| *** ddieterly has joined #openstack-ceilometer | 00:08 | |
| *** ankita_w_ has joined #openstack-ceilometer | 00:11 | |
| *** ankita___ has joined #openstack-ceilometer | 00:13 | |
| *** ankita_wagh has quit IRC | 00:14 | |
| *** ankita_w_ has quit IRC | 00:17 | |
| *** promulo__ has joined #openstack-ceilometer | 00:18 | |
| *** jruano has joined #openstack-ceilometer | 00:19 | |
| *** promulo has quit IRC | 00:22 | |
| *** sbfox has joined #openstack-ceilometer | 00:30 | |
| *** zhenq has joined #openstack-ceilometer | 00:38 | |
| *** zhenq has quit IRC | 00:42 | |
| *** Longgeek has joined #openstack-ceilometer | 00:54 | |
| *** jruano has quit IRC | 00:56 | |
| *** zhenq has joined #openstack-ceilometer | 01:02 | |
| *** amalagon has joined #openstack-ceilometer | 01:04 | |
| *** jruano has joined #openstack-ceilometer | 01:08 | |
| *** zhenq has quit IRC | 01:10 | |
| *** haigang has joined #openstack-ceilometer | 01:13 | |
| *** ankita___ has quit IRC | 01:15 | |
| *** ankita_wagh has joined #openstack-ceilometer | 01:16 | |
| *** zhenq has joined #openstack-ceilometer | 01:16 | |
| *** Longgeek has quit IRC | 01:17 | |
| *** Longgeek has joined #openstack-ceilometer | 01:17 | |
| *** sbfox has quit IRC | 01:17 | |
| *** ankita_wagh has quit IRC | 01:20 | |
| *** mitz has quit IRC | 01:26 | |
| *** jruano has quit IRC | 01:32 | |
| *** mitz has joined #openstack-ceilometer | 01:36 | |
| *** edmondsw has quit IRC | 01:55 | |
| *** Longgeek has quit IRC | 02:09 | |
| *** Longgeek_ has joined #openstack-ceilometer | 02:09 | |
| *** Longgeek_ has quit IRC | 02:20 | |
| *** Longgeek_ has joined #openstack-ceilometer | 02:23 | |
| *** _cjones_ has joined #openstack-ceilometer | 02:30 | |
| *** _cjones_ has quit IRC | 02:33 | |
| *** _cjones_ has joined #openstack-ceilometer | 02:33 | |
| *** jruano has joined #openstack-ceilometer | 02:35 | |
| *** jruano has quit IRC | 02:39 | |
| *** sbfox has joined #openstack-ceilometer | 02:42 | |
| *** rbak has joined #openstack-ceilometer | 02:53 | |
| *** haigang has quit IRC | 02:57 | |
| *** amalagon has quit IRC | 03:00 | |
| *** rbak has quit IRC | 03:02 | |
| *** ddieterly has quit IRC | 03:07 | |
| *** harlowja_ is now known as harlowja_away | 03:07 | |
| *** zhenq has quit IRC | 03:12 | |
| *** alextricity has quit IRC | 03:16 | |
| *** haigang has joined #openstack-ceilometer | 03:22 | |
| *** haigang has quit IRC | 03:27 | |
| *** amalagon has joined #openstack-ceilometer | 03:31 | |
| *** deepthi has joined #openstack-ceilometer | 03:32 | |
| *** rbak has joined #openstack-ceilometer | 03:38 | |
| *** rbak has quit IRC | 03:46 | |
| *** haigang has joined #openstack-ceilometer | 03:55 | |
| *** sbfox has quit IRC | 03:59 | |
| *** ddieterly has joined #openstack-ceilometer | 04:08 | |
| *** ddieterly has quit IRC | 04:12 | |
| *** _cjones_ has quit IRC | 04:15 | |
| *** _cjones_ has joined #openstack-ceilometer | 04:15 | |
| *** yatin has joined #openstack-ceilometer | 04:16 | |
| *** drjones has joined #openstack-ceilometer | 04:16 | |
| *** _cjones_ has quit IRC | 04:20 | |
| *** drjones has quit IRC | 04:21 | |
| *** Longgeek_ has quit IRC | 04:35 | |
| *** _cjones_ has joined #openstack-ceilometer | 04:41 | |
| *** Longgeek has joined #openstack-ceilometer | 04:42 | |
| *** _cjones_ has quit IRC | 04:44 | |
| *** ildikov has quit IRC | 04:55 | |
| *** ddieterly has joined #openstack-ceilometer | 05:09 | |
| *** ddieterly has quit IRC | 05:13 | |
| *** mahipal has joined #openstack-ceilometer | 05:23 | |
| *** exploreshaifali has joined #openstack-ceilometer | 05:40 | |
| *** ildikov has joined #openstack-ceilometer | 05:47 | |
| *** exploreshaifali has quit IRC | 05:53 | |
| *** yatin has quit IRC | 05:58 | |
| gzhai1 | ildikov:hi | 06:00 |
|---|---|---|
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/165703 | 06:06 |
| *** AJaeger has joined #openstack-ceilometer | 06:09 | |
| AJaeger | Good morning ceilometer, could you merge the translations, please? https://review.openstack.org/#/c/165703/ | 06:09 |
| *** ddieterly has joined #openstack-ceilometer | 06:10 | |
| *** ddieterly has quit IRC | 06:14 | |
| *** yatin has joined #openstack-ceilometer | 06:15 | |
| *** sbfox has joined #openstack-ceilometer | 06:17 | |
| *** haigang has quit IRC | 06:27 | |
| *** yatin has quit IRC | 06:27 | |
| *** haigang has joined #openstack-ceilometer | 06:27 | |
| *** haigang has quit IRC | 06:32 | |
| AJaeger | thanks, Mehdi! (Sorry, don't know IRC nic) | 06:33 |
| sileht | AJaeger, it's sileht | 06:33 |
| sileht | AJaeger, welcome | 06:33 |
| AJaeger | sileht, thanks! | 06:33 |
| *** yatin has joined #openstack-ceilometer | 06:41 | |
| *** yprokule has joined #openstack-ceilometer | 06:44 | |
| *** AJaeger has left #openstack-ceilometer | 06:49 | |
| openstackgerrit | ZhiQiang Fan proposed openstack/python-ceilometerclient: WIP: Add timeout for keystoneclient session https://review.openstack.org/169194 | 07:03 |
| *** liusheng has joined #openstack-ceilometer | 07:04 | |
| *** Kennan2 has joined #openstack-ceilometer | 07:08 | |
| gzhai1 | ildikov:hi | 07:08 |
| *** Kennan has quit IRC | 07:09 | |
| *** ddieterly has joined #openstack-ceilometer | 07:10 | |
| ildikov | gzhai1: hi | 07:11 |
