| *** zhenq has joined #openstack-ceilometer | 00:07 | |
| *** Kennan2 has quit IRC | 00:08 | |
| *** Kennan has joined #openstack-ceilometer | 00:08 | |
| *** promulo has joined #openstack-ceilometer | 00:09 | |
| *** promulo__ has quit IRC | 00:12 | |
| *** jaypipes has quit IRC | 00:27 | |
| *** claudiub has quit IRC | 00:43 | |
| *** ankita_wagh has quit IRC | 00:43 | |
| *** ddieterly has joined #openstack-ceilometer | 00:46 | |
| *** Longgeek has joined #openstack-ceilometer | 00:50 | |
| *** Longgeek has quit IRC | 01:07 | |
| *** _cjones_ has quit IRC | 01:08 | |
| *** Longgeek has joined #openstack-ceilometer | 01:08 | |
| *** ddieterly has quit IRC | 01:10 | |
| *** ddieterly has joined #openstack-ceilometer | 01:24 | |
| *** gzhai_mob has joined #openstack-ceilometer | 01:27 | |
| *** gzhai_mob has quit IRC | 01:29 | |
| *** sbfox has joined #openstack-ceilometer | 01:31 | |
| *** llu has joined #openstack-ceilometer | 01:35 | |
| *** sbfox has quit IRC | 01:51 | |
| *** swamireddy has quit IRC | 02:00 | |
| *** swamireddy has joined #openstack-ceilometer | 02:00 | |
| *** boris-42 has quit IRC | 02:02 | |
| *** adam_g has quit IRC | 02:02 | |
| *** jroll has quit IRC | 02:02 | |
| *** boris-42 has joined #openstack-ceilometer | 02:05 | |
| *** jroll has joined #openstack-ceilometer | 02:07 | |
| *** adam_g has joined #openstack-ceilometer | 02:07 | |
| *** adam_g has quit IRC | 02:07 | |
| *** adam_g has joined #openstack-ceilometer | 02:07 | |
| *** jroll has quit IRC | 02:08 | |
| *** jroll has joined #openstack-ceilometer | 02:08 | |
| *** julim has quit IRC | 02:12 | |
| *** botang has quit IRC | 02:12 | |
| *** harlowja is now known as harlowja_away | 02:22 | |
| *** sbfox has joined #openstack-ceilometer | 02:26 | |
| *** jasonamyers_ has joined #openstack-ceilometer | 02:28 | |
| openstackgerrit | Komei Shimamura proposed openstack/ceilometer: Fix a variable name mistake https://review.openstack.org/170644 | 02:32 |
|---|---|---|
| *** fnaval has quit IRC | 02:47 | |
| *** fnaval has joined #openstack-ceilometer | 02:54 | |
| *** sbfox has quit IRC | 03:04 | |
| *** julim has joined #openstack-ceilometer | 03:08 | |
| *** boris-42 has quit IRC | 03:18 | |
| *** zhenq has quit IRC | 03:19 | |
| openstackgerrit | ZhiQiang Fan proposed openstack/ceilometer: WIP: ugly fix: pymongo 3.0 breaks ci gate https://review.openstack.org/171458 | 03:25 |
| *** ddieterly has quit IRC | 03:52 | |
| *** rwsu has quit IRC | 04:07 | |
| *** _cjones_ has joined #openstack-ceilometer | 04:19 | |
| *** _cjones_ has quit IRC | 04:19 | |
| *** _cjones_ has joined #openstack-ceilometer | 04:20 | |
| *** _cjones_ has quit IRC | 04:28 | |
| *** amalagon has joined #openstack-ceilometer | 04:29 | |
| *** llu has quit IRC | 04:29 | |
| *** llu has joined #openstack-ceilometer | 04:30 | |
| *** deepthi has joined #openstack-ceilometer | 04:31 | |
| *** sbfox has joined #openstack-ceilometer | 04:35 | |
| *** ddieterly has joined #openstack-ceilometer | 04:52 | |
| *** ddieterly has quit IRC | 04:58 | |
| *** fnaval has quit IRC | 05:06 | |
| *** botang has joined #openstack-ceilometer | 05:13 | |
| *** julim has quit IRC | 05:15 | |
| *** botang has quit IRC | 05:18 | |
| *** yatin has joined #openstack-ceilometer | 05:24 | |
| *** changbl has joined #openstack-ceilometer | 05:31 | |
| *** jasonamyers_ has quit IRC | 05:33 | |
| *** ddieterly has joined #openstack-ceilometer | 05:54 | |
| *** ildikov has quit IRC | 05:54 | |
| *** ddieterly has quit IRC | 05:58 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 06:36 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 06:36 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Don't touch the sqla orm object https://review.openstack.org/171480 | 06:36 |
| *** cmyster has quit IRC | 06:38 | |
| *** dikonoor has joined #openstack-ceilometer | 06:46 | |
| *** yprokule has joined #openstack-ceilometer | 06:47 | |
| *** Ala has joined #openstack-ceilometer | 06:54 | |
| *** ddieterly has joined #openstack-ceilometer | 06:55 | |
| *** ddieterly has quit IRC | 06:59 | |
| dikonoor | I have a patch that has gone through reviews, but i would like to get some cores attention to this, could you please help? >> https://review.openstack.org/#/c/167370/ | 07:00 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 07:03 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 07:03 |
| *** ildikov has joined #openstack-ceilometer | 07:11 | |
| *** sbfox has quit IRC | 07:12 | |
| *** shardy has joined #openstack-ceilometer | 07:13 | |
| *** _nadya_ has joined #openstack-ceilometer | 07:17 | |
| *** afazekas has joined #openstack-ceilometer | 07:23 | |
| *** _nadya_ has quit IRC | 07:28 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 07:33 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 07:33 |
| *** _nadya_ has joined #openstack-ceilometer | 07:35 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: sqlalchemy: use RDBMS check constraint where available https://review.openstack.org/170544 | 07:37 |
| *** _nadya_ has quit IRC | 07:51 | |
| *** safchain has joined #openstack-ceilometer | 07:51 | |
| *** ildikov has quit IRC | 07:53 | |
| *** ddieterly has joined #openstack-ceilometer | 07:55 | |
| *** eglynn has joined #openstack-ceilometer | 07:59 | |
| *** ddieterly has quit IRC | 08:00 | |
| *** ildikov has joined #openstack-ceilometer | 08:09 | |
| *** IBerezovskiy has joined #openstack-ceilometer | 08:12 | |
| *** llu has quit IRC | 08:25 | |
| *** llu has joined #openstack-ceilometer | 08:25 | |
| *** ildikov has quit IRC | 08:41 | |
| *** amalagon has quit IRC | 08:48 | |
| *** _nadya_ has joined #openstack-ceilometer | 08:55 | |
| *** ddieterly has joined #openstack-ceilometer | 08:56 | |
| *** ildikov has joined #openstack-ceilometer | 08:57 | |
| *** ddieterly has quit IRC | 09:01 | |
| *** theanalyst has quit IRC | 09:04 | |
| *** theanalyst has joined #openstack-ceilometer | 09:07 | |
| dikonoor | ildikov : This is about https://review.openstack.org/#/c/167370/. Did you get a chance to sleep on it ? :) | 09:12 |
| dikonoor | eglynn : Could you also take a look at https://review.openstack.org/#/c/167370/ and check if it's good to go ? | 09:13 |
| ildikov | dikonoor: I still prefer the way of adding tests | 09:14 |
