IRC | 14:59 | |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: Glance::Image get_live_state implementation https://review.openstack.org/250705 | 15:00 |
| *** jasond has joined #heat | 15:00 | |
| *** unlaudable has quit IRC | 15:02 | |
| *** pas-ha has quit IRC | 15:03 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: OSC plugin for stack snapshot create https://review.openstack.org/242938 | 15:04 |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: OSC plugin for stack snapshot show https://review.openstack.org/242888 | 15:04 |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: OSC plugin for stack snapshot delete https://review.openstack.org/275403 | 15:04 |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack cli stack actions https://review.openstack.org/247804 | 15:06 |
| *** vijendar has joined #heat | 15:08 | |
| *** palexster has quit IRC | 15:08 | |
| ryansb | Delair: you could have 2 parameters that you join to form the IP address | 15:09 |
| *** vgridnev has quit IRC | 15:09 | |
| *** NobodyCa1 has joined #heat | 15:09 | |
| ryansb | so you'd have the default gateway be the network address + .1 by default | 15:09 |
| *** vgridnev has joined #heat | 15:10 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add stack hook poll and clear to openstack client https://review.openstack.org/252656 | 15:12 |
| *** permalac has quit IRC | 15:13 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack client stack snapshot restore https://review.openstack.org/267731 | 15:13 |
| *** NobodyCa1 has quit IRC | 15:14 | |
| Delair | @ryansb.. So what do u mean ? | 15:14 |
| ryansb | would you be able to paste me the parameters section of your template? I think it would be easier to show you | 15:15 |
| *** daneyon has joined #heat | 15:16 | |
| *** daneyon__ has joined #heat | 15:17 | |
| *** vgridnev has quit IRC | 15:17 | |
| Delair | @ryansb.. let me send you a private note | 15:18 |
| *** pas-ha has joined #heat | 15:18 | |
| *** gokrokve has quit IRC | 15:19 | |
| *** daneyon_ has quit IRC | 15:20 | |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: Glance::Image get_live_state implementation https://review.openstack.org/250705 | 15:20 |
| *** daneyon has quit IRC | 15:21 | |
| *** fgimenez has quit IRC | 15:22 | |
| *** jaosorior has joined #heat | 15:24 | |
| *** palexster has joined #heat | 15:24 | |
| *** daneyon has joined #heat | 15:24 | |
| *** fgimenez has joined #heat | 15:25 | |
| *** daneyon__ has quit IRC | 15:27 | |
| *** mragupat has joined #heat | 15:30 | |
| *** EricGonczer_ has joined #heat | 15:30 | |
| *** daneyon_ has joined #heat | 15:32 | |
| *** EricGonc_ has joined #heat | 15:32 | |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack client resource type list and show https://review.openstack.org/254330 | 15:32 |
| *** vgridnev has joined #heat | 15:33 | |
| *** pas-ha has quit IRC | 15:33 | |
| *** spotz_zzz is now known as spotz | 15:33 | |
| *** e0ne has quit IRC | 15:34 | |
| *** daneyon has quit IRC | 15:35 | |
| *** EricGonczer_ has quit IRC | 15:35 | |
| *** e0ne has joined #heat | 15:35 | |
| *** jaosorior has quit IRC | 15:36 | |
| *** gokrokve has joined #heat | 15:43 | |
| *** vgridnev has quit IRC | 15:45 | |
| *** vgridnev has joined #heat | 15:45 | |
| *** pas-ha has joined #heat | 15:45 | |
| *** gokrokve has quit IRC | 15:47 | |
| *** palexster has quit IRC | 15:52 | |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: Cinder::VolumeType get_live_state implementation https://review.openstack.org/250451 | 15:56 |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: Cinder::Volume get_live_state implementation https://review.openstack.org/261445 | 15:57 |
| openstackgerrit | Peter Razumovsky proposed openstack/heat: EncryptedVolumeType get_live_state implementation https://review.openstack.org/276730 | 15:57 |
| *** palexster has joined #heat | 15:57 | |
| *** sid14726 has joined #heat | 16:01 | |
| *** gokrokve has joined #heat | 16:04 | |
| *** NobodyCa1 has joined #heat | 16:05 | |
| *** jaosorior_ has joined #heat | 16:09 | |
| *** hdd has joined #heat | 16:09 | |
| *** NobodyCa1 has quit IRC | 16:11 | |
| *** Drago has joined #heat | 16:11 | |
| *** Drago1 has joined #heat | 16:12 | |
| *** dmsimard has left #heat | 16:12 | |
| *** Drago1 has quit IRC | 16:13 | |
| *** Drago1 has joined #heat | 16:13 | |
| *** Drago has quit IRC | 16:15 | |
| *** NobodyCa1 has joined #heat | 16:19 | |
| *** asalkeld has quit IRC | 16:20 | |
