| *** dave-mccowan has quit IRC | 00:13 | |
| *** ianychoi has joined #openstack-oslo | 00:19 | |
| *** dciabrin has joined #openstack-oslo | 00:35 | |
| *** dciabrin_ has quit IRC | 00:35 | |
| lbragstad | goldyfruit technically, you don't need authentication to implement quotas | 00:41 |
|---|---|---|
| lbragstad | oslo.limit will attempt to pull information from keystone, but it won't be authentication per se | 00:41 |
| *** dave-mccowan has joined #openstack-oslo | 01:04 | |
| goldyfruit | lbragstad, I'm in some deployments you could not have keystone at all | 01:26 |
| goldyfruit | Like Bifrost and Cinder | 01:26 |
| *** rcernin has quit IRC | 01:27 | |
| *** rcernin has joined #openstack-oslo | 01:27 | |
| goldyfruit | I mean* | 01:27 |
| lbragstad | interesting - in that case, we'd need to hook into oslo.limit so that it doesn't talk to keystone | 01:34 |
| goldyfruit | yep | 01:37 |
| lbragstad | are you pulling limit information from another system? | 01:37 |
| goldyfruit | Nop | 01:39 |
| lbragstad | so where is the limit information coming from? | 01:39 |
| lbragstad | the unified limit design is pretty specific in that information coming from keystone | 01:39 |
| goldyfruit | from the quotas tables inside the qinling database | 01:39 |
| goldyfruit | yeah, that is why I could not implement oslo.limit within Qinling | 01:40 |
| lbragstad | so qinling has it's own limits API? | 01:41 |
| goldyfruit | Not yet :) | 01:41 |
| goldyfruit | Working on it:https://review.opendev.org/#/c/665015 | 01:41 |
| lbragstad | to qinling, what is a quota? | 01:42 |
| lbragstad | because we define quota and limit here https://docs.openstack.org/keystone/latest/admin/unified-limits.html | 01:42 |
| goldyfruit | Quotas should be applied for functions, runtimes, webhooks, versions, aliases and jobs | 01:43 |
| lbragstad | and those are qinling resources? | 01:43 |
| lbragstad | right? | 01:43 |
| goldyfruit | yep | 01:43 |
| lbragstad | ok - so you want users to be able to limit the amount of runtimes a project can have in qinling? | 01:43 |
| goldyfruit | yeah, same for functions, etc... | 01:44 |
| lbragstad | sure | 01:44 |
| lbragstad | with unified limit, that information would live in keystone | 01:44 |
| goldyfruit | What we would like to avoid is a user to create a function that create infinite functions, etc... | 01:44 |
| lbragstad | sure | 01:45 |
| goldyfruit | I think it's the same for Mistral, Mistral is able to work without Keystone | 01:45 |
| lbragstad | true - cinder is in the same boat | 01:46 |
| goldyfruit | So far Bifrost, Cinder, Qinling, Mistral | 01:46 |
| lbragstad | for now, the unified limit initiative is dependent on services pulling limit information from keystone (since keystone owns projects, associating limits of resources on those projects makes sense to keep in keystone directly) | 01:46 |
| goldyfruit | yeah but maybe we could ask to these projects working without Keystone to add limits tables into the project | 01:47 |
| lbragstad | maybe | 01:48 |
| goldyfruit | and have a way to notice olso.limit to not looking in Keystone | 01:48 |
| goldyfruit | but "locally" | 01:48 |
| lbragstad | sure - you'd need a hook in oslo.limit to handle that | 01:48 |
| goldyfruit | yep | 01:48 |
| lbragstad | and you'd need to make sure each service does something reasonable with that API so that oslo.limit can rely on it regardless of the service employing it | 01:48 |
| goldyfruit | I spent the last 3 days looking how to implement quotas, I checked in Karbor, Mistral Cyborg, etc... none of them have the same implementation, which is why I love the idea of oslo.limit ! | 01:49 |