| eglynn | dikonoor: on a call right now, but will look again soon after | 09:14 |
| ildikov | dikonoor: this is not against you, I annoy others too with this testing topic... :) | 09:16 |
| dikonoor | ildikov: :) I notice that the ceilometer api tests use the pecan.testing module and all the current api tests are written in such a way that the moment I add any rule (for eg. get_meters: "rule:role_admin") , other test cases break.I think it happens because none of them set any header values with role and other data. Am I right about this observation ? | 09:18 |
| *** _nadya_ has quit IRC | 09:18 | |
| dikonoor | ildikov: I see some test cases under ceilometer/tests/api/v2/test_acl_scenarios.py for access related stuff but don't think they really hit the policy.json | 09:21 |
| ildikov | dikonoor: I messed up policy.json yesterday, but the tests did not care about it... | 09:23 |
| ildikov | dikonoor: I mean locally | 09:23 |
| dikonoor | ildikov: well.that's probably because of the bug in the old code. | 09:24 |
| dikonoor | ildikov : Once the proposed changes go in , that might not be the case. The existing policy.json neither has a default rule nor has any entry for a specific rest api . Due to this the current code never hits the policy.json validation | 09:25 |
| ildikov | dikonoor: yeap, true, will check with your patch too | 09:25 |
| dikonoor | ildikov: With the proposed changes, if the rule for a specific rest api is not found, it defaults to the default rule and starts to check for permissions | 09:27 |
| dikonoor | ildikov : If I were to set the default rule to something specific (I have left it as "") ..say like rule : "role:admin" , then all the existing api test cases will fail..Because the test cases don't pass in role information | 09:28 |
| ildikov | dikonoor: well, role:admin as default would be too strict anyway | 09:29 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 09:29 |
| ildikov | dikonoor: well, what's missing is the real unit tests that fix the changed functionality and to check the api tests, which are more functional tests, but the restructirng is still an ongoing process and will finally happen at the beginning of Liberty | 09:30 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 09:31 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 09:31 |
| dikonoor | ildikov : If I have to write few test cases around the changes I have proposed , I should ideally check for things like role , ownership , "!" and so on.. That would mean that I add some rules in the policy.json . However , any addition in policy.json would mean breaking existing test cases | 09:32 |
| ildikov | dikonoor: for instance the unit tests looks like thi sin Nova: https://github.com/openstack/nova/blob/master/nova/tests/unit/test_policy.py | 09:32 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 09:34 |
| dikonoor | ildikov : This would mean writing the entire test suite for ceilometer api permissions for the small bug fix I made :) | 09:40 |
| dikonoor | ildikov : But thanks for pointing me to the test_policy.py | 09:40 |
| *** _nadya_ has joined #openstack-ceilometer | 09:53 | |
| *** ddieterly has joined #openstack-ceilometer | 09:57 | |
| *** ddieterly has quit IRC | 10:01 | |
| *** htruta has quit IRC | 10:05 | |
| *** htruta has joined #openstack-ceilometer | 10:09 | |
| *** cdent has joined #openstack-ceilometer | 10:13 | |
| *** elynn has joined #openstack-ceilometer | 10:26 | |
| dikonoor | eglynn : Please take a look at https://review.openstack.org/#/c/167370/ , when you get a chance | 10:28 |
| *** cdent has quit IRC | 10:30 | |
| eglynn | dikonoor: what's the tl;dr on the discussion above with ildikov on providing a test for this? | 10:31 |
| eglynn | dikonoor: conclusion is that this is too disruptive to accomodation within the current testing framework? | 10:31 |
| eglynn | ildikov: do we have a rough consensus to land this patch for kilo-rc1 and then subsequently revisit the testing logic? | 10:32 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 10:34 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 10:34 |
| dikonoor | egylnn : we haven't arrived at a consensus yet.. It will be better if the testing logic is revisited subsequently.. | 10:35 |
| ildikov | eglynn: dikonoor: I'm not against of fixing this in a subsequent patch (series) | 10:36 |
| eglynn | ildikov: that sounds pragmatic, thanks! :) | 10:37 |
| eglynn | dikonoor: k, let's get this patch landed! ... thank you for your effeorts on this :) | 10:38 |
| eglynn | *efforts | 10:38 |
| dikonoor | ildikov : That can be done : ) I can open another LP bug to get the test cases fixed | 10:39 |
| ildikov | eglynn: I guess the gate is still affected by the pymongo bump though... | 10:39 |
| ildikov | dikonoor: cool, sounds like a plan, thanks :) | 10:40 |
| ildikov | eglynn: thanks for the administration | 10:41 |
| eglynn | ildikov: LOL on the administration | 10:42 |
| ildikov | eglynn: I guess it's a reflex thing, you're too experienced in this :) | 10:44 |
| dikonoor | ildikov , egylnn : Thanks very much Folks ..that was my first OpenStack contribution..phew ! | 10:45 |
| ildikov | dikonoor: welcome on board! :) | 10:46 |
| dikonoor | ildikov , egylnn : Thanks , as promised , I will open a LP bug for the test cases and add you guys as reviewers | 10:47 |
| ildikov | dikonoor: cool, thanks | 10:50 |
| *** claudiub has joined #openstack-ceilometer | 10:51 | |
| *** f13o has joined #openstack-ceilometer | 10:52 | |
| openstackgerrit | Merged openstack/gnocchi: Don't touch the sqla orm object https://review.openstack.org/171480 | 10:53 |
| *** ddieterly has joined #openstack-ceilometer | 10:58 | |
| *** _nadya_ has quit IRC | 10:59 | |
| *** dikonoor has quit IRC | 11:00 | |
| *** ddieterly has quit IRC | 11:02 | |
| *** claudiub has quit IRC | 11:06 | |
| *** noye has joined #openstack-ceilometer | 11:20 | |
| *** noye has quit IRC | 11:28 | |
| *** pauldix has joined #openstack-ceilometer | 11:51 | |
| *** cdent has joined #openstack-ceilometer | 11:56 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 12:01 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 12:01 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Avoid query metric object twice from indexer https://review.openstack.org/171590 | 12:01 |