| *** asalkeld has joined #heat | 16:21 | |
| *** NobodyCa1 has quit IRC | 16:23 | |
| *** abramley has left #heat | 16:25 | |
| *** vijendar has quit IRC | 16:32 | |
| *** hongbin has joined #heat | 16:32 | |
| *** boris-42 has quit IRC | 16:33 | |
| openstackgerrit | Oleksii Chuprykov proposed openstack/heat: Retry on DBDeadlock https://review.openstack.org/278499 | 16:34 |
| *** Raj1 has joined #heat | 16:34 | |
| *** vgridnev has quit IRC | 16:36 | |
| hongbin | Hi Heat team, it looks after this commit https://github.com/openstack/heat/commit/0388da504765f9975aeadc509672c0a47bbcc244 , magnum gate is broken with error: "Failed to validate: resources[0]: Unable to find network with name 'None'" | 16:36 |
| hongbin | Could anyone help to verify it? | 16:36 |
| *** steveg_afk has quit IRC | 16:42 | |
| *** jtomasek_ has quit IRC | 16:42 | |
| hongbin | stevebaker: zaneb: shardy: asalkeld: ping | 16:43 |
| zaneb | hongbin: I take it https://review.openstack.org/#/c/275832/ doesn't help? | 16:45 |
| hongbin | zaneb: looking at it | 16:46 |
| hongbin | zaneb: It looks that doesn't help | 16:47 |
| hongbin | zaneb: the gate already include that commit, still break... | 16:47 |
| hongbin | Here is the log: http://logs.openstack.org/61/277961/1/gate/gate-functional-dsvm-magnum-api/db02160/logs/screen-h-api.txt | 16:48 |
| ramishra | hongbin: hi, why it's using a network like 'None' I mean network name None as a string? Are you using it somewhere in your templates? | 16:48 |
| hongbin | ramishra: No idea, I cannot reproduce the error locally | 16:50 |
| *** vijendar has joined #heat | 16:50 | |
| hongbin | It only happened on gate | 16:50 |
| ramishra | hongbin: you mean using the patch you mentioned, that broke your gate | 16:50 |
| ramishra | hongbin: are you using latest heat master locally? | 16:51 |
| hongbin | ramishra: I was not using the latest heat locally, gate used the latest heat | 16:52 |
| hongbin | That is the difference I guess | 16:52 |
| ramishra | yes, I think you're using 'None' (string) somewhere, may be coming from the parameter to the network property. | 16:54 |
| hongbin | The gate broke right after this commit (https://github.com/openstack/heat/commit/0388da504765f9975aeadc509672c0a47bbcc244) merged, which looks like the culprit from me | 16:54 |
| hongbin | sure. will check that | 16:54 |
| ramishra | hongbin: that's my patch;) | 16:54 |
| *** achanda has joined #heat | 16:55 | |
| hongbin | oh | 16:55 |
| hongbin | Yes, I will try to investigate on my side about how the "None" get passed | 16:57 |
| ramishra | ok | 16:57 |
| hongbin | but it would be great that Heat can tolerate this inputs as before. | 16:58 |
| *** jaosorior_ has quit IRC | 16:58 | |
| zaneb | it's probably "None" (the singleton), not "'None'" the string | 16:59 |
| zaneb | they look the same when you print 'em | 16:59 |
| *** achanda has quit IRC | 16:59 | |
| *** jaosorior_ has joined #heat | 17:00 | |
| hongbin | zaneb: ack | 17:01 |
| *** Raj1 has quit IRC | 17:01 | |
| hongbin | zaneb: would appreciate a quick fix if it can be done | 17:02 |
| hongbin | I can help if anything needs to be fixed in Magnum side | 17:02 |
| zaneb | hongbin: I don't have time right now, but I think documenting the template that was used and any parameters that were passed would be helpful | 17:03 |
| zaneb | feel free to raise a Heat bug with those details and link it here | 17:03 |
| hongbin | sure | 17:04 |
| zaneb | I'm sure someone will be able to get on it | 17:04 |
| *** jaime49 has joined #heat | 17:05 | |
| *** sdake has quit IRC | 17:08 | |
| *** e0ne has quit IRC | 17:08 | |
| *** sid14726 has quit IRC | 17:09 | |
| *** steveg_afk has joined #heat | 17:09 | |
| *** jaime49 has quit IRC | 17:12 | |
| *** jreeves has joined #heat | 17:13 | |
| *** jcoufal has quit IRC | 17:13 | |
| therve | ramishra, Isn't the issue this: https://github.com/openstack/heat/blob/master/heat/engine/resources/openstack/neutron/neutron.py#L104 ? | 17:14 |
| therve | IE the thing with https://bugs.launchpad.net/heat/+bug/1533356 | 17:15 |
| openstack | Launchpad bug 1533356 in heat "Stack update preview fails with "NeutronClientException: Unable to find network with name 'None'"" [Undecided,In progress] - Assigned to Jason Dunsmore (jasondunsmore) | 17:15 |
| *** vgridnev has joined #heat | 17:15 | |
| *** yassine has quit IRC | 17:16 | |
| ramishra | therve: looks like, but I was wondering why it started happening for them after the above patch landed. | 17:16 |