| lbragstad | well... that's part of the reason why we want to pull the implementations out of each service and into keystone | 01:50 |
| goldyfruit | :) | 01:50 |
| lbragstad | a lot of services are rebuilding the same wheel | 01:50 |
| goldyfruit | Thanks for your time/information lbragstad :) | 01:51 |
| lbragstad | so - it's a double-edged sword... you push the commonality into keystone and have keystone deal with associating limits to the projects (since it's the authority of that resource) or you have to make sure 60+ services implement the same exact API | 01:51 |
| lbragstad | sure thing goldyfruit | 01:52 |
| goldyfruit | yeah :/ | 01:52 |
| *** dmellado_ has joined #openstack-oslo | 03:08 | |
| *** dmellado has quit IRC | 03:08 | |
| *** dmellado_ is now known as dmellado | 03:08 | |
| *** hberaud|gone has quit IRC | 03:22 | |
| *** zaneb has quit IRC | 03:22 | |
| *** HenryG has quit IRC | 03:22 | |
| *** mugsie has quit IRC | 03:22 | |
| *** seba has quit IRC | 03:22 | |
| *** jbadiapa has quit IRC | 03:22 | |
| *** szaher has quit IRC | 03:22 | |
| *** mgagne has quit IRC | 03:22 | |
| *** dasp has quit IRC | 03:22 | |
| *** sean-k-mooney has quit IRC | 03:22 | |
| *** d34dh0r53 has quit IRC | 03:22 | |
| *** hberaud|gone has joined #openstack-oslo | 03:38 | |
| *** zaneb has joined #openstack-oslo | 03:38 | |
| *** HenryG has joined #openstack-oslo | 03:38 | |
| *** mugsie has joined #openstack-oslo | 03:38 | |
| *** seba has joined #openstack-oslo | 03:38 | |
| *** jbadiapa has joined #openstack-oslo | 03:38 | |
| *** szaher has joined #openstack-oslo | 03:38 | |
| *** mgagne has joined #openstack-oslo | 03:38 | |
| *** dasp has joined #openstack-oslo | 03:38 | |
| *** sean-k-mooney has joined #openstack-oslo | 03:38 | |
| *** d34dh0r53 has joined #openstack-oslo | 03:38 | |
| *** irclogbot_3 has quit IRC | 03:41 | |
| *** irclogbot_0 has joined #openstack-oslo | 03:42 | |
| *** dave-mccowan has quit IRC | 03:49 | |
| *** dasp has quit IRC | 03:58 | |
| *** dasp has joined #openstack-oslo | 04:21 | |
| *** lbragstad has quit IRC | 04:43 | |
| *** pcaruana has joined #openstack-oslo | 05:09 | |
| *** e0ne has joined #openstack-oslo | 05:21 | |
| *** jbadiapa has quit IRC | 05:49 | |
| openstackgerrit | Merged openstack/oslo.vmware master: Sync Sphinx requirement https://review.opendev.org/662676 | 06:22 |
| *** takamatsu has joined #openstack-oslo | 06:34 | |
| *** iurygregory has joined #openstack-oslo | 06:58 | |
| *** gibi has joined #openstack-oslo | 07:08 | |
| *** tesseract has joined #openstack-oslo | 07:08 | |
| *** tosky has joined #openstack-oslo | 07:15 | |
| *** rcernin has quit IRC | 07:22 | |
| *** jbadiapa has joined #openstack-oslo | 07:27 | |
| *** e0ne has quit IRC | 07:35 | |
| openstackgerrit | Natal Ngétal proposed openstack/oslo.log master: Modify the constraints url in tox. https://review.opendev.org/661100 | 07:35 |
| *** hberaud|gone is now known as hberaud | 07:52 | |
| *** ralonsoh has joined #openstack-oslo | 08:20 | |
| openstackgerrit | Gabriele Santomaggio proposed openstack/oslo.messaging master: Add transport_options parameter https://review.opendev.org/660373 | 08:31 |
| openstackgerrit | Gabriele Santomaggio proposed openstack/oslo.messaging master: Add transport_options parameter https://review.opendev.org/660373 | 08:37 |
| *** e0ne has joined #openstack-oslo | 09:08 | |
| *** Luzi has joined #openstack-oslo | 09:29 | |
| openstackgerrit | Merged openstack/osprofiler master: Support standalone placement in the devstack https://review.opendev.org/665135 | 09:44 |
| *** hberaud is now known as hberaud|lunch | 09:56 | |
