| *** thrash is now known as thrash|g0ne | 01:21 | |
| SamYaple | hey mordred, for the salt modules i am considering defaulting to using OS_CLOUD=default for auth | 02:20 |
|---|---|---|
| SamYaple | this would prevent the need to specify and 'auth' line for every state run | 02:20 |
| SamYaple | thoughts? | 02:20 |
| SamYaple | to make it clear, the user would need to lay down /etc/openstack/clouds.yaml (or another default path) with the cloud named 'default' to use this | 02:21 |
| -openstackstatus- NOTICE: Gerrit is being restarted to feed its insatiable memory appetite | 02:41 | |
| *** Shrews has quit IRC | 06:05 | |
| *** dhellmann has quit IRC | 06:05 | |
| *** spredzy has quit IRC | 06:05 | |
| *** cmurphy has quit IRC | 06:05 | |
| *** melwitt has quit IRC | 06:05 | |
| *** melwitt has joined #openstack-shade | 06:05 | |
| *** dhellmann has joined #openstack-shade | 06:05 | |
| *** Shrews has joined #openstack-shade | 06:05 | |
| *** melwitt is now known as Guest36641 | 06:06 | |
| *** cmurphy has joined #openstack-shade | 06:06 | |
| *** spredzy has joined #openstack-shade | 06:10 | |
| *** ioggstream has joined #openstack-shade | 07:13 | |
| -openstackstatus- NOTICE: Post jobs are not executed currently, do not tag any releases | 07:23 | |
| *** ChanServ changes topic to "Post jobs are not executed currently, do not tag any releases" | 07:23 | |
| *** slaweq has quit IRC | 09:04 | |
| *** slaweq has joined #openstack-shade | 09:07 | |
| *** thrash|g0ne is now known as thrash | 10:50 | |
| *** gouthamr has joined #openstack-shade | 13:13 | |
| *** Guest36641 is now known as melwitt | 14:03 | |
| *** thrash has quit IRC | 14:04 | |
| *** thrash has joined #openstack-shade | 14:08 | |
| *** thrash has quit IRC | 14:08 | |
| *** thrash has joined #openstack-shade | 14:08 | |
| mordred | SamYaple: I recommend not doing that. if there is only one cloud, shade will already just use it - if there is more than one, the user should select it | 14:16 |
| mordred | SamYaple: if we want to add a default cloud concept, we should add it to os-client-config - perhaps as a "default: True" option to stick on the cloud entry | 14:17 |
| *** slaweq has quit IRC | 15:16 | |
| *** slaweq_ has quit IRC | 15:27 | |
| *** slaweq has joined #openstack-shade | 15:28 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade master: Add method to set bootable flag on volumes https://review.openstack.org/502479 | 15:33 |
| mordred | slaweq: ^^ updated that | 15:35 |
| mordred | slaweq: also, in the process, I discovered we have ZERO unittests around create_volume :) | 15:35 |
| mordred | slaweq: and that we have an open/passthrough kwargs | 15:35 |
| *** thrash is now known as thrash|lunch | 16:32 | |
| openstackgerrit | Merged openstack/os-client-config master: Fix requires_floating_ip https://review.openstack.org/504616 | 16:41 |
| *** gouthamr has quit IRC | 17:01 | |
| *** gouthamr has joined #openstack-shade | 17:02 | |
| SamYaple | mordred: what do you mean "if there is only one cloud, shade will already just use it"? | 17:12 |
| SamYaple | without specifying OS_CLOUD or similar? | 17:13 |
| mordred | SamYaple: yah | 17:26 |
| mordred | SamYaple: if you don't specify a cloud argument and os-client-config finds only one cloud in your config, it will use it | 17:27 |
| mordred | SamYaple: this makes envvar consumption in the absence of a clouds.yaml work like people expect | 17:27 |
| mordred | SamYaple: since in that case the only cloud around would be "envvars" | 17:27 |
| mordred | SamYaple: the only time it gets squirrely for people is if they have one cloud defined in their clouds.yaml and ALSO have OS_ envvars set (OS_CLOUD and OS_REGION_NAME don't count - they are used to select which cloud you want to use) | 17:28 |