| ramishra | therve: we don't see that error in our gate | 17:17 |
| therve | Presumably because we're missing a test :) | 17:17 |
| therve | I don't know if it's related to your patch, FWIW. Just pointing that there is indeed an issue. | 17:19 |
| *** zeih_ has quit IRC | 17:20 | |
| *** jcoufal has joined #heat | 17:20 | |
| ramishra | therve: I assume we've neutron resource references in many of out templates used for func/scenario tests | 17:21 |
| *** thumpba has joined #heat | 17:21 | |
| therve | Some, sure. Not covering the whole thing. | 17:21 |
| *** steveg_afk has quit IRC | 17:22 | |
| *** prazumovsky has quit IRC | 17:23 | |
| *** jcoufal has quit IRC | 17:23 | |
| *** vijendar has quit IRC | 17:24 | |
| ramishra | therve: magnum guys do some funky stuff with dynamic parameters to templates they use ex. https://github.com/openstack/magnum/blob/master/magnum/conductor/template_definition.py#L70 | 17:27 |
| *** jaosorior_ has quit IRC | 17:28 | |
| *** Raj1 has joined #heat | 17:30 | |
| openstackgerrit | Zane Bitter proposed openstack/heat: Ignore errors on old properties during update https://review.openstack.org/277984 | 17:31 |
| *** Raj1 has quit IRC | 17:32 | |
| *** vijendar has joined #heat | 17:32 | |
| *** sdake has joined #heat | 17:33 | |
| *** EricGonc_ has quit IRC | 17:34 | |
| *** boris-42 has joined #heat | 17:39 | |
| therve | Can we do something like donate to a charity when you do a dumb recheck? | 17:45 |
| therve | I'd like a better punishment than a slap in the face | 17:45 |
| *** jtomasek has joined #heat | 17:48 | |
| *** fgimenez has quit IRC | 17:51 | |
| *** sid14726 has joined #heat | 17:55 | |
| openstackgerrit | Zane Bitter proposed openstack/heat: Ignore errors on old properties during update https://review.openstack.org/277984 | 17:57 |
| *** achanda has joined #heat | 17:57 | |
| *** derekh has quit IRC | 18:01 | |
| *** achanda has quit IRC | 18:01 | |
| *** sridhar_ram has joined #heat | 18:04 | |
| *** zeih_ has joined #heat | 18:05 | |
| *** vijendar has quit IRC | 18:05 | |
| *** vijendar has joined #heat | 18:06 | |
| *** achanda has joined #heat | 18:12 | |
| *** steveg_afk has joined #heat | 18:13 | |
| *** achanda has quit IRC | 18:15 | |
| *** bana_k has joined #heat | 18:18 | |
| *** Raj1 has joined #heat | 18:18 | |
| *** e0ne has joined #heat | 18:20 | |
| *** Marga_ has joined #heat | 18:21 | |
| *** zeih_ has quit IRC | 18:25 | |
| *** dmsimard has joined #heat | 18:28 | |
| dmsimard | therve: heya | 18:28 |
| dmsimard | therve: so I think we're close with the revert and your patch, there seems to be one more thing missing | 18:29 |
| dmsimard | see: http://logs.openstack.org/20/278520/1/check/gate-puppet-openstack-integration-scenario002-tempest-dsvm-centos7/6f82bf6/console.html#_2016-02-10_17_56_41_885 | 18:30 |
| *** russellb has joined #heat | 18:32 | |
| russellb | quick question ... what's the status of heat "convergence" ? | 18:32 |
| russellb | shardy: around? ^ | 18:33 |
| shardy | russellb: much of the initial internal implementation has landed, and we have CI tests running against it | 18:33 |
| russellb | great, thanks. | 18:33 |
| shardy | however it's still viewed as somewhat experimental and isn't yet enabled by default | 18:33 |
| russellb | fair enough. I ask because I see overlap in a new project proposal that's already using heat | 18:34 |
| russellb | and this project is new and experimental too | 18:34 |
| russellb | (tacker fwiw) | 18:34 |
| shardy | russellb: interesting, got a link handy? | 18:34 |
| russellb | it has drivers for different methods of monitoring VMs, and then configurable policy for reacting to failures | 18:34 |
| russellb | i thought of heat convergence | 18:34 |
| russellb | since they're already using heat for VM lifecycle management | 18:34 |
| dmsimard | ramishra: Can you see what I sent to therve above ? | 18:34 |
| russellb | shardy: https://review.openstack.org/#/c/276417/3 | 18:35 |
| russellb | shardy: though i'm in the middle of writing up a pretty lengthy response where this is a part of it | 18:35 |
| *** vijendar has quit IRC | 18:35 | |
| russellb | i can ping again when my comments are posted | 18:36 |
| shardy | oh wow they use TOSCA templates too | 18:36 |
| russellb | yeah.. | 18:36 |
| *** adrian_otto has joined #heat | 18:37 | |
| shardy | russellb: the heat weekly meeting is at 20.00 UTC today, I'll mention this then | 18:37 |