| openstackgerrit | Merged openstack/oslo.log master: Add logging guidelines based on previous spec https://review.opendev.org/665112 | 10:18 |
| *** jbadiapa has quit IRC | 10:57 | |
| *** hberaud|lunch is now known as hberaud | 11:09 | |
| *** jaosorior has joined #openstack-oslo | 11:20 | |
| *** raildo has joined #openstack-oslo | 11:20 | |
| *** goldyfruit has quit IRC | 11:40 | |
| *** Luzi has quit IRC | 11:49 | |
| *** jaosorior has quit IRC | 12:09 | |
| gsantomaggio | @hberaud the problem is that: | 12:10 |
| gsantomaggio | ``` | 12:10 |
| gsantomaggio | result = self.transport._send(self.target, msg_ctxt, msg, | 12:10 |
| gsantomaggio | wait_for_reply=True, timeout=timeout, | 12:10 |
| gsantomaggio | call_monitor_timeout=cm_timeout, | 12:10 |
| gsantomaggio | retry=self.retry, | 12:10 |
| gsantomaggio | transport_options=self.transport_options) | 12:10 |
| gsantomaggio | ``` | 12:10 |
| gsantomaggio | is > 80 columns | 12:10 |
| gsantomaggio | We did not have this problem before because it was: | 12:11 |
| gsantomaggio | ``` | 12:11 |
| gsantomaggio | result = self.transport._send(self.target, msg_ctxt, msg, | 12:11 |
| gsantomaggio | wait_for_reply=True, timeout=timeout, | 12:11 |
| gsantomaggio | call_monitor_timeout=cm_timeout, | 12:11 |
| gsantomaggio | retry=self.retry, | 12:11 |
| gsantomaggio | options=self.options) | 12:11 |
| gsantomaggio | ``` | 12:11 |
| gsantomaggio | so, with the rename is < 80 col : ) | 12:11 |
| gsantomaggio | so, with the rename is > 80 col : ) | 12:11 |
| hberaud | gsantomaggio: oh right good catch | 12:12 |
| gsantomaggio | about the blue print ... | 12:12 |
| hberaud | gsantomaggio: sorry I doesn't take care about the indent on transport_otions | 12:12 |
| gsantomaggio | have no idea | 12:12 |
| hberaud | gsantomaggio: me too | 12:13 |
| hberaud | gsantomaggio: maybe the dash on the name or something like that | 12:13 |
| hberaud | apparently the "Implements: blueprint transport-options" will trigger a search on LP | 12:14 |
| hberaud | gsantomaggio: and not only redirect to an url like "+spec/transport-options" | 12:14 |
| hberaud | gsantomaggio: I guess the issue is that this functionality trigger a search and it only search in the content or tags on the LP blueprints or something like that | 12:15 |
| hberaud | gsantomaggio: but I'm not sure | 12:16 |
| gsantomaggio | @hberaud the same problem here | 12:17 |
| gsantomaggio | https://review.opendev.org/#/c/661698/ | 12:17 |
| hberaud | gsantomaggio: yeah good catch | 12:17 |
| gsantomaggio | I got the problem | 12:21 |
| *** dave-mccowan has joined #openstack-oslo | 12:28 | |
| *** dave-mccowan has quit IRC | 12:35 | |
| *** lbragstad has joined #openstack-oslo | 12:59 | |
| *** goldyfruit has joined #openstack-oslo | 13:08 | |
| *** boden has joined #openstack-oslo | 13:09 | |
| *** goldyfruit has quit IRC | 13:23 | |
| *** goldyfruit has joined #openstack-oslo | 13:24 | |
| *** bobh has joined #openstack-oslo | 13:24 | |
| gsantomaggio | I have a problem with blue print reference. | 13:29 |
| gsantomaggio | I added this `Implements: blueprint transport-options` on | 13:29 |
| gsantomaggio | https://review.opendev.org/#/c/660373/ | 13:29 |
| gsantomaggio | the problem is that it points to: | 13:29 |
| gsantomaggio | ``` | 13:29 |
| gsantomaggio | https://blueprints.launchpad.net/openstack/?searchtext=transport-options | 13:29 |
| gsantomaggio | ``` | 13:29 |
| gsantomaggio | but I need this: | 13:29 |
| gsantomaggio | ``` | 13:29 |
| gsantomaggio | https://blueprints.launchpad.net/oslo.messaging?searchtext=transport-options | 13:29 |
| gsantomaggio | ``` | 13:29 |
| gsantomaggio | do you know why? | 13:29 |