| *** claudiub has joined #openstack-ceilometer | 12:02 | |
| *** deepthi has quit IRC | 12:06 | |
| *** cmyster has joined #openstack-ceilometer | 12:19 | |
| *** cmyster has joined #openstack-ceilometer | 12:19 | |
| *** boris-42 has joined #openstack-ceilometer | 12:23 | |
| *** jdandrea has quit IRC | 12:23 | |
| *** pauldix has quit IRC | 12:32 | |
| *** cmyster has quit IRC | 12:35 | |
| *** gordc has joined #openstack-ceilometer | 12:36 | |
| *** jdandrea has joined #openstack-ceilometer | 12:38 | |
| sileht | jd__, I have added the tests ^^ | 12:38 |
| sileht | jd__, I'm preparing a new review with some docs/examples | 12:38 |
| jd__ | sileht: cool | 12:38 |
| jd__ | I saw the tests for the timespam | 12:39 |
| *** jasonamyers_ has joined #openstack-ceilometer | 12:39 | |
| sileht | jd__, I have finished to fight with sqlalchemy to make my union query working with the 'orm joinedquery' | 12:39 |
| * jd__ papats sileht | 12:39 | |
| jd__ | joy of ORM :) | 12:40 |
| *** jasonamyers has quit IRC | 12:40 | |
| *** prad has quit IRC | 12:41 | |
| *** jasonamyers_ has quit IRC | 12:42 | |
| *** jasonamyers has joined #openstack-ceilometer | 12:42 | |
| *** _nadya_ has joined #openstack-ceilometer | 12:42 | |
| openstackgerrit | Merged openstack/gnocchi: gabbi: merge metric and metrics test files https://review.openstack.org/171254 | 12:45 |
| *** yatin has quit IRC | 12:46 | |
| *** jdandrea has quit IRC | 12:46 | |
| *** jdandrea has joined #openstack-ceilometer | 12:51 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: add support for filtering by duration https://review.openstack.org/171606 | 12:58 |
| *** ddieterly has joined #openstack-ceilometer | 12:59 | |
| *** cmyster has joined #openstack-ceilometer | 13:02 | |
| *** cmyster has joined #openstack-ceilometer | 13:02 | |
| *** ddieterly has quit IRC | 13:04 | |
| *** cdent has quit IRC | 13:11 | |
| *** julim has joined #openstack-ceilometer | 13:11 | |
| *** edmondsw has joined #openstack-ceilometer | 13:12 | |
| *** jmank has joined #openstack-ceilometer | 13:12 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: Move JSON encoding routing in its own module https://review.openstack.org/171616 | 13:14 |
| *** pauldix has joined #openstack-ceilometer | 13:15 | |
| *** jmanko has quit IRC | 13:21 | |
| *** jmank has quit IRC | 13:21 | |
| *** jmank has joined #openstack-ceilometer | 13:21 | |
| sileht | jd__, any reason to have PATCH resource that return 200 with no data ? | 13:22 |
| *** zul has quit IRC | 13:22 | |
| *** zul has joined #openstack-ceilometer | 13:22 | |
| sileht | jd__, it's return 204, but why do we not return 200 with the new resource data ? | 13:23 |
| jd__ | sileht: to avoid fetching the data once updated I'd say | 13:23 |
| *** Longgeek has quit IRC | 13:23 | |
| jd__ | though we may have it available so I don't know otherwise | 13:24 |
| sileht | jd__, the indexer return the full resource | 13:24 |
| jd__ | yeah so that's not the reason | 13:24 |
| *** cdent has joined #openstack-ceilometer | 13:24 | |
| jd__ | sileht: there might be no reason! | 13:24 |
| sileht | jd__, ok I will send something, so | 13:25 |
| sileht | jd__, I need your https://review.openstack.org/#/c/171606/ to add the same thing for lifetime | 13:26 |
| jd__ | hint hint cdent ^^^^ | 13:26 |
| cdent | pardon what jd__ ? | 13:27 |
| jd__ | cdent: you should review that patch :-) | 13:27 |
| * cdent has been doing paperwork :( | 13:28 | |
| jd__ | I feel your pain cdent | 13:30 |
| *** thomasem has joined #openstack-ceilometer | 13:34 | |
| cdent | jd__: interesting test failure on that review | 13:34 |
| cdent | http://logs.openstack.org/06/171606/1/check/gate-gnocchi-python27/f87b647/console.html#_2015-04-08_13_16_03_441 | 13:35 |
| *** zhenq has joined #openstack-ceilometer | 13:35 | |
| *** thomasem has quit IRC | 13:36 | |
| *** thomasem has joined #openstack-ceilometer | 13:41 | |
| *** jaypipes has joined #openstack-ceilometer | 13:48 | |
| *** noye has joined #openstack-ceilometer | 13:49 | |
| *** ddieterly has joined #openstack-ceilometer | 13:52 | |
| *** jmatthew_ has joined #openstack-ceilometer | 13:59 | |
| gordc | ildikov: i forgot this yesterday but these are the equivalent functions: http://paste.openstack.org/show/200439/ | 14:00 |
| ildikov | gordc: cool, thank you very much! | 14:02 |
| gordc | np... re: meter-list there isn't one because well, the meter shouldn't reallly exist on its own. | 14:03 |
| ildikov | gordc: and these are supported on all back ends now, right? | 14:03 |
| ildikov | the meter-list is not an issue, I realized it too after | 14:04 |
| ildikov | if the deprecation will be marked, then it will be totally fine to not have this in meter-list | 14:04 |
| ildikov | ... I mean when we finally remove it | 14:04 |
| gordc | there's patches for mongo and sql... elasticsearch is not applicable. | 14:05 |
| ildikov | gordc: hbase | 14:06 |
| ildikov | ? | 14:06 |
| gordc | it's not available in hbase... because hbase is not close to feature parity | 14:07 |
| gordc | i'd actually like to know if anyone is using it... because i'm not even sure it works. | 14:08 |
| ildikov | ok, so it is acceptable that it will not be available there | 14:08 |
| gordc | i personally think so but we can reach out to community on that? | 14:09 |
| ildikov | I'm not against your viewpoint, I just ask these many questions to be sure we don't miss anything major | 14:10 |
| *** exploreshaifali has joined #openstack-ceilometer | 14:10 | |
| ildikov | I don't think either that it would be widely used, but maybe it worth a question | 14:10 |
| gordc | ildikov: thanks for keeping me honest.lol | 14:11 |
| ildikov | gordc: LOL :), you're welcome | 14:11 |
| ildikov | gordc: BTW, just like all our back end drivers it is also marked as production ready | 14:12 |
| gordc | hbase? | 14:12 |
| ildikov | yes | 14:13 |
| gordc | errr... we should probably look into that. DinaBelova, _nadya_, could probably correct me but i don't think even mirantis is using hbase... just playing with it. | 14:14 |
| ildikov | gordc: I think I owe you some beers or smth because of the backward compat stuff and other reminders... :) | 14:15 |
| gordc | i won't turn down beers. | 14:15 |