| ildikov | gzhai1: sorry, I was afk, I just always forget to change my nick | 07:11 |
| ildikov | gzhai1: what's up? | 07:11 |
| *** haigang has joined #openstack-ceilometer | 07:12 | |
| *** yatin has quit IRC | 07:13 | |
| *** ddieterly has quit IRC | 07:15 | |
| *** ala_ has joined #openstack-ceilometer | 07:19 | |
| *** yatin has joined #openstack-ceilometer | 07:30 | |
| sileht | :bn | 07:31 |
| *** mahipal has quit IRC | 07:33 | |
| *** liusheng has quit IRC | 07:33 | |
| *** liusheng has joined #openstack-ceilometer | 07:34 | |
| gzhai1 | ildikov::could you pls. review my doc changes? | 07:44 |
| gzhai1 | ildikov:https://review.openstack.org/#/c/165374/ | 07:44 |
| *** yatin has quit IRC | 07:44 | |
| ildikov | gzhai1: sure, I will check soon, thanks | 07:45 |
| *** safchain has joined #openstack-ceilometer | 07:49 | |
| gzhai1 | ildikov:Thanks! | 07:53 |
| *** eglynn has joined #openstack-ceilometer | 07:56 | |
| *** Ephur has quit IRC | 08:01 | |
| *** yatin has joined #openstack-ceilometer | 08:01 | |
| *** liusheng has quit IRC | 08:03 | |
| *** liusheng has joined #openstack-ceilometer | 08:04 | |
| openstackgerrit | Merged openstack/gnocchi: rest: add more operators in complex queries https://review.openstack.org/166920 | 08:05 |
| openstackgerrit | Merged openstack/gnocchi: storage: pass query rather than predicate in value_search https://review.openstack.org/166921 | 08:10 |
| *** ddieterly has joined #openstack-ceilometer | 08:11 | |
| *** ddieterly has quit IRC | 08:15 | |
| *** sbfox has quit IRC | 08:17 | |
| jd__ | gosh I thought I could avoid that crap requirement check | 08:22 |
| jd__ | someone added if on Gnocchi too | 08:22 |
| jd__ | s/if/it/ | 08:22 |
| jd__ | now I'm stuck to list dependencies versions that are false :( | 08:22 |
| *** KrustyB has joined #openstack-ceilometer | 08:22 | |
| jd__ | removal requested, let's cross fingers | 08:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: allow to search for metric value https://review.openstack.org/167164 | 08:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: return objects rather than dict https://review.openstack.org/167965 | 08:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: Bump to oslo.serialization 1.4.0 https://review.openstack.org/166922 | 08:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: encode all timestamps with timezone https://review.openstack.org/166923 | 08:26 |
| openstackgerrit | Merged openstack/gnocchi: Use modern versions of the oslo packages https://review.openstack.org/168998 | 08:34 |
| *** haigang has quit IRC | 08:38 | |
| *** amalagon has quit IRC | 08:41 | |
| *** amalagon has joined #openstack-ceilometer | 08:42 | |
| *** amalagon has quit IRC | 08:46 | |
| -openstackstatus- NOTICE: CI Check/Gate pipelines currently stuck due to a bad dependency creeping in the system. No need to recheck your patches at the moment. | 08:52 | |
| *** ChanServ changes topic to "CI Check/Gate pipelines currently stuck due to a bad dependency creeping in the system. No need to recheck your patches at the moment." | 08:52 | |
| *** flcn___ has joined #openstack-ceilometer | 08:56 | |
| *** flcn___ has quit IRC | 08:59 | |
| *** flcn___ has joined #openstack-ceilometer | 08:59 | |
| *** ildikov has quit IRC | 09:06 | |
| *** afazekas has joined #openstack-ceilometer | 09:11 | |
| *** ddieterly has joined #openstack-ceilometer | 09:12 | |
| *** Longgeek has quit IRC | 09:16 | |
| *** ddieterly has quit IRC | 09:16 | |
| *** ildikov has joined #openstack-ceilometer | 09:19 | |
| *** flcn___ has quit IRC | 09:45 | |
| *** cdent has joined #openstack-ceilometer | 09:52 | |
| *** ildikov has quit IRC | 09:54 | |
| *** liusheng has quit IRC | 10:00 | |
| *** liusheng has joined #openstack-ceilometer | 10:01 | |
| *** ildikov has joined #openstack-ceilometer | 10:09 | |
| *** ddieterly has joined #openstack-ceilometer | 10:13 | |
| *** Longgeek has joined #openstack-ceilometer | 10:17 | |
| *** ddieterly has quit IRC | 10:17 | |
| *** KrustyB has quit IRC | 10:28 | |
| *** Longgeek has quit IRC | 10:30 | |
| *** deepthi has quit IRC | 10:30 | |
| *** claudiub has joined #openstack-ceilometer | 10:52 | |
| *** IBerezovskiy has joined #openstack-ceilometer | 10:53 | |
| *** nsaje_ has quit IRC | 10:57 | |
| *** yatin has quit IRC | 11:02 | |
| *** ildikov has quit IRC | 11:06 | |
| *** Kennan has joined #openstack-ceilometer | 11:10 | |
| *** Kennan2 has quit IRC | 11:11 | |
| *** ddieterly has joined #openstack-ceilometer | 11:13 | |
| *** Kennan2 has joined #openstack-ceilometer | 11:14 | |
| *** Kennan has quit IRC | 11:14 | |
| *** Longgeek has joined #openstack-ceilometer | 11:17 | |
| *** ddieterly has quit IRC | 11:18 | |
| *** _nadya_ has joined #openstack-ceilometer | 11:19 | |
| *** ildikov has joined #openstack-ceilometer | 11:22 | |
| *** changbl has quit IRC | 11:30 | |
| *** yatin has joined #openstack-ceilometer | 11:32 | |
| *** amalagon has joined #openstack-ceilometer | 11:41 | |
| -openstackstatus- NOTICE: Check/Gate unstuck, feel free to recheck your abusively-failed changes. | 11:48 | |
| *** ChanServ changes topic to "OpenStack Telemetry (Ceilometer) | https://wiki.openstack.org/wiki/Ceilometer" | 11:48 | |