| russellb | thanks | 18:38 |
| adrian_otto | shardy: It looks like this commit broke our gate: https://review.openstack.org/265137 | 18:38 |
| russellb | the comment i'm posting will have a less telco heavy explanation of what the project does that might be useful, heh | 18:38 |
| shardy | adrian_otto: Yes, sorry about that ramishra is looking into it | 18:38 |
| adrian_otto | ok | 18:38 |
| *** sridhar_ram is now known as sridhar_ram_afk | 18:40 | |
| shardy | therve: Hey were you also looking at the magnum breakage? | 18:41 |
| shardy | I guess it's pretty late for ramishra now, perhaps we should revert and work out the fix afterwards? | 18:42 |
| *** vijendar has joined #heat | 18:43 | |
| shardy | russellb: thanks - when we have that summary, we can probably help with a summary of the current heat scope, and in particular to what extent convergence will eventually address things like auto-healing | 18:43 |
| *** palexster has quit IRC | 18:43 | |
| russellb | thanks | 18:43 |
| *** sridhar_ram_afk has quit IRC | 18:44 | |
| shardy | adrian_otto: Do you have an existing bug for this? | 18:46 |
| shardy | I was going to add heat, and propose a revert if we can't figure out another solution very quickly | 18:46 |
| adrian_otto | shardy: Not that I am aware of. Maybe hongbin can add details? | 18:47 |
| hongbin | I can create a bug | 18:47 |
| shardy | hongbin: thanks, if you can raise one with details of the breakage that would be very helpful | 18:47 |
| *** sid14726 has quit IRC | 18:49 | |
| *** Drago1 has quit IRC | 18:49 | |
| *** Drago has joined #heat | 18:50 | |
| *** neelashah has quit IRC | 18:50 | |
| *** pas-ha has quit IRC | 18:50 | |
| hongbin | shardy: adrian_otto : Here is the bug report: https://bugs.launchpad.net/heat/+bug/1544227 | 18:51 |
| openstack | Launchpad bug 1544227 in Magnum "Magnum gate broke with error "Unable to find network with name 'None'"" [Undecided,New] | 18:51 |
| *** Raj1 has quit IRC | 18:52 | |
| adrian_otto | thanks hongbin | 18:55 |
| hongbin | np | 18:55 |
| *** Guest66498 has quit IRC | 18:56 | |
| *** jonesbr has quit IRC | 18:58 | |
| *** achanda has joined #heat | 19:00 | |
| russellb | shardy: commented on https://review.openstack.org/#/c/276417/3 | 19:01 |
| *** jonesbr has joined #heat | 19:01 | |
| russellb | shardy: I mention heat in the latter half, in "3) ..." | 19:01 |
| *** EricGonczer_ has joined #heat | 19:02 | |
| *** thumpba has quit IRC | 19:04 | |
| *** zeih_ has joined #heat | 19:04 | |
| *** EricGonc_ has joined #heat | 19:07 | |
| *** sridhar_ram has joined #heat | 19:07 | |
| *** zeih_ has quit IRC | 19:08 | |
| *** thumpba has joined #heat | 19:10 | |
| *** EricGonczer_ has quit IRC | 19:10 | |
| *** neelashah has joined #heat | 19:10 | |
| *** skraynev_ has joined #heat | 19:12 | |
| *** dimtruck has joined #heat | 19:18 | |
| *** jprovazn has quit IRC | 19:20 | |
| *** Marga_ has quit IRC | 19:21 | |
| therve | shardy, I wasn't | 19:24 |
| therve | And I'm not :) | 19:24 |
| shardy | therve: Ok, I'll propose a revert and we can follow up tomorrow | 19:24 |
| shardy | ramishra pointed out https://github.com/openstack/magnum/blob/master/magnum/conductor/template_definition.py#L70 | 19:25 |
| therve | dmsimard, That's just the same error as before the revert | 19:26 |
| shardy | and not being able to find a network with name "None" sounds kind of expected, but I assume the template generation provides parameters assigned to None which are somehow not ignored | 19:26 |
| dmsimard | therve: Hmm, you think ? I believe this test ran after the revert but I could be mistaken. I got help this morning to get the tempest revert merged. | 19:27 |
| therve | dmsimard, The revert didn't get merge, AFAICT? | 19:27 |
| * dmsimard looks | 19:27 | |
| therve | It just got | 19:28 |
| therve | Actually no, jenkins failure | 19:28 |
| dmsimard | doh it failed merge ci | 19:28 |
| *** coreyob has joined #heat | 19:28 | |
| dmsimard | okay, my bad | 19:28 |
| therve | I prefer that :) | 19:28 |
| openstackgerrit | Steven Hardy proposed openstack/heat: Revert "Resolve name/id using translation rules" https://review.openstack.org/278575 | 19:29 |
| shardy | hongbin, adrian_otto, ramishra ^^ | 19:29 |
| * shardy goes for dinner | 19:29 | |
| dmsimard | therve: fyi I submitted another bug the CI ran into https://bugs.launchpad.net/heat/+bug/1544220 | 19:29 |