| ildikov | ok :) | 14:16 |
| *** ddieterly has quit IRC | 14:18 | |
| *** ddieterly has joined #openstack-ceilometer | 14:18 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 14:19 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 14:19 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Avoid query metric object twice from indexer https://review.openstack.org/171590 | 14:19 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparaison of timestamp field to None https://review.openstack.org/171669 | 14:19 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 14:19 |
| gordc | sileht: was there a recent oslo.messaging patch that resolved tihs? https://bugs.launchpad.net/ceilometer/+bug/1393708 | 14:21 |
| openstack | Launchpad bug 1393708 in Ceilometer "ceilometer collector can not reconnect to rabbitmq after restarting rabbitmq" [Undecided,New] | 14:21 |
| ildikov | gordc: anyway, I will update the docs with the deprecation info and then we will see what to do with the hbase case. does it sound ok? | 14:22 |
| gordc | ildikov: sounds good to me. | 14:23 |
| sileht | gordc, I don't known, the reconnection code was a really crap before | 14:23 |
| ildikov | gordc: cool, tnx | 14:23 |
| sileht | gordc, but the kilo version have improve a lot that | 14:23 |
| gordc | sileht: hmm... i'll try simulating it. | 14:24 |
| _nadya_ | gordc: yep, you're right | 14:26 |
| _nadya_ | gordc: what's the question about hbase :)? | 14:27 |
| sileht | gordc, the error message looks like a issue with the ssl version negociation | 14:27 |
| jd__ | sileht: I'm pondering about removing the PreciseTimestamp sometime… | 14:27 |
| sileht | jd__, the precise timestamp can be useful for ETAG | 14:28 |
| *** prad has joined #openstack-ceilometer | 14:28 | |
| *** rbak has joined #openstack-ceilometer | 14:29 | |
| sileht | jd__, because 'revision' is not present for the latest version and allows an attackers to known the number of records for your databases | 14:29 |
| jd__ | sileht: ok but can we use that http://dev.mysql.com/doc/refman/5.6/en/fractional-seconds.html instead? | 14:29 |
| sileht | jd__, and 'id+lifetime_from' make a resource history row unique, so it can be used for etag | 14:30 |
| gordc | _nadya_: i was just wondering since mirantis did most/all contributions on hbase, if you actually test against it actively? because i have no idea if it's really production ready. | 14:30 |
| sileht | jd__, that looks good, but I don't remember why I have not used that for ceilometer | 14:31 |
| jd__ | sileht: I think that's because we support old MySQL version | 14:31 |
| sileht | jd__, mariadb support that since 5.3 | 14:33 |
| jd__ | sileht: ok I'll work on that | 14:34 |
| sileht | jd__, https://mariadb.com/kb/en/mariadb/microseconds-in-mariadb/ | 14:34 |
| _nadya_ | gordc: yep, we did some tests, but it was a release ago | 14:35 |
| gordc | _nadya_: ah cool cool. so i guess it's still a viable backend? except it has less functionality. | 14:38 |
| _nadya_ | gordc: I believe yes | 14:38 |
| gordc | _nadya_: kk | 14:42 |
| ildikov | _nadya_: gordc: as for the Kilo release it still does not sound good that it is not really tested :( | 14:42 |
| ildikov | same for DB2, even if it is similar to Mongo... | 14:43 |
| _nadya_ | ildikov: gordc: we will discuss it today with the team | 14:43 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 14:45 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparaison of timestamp field to None https://review.openstack.org/171669 | 14:45 |
| ildikov | _nadya_: cool, sounds good, please let us know about the outcome :) | 14:46 |
| gordc | ildikov: yeah, the db2 stuff is really sketchy to me. it's definitely something we should follow up on. | 14:46 |
| gordc | _nadya_: thanks! | 14:46 |
| ildikov | gordc: yeap, we have so many issues with back ends, would be nice to have some priority on clarifying the current state in the next cycle | 14:48 |
| gordc | ildikov: agreed | 14:49 |
| gordc | who understands this english? " Accept the default write concern, acknowledged writes, instead of setting safe=True." | 14:50 |
| ildikov | LOL :) | 14:50 |
| *** claudiub has quit IRC | 14:50 | |
| ildikov | you definitely need some background to understand it... | 14:51 |
| gordc | http://api.mongodb.org/python/current/changelog.html?highlight=safe | 14:51 |
| gordc | i can't tell if that's a real sentence and i'm stilling waking up or if it makes no sense at all.lol | 14:52 |
| cdent | it makes no sense | 14:53 |
| ildikov | gordc: this explains the things better: http://emptysqua.re/blog/pymongos-new-default-safe-writes/ | 14:54 |
| gordc | cdent: yeah... ... ... lol | 14:54 |
| gordc | ildikov: ah cool. | 14:55 |
| ildikov | gordc: I still think that we should limit the version in requirements... | 14:55 |
| ildikov | I rechecked ZhiQiang's WIP patch to see if it fails with the same issue or a different one as it fails randomly for me locally... | 14:57 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Avoid query metric object twice from indexer https://review.openstack.org/171590 | 15:00 |
| *** jaypipes has quit IRC | 15:03 | |
| gordc | ildikov: we could do something similar to keystone and check if safe is valid attr in MongoClient. | 15:03 |
| gordc | if not we can just not pass it in? | 15:03 |
| *** jaypipes has joined #openstack-ceilometer | 15:05 | |
| ildikov | gordc: sounds good to me | 15:06 |
| *** jaypipes has quit IRC | 15:06 | |
| ildikov | gordc: the only issue is that two tests fail with the w=1 setting on the gate currently... | 15:06 |
| *** exploreshaifali has quit IRC | 15:08 | |
| ildikov | gordc: logs: http://logs.openstack.org/58/171458/1/check/gate-ceilometer-python27/f7d5bb0/console.html | 15:09 |
| sileht | cdent, jd__ gordc I try to find a better name for the 'lifetime_to/from' columns in my gnocchi history patch | 15:12 |
| sileht | to ensure people make the difference between lifetime_to/from and started_at/ended_at | 15:12 |
| cdent | sileht: can you explain to me in more words what it is supposed to represent, then maybe I can think of a different term? | 15:12 |
| gordc | ildikov: dang pymongo. | 15:13 |
| sileht | cdent, lifetime_to/from is related to the history | 15:14 |
| sileht | cdent, attributes of the resource was that between lifetime_to/from | 15:14 |