| *** ildikov has quit IRC | 12:03 | |
| *** nsaje_ has joined #openstack-ceilometer | 12:11 | |
| *** _nadya_ has quit IRC | 12:12 | |
| *** ddieterly has joined #openstack-ceilometer | 12:14 | |
| *** ildikov has joined #openstack-ceilometer | 12:18 | |
| *** ddieterly has quit IRC | 12:19 | |
| *** jruano has joined #openstack-ceilometer | 12:23 | |
| *** zhenq has joined #openstack-ceilometer | 12:24 | |
| *** claudiub has quit IRC | 12:27 | |
| *** claudiub has joined #openstack-ceilometer | 12:27 | |
| *** zhenq has quit IRC | 12:36 | |
| *** yatin has quit IRC | 12:38 | |
| *** jruano has quit IRC | 12:38 | |
| *** jruano has joined #openstack-ceilometer | 12:45 | |
| *** tongli has joined #openstack-ceilometer | 12:49 | |
| *** gordc has joined #openstack-ceilometer | 12:50 | |
| gordc | jd__: now that gnocchi is under openstack, we have plans to merge teams? | 12:54 |
| *** eglynn has quit IRC | 12:54 | |
| jd__ | gordc: no | 12:58 |
| *** ildikov_ has joined #openstack-ceilometer | 12:58 | |
| jd__ | gordc: we just keep ceilometer-core and gnocchi-core and manage them as people contribute to one or another | 12:58 |
| *** ddieterly has joined #openstack-ceilometer | 12:58 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: WIP: Keep history of updated resource attributes https://review.openstack.org/168405 | 13:00 |
| *** ildikov has quit IRC | 13:00 | |
| *** maishsk has joined #openstack-ceilometer | 13:00 | |
| *** maishsk has quit IRC | 13:04 | |
| cdent | jd__: on https://review.openstack.org/#/c/169001/ you want a test that confirms the empty list case? or something else? | 13:05 |
| jd__ | cdent: yes just that | 13:05 |
| cdent | roger, will do | 13:06 |
| jd__ | sileht: why don't you use autoincrement? | 13:06 |
| jasonamyers | hello all, would any of you be up to talking with the Ops group at the Summit in Vancouver? They've got a proposal on their etherpad https://etherpad.openstack.org/p/YVR-ops-meetup | 13:08 |
| openstackgerrit | Romain Soufflet proposed openstack/gnocchi: rest: Add list of resources types URL on /v1/resource https://review.openstack.org/169294 | 13:08 |
| jasonamyers | it is the most +1ed thing on their agenda | 13:08 |
| gordc | jd__: cool cool. | 13:10 |
| jd__ | jasonamyers: I'm volunteer to represent Gnocchi at least | 13:11 |
| jasonamyers | awesome jd__ | 13:11 |
| gordc | jd__: was just going to volunteer you.lol | 13:11 |
| jd__ | jasonamyers: just keep me in the loop so I don't miss it | 13:12 |
| jd__ | gordc: what about you represent Ceilometer? :) | 13:12 |
| *** rbak has joined #openstack-ceilometer | 13:12 | |
| jasonamyers | will do jd__ there is a thread going around now on the Ops list | 13:12 |
| gordc | jd__: well played... well played. | 13:12 |
| jasonamyers | I'll keep an eye on it | 13:12 |
| jasonamyers | and share updates | 13:12 |
| jd__ | jasonamyers: ok, I'm not subscribed to it so don't hesitate to Cc me :) | 13:12 |
| jd__ | thanks jasonamyers :) | 13:13 |
| gordc | jasonamyers: i think a few of can join so it should be good. | 13:13 |
| jasonamyers | excellent | 13:13 |
| cdent | "what needs fixing" | 13:13 |
| cdent | people's grammar! ;) | 13:14 |
| * cdent is the last person to complain about such things | 13:14 | |
| jasonamyers | etherpads are often stream of consciousness | 13:14 |
| cdent | indeed | 13:14 |
| gordc | what be grammar? | 13:14 |
| jasonamyers | Tim Bell from Cern said: I think it is important to understand what Gnocchi does and what it will not do in case people feel it will solve all of the problems. | 13:15 |
| cdent | very good point | 13:15 |
| cdent | gordc grammar be other people stuff | 13:15 |
| cdent | not you stuff | 13:15 |
| *** ddieterly has quit IRC | 13:16 | |
| *** logan2 has quit IRC | 13:16 | |
| *** ddieterly has joined #openstack-ceilometer | 13:17 | |
| gordc | word | 13:18 |
| *** logan2 has joined #openstack-ceilometer | 13:19 | |
| *** ddieterly has quit IRC | 13:21 | |
| openstackgerrit | Romain Soufflet proposed openstack/gnocchi: rest: Add list of resources types URL on /v1/resource https://review.openstack.org/169294 | 13:23 |
| *** wayneeseguin has quit IRC | 13:24 | |
| *** eglynn has joined #openstack-ceilometer | 13:24 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: allow to search for metric value https://review.openstack.org/167164 | 13:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: return objects rather than dict https://review.openstack.org/167965 | 13:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: Bump to oslo.serialization 1.4.0 https://review.openstack.org/166922 | 13:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: encode all timestamps with timezone https://review.openstack.org/166923 | 13:26 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: return resource objects rather than dict https://review.openstack.org/169302 | 13:26 |
| sileht | jd__, autoincrement doesn't forbid the insertion of a new raw without outdated attributes | 13:26 |
| jd__ | sileht: "without outdated attributes", WDYM? | 13:27 |
| sileht | jd__, imagine to concurency calls, one change the user, the other the project | 13:27 |
| *** jruano has quit IRC | 13:27 | |