| mordred | SamYaple: in any case- long story short - the "which cloud config is going to get used" is already slightly complex, so I'd like to fix issues in it in os-client-config directly so that salt, ansible, shade and python-openstackclient all behave the same - if we can manage it | 17:29 |
| *** olaph has quit IRC | 17:30 | |
| SamYaple | mordred: thats not the behaviour i see | 17:31 |
| *** olaph has joined #openstack-shade | 17:32 | |
| SamYaple | i have no ENVVARS set and i have a single cloud in clouds.yaml: `t = shade.operator_cloud(); t.list_domains()` fails | 17:32 |
| SamYaple | `t = shade.operator_cloud(cloud='default'); t.list_domains()`works | 17:33 |
| mordred | SamYaple: ok. I can verify that is happening locally - that is a bug | 17:35 |
| SamYaple | has it ever worked the other way? its worked this way for as long as ive known | 17:36 |
| SamYaple | otherwise something like `openstack domain list` should 'just work' with a single cloud too, and ive never seen that happen without OS_CLOUD or --os-cloud | 17:37 |
| mordred | it's definitely the intent that in the presence of only one cloud config that you don't have to specify the cloud | 17:37 |
| SamYaple | ok cool. that makes sense | 17:37 |
| mordred | SamYaple: ah- I think I see it - it works if the only cloud you have is envvars :) | 17:38 |
| SamYaple | yea that makes complete sense and is the behaviour i know | 17:38 |
| SamYaple | its ok, i found a way to share auth params using saltstack proper without needing to set a default cloud name | 17:38 |
| SamYaple | so i can scrap that idea | 17:39 |
| *** ioggstream has quit IRC | 17:56 | |
| *** thrash|lunch is now known as thrash | 18:17 | |
| *** slaweq has quit IRC | 19:14 | |
| *** slaweq has joined #openstack-shade | 19:14 | |
| *** slaweq_ has joined #openstack-shade | 19:33 | |
| *** slaweq has quit IRC | 19:37 | |
| openstackgerrit | Sam Yaple proposed openstack-infra/shade master: Allow domain_id for roles https://review.openstack.org/496992 | 19:38 |
| openstackgerrit | Sam Yaple proposed openstack-infra/shade master: Move role normalization to normalize.py https://review.openstack.org/500170 | 19:38 |
| slaweq_ | mordred: hello, thx for changes in Your patch https://review.openstack.org/502479 :) | 19:38 |
| openstackgerrit | Monty Taylor proposed openstack/os-client-config master: Treat clouds.yaml with one cloud like envvars https://review.openstack.org/505388 | 19:41 |
| mordred | SamYaple: remote: https://review.openstack.org/505388 Treat clouds.yaml with one cloud like envvars | 19:42 |
| mordred | oh - hah. didn't need to paste | 19:42 |
| SamYaple | mordred: awesome! | 19:47 |
| SamYaple | mordred: say... in a theoretical shade/openstacksdk merger will openstacksdk make the same commitment shade does as far as "no breaking changes ever" | 19:48 |
| mordred | SamYaple: yup | 19:49 |
| SamYaple | hmm ok. i feel if we do that we might end up wanting an openstacksdk2 at some point to fix random issues that have creeped in | 19:49 |
| mordred | SamYaple: let's put it this way - I will not accept any changes to shade's organization that cause that comittement to go away | 19:49 |
| mordred | SamYaple: openstacksdk has not yet released a 1.0 | 19:49 |
| SamYaple | we will be able to at least fix all the positional args and make a unified way to pass domain/domain_id etc | 19:50 |
| mordred | SamYaple: oh - there's a missing piece here, which I was tlaking to dtroyer about at the PTG ... | 19:50 |
| SamYaple | yea sorry i didnt chat with you there about this. i kept only seeing oyu when i was busy | 19:50 |
| mordred | SamYaple: which is that if we take the shade.OpenStackCloud object and move it in to openstacksdk (*handwave*) as openstack.cloud.OpenStackCloud for instance... | 19:50 |