| cdent | is it a value that is per revision of the resource, or per the entire history? | 15:14 |
| sileht | cdent, per revision | 15:15 |
| gordc | ildikov: there's this patch https://review.openstack.org/#/c/171438/ | 15:15 |
| sileht | cdent, lifetime_to/from is handled by gnocchi , started_at/ended_at is user defined | 15:15 |
| ildikov | gordc: agreed | 15:15 |
| * cdent nods and thinks | 15:15 | |
| ildikov | gordc: cool, I saw the fixes on the stable branches already | 15:16 |
| ildikov | gordc: I think it's the best for the master too | 15:16 |
| * cdent looks at code | 15:16 | |
| gordc | sileht: me thinking too. | 15:16 |
| cdent | sileht: is the reason for having a lifetime_to to avoid looking at the next revision in the history? presumably the window of a lifetime is from when this one changed to when the next one (forward in time) changed | 15:20 |
| *** _cjones_ has joined #openstack-ceilometer | 15:20 | |
| sileht | cdent, yes | 15:20 |
| cdent | another question: What we expecting someone to want to do with this info? | 15:20 |
| gordc | inception/death? | 15:21 |
| gordc | we need our namer of things | 15:21 |
| * cdent plays the trombone of doom | 15:22 | |
| sileht | cdent, for example: a user want to know all the instances that have the flavor 'mini' during the last month, with how many times | 15:22 |
| cdent | ah | 15:22 |
| cdent | so if I resize my instance multiple times during the last month, that's going to make for resource history messiness...got it | 15:23 |
| * cdent continues thinking | 15:23 | |
| cdent | what about instead of a to and from a start and duration? | 15:24 |
| sileht | cdent, you have an example of this query here: https://review.openstack.org/#/c/171670/2/doc/source/rest.yaml | 15:24 |
| *** _cjones_ has quit IRC | 15:24 | |
| sileht | line 236 | 15:24 |
| * cdent nods | 15:25 | |
| cdent | so duration is semantically awkward for those kinds of queries | 15:25 |
| sileht | yes | 15:25 |
| sileht | it's simpler to store utcnow(), than computing thing with duration | 15:26 |
| * cdent nods | 15:26 | |
| cdent | so the very long form of the names are: | 15:26 |
| cdent | resource revision start time | 15:27 |
| cdent | resourc revision end time | 15:27 |
| sileht | yes :) | 15:27 |
| cdent | is there anything wrong with revision_start and revision_end? | 15:27 |
| cdent | I find that a bit more comprehensible than "lifetime" | 15:27 |
| sileht | no that looks good | 15:27 |
| cdent | \o/ | 15:28 |
| sileht | jd__, gordc wdyt ? ^ | 15:28 |
| sileht | lets change to that, that looks better than lifetime and see what happend | 15:30 |
| sileht | cdent, thx | 15:31 |
| gordc | sileht: revision seems ok to me. | 15:31 |
| cdent | you're welcome, we got there in the end | 15:31 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 15:34 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 15:34 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 15:34 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparison of timestamp field to None https://review.openstack.org/171669 | 15:34 |
| *** claudiub has joined #openstack-ceilometer | 15:36 | |
| *** rjaiswal has joined #openstack-ceilometer | 15:37 | |
| *** rwsu has joined #openstack-ceilometer | 15:37 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 15:38 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 15:38 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 15:38 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparison of timestamp field to None https://review.openstack.org/171669 | 15:38 |
| *** afazekas has quit IRC | 15:43 | |
| *** ildikov has quit IRC | 15:44 | |
| *** ozialien has joined #openstack-ceilometer | 15:44 | |
| *** ozialien has quit IRC | 15:47 | |
| *** Ephur_ is now known as Ephur | 15:48 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 15:51 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 15:51 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 15:51 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparison of timestamp field to None https://review.openstack.org/171669 | 15:51 |
| openstackgerrit | gordon chung proposed openstack/ceilometer-specs: Enable Liberty specs https://review.openstack.org/168321 | 15:54 |
| *** rjaiswal has quit IRC | 15:55 | |
| *** rjaiswal has joined #openstack-ceilometer | 15:55 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer-specs: Enable Liberty specs https://review.openstack.org/168321 | 15:55 |
| *** _cjones_ has joined #openstack-ceilometer | 15:57 | |
| *** ozialien has joined #openstack-ceilometer | 15:59 | |
| *** ozialien has quit IRC | 16:01 | |
| *** theanalyst has quit IRC | 16:04 | |
| *** ozialien has joined #openstack-ceilometer | 16:05 | |
| *** ozialien has quit IRC | 16:06 | |
| *** theanalyst has joined #openstack-ceilometer | 16:07 | |
| *** julim has quit IRC | 16:14 | |
| *** ozialien has joined #openstack-ceilometer | 16:19 | |
| *** ozialien has quit IRC | 16:23 | |
| *** Longgeek has joined #openstack-ceilometer | 16:25 | |
| *** jruano has joined #openstack-ceilometer | 16:25 | |
| *** claudiub has quit IRC | 16:28 | |
| *** Longgeek has quit IRC | 16:30 | |
| *** rjaiswal has quit IRC | 16:34 | |
| *** ankita_wagh has joined #openstack-ceilometer | 16:48 | |
| *** julim has joined #openstack-ceilometer | 16:54 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer-specs: Enable Liberty specs https://review.openstack.org/168321 | 17:00 |
| *** rbak has quit IRC | 17:02 | |
| *** claudiub has joined #openstack-ceilometer | 17:05 | |
| *** _cjones_ has quit IRC | 17:06 | |
| *** _cjones_ has joined #openstack-ceilometer | 17:07 | |
| *** drjones has joined #openstack-ceilometer | 17:09 | |
| *** _cjones_ has quit IRC | 17:10 | |
| *** yatin has joined #openstack-ceilometer | 17:10 | |
| *** ddieterly has quit IRC | 17:17 | |
| *** safchain has quit IRC | 17:18 | |
| *** ddieterly has joined #openstack-ceilometer | 17:18 | |
| *** megh has joined #openstack-ceilometer | 17:19 | |
| *** megh_ has joined #openstack-ceilometer | 17:20 | |
| *** megh has quit IRC | 17:20 | |
| *** harlowja_away is now known as harlowja | 17:21 | |
| *** ozialien has joined #openstack-ceilometer | 17:22 | |
| *** megh_ has quit IRC | 17:22 | |