| sileht | jd__, with autoincrement, the row created by the latest that commit the transaction will not have the correct value from the first one that have commit the transaction | 13:28 |
| *** yprokule1 has joined #openstack-ceilometer | 13:28 | |
| *** Ephur has joined #openstack-ceilometer | 13:29 | |
| jd__ | sileht: shit you blew my mind indeed | 13:30 |
| sileht | jd__, because when the lastest version of resource is retrieved, both transaction get the same row, each transassion updated their fields, both successfuly insert the new row, but have one of the field not updated | 13:31 |
| sileht | jd__, that make sense ? | 13:31 |
| *** yprokule has quit IRC | 13:31 | |
| *** rbak has quit IRC | 13:32 | |
| jd__ | sileht: so that's a problem when using PATCH right? | 13:32 |
| sileht | jd__, yes | 13:33 |
| jd__ | and update_resource()… it wouldn't happen with PUT (that we don't have) | 13:33 |
| jd__ | yeah it makes sense | 13:33 |
| jd__ | sileht: and what about doing INSERT INTO (SELECT) ? | 13:35 |
| jd__ | sileht: something along http://docs.sqlalchemy.org/en/latest/changelog/migration_09.html#insert-from-select | 13:35 |
| jd__ | so you just insert default value with a select from MAX(revision) | 13:35 |
| sileht | jd__, that could work but you have to list all the columns you want, | 13:38 |
| jd__ | sileht: well that's all the columns except the one updated | 13:38 |
| sileht | jd__, sqlalchemy doesn't allow '*!col1' | 13:38 |
| jd__ | sileht: well we can iterate the class to know the column | 13:39 |
| sileht | jd__, and the column list is pluggable | 13:39 |
| jd__ | sileht: that sill sounds like a better way of doing it no? | 13:39 |
| jd__ | just run one query in SQL | 13:39 |
| sileht | jd__, sure | 13:40 |
| jd__ | I'm pretty sure we can introspect a class to have the list of column | 13:40 |
| sileht | jd__, this is the tricky part, I will try that | 13:40 |
| jd__ | sileht: ok let me know if you need help | 13:40 |
| jd__ | I've been digging more and more deeply in sqlalchemy recently so… | 13:41 |
| jd__ | like I've now understand lazy/eager loading, imagine that | 13:41 |
| jd__ | I'm becoming smartsql | 13:41 |
| sileht | jd__, also I have start another patch to keep history of what metrics was attached to a resource, does It make sense to track that ? | 13:41 |
| jd__ | sileht: I don't think so | 13:42 |
| sileht | jd__, in my current patch, when you retreive a old version of a resource, "metrics" is a empty dict | 13:43 |
| jd__ | sileht: it should be the current list of metrics I think | 13:43 |
| jd__ | metrics isn't a real attribute | 13:43 |
| sileht | jd__, ok | 13:43 |
| jd__ | I really don't think it's necessary to version metrics, it'll add a lot of cost for no gain | 13:44 |
| jd__ | and probably confusion | 13:44 |
| jd__ | if you version metrics, then you retrive the object history 2 years ago… with metrics that have data from maybe 2 days ago because they didn't change | 13:44 |
| jd__ | you can't always go back in time :D | 13:44 |
| sileht | jd__, yeah I have written down the schema for that and it looks ugly, | 13:45 |
| *** ddieterly has joined #openstack-ceilometer | 13:45 | |
| jd__ | yeah so please let it go :D | 13:45 |
| *** logan2 has quit IRC | 13:53 | |
| *** ddieterly has quit IRC | 13:54 | |
| *** logan2 has joined #openstack-ceilometer | 13:56 | |
| *** ildikov_ has quit IRC | 13:57 | |
| *** rbak has joined #openstack-ceilometer | 14:05 | |
| *** yprokule has joined #openstack-ceilometer | 14:05 | |
| *** yprokule1 has quit IRC | 14:08 | |
| *** _nadya_ has joined #openstack-ceilometer | 14:11 | |
| *** ildikov_ has joined #openstack-ceilometer | 14:13 | |
| *** jruano has joined #openstack-ceilometer | 14:25 | |
| openstackgerrit | Chris Dent proposed openstack/gnocchi: Only query metrics by uuid if we have uuids https://review.openstack.org/169001 | 14:25 |
| *** Ephur has quit IRC | 14:57 | |
| openstackgerrit | Merged openstack/gnocchi: Bump to oslo.serialization 1.4.0 https://review.openstack.org/166922 | 15:01 |
| *** claudiub has quit IRC | 15:01 | |
| *** cdent has quit IRC | 15:02 | |
| *** Ephur has joined #openstack-ceilometer | 15:05 | |
| sileht | jd__, sqlalchemy+postgres can make the history for us: http://docs.sqlalchemy.org/en/latest/orm/versioning.html#server-side-version-counters | 15:06 |
| sileht | jd__, but mysql doesn't support thaht ... | 15:07 |
| openstackgerrit | Merged openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/165703 | 15:07 |
| jd__ | sileht: I never thought about MySQL… | 15:08 |
| jd__ | sileht: is there a way to mimic that in MySQL, even if it's not efficient? | 15:11 |
| jd__ | sileht: I'd say let's NotImplementedError in MySQL but I don't think it's going to be popular | 15:11 |
| sileht | lol | 15:11 |
| openstackgerrit | Merged openstack/gnocchi: rest: encode all timestamps with timezone https://review.openstack.org/166923 | 15:12 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: enhance metric retrieval in controllers https://review.openstack.org/169375 | 15:15 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: do not use expect_error in tests https://review.openstack.org/169376 | 15:15 |
| *** claudiub has joined #openstack-ceilometer | 15:16 | |