| openstack | Launchpad bug 1544220 in heat "heat-engine throws errors when trying to import things that may not be installed" [Undecided,New] | 19:29 |
| hongbin | shardy: thx | 19:30 |
| openstackgerrit | Rabi Mishra proposed openstack/heat: [WIP]Do not convert resource_id to string https://review.openstack.org/278576 | 19:30 |
| ramishra | hongbin: can you try with the above patch, locally/gate, whether it fixes your issue? | 19:31 |
| hongbin | ramishra: trying | 19:31 |
| ramishra | shardy: we'll give the above patch a try before doing a revert. | 19:31 |
| adrian_otto | thanks shardy and ramishra | 19:31 |
| therve | dmsimard, Not sure it's a bug. Those libs are dependencies. | 19:31 |
| therve | We can talk about it in LP | 19:31 |
| * therve away for a bit | 19:31 | |
| dmsimard | therve: I don't know, glance has a ton of backends too and it tries to import them (i.e, sheepdog or other things) but it doesn't throw errors | 19:32 |
| hongbin | ramishra: worked in my local machine! | 19:32 |
| ramishra | Did you try this https://review.openstack.org/278576? | 19:32 |
| hongbin | ramishra: yes | 19:33 |
| *** neelashah1 has joined #heat | 19:34 | |
| ramishra | ok, There are a quite a few unit tests in heat to be fixed before this can land. I've not fixed them yet. I'll try and fix them. | 19:34 |
| *** neelashah has quit IRC | 19:35 | |
| ramishra | shardy: ^^^ | 19:35 |
| *** achanda has quit IRC | 19:36 | |
| *** randallburt has joined #heat | 19:38 | |
| *** sid14726 has joined #heat | 19:38 | |
| *** hdd has quit IRC | 19:39 | |
| *** jaime49 has joined #heat | 19:39 | |
| hongbin | ramishra: fyi: setting up a patch to test the gate: https://review.openstack.org/#/c/278578/ . Let's see if it passes. | 19:42 |
| *** achanda has joined #heat | 19:45 | |
| *** krotscheck is now known as krotscheck_dcm | 19:47 | |
| *** pt_15 has joined #heat | 19:48 | |
| *** jasond has quit IRC | 19:51 | |
| *** jasond has joined #heat | 19:51 | |
| *** pas-ha has joined #heat | 19:55 | |
| shardy | hongbin, ramishra: thanks, if that works I'll abandon the revert | 19:55 |
| skraynev_ | meeting time | 20:00 |
| *** jaime49 has quit IRC | 20:04 | |
| *** jtomasek has quit IRC | 20:10 | |
| *** achanda has quit IRC | 20:16 | |
| *** hdd has joined #heat | 20:16 | |
| *** apuch_ has joined #heat | 20:17 | |
| apuch_ | test for echo ? | 20:17 |
| *** prazumovsky_ has joined #heat | 20:18 | |
| apuch_ | I have a heat template where I am ccreating a bash script to use -> myTotalSleepTime=`echo $MaxNumOfReTrys $SleepTime undefined*p | dc ` | 20:18 |
| apuch_ | vs myTotalSleepTime=`echo $MaxNumOfReTrys $SleepTime \*p | dc ` | 20:19 |
| apuch_ | so how do I escape a \* in heat ? | 20:21 |
| *** achanda has joined #heat | 20:23 | |
| *** Marga_ has joined #heat | 20:23 | |
| *** EricGonc_ has quit IRC | 20:29 | |
| *** e0ne has quit IRC | 20:32 | |
| *** dims has quit IRC | 20:34 | |
| *** dims has joined #heat | 20:35 | |
| *** sid14726 has quit IRC | 20:37 | |
| *** gokrokve has quit IRC | 20:40 | |
| *** sdake has quit IRC | 20:43 | |
| *** EricGonczer_ has joined #heat | 20:43 | |
| *** jpeeler has quit IRC | 20:44 | |
| *** alexpilotti_ has joined #heat | 20:44 | |
| *** hdd has quit IRC | 20:47 | |
| *** EricGonczer_ has quit IRC | 20:48 | |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OpenstackClient plugin for stack delete https://review.openstack.org/242302 | 20:52 |
| *** zeih_ has joined #heat | 20:53 | |
| *** e0ne has joined #heat | 20:53 | |
| *** pas-ha has quit IRC | 20:59 | |
| *** pas-ha has joined #heat | 20:59 | |
| *** skraynev_ has quit IRC | 21:00 | |
| *** pas-ha has quit IRC | 21:00 | |
| *** mallya has joined #heat | 21:00 | |
| *** prazumovsky_ has quit IRC | 21:01 | |
| *** zeih_ has quit IRC | 21:03 | |
| *** jlk has quit IRC | 21:05 | |
| openstackgerrit | Merged openstack/python-heatclient: OSC plugin for stack snapshot show https://review.openstack.org/242888 | 21:06 |
| *** e0ne has quit IRC | 21:09 | |
| *** jlk has joined #heat | 21:10 | |
| *** jlk has quit IRC | 21:10 | |
| *** jlk has joined #heat | 21:10 | |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OpenstackClient plugin for stack delete https://review.openstack.org/242302 | 21:12 |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OpenstackClient plugin for event list https://review.openstack.org/245308 | 21:12 |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OpenstackClient plugin for event show https://review.openstack.org/245307 | 21:12 |