| *** megh has joined #openstack-ceilometer | 17:22 | |
| cdent | gordc: do we want to moot specs such as "extract alarming to its own repo" or "extract polling agent to its own repo" (even if just for sake of discussion)? | 17:23 |
| *** megh has quit IRC | 17:23 | |
| *** megh has joined #openstack-ceilometer | 17:23 | |
| *** IBerezovskiy has quit IRC | 17:24 | |
| gordc | cdent: what do you mean by 'moot' | 17:24 |
| cdent | bring to the meeting to talk about | 17:24 |
| *** megh has quit IRC | 17:24 | |
| *** megh_ has joined #openstack-ceilometer | 17:24 | |
| cdent | consider for our consideration | 17:24 |
| cdent | that kind of thing | 17:24 |
| *** ildikov has joined #openstack-ceilometer | 17:25 | |
| gordc | cdent: ah, yeah, it's probably best to give a heads up on that. | 17:25 |
| *** megh_ has quit IRC | 17:26 | |
| *** megh has joined #openstack-ceilometer | 17:26 | |
| gordc | cdent: i'd assume there will be some people that would want to discuss how/what we split. | 17:26 |
| cdent | gordc: I think it should happen, but I don't know if I want to go to the trouble of writing the specs if people are opposed in principle | 17:26 |
| *** claudiub has quit IRC | 17:26 | |
| gordc | cdent: hahah. i'm going to assume we'll need to write something detailed to explain what 'extract alarming' exactly entails. | 17:27 |
| *** ddieterly has quit IRC | 17:28 | |
| cdent | true | 17:28 |
| *** ddieterly has joined #openstack-ceilometer | 17:28 | |
| ildikov | do we have an exact plan regarding what to extract and why, etc? | 17:30 |
| *** exploreshaifali has joined #openstack-ceilometer | 17:31 | |
| cdent | ildikov: not yet, no | 17:31 |
| cdent | I guess my question is: do we want to make a plan in advance of specs, or use specs as the place where discussion happens | 17:32 |
| ildikov | cdent: where did the question come from? | 17:32 |
| ildikov | I guess specs should be the place or maybe a plan spec in advance of specs can happen :) | 17:33 |
| cdent | just before you joined I said: gordc: do we want to moot specs such as "extract alarming to its own repo" or "extract polling agent to its own repo" (even if just for sake of discussion)? | 17:33 |
| ildikov | yeap, I quickly checked the logs, that is why I asked if we or you have a plan or desired state to reach? | 17:34 |
| *** yatin has quit IRC | 17:35 | |
| cdent | the desired state is for each of the several unique value propositions provided by the current ceilometer to be serviced by decoupled pieces | 17:35 |
| cdent | this would make them easier to maintain and improve and also allow greater diversity in the ecosystem: they could evolve independently more easily | 17:35 |
| *** drjones has quit IRC | 17:36 | |
| gordc | cdent: we'd probably need some strategy on how to do testing between the separated services. | 17:36 |
| ildikov | well, as for the pieces it's more architectural question than repo question I think, but well maybe from one side this direction could still help | 17:38 |
| cdent | the different repos help reinforce what ought to be strong architectural boundaries. it's sort of social cue or clue on how to manage stuff. | 17:39 |
| *** _cjones_ has joined #openstack-ceilometer | 17:39 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Smarter list_resources for history https://review.openstack.org/170385 | 17:40 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Keep history of updated resource attributes https://review.openstack.org/168405 | 17:40 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Add history documentation https://review.openstack.org/171670 | 17:40 |
| openstackgerrit | Mehdi Abaakouk proposed openstack/gnocchi: Allow comparison of timestamp field to None https://review.openstack.org/171669 | 17:40 |
| edmondsw | getting gate failures for icehouse and juno on https://review.openstack.org/#/c/169442/ ... any ideas here? | 17:41 |
| edmondsw | I see the following in the logs... but if that was the problem, why wouldn't we have hit this before? | 17:41 |
| edmondsw | pkg_resources.ContextualVersionConflict: (python-ceilometerclient 1.0.13.post24 (/opt/stack/new/python-ceilometerclient), Requirement.parse('python-ceilometerclient<=1.0.12,>=1.0.6'), set(['ceilometer'])) | 17:41 |
| edmondsw | eglynn, gordc, ^^ any ideas? | 17:42 |
| *** jruano has quit IRC | 17:43 | |
| ildikov | cdent: hmm, maybe you're right, but it would be better to see an architecture diagram with proposed boundaries as repos to have the same big picture | 17:43 |
| *** srsakhamuri has quit IRC | 17:44 | |
| cdent | ildikov: yeah, some diagrams would be useful. I haven't had a chance to think about it deeply enough, yet, but plan to | 17:45 |
| ildikov | edmondsw: we had issues with heat and the latest ceilo client, maybe a restriction has just landed as requirements patches landed today on stable branches because of the latest pymongo client | 17:45 |
| edmondsw | ildikov, tx... so what can I do about that? | 17:46 |
| ildikov | cdent: cool, it can worth some time on the summit too, if you have diagrams to sketch on a whiteboard | 17:47 |
| * cdent nods | 17:47 | |
| ildikov | cdent: but we will see what we will have room for... | 17:47 |
| ildikov | edmondsw: hmm, as we are talking about the gate maybe the infra guys can give you a better answer than me | 17:49 |
| *** jruano has joined #openstack-ceilometer | 17:49 | |
| edmondsw | ildikov, yeah... headed there | 17:49 |
| ildikov | edmondsw: cool, tnx, I will try to follow the discussion to be prepared if it happens again in the future | 17:51 |
| *** ddieterly has quit IRC | 17:52 | |
| *** _nadya_ has quit IRC | 17:53 | |
| *** ddieterly has joined #openstack-ceilometer | 17:53 | |
| *** yprokule has quit IRC | 18:04 | |
| *** jruano has quit IRC | 18:13 | |
| edmondsw | eglynn, ildikov, the infra team is telling me and gordc they think the problem on https://review.openstack.org/#/c/169442/ is caused by not having stable branches for ceilometerclient | 18:17 |
| edmondsw | http://specs.openstack.org/openstack/openstack-specs/specs/library-stable-branches.html | 18:17 |
| ildikov | edmondsw: yeap, I saw, thanks for heads up | 18:18 |
| edmondsw | dhellmann ^ | 18:18 |
| edmondsw | what do we do about that? | 18:18 |