| *** zigo__ is now known as zigo | 15:23 | |
| *** exploreshaifali has joined #openstack-ceilometer | 15:25 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: pass the whole metric to aggregation https://review.openstack.org/169380 | 15:27 |
| *** ildikov_ has quit IRC | 15:49 | |
| *** amalagon has quit IRC | 15:52 | |
| *** ala_ has quit IRC | 15:53 | |
| *** claudiub has quit IRC | 16:01 | |
| *** _cjones_ has joined #openstack-ceilometer | 16:05 | |
| *** jruano has quit IRC | 16:12 | |
| *** claudiub has joined #openstack-ceilometer | 16:19 | |
| *** ildikov_ has joined #openstack-ceilometer | 16:22 | |
| openstackgerrit | Matthew Edmonds proposed openstack/python-ceilometerclient: ceilometerclient fails with keystone v3 auth https://review.openstack.org/169409 | 16:24 |
| openstackgerrit | Merged openstack/gnocchi: Only query metrics by uuid if we have uuids https://review.openstack.org/169001 | 16:36 |
| *** sbfox has joined #openstack-ceilometer | 16:43 | |
| *** harlowja_away is now known as harlowja_ | 16:45 | |
| sileht | jd__, I have played with insert from select, the main issue is that mysql lock the entire table on each update | 16:46 |
| sileht | jd__, and can raise 'ERROR 1205 (HY000): Lock wait timeout exceeded; try restarting transaction' | 16:47 |
| sileht | jd__, otherwise that works | 16:47 |
| *** ankita_wagh has joined #openstack-ceilometer | 16:54 | |
| *** sbfox has quit IRC | 16:58 | |
| *** sbfox has joined #openstack-ceilometer | 16:59 | |
| openstackgerrit | Merged openstack/ceilometer: Avoid a error when py27 and py-mysql tests run in sequence https://review.openstack.org/169005 | 17:03 |
| *** amalagon has joined #openstack-ceilometer | 17:03 | |
| jd__ | sileht: ok let's do that and just advice PG :D | 17:03 |
| *** claudiub has quit IRC | 17:05 | |
| *** amalagon has quit IRC | 17:08 | |
| *** yprokule has quit IRC | 17:12 | |
| *** dikonoor has joined #openstack-ceilometer | 17:13 | |
| dikonoor | gordc , this is about https://review.openstack.org/#/c/167370/ | 17:14 |
| dikonoor | i have added a comment on what it intends to do | 17:17 |
| *** wayneeseguin has joined #openstack-ceilometer | 17:19 | |
| dikonoor | http://lists.openstack.org/pipermail/openstack-dev/2015-March/060209.html | 17:20 |
| *** msly has quit IRC | 17:20 | |
| gordc | dikonoor: cool cool. just taking a look now. | 17:21 |
| dikonoor | ok.. do you know from where the tempest UT picks up the config files? | 17:22 |
| gordc | dikonoor: but basically if your policy file doesn't have 'default' in it, it will break. | 17:22 |
| dikonoor | yeah | 17:22 |
| gordc | is this something the is explicitly stated in oslo.policy? it seems like this should really be handled there. | 17:23 |
| dikonoor | the oslo.policy handles default..so all we are trying to do with the fix is let oslo.policy do the validation | 17:24 |
| gordc | dikonoor: the policy file is part of devstack install process, https://github.com/openstack-dev/devstack/blob/master/lib/ceilometer#L176-L178 | 17:24 |
| gordc | it's just a copy of the one that exists here: https://github.com/openstack/ceilometer/blob/master/etc/ceilometer/policy.json | 17:24 |
| dikonoor | i don't think it's a copy | 17:25 |
| dikonoor | The policy.json that's used by tempest looks pretty empty compared to the one at https://github.com/openstack/ceilometer/blob/master/etc/ceilometer/policy.json | 17:26 |
| dikonoor | http://logs.openstack.org/70/167370/5/check/check-grenade-dsvm/7767ba3/logs/etc/ceilometer/policy.json.txt.gz | 17:27 |
| dikonoor | looks like this : | 17:27 |
| dikonoor | { | 17:27 |
| dikonoor | "context_is_admin": [["role:admin"]] | 17:27 |
| dikonoor | } | 17:27 |
| gordc | dikonoor: right. so grenade actually installs the previous cycle's build first and then installs current cycle build... but doesn't change any config files | 17:28 |
| gordc | it's to test backward compat | 17:28 |
| gordc | so in juno it's: https://github.com/openstack/ceilometer/blob/stable/juno/etc/ceilometer/policy.json | 17:28 |
| *** ankita_w_ has joined #openstack-ceilometer | 17:28 | |
| dikonoor | then i guess i should change the rbac.py a little further.. | 17:30 |
| *** ankita___ has joined #openstack-ceilometer | 17:30 | |
| dikonoor | With my current changes, it throws 403 , if there is no rule key and default defined | 17:31 |
| dikonoor | If a rule key is not defined and a default is also not defined , then I think it should just let the rest call go through without a rbac validation | 17:32 |
| *** ankita_wagh has quit IRC | 17:32 | |
| *** ankita_w_ has quit IRC | 17:33 | |
| dikonoor | The current rbac code https://github.com/openstack/ceilometer/blob/master/ceilometer/api/rbac.py#L49-L56 could be replaced by | 17:35 |
| dikonoor | _ENFORCER.enforce( | 17:35 |
| dikonoor | rule_name, | 17:35 |
| dikonoor | {}, | 17:35 |
| dikonoor | policy_dict) | 17:35 |
| gordc | dikonoor: hmm.. i mean it sort of does make sense to block (return 403) if there's no default | 17:35 |
| dikonoor | also depends on the behavior implemented in the other services. | 17:36 |
| dikonoor | I mean to be consistent | 17:36 |
| *** safchain has quit IRC | 17:36 | |
| dikonoor | I agree with you that ideally all calls defined should be blocked.That makes more sense | 17:37 |