| *** sdake has joined #heat | 21:12 | |
| shardy | stevebaker: Hey, was wondering if we'd reached a consensus on https://review.openstack.org/#/c/274101/ (the heat.conf user_data_update_policy) | 21:13 |
| shardy | there were some conflicting reviews and various bug discussions | 21:13 |
| *** Marga_ has quit IRC | 21:16 | |
| *** sdake has quit IRC | 21:17 | |
| randallburt | shardy: fwiw I'm pro config option as it lets me set the default rather than having the default coded. It lets me allow users update pre-config stacks in a safe manner | 21:19 |
| *** alexpilotti_ has quit IRC | 21:19 | |
| randallburt | shardy: I'll add that very comment to the review and +1 (gates are failing atm) | 21:19 |
| shardy | randallburt: thanks, yeah the gate failure is Zaqar not starting, I don't think it's due to the patch | 21:20 |
| randallburt | shardy: most of the failures these last few days haven't been | 21:20 |
| randallburt | shardy: oh, which was another thing I forgot to bring up at the meeting re convergence. I think we've been having gate failures for the convergence job due to lots of timeouts. | 21:21 |
| shardy | randallburt: Yeah, that's one reason I'm concerned | 21:21 |
| shardy | we've been hitting TripleO timeouts without convergence because hitting heat with a large number of stacks creates some major scalability problems | 21:22 |
| openstackgerrit | Mark Vanderwiel proposed openstack/python-heatclient: Add openstack cli stack actions https://review.openstack.org/247804 | 21:22 |
| randallburt | and the number of data reads don't seem to be going down either | 21:22 |
| shardy | it's not yet clear to me if convergence will make that better or worse | 21:22 |
| shardy | my initial tests seemed to indicate, at the DB layer at least, it's the latter | 21:22 |
| stevebaker | shardy: actually those timeouts were a regression in the puppet modules to configure the rpc response timeout | 21:22 |
| randallburt | shardy: not sure tbh. if you're hitting heat on a single node, even with multiple processes, you're not going to get a ton of throughput | 21:23 |
| shardy | stevebaker: Hmm, OK, I wan't seeing RPC timeouts, but tbh I need to re-test with the latest fixes in | 21:25 |
| shardy | stevebaker: I was thinking about the patch where zaneb had to unroll a bunch of stacks because the combination of heat and the DB couldn't take ~35 stacks concurrently | 21:26 |
| shardy | I'm imagining what happens when you throw 100 stacks containing 2000 resources at it, and have all of them broken out via RPC | 21:26 |
| stevebaker | shardy: yeah, we do have scaling issues with the number of nested stacks | 21:27 |
| stevebaker | shardy: in a meeting currently, won't be long | 21:27 |
| randallburt | shardy /me shudders | 21:28 |
| randallburt | shardy: the 100 stacks isn't a problem if Heat is scaled properly, but the thought of doing an update on all of them at the same time using convergence makes me a little queezy. | 21:29 |
| shardy | I'm actually considering making an optional mode to run with only 1 engine process, e.g like we used to, where you explicitly don't RPC anything engine-to-engine | 21:30 |
| *** vijendar has quit IRC | 21:30 | |
| shardy | Only for testing obviously, but recent TripleO issues demonstrated how counter productive it is to round-robin a ton of work to yourself ;) | 21:31 |
| randallburt | indeed | 21:31 |
| *** vijendar has joined #heat | 21:34 | |
| *** hongbin_ has joined #heat | 21:39 | |
| *** al_indigo has joined #heat | 21:39 | |
| *** hongbin_ has quit IRC | 21:39 | |
| *** hongbin_ has joined #heat | 21:39 | |
| *** achanda_ has joined #heat | 21:40 | |
| stevebaker | shardy: ok, back | 21:40 |
| *** dprince has quit IRC | 21:40 | |
| *** mragupat_ has joined #heat | 21:41 | |
| stevebaker | shardy: regarding user_data_update_policy, regardless of what we do there I really like to see that heat-admin user_data change reverted and change the image building to create the heat-admin user | 21:42 |
| stevebaker | shardy: that change landed before we had the appreciation for upgrades that we do now ;) | 21:43 |
| *** Dave_____ has joined #heat | 21:43 | |
| shardy | stevebaker: I guess - baking changes into the image makes them even harder to modify than in user_data tho | 21:44 |
| shardy | maybe that's justified in the admin user case tho | 21:44 |
| stevebaker | I think it is, the user expects a heat-admin user, the image should have a heat-admin user | 21:44 |