| dhellmann | edmondsw: someone with permission needs to create a stable branch at the appropriate commit in the repo | 18:19 |
| edmondsw | yeah... which isn't me, so... Guess I'm asking WHO is going to do something about that :) | 18:19 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: add support for filtering by duration https://review.openstack.org/171606 | 18:20 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: sqlalchemy: use precise DATETIME rather than DECIMAL in MySQL https://review.openstack.org/171764 | 18:20 |
| edmondsw | and what do we do about my patch and others in the meantime... find a way to bypass that check in the gate, or will they just wait? | 18:21 |
| dhellmann | edmondsw: I don't know if I have permission, and I'm travelling this week so I wouldn't want to do it and disappear from conference wifi if there was an issue. gordc or eglynn should be able to take care of it | 18:21 |
| edmondsw | dhellmann, tx | 18:22 |
| ildikov | edmondsw: as much as I know eglynn is our stable branch expert on board | 18:22 |
| eglynn | edmondsw: I can look into this tomorrow, I need to run out the office door very shortly (after 7pm here) | 18:23 |
| edmondsw | eglynn, tx | 18:23 |
| edmondsw | always another thing for the todo list :) | 18:24 |
| *** yeungp has joined #openstack-ceilometer | 18:24 | |
| gordc | sorry, i don't have permission either | 18:25 |
| *** _nadya_ has joined #openstack-ceilometer | 18:25 | |
| dhellmann | gordc: I added you to https://review.openstack.org/#/admin/groups/533,members so you should now. Do you know how to use gerrit to create the branch? | 18:28 |
| *** boris-42 has quit IRC | 18:28 | |
| dhellmann | eglynn: fyi ^^ | 18:28 |
| *** jruano has joined #openstack-ceilometer | 18:29 | |
| gordc | dhellmann: nope... follow the steps in the link you gave? | 18:29 |
| dhellmann | gordc: fill out the form at https://review.openstack.org/#/admin/projects/openstack/python-ceilometerclient,branches -- be careful, though! :-) | 18:30 |
| dhellmann | gordc: the branch name should be stable/something, probably stable/kilo at this point but you might need more branches for older releases | 18:31 |
| dhellmann | gordc: you'll want to find the sha from the tag for the version meant to be used with kilo | 18:31 |
| gordc | dhellmann: i don't have a form on that link. | 18:33 |
| *** _nadya_ has quit IRC | 18:34 | |
| gordc | edmondsw: is this critical? or we ok with letting eglynn handle it later? | 18:34 |
| *** jmatthew_ has quit IRC | 18:39 | |
| edmondsw | gordc, I think it can wait until tomorrow | 18:39 |
| gordc | cool cool | 18:39 |
| edmondsw | then once that's in we need to talk about cutting 1.0.14 to pick up it and the other recent changes | 18:40 |
| dhellmann | gordc: hrm. I wonder if I gave permission to the wrong gordon. What is your gerrit id? | 18:40 |
| dhellmann | gordc: gord@live.ca? | 18:40 |
| gordc | yep. chungg | 18:41 |
| gordc | dhellmann: i can see i'm part of ceilometer-stable-maint | 18:41 |
| dhellmann | gordc: ah, you may need to be in https://review.openstack.org/#/admin/groups/151,members instead -- jd__ can you help with permissions for gordc to make a stable branch of ceilometer-client? | 18:42 |
| dhellmann | I'm not in that group, so I can't add anyone | 18:42 |
| jd__ | dhellmann: sure | 18:42 |
| jd__ | you're in gordc | 18:42 |
| dhellmann | jd__: thanks | 18:43 |
| edmondsw | ildikov, you mentioned "requirements patches landed today on stable branches because of the latest pymongo client"... could that be causing the py27 errors on https://review.openstack.org/#/c/167370/ ? | 18:44 |
| gordc | jd__: cool cool. | 18:44 |
| ildikov | edmondsw: yes, we still have the issue on master | 18:44 |
| * gordc sees same list on https://review.openstack.org/#/admin/projects/openstack/python-ceilometerclient,branches as before. | 18:45 | |
| edmondsw | ildikov, have a bug number? | 18:45 |
| gordc | dhellmann, i'm assuming i'm doing something wrong. will check with eglynn tomorrow | 18:45 |
| ildikov | edmondsw: one sec, need to fing the links | 18:45 |
| gordc | https://bugs.launchpad.net/ceilometer/+bug/1441393 | 18:46 |
| openstack | Launchpad bug 1441393 in Keystone juno "Keystone and Ceilometer unit tests fail with pymongo 3.0" [High,In progress] - Assigned to Ihar Hrachyshka (ihar-hrachyshka) | 18:46 |
| gordc | edmondsw: pymongo3 broke the world | 18:46 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: Move JSON encoding routing in its own module https://review.openstack.org/171616 | 18:46 |
| *** eglynn has quit IRC | 18:46 | |
| edmondsw | gordc, tx ... ildikov ^ | 18:47 |
| ildikov | edmondsw: requirements patch on master: https://review.openstack.org/#/c/171438 | 18:47 |
| ildikov | gordc: thanks :) I was already too optimistic and closed that tab already :) | 18:48 |
| *** _nadya_ has joined #openstack-ceilometer | 18:49 | |
| ildikov | edmondsw: the plan is to limit pymongo for kilo for now | 18:49 |
| edmondsw | works for me | 18:50 |
| ildikov | edmondsw: you mean locally? | 18:51 |
| *** _nadya_ has quit IRC | 18:52 | |
| edmondsw | ildikov, no, meant "that sounds good" | 18:52 |
| edmondsw | sorry for the confusion | 18:52 |
| ildikov | edmondsw: a-ha, ok, I've just seen that you rechecked that patch I think that is why I associated to the tests :) | 18:53 |
| edmondsw | wanted to associate the bugs with it for dikonoor doesn't have to look those up herself | 18:54 |
| ildikov | cool, got it | 18:54 |
| *** exploreshaifali has quit IRC | 18:58 | |
| *** Ala has quit IRC | 19:01 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/gnocchi: WIP: Add support for ArchivePolicy Rule Endpoints https://review.openstack.org/171288 | 19:01 |
| *** rbak has joined #openstack-ceilometer | 19:02 | |
| jd__ | sileht: HAHAHA we have MySQL 5.5 on the gate which does not support decimal with timestamp | 19:03 |
| *** megh has quit IRC | 19:04 | |
| * jd__ is going berserk | 19:04 | |
| * cdent gives jd__ a cookie | 19:04 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer-specs: Enable Liberty specs https://review.openstack.org/168321 | 19:08 |
| openstackgerrit | gordon chung proposed openstack/ceilometer: use oslo.messaging dispatch filter https://review.openstack.org/171347 | 19:11 |
