| *** charlesw has joined #magnetodb | 01:05 | |
| *** vivekd has joined #magnetodb | 03:28 | |
| *** jeromatron has joined #magnetodb | 03:33 | |
| *** jeromatron has quit IRC | 03:42 | |
| *** charlesw has quit IRC | 03:48 | |
| *** rushiagr_away is now known as rushiagr | 04:01 | |
| *** rushiagr is now known as rushiagr_away | 04:51 | |
| *** rushiagr_away is now known as rushiagr | 05:32 | |
| *** ajayaa has joined #magnetodb | 05:45 | |
| *** ajayaa has quit IRC | 06:17 | |
| *** k4n0 has joined #magnetodb | 06:27 | |
| *** vivekd_ has joined #magnetodb | 06:35 | |
| *** vivekd has quit IRC | 06:38 | |
| *** vivekd_ is now known as vivekd | 06:38 | |
| *** romainh has joined #magnetodb | 06:59 | |
| *** ajayaa has joined #magnetodb | 07:02 | |
| *** romainh has quit IRC | 07:07 | |
| *** romainh has joined #magnetodb | 07:07 | |
| isviridov | GM everybody | 07:19 |
|---|---|---|
| idegtiarov | Hi! | 07:25 |
| isviridov | Hey idegtiarov | 07:30 |
| isviridov | How it is going in celiometer? | 07:30 |
| idegtiarov | all run its course | 07:32 |
| ajayaa | Hi all. | 08:11 |
| isviridov | Hello ajayaa | 08:11 |
| ajayaa | Hi isviridov. | 08:13 |
| isviridov | ajayaa, how are your progress with celiometer? | 08:28 |
| ajayaa | isviridov, I am preparing a spec to ceilometer spec repo right now. | 08:29 |
| ajayaa | I have a working code also.(POC kind of thing) | 08:29 |
| ajayaa | Need to refine it though. | 08:29 |
| isviridov | Great to know. We have a guys from celio here as well | 08:30 |
| isviridov | I mean idegtiarov | 08:30 |
| isviridov | Don't hesistate to share the spec here also | 08:31 |
| ajayaa | Sure. | 08:31 |
| ajayaa | I have a little doubt regarding to resource_id field in ceilometer samples. | 08:31 |
| ajayaa | idegtiarov, | 08:31 |
| ajayaa | ^^ | 08:31 |
| ajayaa | In mdb we don't have a UUID for a table or ant other identifier which could be used for a resource field. | 08:32 |
| isviridov | Is i mandatory? | 08:33 |
| isviridov | *it | 08:33 |
| ajayaa | Yes. | 08:34 |
| idegtiarov | Yep its mandatory | 08:34 |
| isviridov | There is a similar case with HEAT | 08:34 |
| isviridov | In AWS the stack name is used for indentification, but in OpenStack it is UUID | 08:35 |
| isviridov | They solved it with allowing to access by both. | 08:35 |
| isviridov | idegtiarov, do you know any other usecases where UUID is not available? | 08:36 |
| isviridov | ajayaa, good point! I think it should be discussed with celio cores. | 08:40 |
| ajayaa | Yes. I need to check in celio documentation regarding use of resource_id. | 08:41 |
| ajayaa | If it it something not exposed by REST api then we can use a dummy value. | 08:42 |
| idegtiarov | isviridov: sorry I don't get the meaning of your question | 08:47 |
| isviridov | ajayaa, however I think that having UUID for table is a good practice | 08:53 |
| ajayaa | It is not too much of a overhead I guess to have a UUID for a table. | 08:54 |
| ajayaa | isviridov, Do we have other cases where a UUID could be useful? | 08:55 |
| ajayaa | I can think of one use case, In future we could expose an api to access table with "/UUID". | 08:56 |
| ajayaa | That way user won't have to keep track of both project name and table name. | 08:56 |
| isviridov | Sorry having troubles with connection it seems. Loosing your messages | 08:57 |
| isviridov | I think that UUID is always good thing. | 08:58 |
| isviridov | ajayaa, will you create a BP for this? | 08:58 |
| ajayaa | isviridov, sure. | 08:59 |
| *** ajayaa has quit IRC | 10:05 | |
| *** romainh has quit IRC | 10:13 | |
| *** ajayaa has joined #magnetodb | 10:29 | |
| ajayaa | isviridov, idegtiarov here is the spec I have written for mdb-ceilo integration. | 10:38 |
| ajayaa | https://github.com/ajayaa/python-akhada/blob/master/magnetodb-notifications.rst | 10:38 |
| isviridov | ajayaa, i like it. | 10:57 |
| isviridov | Have you discussed it with celiometer team? I mean the prioritization | 10:58 |
| isviridov | BTW here is BP I've registered some time ago for monitoring of mdb by celiometer https://blueprints.launchpad.net/ceilometer/+spec/support-magnetodb | 10:59 |