| *** markvan_ has joined #heat | 21:45 | |
| shardy | the nice thing about the current approach is it's easy to use the same image and adjust the admin user configuration for each environment | 21:45 |
| *** mragupat has quit IRC | 21:45 | |
| stevebaker | and I assume we can make it work with standard cloud-init key injection, so we do nothing special | 21:45 |
| shardy | e.g you'd wire it in differently with dev/preprod than you would in a locked-down production environment | 21:45 |
| stevebaker | shardy: does anybody do that though? you can still have different keys in different environments | 21:46 |
| shardy | stevebaker: I'm not sure - I'm thinking of banks I've worked in, where they'd have dev/pre-prod stuff wired in to a user directory, or deploy credentials for the whole unix team, then have a completely different setup for production where access was controlled via change requests | 21:47 |
| zaneb | shardy, stevebaker: I think it was actually memory that was the problem there | 21:48 |
| shardy | stevebaker: I guess all that can be documented via virt-customize if need be | 21:48 |
| *** achanda has quit IRC | 21:48 | |
| *** hongbin has quit IRC | 21:48 | |
| *** dixiaoli_ has quit IRC | 21:48 | |
| *** fbo_ has quit IRC | 21:48 | |
| *** miguelgrinberg has quit IRC | 21:48 | |
| *** Dave has quit IRC | 21:48 | |
| *** jrist has quit IRC | 21:48 | |
| *** zhenguo_ has quit IRC | 21:48 | |
| *** al_indig_ has quit IRC | 21:48 | |
| *** dhellmann has quit IRC | 21:48 | |
| *** ramishra has quit IRC | 21:48 | |
| *** larsks has quit IRC | 21:48 | |
| *** markvan has quit IRC | 21:48 | |
| *** miguelgrinberg_ has joined #heat | 21:48 | |
| stevebaker | shardy: so the fallback there could still avoid user_data by creating the user in pre-config | 21:48 |
| *** miguelgrinberg_ is now known as miguelgrinberg | 21:48 | |
| *** larsks has joined #heat | 21:49 | |
| *** jasond has quit IRC | 21:49 | |
| *** dhellmann has joined #heat | 21:49 | |
| shardy | stevebaker: the *ExtraConfigPre (if that's what you mean by pre-config) is too late, we need something which runs before NetworkDeployment | 21:49 |
| *** fbo has joined #heat | 21:49 | |
| shardy | stevebaker: I think I'd prefer either an image build/customize step which still uses cloud-init to setup the user (e.g by injecting a script or local userdata) | 21:51 |
| stevebaker | shardy: they can use user_data if they policy dictates different usernames in different environments, but we don't necessarily have to help them. Having a default heat-admin user on the image which just works seems like a reasonable approach | 21:52 |
| zaneb | shardy: IMHO tripleo should never put anything in UserData. we know what a PITA that is w.r.t. updates, that's largely why we had to implement SW deployments | 21:52 |
| shardy | stevebaker: Ok - so you're still -1 on the heat.conf option then I take it | 21:53 |
| stevebaker | shardy: we can always wait for the RFE before we enable customisable users | 21:53 |
| stevebaker | shardy: If we leave UserData alone then the backport becomes unnecessary | 21:53 |
| shardy | stevebaker: https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud-resource-registry-puppet.yaml#L32 | 21:54 |
| shardy | what about RDO? | 21:54 |
| shardy | we'll be removing a feature that already exists | 21:54 |
| *** jrist has joined #heat | 21:54 | |
| *** dixiaoli_ has joined #heat | 21:55 | |
| stevebaker | ugh, we're utterly wedged then | 21:55 |
| shardy | honestly, I still see my patch as a perfectly reasonable and low risk solution | 21:56 |
| shardy | if we move to images long term, that's fine | 21:56 |
| *** ramishra has joined #heat | 21:57 | |
| shardy | but for existing deployments, the last thing they want is to roll out a new image, or hack on the one they have just to maintain access to new nodes | 21:57 |
| zaneb | it's kind of pure evil for interop because it makes the same template behave in different ways on different clouds | 21:57 |
| shardy | there are other use-cases, such as os-net-config mappings where adding stuff to user-data for new nodes would be helpful too | 21:57 |
| shardy | e.g if you need that data to set up the network, and it's node specific | 21:57 |
| shardy | zaneb: well, that's why the very next patch makes it explict via a template level interface | 21:58 |
| *** vgridnev has quit IRC | 21:58 | |
| shardy | and the config file option leaves the default as it is now | 21:58 |