| dikonoor | sorry..meant not defined | 17:37 |
| dikonoor | all calls that are NOT defined should be blocked , which is what my current changes do | 17:38 |
| dikonoor | if the tempest config files cannot be changed , then the tempest UT cases need to be fixed | 17:39 |
| dikonoor | Currently they expect a 200 but without any entries in the policy.json, all of them will return 403 | 17:39 |
| gordc | dikonoor: how did policy work in juno? was default just assumed? | 17:40 |
| gordc | if we do need a patch, you'd probably want something like this: https://github.com/openstack-dev/grenade/blob/master/from-icehouse/upgrade-ceilometer | 17:41 |
| gordc | which can allow you to refresh the file. | 17:42 |
| dikonoor | I am not sure if there was rbac implemented for ceilometer in juno.. | 17:43 |
| *** ankita_wagh has joined #openstack-ceilometer | 17:44 | |
| dikonoor | When we tried few months back with juno, it wasn't implemented | 17:44 |
| dikonoor | it was probably implemented as part of https://review.openstack.org/#/c/115717 | 17:45 |
| dikonoor | and this implementation disregards default values and skips RBAC for rules not defined in the policy.json | 17:46 |
| dikonoor | gordc, I did not quite follow what this does >>https://github.com/openstack-dev/grenade/blob/master/from-icehouse/upgrade-ceilometer | 17:46 |
| dikonoor | I mean from the context of what we are discussing | 17:46 |
| *** ankita___ has quit IRC | 17:47 | |
| gordc | dikonoor: basically what grenade is doing right now is install juno, test juno, install kilo, test kilo with juno config files | 17:47 |
| gordc | the upgrade part is when you need to force a change ... ie. if you add a new option or config file between cycles. | 17:48 |
| gordc | so in example, it's copying in a new config file before testing the kilo against juno config files (or in examples cases, testing juno against icehouse config files) | 17:49 |
| dikonoor | ok.. so you mean to say that we would need to update the policy.json and push it in so that tempest/grenade picks this new one instead of the juno one | 17:50 |
| dikonoor | ok..so i guess there should be something like https://github.com/openstack-dev/grenade/blob/master/from-juno/upgrade-ceilometer that upgrades the policy.json | 17:53 |
| *** ankita_w_ has joined #openstack-ceilometer | 17:55 | |
| gordc | right... or we modify the policy in juno branch. | 17:55 |
| dikonoor | ok..let me try the grenade change.. I had set the default as "" in the policy json file so that all UTs pass like before | 17:57 |
| *** ankita_wagh has quit IRC | 17:58 | |
| gordc | i just find it strange that default is missing juno but it's fine... i understand we implemented rbac but i guess oslo.policy requires default value now? | 17:58 |
| openstackgerrit | Matthew Edmonds proposed openstack/python-ceilometerclient: ceilometerclient insecure argument no longer works https://review.openstack.org/169442 | 18:00 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: return resource objects rather than dict https://review.openstack.org/169302 | 18:01 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: enhance metric retrieval in controllers https://review.openstack.org/169375 | 18:01 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: do not use expect_error in tests https://review.openstack.org/169376 | 18:01 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: pass the whole metric to aggregation https://review.openstack.org/169380 | 18:01 |
| dikonoor | I am not sure if oslo.policy was implemented differently before. But in general , any rbac logic should first check if the rule is present , if no fall back on the default rule and that's how it works for all other services | 18:01 |
| dikonoor | the policy.json doesn't neccessarily need to have a default value with oslo.policy but if a specific rule is not defined , it uses the default | 18:02 |
| gordc | i guess before it would make up it's own default if not present. | 18:03 |
| dikonoor | no , earlier the default was to allow all calls (and ignore the default value in policy.json) | 18:04 |
| dikonoor | with the proposed changes , we are trying to get the default value from the policy.json to be used | 18:05 |
| *** jruano has joined #openstack-ceilometer | 18:07 | |
| dikonoor | so i will leave the proposed rbac.py changes the way it is | 18:09 |
| dikonoor | and try to push in policy.json related changes to grenade (the other option is to modify the grenade test cases) | 18:10 |
| dikonoor | Thanks gordc | 18:11 |
| gordc | yeah, i meant in juno, pre-rbac change... regardless, thanks for lookin at it. | 18:16 |
| *** Longgeek has quit IRC | 18:16 | |
| *** Longgeek has joined #openstack-ceilometer | 18:17 | |
| *** edmondsw has joined #openstack-ceilometer | 18:17 | |
| *** msly has joined #openstack-ceilometer | 18:17 | |
| edmondsw | gordc: proposed a couple simple patches fixing keystone v3 and insecure support in ceilometerclient, if you can take a look | 18:18 |
| edmondsw | or anyone else... | 18:18 |
| edmondsw | https://review.openstack.org/#/c/169409/ | 18:19 |
| edmondsw | https://review.openstack.org/#/c/169442/ | 18:19 |