| cdent | that's it for me | 19:11 |
| cdent | night | 19:11 |
| *** cdent has quit IRC | 19:11 | |
| *** fnaval has joined #openstack-ceilometer | 19:23 | |
| *** eglynn has joined #openstack-ceilometer | 19:25 | |
| *** jruano has quit IRC | 19:40 | |
| *** _cjones_ has quit IRC | 19:50 | |
| *** jruano has joined #openstack-ceilometer | 19:53 | |
| *** ddieterly has quit IRC | 19:58 | |
| *** ozialien has quit IRC | 19:59 | |
| *** _cjones_ has joined #openstack-ceilometer | 20:04 | |
| *** ddieterly has joined #openstack-ceilometer | 20:04 | |
| *** megh has joined #openstack-ceilometer | 20:05 | |
| *** megh has quit IRC | 20:05 | |
| *** megh has joined #openstack-ceilometer | 20:05 | |
| *** _nadya_ has joined #openstack-ceilometer | 20:09 | |
| *** _nadya_ has quit IRC | 20:12 | |
| *** msly has joined #openstack-ceilometer | 20:17 | |
| msly | anyone have additional comments on this bug? https://bugs.launchpad.net/ceilometer/+bug/1421584 | 20:18 |
| openstack | Launchpad bug 1421584 in Ceilometer "cpu_util volume won't report right for cpu over commit case" [Medium,Triaged] - Assigned to Ilya Tyaptin (ityaptin) | 20:18 |
| *** prad has quit IRC | 20:19 | |
| gordc | msly: sorry, haven't looked at it lately, putting out gate fires. | 20:20 |
| gordc | will look in a bit. | 20:20 |
| *** prad has joined #openstack-ceilometer | 20:30 | |
| *** rbak has quit IRC | 20:33 | |
| *** rbak has joined #openstack-ceilometer | 20:34 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: WIP: ugly fix: pymongo 3.0 breaks ci gate https://review.openstack.org/171458 | 20:36 |
| *** rbak has quit IRC | 20:36 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: pymongo 3.0 breaks ci gate https://review.openstack.org/171458 | 20:36 |
| *** rbak has joined #openstack-ceilometer | 20:36 | |
| *** rbak has quit IRC | 20:36 | |
| openstackgerrit | Komei Shimamura proposed openstack/ceilometer: Fix a variable name mistake https://review.openstack.org/170644 | 20:37 |
| *** eglynn has quit IRC | 20:42 | |
| gordc | msly: do you have a solution? | 20:44 |
| *** sbfox has joined #openstack-ceilometer | 20:47 | |
| *** exploreshaifali has joined #openstack-ceilometer | 20:55 | |
| *** sbfox has quit IRC | 21:00 | |
| *** ddieterly has quit IRC | 21:02 | |
| *** ddieterly has joined #openstack-ceilometer | 21:02 | |
| *** jruano has quit IRC | 21:03 | |
| msly | I do not... I tried digging into libvirt to see if I could get the vm util from it, but couldn't find it | 21:05 |
| *** jruano has joined #openstack-ceilometer | 21:06 | |
| *** megh has quit IRC | 21:07 | |
| gordc | msly: yeah... i need to look into this further but from what i've been told, the only real way is to have an agent on each guest. | 21:07 |
| gordc | there is apparently a libvirt agent but only works based a list of conditions (ie. must be linux, might not be installed, etc...) | 21:08 |
| msly | gordc: I have thought about that but.. that doesn't seem like a solution the community would accept to me | 21:08 |
| gordc | msly: agreed. especially if the only thing you get is cpu_util. | 21:09 |
| gordc | msly: we could multiple the value by number of cpus? but i think that gives only a rough estimate... and i haven't really tried it to verify if that's even valid. | 21:10 |
| msly | gordc: The hard part about generating a formula for it with the current info we can get is figuring out when it is 'overcommitting' | 21:12 |
| *** edmondsw has quit IRC | 21:15 | |
| gordc | msly: yeah... tbh, i'm not a libvirt expert so i'll need to chat more with Nova contact. maybe there is some obvious option we're missing. | 21:16 |
| *** ddieterly has quit IRC | 21:30 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: add support for filtering by duration https://review.openstack.org/171606 | 21:33 |
| *** megh has joined #openstack-ceilometer | 21:35 | |
| openstackgerrit | Rohit Jaiswal proposed openstack/ceilometer-specs: configuration via data store https://review.openstack.org/171826 | 21:39 |
| openstackgerrit | Rohit Jaiswal proposed openstack/ceilometer-specs: configuration via data store https://review.openstack.org/171826 | 21:40 |
| *** noye has quit IRC | 21:43 | |
| *** noye has joined #openstack-ceilometer | 21:47 | |
| openstackgerrit | Rohit Jaiswal proposed openstack/ceilometer-specs: configuration via data store https://review.openstack.org/171826 | 21:49 |
| *** exploreshaifali has quit IRC | 21:57 | |
| *** exploreshaifali has joined #openstack-ceilometer | 22:02 | |
| *** noye has quit IRC | 22:03 | |
| *** noye has joined #openstack-ceilometer | 22:03 | |
| *** noye has quit IRC | 22:13 | |
| *** jruano has quit IRC | 22:20 | |
| *** prad has quit IRC | 22:27 | |
| openstackgerrit | Rohit Jaiswal proposed openstack/ceilometer-specs: configuration via data store https://review.openstack.org/171826 | 22:28 |
| *** zhenq has quit IRC | 22:31 | |
| *** exploreshaifali has quit IRC | 22:32 | |
| *** alexpilotti has quit IRC | 22:39 | |
| *** Challa has joined #openstack-ceilometer | 22:40 | |
| *** Challa has left #openstack-ceilometer | 22:44 | |
| *** gordc has quit IRC | 22:45 | |
| *** challa has joined #openstack-ceilometer | 22:45 | |
| *** noye has joined #openstack-ceilometer | 22:47 | |
| ankita_wagh | Hi, Did anyone face any issues in running tox for juno release on centos 7 ? | 22:49 |
| *** sbfox has joined #openstack-ceilometer | 22:54 | |
| challa | Hi, i am having issues with PartitionCoordination when i enable HA for Central agent. Members of group: set([]) & My subset: [] extract_my_subset are always empty, can some point me if my zookeeper configuration is indeed correct? | 22:56 |
| *** noye has quit IRC | 22:58 | |
| challa | [coordination] backend_url = kazoo://10.22.152.47:2181 heartbeat = 1.0 | 23:01 |
| *** sbfox has quit IRC | 23:09 | |
| *** pauldix has quit IRC | 23:16 | |
| *** sbfox has joined #openstack-ceilometer | 23:32 | |
| *** Ephur has quit IRC | 23:34 | |
| *** yeungp has quit IRC | 23:38 | |
| *** sbfox has quit IRC | 23:43 | |
| *** sbfox has joined #openstack-ceilometer | 23:46 | |
| openstackgerrit | Komei Shimamura proposed openstack/ceilometer: Fix a variable name mistake https://review.openstack.org/170644 | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!