| isviridov | I think the description should be corrected in order to keep scope smaller and finish first with processing of notifications. | 11:00 |
| idegtiarov | ajayaa: I'll look at the spec | 11:03 |
| ajayaa | isviridov, Yesterday I was talking to ceilometer cores. | 11:04 |
| ajayaa | They are fine with notifications from mdb. | 11:04 |
| isviridov | Great! | 11:05 |
| ajayaa | isviridov, Please assign that blueprint to me. | 11:06 |
| isviridov | Done | 11:07 |
| isviridov | I also think that it will be useful to describe the metrics and events you are going to store to celio. | 11:08 |
| isviridov | F.e. like here https://github.com/openstack/ceilometer/commit/b27ac819fb2b6cd44d41dbaf71ba72805c6fe7d2 | 11:08 |
| isviridov | The more detailed and clear the spec is, the esier it will be approved | 11:08 |
| * isviridov or rejected | 11:08 | |
| ajayaa | isviridov, I am yet to find out how to do aggregation in ceilometer. | 11:09 |
| ajayaa | I was hoping to get help from ceilometer team on that. | 11:09 |
| ajayaa | And this is just the initial draft of the spec. I plan on adding more stuff. :) | 11:10 |
| isviridov | ajayaa, have you agreed any timeline for implementation? There is end if cycle now | 11:10 |
| ajayaa | Yes. Kilo | 11:10 |
| ajayaa | Not in juno | 11:10 |
| isviridov | Feel free to ping me if any review needed or something | 11:11 |
| ajayaa | sure. Thank you. | 11:11 |
| isviridov | BTW what are your plans about https://blueprints.launchpad.net/magnetodb/+spec/support-roles? | 11:14 |
| isviridov | ajayaa, I'm thinking if we can do it during juno, but seems there is no spec at all ATM. Could you please share your plans on it? | 11:15 |
| ajayaa | isviridov, We have a spec for that. | 11:15 |
| ajayaa | https://wiki.openstack.org/wiki/MagnetoDB/specs/rbac | 11:16 |
| ajayaa | I am doing that in my free time, besides office hours. | 11:16 |
| isviridov | Ah, i remember, could you please add it to BP? | 11:16 |
| isviridov | ajayaa, WOW work=hobby=OpenStack | 11:17 |
| ajayaa | isviridov, :) | 11:17 |
| ajayaa | isviridov, You could similarly assign that bp to me. :) | 11:18 |
| isviridov | ajayaa, does it mean different SLAs for these patches? Do you think we have to introduce the tag 'hobby' or 'in-my-free-time'? | 11:18 |
| ajayaa | isviridov, if you feel the need for it! | 11:19 |
| ajayaa | I don't know. | 11:19 |
| isviridov | Sure, done. You are drafter of it as well. | 11:19 |
| isviridov | The only thing, please pay attention that according to priorities the first things to discuss and review are items in current development focus | 11:20 |
| isviridov | In other words active milestone. Now it is https://launchpad.net/magnetodb/+milestone/juno-rc1 | 11:20 |
| isviridov | With growing the team we are getting more capacity, so can extend the scope. But it should be agreed | 11:21 |
| ajayaa | When does juno milestone finish? | 11:22 |
| isviridov | 16 of Oct | 11:22 |
| ajayaa | We can put the role support in there, I should be able to finish it by then. | 11:23 |
| ajayaa | Only the unittest part is there. | 11:24 |
| isviridov | I really would love to see it in juno. | 11:24 |
| ajayaa | I also would need to modify current unitests to be accomodating of policy files. | 11:24 |
| ajayaa | Or figure out a way to keep policy out of current unittests! | 11:25 |
| isviridov | I've added it to juno. | 11:27 |
| isviridov | Next step the spec should be approved by other core developers. | 11:27 |
| isviridov | It the easiest way to get +2 for patch. | 11:28 |
| isviridov | I have some comments to spec right now | 11:29 |
| ajayaa | How do I propose it for approval? | 11:30 |
| ajayaa | Feel free to express your concerns. :) | 11:31 |
| isviridov | From first look I think that all supported actions should be enumerated. | 11:38 |
| isviridov | Also, please keep the template structure. If there is nothing to add out None https://wiki.openstack.org/wiki/MagnetoDB/specs/template | 11:38 |