| mordred | SamYaple: then we can cut a release of shade that depends on sdk and is a very thin wrapper around that class- ... BUT ... | 19:51 |
| mordred | we can fix the positional args and whatnot in openstack.cloud.OpenStackCloud | 19:51 |
| SamYaple | yea we talked about that right before the ptg | 19:51 |
| mordred | SamYaple: and just have the thin wraper in the shade repo be about maintaining backwards compat for argument ordering and whatnot | 19:51 |
| mordred | ok. cool | 19:51 |
| SamYaple | ok cool. thats what im looking for. | 19:51 |
| SamYaple | we should be able to get it really clean looking | 19:52 |
| mordred | yah - we can use openstack.cloud.OpenStackCloud as the 'v2' of shade - since the v1 of shade will still be there and can still be easily maintained | 19:52 |
| SamYaple | i would love to see the ability to pass domain everywhere you can pass domain_id and shade/openstacksdk will look it up | 19:52 |
| SamYaple | same goes for project/project_id etc | 19:53 |
| mordred | yup. I'd also like to see a general mechanism for "operate in this other context for a second" | 19:53 |
| SamYaple | yea | 19:53 |
| mordred | I've been noodling on it as something like with cloud.in_project(id='my-project-id') as other_project: other_project.create_user('foo') | 19:53 |
| mordred | SamYaple: but I haven't tried writing that yet, so I don't know if it would be good or bad :) | 19:54 |
| SamYaple | that would work, but might get hairy with something like this role from domain1 applies to this group in domain2 for this project in domain3 | 19:55 |
| SamYaple | cloud.in_domain() wouldn't really make sense there | 19:55 |
| SamYaple | i think i might be happier with a dict that has things you might be interested in. {'project':<projectobj>,'project_id':123,'domain':<domobj>,'domain_id':123} etc etc | 20:01 |
| SamYaple | so we could just check that dict if the thing we care about is in there, if not go look it up | 20:01 |
| *** gouthamr has quit IRC | 20:03 | |
| -openstackstatus- NOTICE: Zuul and Gerrit are being restarted to address issues discovered with the Gerrit 2.13 upgrade. review.openstack.org will be inaccessible for a few minutes while we make these changes. Currently running jobs will be restarted for you once Zuul and Gerrit are running again. | 20:27 | |
| *** thrash is now known as thrash|g0ne | 20:48 | |
| mordred | SamYaple: yah. agree | 21:22 |
| *** gouthamr has joined #openstack-shade | 21:29 | |
| *** gouthamr has quit IRC | 21:33 | |
| *** gouthamr has joined #openstack-shade | 21:35 | |
| *** gouthamr has quit IRC | 22:21 | |
| -openstackstatus- NOTICE: Gerrit is being restarted to address some final issues, review.openstack.org will be inaccessible for a few minutes while we restart | 22:33 | |
| *** slaweq_ has quit IRC | 22:36 | |
| *** openstackgerrit has quit IRC | 22:47 | |
| rods | mordred Hi, what's the plan for the shade/openstackskd merge? anything I can help with? | 22:53 |
| *** gouthamr has joined #openstack-shade | 23:01 | |
| mordred | rods: had some good discussions with folks at the PTG - I need to write those up ... | 23:05 |
| mordred | rods: but yes, absolutely! :) | 23:05 |
| *** slaweq has joined #openstack-shade | 23:06 | |
| *** slaweq has quit IRC | 23:12 | |
| *** gouthamr has quit IRC | 23:40 | |
| *** ChanServ changes topic to "#openstack-shade" | 23:40 | |
| -openstackstatus- NOTICE: Gerrit is once again part of normal puppet config management. Problems with Gerrit gitweb links and Zuul post jobs have been addressed. We currently cannot create new gerrit projects (fixes in progress) and email sending is slow (being debugged). | 23:40 | |
| *** gouthamr has joined #openstack-shade | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!