| edmondsw | gordc: is there a plan/schedule for a new ceilometerclient release that could include these? | 18:20 |
| *** _nadya_ has quit IRC | 18:21 | |
| gordc | edmondsw: i don't have ability to cut releases but i can let eglynn know. | 18:22 |
| edmondsw | gordc, sure... wasn't asking for you to cut it, just whether there was already something in plan or I needed to go ask for one | 18:22 |
| eglynn | edmondsw: we usually bring it up at the weekly ceilometer meeting (Thurs @1500UTC) if needed | 18:23 |
| * gordc deletes what i was typing | 18:23 | |
| eglynn | gordc: sorry :) | 18:23 |
| gordc | :) | 18:24 |
| *** packet has joined #openstack-ceilometer | 18:24 | |
| edmondsw | gordc, I'm not certain these are the last two issues we'll find... looked like some other places were missing things they should have, but didn't notice them causing a problem | 18:24 |
| edmondsw | gordc, eglynn, on a separate topic... y'all see http://lists.openstack.org/pipermail/openstack-dev/2015-March/060209.html ? | 18:24 |
| eglynn | edmondsw: I'll respond on the ML | 18:27 |
| edmondsw | eglynn, thanks | 18:27 |
| *** dikonoor has quit IRC | 18:28 | |
| gordc | edmondsw: we were discussing it ^^ | 18:28 |
| gordc | hmm..never just noticed we stopped right before you came joined. | 18:29 |
| edmondsw | gordc, that's what I get for forgetting to login to irc this morning... | 18:29 |
| gordc | edmondsw: http://eavesdrop.openstack.org/irclogs/%23openstack-ceilometer/%23openstack-ceilometer.2015-03-31.log | 18:30 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements https://review.openstack.org/161571 | 18:30 |
| *** zhenq has joined #openstack-ceilometer | 18:35 | |
| eglynn | gordc, edmondsw: so as I understood it, the aspect of fabiog's original RBAC that allowed API calls by default was deliberate | 18:37 |
| eglynn | this was specifically to provide for backward compatability, i.e. an empty policy.json to work as before | 18:38 |
| edmondsw | eglynn, sort of... I think there was a flaw in his logic, as I outlined in the email | 18:38 |
| * eglynn re-reads | 18:39 | |
| edmondsw | the code fabiog wrote expects a "segregation" rule in the policy.json, which won't be there with an upgrade from juno | 18:39 |
| eglynn | one sec | 18:40 |
| *** afazekas has quit IRC | 18:43 | |
| edmondsw | finished reading back on the chat with dikonoor earlier | 18:43 |
| edmondsw | the qa guys told me we'd need to add something in the release notes if we're going to put something in upgrade-ceilometer for this | 18:44 |
| edmondsw | I think we probably have to, though | 18:44 |
| eglynn | edmondsw: OK, in order to fully be backward compat to old configs, we'd have to explicitly deal with the absence of a 'segregation' rule and fallback to 'context_id_admin' as before | 18:50 |
| eglynn | edmondsw: so say we take https://github.com/openstack/ceilometer/blob/master/ceilometer/api/rbac.py#L80-82 | 18:50 |
| eglynn | edmondsw: and check first for ... segregation_rule = 'segregation' if 'segregation' in _ENFORCER.rules.keys() else 'context_is_admin' | 18:52 |
| edmondsw | eglynn, yep, I think we could do that | 18:52 |
| edmondsw | eglynn, and something similar with the default rule dikonoor is trying to add | 18:53 |
| *** jruano has quit IRC | 18:54 | |
| eglynn | edmondsw: yeah, we'd still have broken expectations around the the default rule handling | 18:55 |
| eglynn | edmondsw: ... so in that case, we'd need the absence of both an explicit *and* a default rule to still boil down to allow | 18:55 |
| edmondsw | eglynn, could change dikonoor's proposal in https://review.openstack.org/#/c/167370/ to act the original way if default rule doesn't exist, or the way she's written if it does | 18:56 |
| edmondsw | eglynn, gordc, I still think it would be better to require that a default rule is set, though... | 18:57 |
| edmondsw | eglynn, gord, I'm not sure we should be trying to maintain backward compat with the lack of rbac support we had in juno | 18:58 |
| eglynn | edmondsw: "better to require that a default rule is set" ==> but that automatically breaks the upgrade-with-no-config-change requirement, amiright? | 18:58 |
| edmondsw | eglynn, yes | 18:58 |
| edmondsw | eglynn, but is that really a requirement? Exceptions can be made, that's what the upgrade-ceilometer file is for, and this seems like a case where it may be warranted | 18:59 |
| edmondsw | eglynn, if not now, eventually we'll want to start acting like all the other projects here, right? | 19:00 |
| eglynn | edmondsw: the idea was to allow distros selectively backport that RBAC feature or not ... and for it not to break things if an old policy.json was used | 19:00 |
| *** _cjones_ has quit IRC | 19:00 | |
| *** jruano has joined #openstack-ceilometer | 19:03 | |
| eglynn | edmondsw: do you want me to summarize the discussion above on the ML? | 19:12 |
| edmondsw | eglynn, yes please... I think we can make it work in a way that is backward compatible if that's what we want, and maybe it is for now, but eventually we might want to cut that tie and require a normal policy.json | 19:13 |
| eglynn | edmondsw: yep, in liberty we could do so | 19:14 |
| *** vishwanathj has quit IRC | 19:14 | |