| gsantomaggio | ty | 13:29 |
| *** goldyfruit has quit IRC | 13:47 | |
| openstackgerrit | Gabriele Santomaggio proposed openstack/oslo.messaging master: Add transport_options parameter https://review.opendev.org/660373 | 13:49 |
| *** jbadiapa has joined #openstack-oslo | 13:50 | |
| gsantomaggio | @hberaud added the explicit link | 13:52 |
| gsantomaggio | everything should be fine now :) I hope! | 13:52 |
| *** goldyfruit has joined #openstack-oslo | 13:53 | |
| hberaud | gsantomaggio: w00t | 13:54 |
| *** e0ne has quit IRC | 14:13 | |
| *** stephenfin is now known as finucannot | 14:26 | |
| *** lpetrut has joined #openstack-oslo | 14:46 | |
| *** bnemec has joined #openstack-oslo | 14:48 | |
| lpetrut | Hi, could anyone please take a look over this simple patch fixing Windows pbr support? https://review.opendev.org/#/c/662450/ | 14:50 |
| *** e0ne has joined #openstack-oslo | 14:57 | |
| *** e0ne has quit IRC | 15:00 | |
| *** bnemec is now known as beekneemech | 15:11 | |
| *** e0ne has joined #openstack-oslo | 15:15 | |
| *** e0ne has quit IRC | 15:21 | |
| *** e0ne has joined #openstack-oslo | 15:24 | |
| *** e0ne has quit IRC | 15:27 | |
| *** e0ne has joined #openstack-oslo | 15:29 | |
| *** e0ne has quit IRC | 15:33 | |
| *** hberaud is now known as hberaud|school-r | 15:37 | |
| *** raildo has quit IRC | 15:40 | |
| *** raildo has joined #openstack-oslo | 15:40 | |
| *** raildo has quit IRC | 15:49 | |
| *** raildo has joined #openstack-oslo | 15:49 | |
| *** hberaud|school-r is now known as hberaud | 15:51 | |
| *** takamatsu has quit IRC | 16:07 | |
| *** goldyfruit has quit IRC | 16:24 | |
| *** goldyfruit has joined #openstack-oslo | 16:32 | |
| *** iurygregory has quit IRC | 16:53 | |
| *** lbragstad is now known as lbragstad_runnin | 17:04 | |
| *** iurygregory has joined #openstack-oslo | 17:15 | |
| *** iurygregory has quit IRC | 17:23 | |
| *** lpetrut has quit IRC | 17:30 | |
| *** ralonsoh has quit IRC | 17:31 | |
| *** iurygregory has joined #openstack-oslo | 17:36 | |
| *** iurygregory has quit IRC | 17:38 | |
| *** e0ne has joined #openstack-oslo | 17:41 | |
| *** lbragstad_runnin is now known as lbragstad | 17:43 | |
| *** iurygregory has joined #openstack-oslo | 17:44 | |
| *** e0ne has quit IRC | 17:46 | |
| *** hberaud is now known as hberaud|gone | 17:53 | |
| openstackgerrit | Merged openstack/oslo.config master: Add unit tests for validator https://review.opendev.org/662816 | 18:07 |
| *** tesseract has quit IRC | 18:18 | |
| *** e0ne has joined #openstack-oslo | 18:42 | |
| *** iurygregory has quit IRC | 19:07 | |
| *** e0ne has quit IRC | 19:16 | |
| *** e0ne has joined #openstack-oslo | 19:16 | |
| *** goldyfruit has quit IRC | 19:17 | |
| *** goldyfruit has joined #openstack-oslo | 19:17 | |
| *** e0ne has quit IRC | 19:21 | |
| *** goldyfruit has quit IRC | 19:22 | |
| *** e0ne has joined #openstack-oslo | 19:23 | |
| *** e0ne has quit IRC | 19:33 | |
| *** logan- has joined #openstack-oslo | 19:33 | |
| logan- | is there a way to debug what files oslopolicy-policy-generator is looking at? or does anyone know why I am seeing output on rocky that is ignoring the /etc/cinder/policy.json? http://paste.openstack.org/raw/752968/ | 19:34 |
| smcginnis | logan-: Is that supposed to be "role:support_technician", not "rule:support_technician"? | 19:37 |
| logan- | nope, the role is named 'support' | 19:37 |
| *** pcaruana has quit IRC | 19:39 | |
| *** jbadiapa has quit IRC | 19:40 | |
| *** harlowja has joined #openstack-oslo | 19:40 | |
| logan- | simplifying the example more, I changed the context_is_admin rule to "role:admin or rule:support_technician" and the output from oslopolicy-policy-generator still renders as "context_is_admin": "role:admin" | 19:42 |