| isviridov | You can propose your BP or spec via mail list (preferable), IRC weekly meeting, here is IRC channel | 11:40 |
| isviridov | * in | 11:40 |
| ajayaa | I will make some changes to spec. | 11:41 |
| aostapenko | https://wiki.openstack.org/w/index.php?title=MagnetoDB/specs/monitoring-health-check | 11:43 |
| aostapenko | Hi guys, take a look please https://wiki.openstack.org/w/index.php?title=MagnetoDB/specs/monitoring-health-check | 11:44 |
| isviridov | Looks better now. What is the query to C* are yo going to use for? | 11:46 |
| * isviridov feels sleepy | 11:49 | |
| isviridov | aostapenko, could you specify it please in spec | 11:50 |
| isviridov | aostapenko, thank you | 11:50 |
| isviridov | aostapenko, having a meeting now. See you in 30mins | 11:50 |
| aostapenko | isviridov, ok | 11:50 |
| *** isviridov is now known as isviridov_away | 11:50 | |
| *** romainh has joined #magnetodb | 11:51 | |
| aostapenko | isviridov, done | 11:52 |
| *** rushiagr is now known as rushiagr_away | 12:51 | |
| *** k4n0 has quit IRC | 13:01 | |
| *** isviridov_away is now known as isviridov | 13:06 | |
| isviridov | aostapenko, with growing the number of tables, the request will be slower. What about select * from magnetodb.table_info limit 1 ? | 13:19 |
| *** rushiagr_away is now known as rushiagr | 13:24 | |
| *** rushiagr is now known as rushiagr_away | 13:36 | |
| *** charlesw has joined #magnetodb | 13:38 | |
| *** vivekd has quit IRC | 13:47 | |
| aostapenko | isviridov: great idea. thank you | 13:57 |
| aostapenko | isviridov, ikhudoshyn, dukhlov, cwang look please and approve if you have no objections https://wiki.openstack.org/w/index.php?title=MagnetoDB/specs/monitoring-health-check | 13:59 |
| *** rushiagr_away is now known as rushiagr | 14:19 | |
| *** ajayaa has quit IRC | 14:24 | |
| *** openstackgerrit has quit IRC | 14:42 | |
| *** jeromatron has joined #magnetodb | 15:23 | |
| *** openstackgerrit has joined #magnetodb | 15:25 | |
| openstackgerrit | Andrei V. Ostapenko proposed a change to stackforge/magnetodb: (WIP) Monitoring health check request https://review.openstack.org/125108 | 16:13 |
| *** jeromatron has quit IRC | 16:24 | |
| *** jeromatron has joined #magnetodb | 16:28 | |
| *** jeromatron has quit IRC | 16:30 | |
| *** jeromatron has joined #magnetodb | 16:30 | |
| charlesw | aostapenko, the spec looks good mostly. A couple of things. healthcheck is a verb. Maybe health is more RESTful? Second is when status is ERROR, do you still return 200 HTTP response code? A different error code such as 500 can be potentially used by haproxy or other load balancer to mark the impacted API server as DOWN. | 16:32 |
| *** jeromatron has quit IRC | 16:36 | |
| *** jeromatron has joined #magnetodb | 16:37 | |
| *** romainh has left #magnetodb | 16:48 | |
| *** romainh has joined #magnetodb | 16:49 | |
| *** jeromatron has quit IRC | 17:20 | |
| *** romainh has left #magnetodb | 17:33 | |
| *** rushiagr is now known as rushiagr_away | 17:48 | |
| *** rushiagr_away is now known as rushiagr | 17:51 | |
| *** openstackgerrit has quit IRC | 18:47 | |
| *** openstackgerrit has joined #magnetodb | 18:49 | |
| *** rushiagr is now known as rushiagr_away | 19:02 | |
| *** jeromatron has joined #magnetodb | 20:12 | |
| *** jeromatron has quit IRC | 20:23 | |
| *** jeromatron has joined #magnetodb | 20:23 | |
| *** jeromatron has quit IRC | 20:48 | |
| *** jeromatron has joined #magnetodb | 20:52 | |
| *** jeromatron has quit IRC | 21:10 | |
| *** jeromatron has joined #magnetodb | 21:11 | |
| *** jeromatron has quit IRC | 21:16 | |
| *** jeromatron has joined #magnetodb | 21:16 | |
| openstackgerrit | Charles Wang proposed a change to stackforge/magnetodb: Handle non-existent item when calling update_item https://review.openstack.org/123927 | 22:12 |
| *** charlesw has quit IRC | 22:44 | |
| *** jeromatron has quit IRC | 23:13 | |
| *** jeromatron has joined #magnetodb | 23:17 | |
| *** jeromatron has quit IRC | 23:18 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!