| edmondsw | eglynn, I think that's fine... I'm testing a change to dikonoor's work that I think should work for now | 19:14 |
| eglynn | coolness | 19:14 |
| edmondsw | tx | 19:15 |
| *** sbfox has quit IRC | 19:19 | |
| *** cdent has joined #openstack-ceilometer | 19:22 | |
| *** sbfox has joined #openstack-ceilometer | 19:26 | |
| eglynn | edmondsw: done | 19:27 |
| edmondsw | eglynn, tx | 19:27 |
| eglynn | edmondsw: I'm can target https://bugs.launchpad.net/ceilometer/+bug/1435855 at kilo-rc1 if those proposed changes to the patch will fly? | 19:28 |
| openstack | Launchpad bug 1435855 in Ceilometer "Default rule does not work in ceilometer policy.json" [High,In progress] - Assigned to Divya K Konoor (dikonoor) | 19:28 |
| edmondsw | eglynn, my tests worked... I'm writing up an email to Divya on how she can modify her patch | 19:28 |
| eglynn | edmondsw: coolness, thank you for your help on this! | 19:28 |
| edmondsw | eglynn, np... tx for your time as well | 19:29 |
| *** alexpilotti has joined #openstack-ceilometer | 19:41 | |
| *** jruano has quit IRC | 19:44 | |
| *** vishwanathj has joined #openstack-ceilometer | 19:46 | |
| *** jruano has joined #openstack-ceilometer | 19:49 | |
| openstackgerrit | Merged openstack/ceilometer: support ability to skip message signing https://review.openstack.org/167853 | 19:51 |
| openstackgerrit | Merged openstack/ceilometer: remove instance:<flavor> meter https://review.openstack.org/168466 | 19:52 |
| edmondsw | Jenkins failed on gate-tempest-dsvm-neutron-src-python-ceilometerclient-icehouse with one of my proposed ceilometerclient changes... for reasons that don't appear to have anything to do with my changes | 19:52 |
| edmondsw | is there a way to just recheck that, rather than recheck everything? | 19:53 |
| cdent | not really edmondsw, a “recheck” response is all or nothing | 19:55 |
| cdent | pretty much what you gotta do | 19:55 |
| edmondsw | cdent, I'll bit the bullet then... tx | 19:55 |
| cdent | good luck | 19:57 |
| * cdent bows out | 19:57 | |
| *** cdent has quit IRC | 19:57 | |
| *** _cjones_ has joined #openstack-ceilometer | 20:01 | |
| *** eglynn has quit IRC | 20:11 | |
| *** jruano has quit IRC | 20:19 | |
| *** alexpilotti has quit IRC | 20:21 | |
| *** _cjones_ has quit IRC | 20:22 | |
| *** maishsk has joined #openstack-ceilometer | 20:26 | |
| *** ankita_wagh has joined #openstack-ceilometer | 20:26 | |
| *** ankita_w_ has quit IRC | 20:29 | |
| *** _cjones_ has joined #openstack-ceilometer | 20:33 | |
| *** _cjones_ has quit IRC | 20:40 | |
| *** david-lyle has quit IRC | 20:40 | |
| *** promulo has joined #openstack-ceilometer | 20:41 | |
| *** promulo__ has quit IRC | 20:42 | |
| *** _cjones_ has joined #openstack-ceilometer | 20:42 | |
| *** Ephur has quit IRC | 20:45 | |
| *** claudiub has joined #openstack-ceilometer | 20:51 | |
| *** eglynn has joined #openstack-ceilometer | 20:53 | |
| *** maishsk_ has joined #openstack-ceilometer | 21:04 | |
| *** maishsk has quit IRC | 21:07 | |
| *** packet has quit IRC | 21:08 | |
| *** packet has joined #openstack-ceilometer | 21:10 | |
| *** sbfox has quit IRC | 21:16 | |
| *** jruano has joined #openstack-ceilometer | 21:17 | |
| openstackgerrit | Matthew Edmonds proposed openstack/python-ceilometerclient: ceilometerclient insecure argument no longer works https://review.openstack.org/169442 | 21:17 |
| *** _nadya_ has joined #openstack-ceilometer | 21:21 | |
| *** exploreshaifali has quit IRC | 21:25 | |
| *** _nadya_ has quit IRC | 21:25 | |
| *** sbfox has joined #openstack-ceilometer | 21:26 | |
| *** gordc has quit IRC | 21:34 | |
| *** sbfox has quit IRC | 21:38 | |
| *** maishsk_ has quit IRC | 21:40 | |
| *** sbfox has joined #openstack-ceilometer | 21:42 | |
| *** parkerr_ has joined #openstack-ceilometer | 21:44 | |
| *** eglynn has quit IRC | 21:55 | |
| *** jruano has quit IRC | 22:00 | |
| *** harlowja_ is now known as harlowja_away | 22:04 | |
| *** harlowja_away is now known as harlowja_ | 22:07 | |
| *** harlowja_ has quit IRC | 22:21 | |
| *** ankita_w_ has joined #openstack-ceilometer | 22:22 | |
| *** ankita___ has joined #openstack-ceilometer | 22:22 | |
| *** ankit____ has joined #openstack-ceilometer | 22:25 | |
| *** ankita_wagh has quit IRC | 22:25 | |
| *** ankita_w_ has quit IRC | 22:26 | |
| *** harlowja has joined #openstack-ceilometer | 22:26 | |
| *** ankita___ has quit IRC | 22:28 | |
| *** ankita_wagh has joined #openstack-ceilometer | 22:29 | |
| *** ankit____ has quit IRC | 22:32 | |
| *** ankita_w_ has joined #openstack-ceilometer | 22:35 | |
| *** packet has quit IRC | 22:36 | |
| *** promulo__ has joined #openstack-ceilometer | 22:37 | |
| *** promulo has quit IRC | 22:37 | |
| *** ankita_wagh has quit IRC | 22:39 | |
| *** rbak has quit IRC | 22:47 | |
| *** devlaps has joined #openstack-ceilometer | 22:50 | |
| *** packet has joined #openstack-ceilometer | 22:50 | |
| *** devlaps has quit IRC | 22:54 | |
| *** claudiub has quit IRC | 23:04 | |
| *** sbfox has quit IRC | 23:11 | |
| *** edmondsw has quit IRC | 23:24 | |
| *** tongli has quit IRC | 23:30 | |
| *** changbl has joined #openstack-ceilometer | 23:38 | |
| *** zhenq has quit IRC | 23:51 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!