| stevebaker | ok, the only option which doesn't result in somebody's nodes getting replaced is the heat.conf change | 21:58 |
| zaneb | shardy: but that doesn't help because the default varies with the config option | 21:58 |
| shardy | so you have to try pretty hard to make it work in an evil-from-interop perspective | 21:58 |
| zaneb | so you have to explicitly specify it in the template to get the same behaviour anywhere | 21:59 |
| * shardy sighs | 21:59 | |
| shardy | I guess that's true | 21:59 |
| zaneb | something that approximately no-one will do until after they have been bitten by it | 21:59 |
| stevebaker | zaneb: honestly, this flag will only ever be enabled on the undercloud | 21:59 |
| zaneb | yeah, I was about to say, maybe if it came with a huge warning THIS IS FOR TRIPLEO ONLY DO NOT CHANGE IT ON PAIN OF DEATH | 22:00 |
| zaneb | then maybe | 22:00 |
| shardy | lol | 22:00 |
| *** zhenguo_ has joined #heat | 22:00 | |
| stevebaker | how about a custom OS::Nova::Server subclass just for undercloud? | 22:00 |
| zaneb | stevebaker: I like that but I'm volunteering you to do the packaging | 22:02 |
| stevebaker | hack it into instack! | 22:02 |
| shardy | I guess that would be OK - it'd be pretty horrible if we have to maintain it out of the heat tree tho | 22:02 |
| *** swygue has joined #heat | 22:04 | |
| shardy | Ok, I'm going to abandon my patches - if you guys can figure out and propose an alternative (either to heat, or TripleO, I don't really mind), that would be great | 22:04 |
| shardy | AFAIK this is one of the blockers for jistr and others working on TripleO upgrades, so we do need some solution pretty fast | 22:05 |
| *** jasond has joined #heat | 22:05 | |
| openstackgerrit | Merged openstack/heat: Release notes for template validation improvements https://review.openstack.org/276633 | 22:05 |
| randallburt | well poop. this is what I get for not watching this chat. | 22:12 |
| shardy | I'm off for some sleep - we also need to land either https://review.openstack.org/#/c/278575/ or https://review.openstack.org/278576 because we broke the magnum gate, but it seems ref bug #1544292 that Zaqar has just broken ours | 22:14 |
| openstack | bug 1544292 in zaqar "Zaqar fails to start in heat gate tests" [Undecided,New] https://launchpad.net/bugs/1544292 | 22:14 |
| *** shardy has quit IRC | 22:14 | |
| *** hongbin has joined #heat | 22:16 | |
| *** Marga_ has joined #heat | 22:16 | |
| *** spotz is now known as spotz_zzz | 22:17 | |
| *** boris-42_ has joined #heat | 22:17 | |
| *** sthillma has joined #heat | 22:20 | |
| *** dyasny_ has joined #heat | 22:20 | |
| *** chlong_ has joined #heat | 22:22 | |
| *** boris-42 has quit IRC | 22:22 | |
| *** dhellmann has quit IRC | 22:22 | |
| *** dyasny has quit IRC | 22:22 | |
| *** chlong has quit IRC | 22:22 | |
| *** hongbin_ has quit IRC | 22:22 | |
| *** dhellmann has joined #heat | 22:23 | |
| *** boris-42_ is now known as boris-42 | 22:24 | |
| *** mallya has quit IRC | 22:29 | |
| *** mallya has joined #heat | 22:30 | |
| openstackgerrit | Bryan Jones proposed openstack/python-heatclient: OSC plugin for stack resource show and list https://review.openstack.org/244927 | 22:34 |
| *** mallya has quit IRC | 22:35 | |
| *** jonesbr has left #heat | 22:36 | |
| *** dimtruck is now known as zz_dimtruck | 22:36 | |
| *** mallya has joined #heat | 22:37 | |
| *** mallya has quit IRC | 22:40 | |
| *** zz_dimtruck is now known as dimtruck | 22:45 | |
| *** neelashah1 has quit IRC | 22:47 | |
| *** achanda_ has quit IRC | 22:49 | |
| *** alex_xu has quit IRC | 22:52 | |
| *** crose has quit IRC | 22:53 | |
| *** kebray has joined #heat | 22:54 | |
| *** alex_xu has joined #heat | 22:55 | |
| *** rpothier has quit IRC | 22:56 | |
| *** adrian_otto has quit IRC | 23:02 | |
| *** rwsu has quit IRC | 23:06 | |
| *** vijendar has quit IRC | 23:06 | |
| *** jasond has quit IRC | 23:07 | |
| *** Delair has quit IRC | 23:07 | |
| *** kebray has quit IRC | 23:09 | |
| *** sid14726 has joined #heat | 23:12 | |
| *** achanda has joined #heat | 23:31 | |
| *** mragupat_ has quit IRC | 23:35 | |
| *** chlong_ has quit IRC | 23:46 | |
| *** dimtruck is now known as zz_dimtruck | 23:49 | |
| *** swygue has quit IRC | 23:52 | |
| *** jlk has quit IRC | 23:54 | |
| *** jlk has joined #heat | 23:57 | |
| *** jlk has quit IRC | 23:57 | |
| *** jlk has joined #heat | 23:57 | |
| *** thumpba has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!