| logan- | er correction I simplified the example policy to "role:admin or role:support" | 19:42 |
| logan- | also tested dumping the same policy in a nova rocky venv and it produces the expected "context_is_admin": "role:admin or role:support" output | 19:43 |
| beekneemech | logan-: It looks like the generator hard-codes the log level to warning, so you may need to change that locally to get more log output. | 19:50 |
| beekneemech | https://github.com/openstack/oslo.policy/blob/master/oslo_policy/generator.py#L335 | 19:50 |
| beekneemech | Which makes me wonder why we have log messages at the info level in these tools: https://github.com/openstack/oslo.policy/blob/master/oslo_policy/generator.py#L352 | 19:50 |
| beekneemech | Otherwise I don't see anything wrong with what you're doing. That looks like it should work to me. | 19:53 |
| beekneemech | I've never actually run this tool though. Maybe lbragstad will know more? | 19:54 |
| openstackgerrit | Lance Bragstad proposed openstack/oslo.limit master: WIP: Rewrite oslo.limit https://review.opendev.org/665465 | 19:57 |
| lbragstad | logan- hmmm | 19:58 |
| lbragstad | you're trying to generate a new policy file with the overrides you have in /etc/cinder/policy.json? | 20:00 |
| logan- | yep | 20:04 |
| lbragstad | interesting... you're able to get policy files with overrides and default with nova, but not cinder? | 20:07 |
| lbragstad | defaults* | 20:11 |
| lbragstad | logan- i'm able to generate it locally | 20:14 |
| lbragstad | logan- v | 20:15 |
| lbragstad | bah | 20:15 |
| lbragstad | https://pasted.tech/pastes/ef1ae68a444a08962c07db989ff55e38d1c7dee0.raw | 20:15 |
| lbragstad | logan- is your cinder policy file defined in cinder.conf? | 20:15 |
| logan- | nope, no [oslo_policy] section in cinder.conf | 20:19 |
| lbragstad | if you had it, does it work? | 20:20 |
| logan- | checking | 20:20 |
| smcginnis | Is this a side effect of moving the default policy in code? | 20:20 |
| lbragstad | even if you are just specifying the default? | 20:20 |
| lbragstad | i've noticed some weirdness with that in the past | 20:20 |
| logan- | yep, that fixed it | 20:20 |
| lbragstad | nice | 20:20 |
| lbragstad | i'm not sure why it has to be that way | 20:21 |
| lbragstad | i haven't dug into it in a while | 20:21 |
| lbragstad | i thought nova was having a similar issue once | 20:21 |
| logan- | yeah that's strange, I don't have that section in keystone or nova yet both have worked since we switched to policy in code for those services in pike | 20:21 |
| lbragstad | huh | 20:21 |
| lbragstad | strange | 20:22 |
| logan- | thanks for the help! | 20:22 |
| lbragstad | anytime | 20:22 |
| *** goldyfruit has joined #openstack-oslo | 20:28 | |
| *** raildo has quit IRC | 20:42 | |
| *** goldyfruit_ has joined #openstack-oslo | 20:54 | |
| *** goldyfruit has quit IRC | 20:56 | |
| *** goldyfruit_ has quit IRC | 21:00 | |
| *** goldyfruit_ has joined #openstack-oslo | 21:01 | |
| *** boden has quit IRC | 21:21 | |
| *** goldyfruit___ has joined #openstack-oslo | 21:57 | |
| *** goldyfruit_ has quit IRC | 22:00 | |
| *** bobh has quit IRC | 22:38 | |
| *** bobh has joined #openstack-oslo | 22:39 | |
| *** bobh has quit IRC | 22:43 | |
| *** tosky has quit IRC | 22:48 | |
| openstackgerrit | Merged openstack/oslo.cache master: Pass server_max_value_length to memcached client https://review.opendev.org/661553 | 22:50 |
| *** harlowja has quit IRC | 23:16 | |
| *** goldyfruit___ has quit IRC | 23:19 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!