| *** julim has quit IRC | 00:05 | |
| *** shardy has quit IRC | 00:11 | |
| *** fandikurnia01 has quit IRC | 00:13 | |
| *** matsuhashi has joined #openstack-metering | 00:30 | |
| *** __nikhil__ has joined #openstack-metering | 00:49 | |
| *** changbl has joined #openstack-metering | 00:51 | |
| *** fandikurnia01 has joined #openstack-metering | 01:02 | |
| *** shakayumi has joined #openstack-metering | 01:21 | |
| *** gordc has joined #openstack-metering | 01:30 | |
| *** fnaval_ has joined #openstack-metering | 01:48 | |
| *** ryanpetrello has joined #openstack-metering | 01:52 | |
| *** evanjfraser has quit IRC | 01:59 | |
| *** peterinfra has quit IRC | 02:13 | |
| *** shakayumi has quit IRC | 02:28 | |
| *** ryanpetrello has quit IRC | 02:33 | |
| *** ryanpetrello has joined #openstack-metering | 02:35 | |
| *** openstack has joined #openstack-metering | 03:23 | |
| *** hobana.freenode.net sets mode: +ns | 03:23 | |
| *** shakayumi has joined #openstack-metering | 03:26 | |
| *** hobana.freenode.net sets mode: -o openstack | 03:29 | |
| -hobana.freenode.net- *** Notice -- TS for #openstack-metering changed from 1381289023 to 1335337574 | 03:29 | |
| *** hobana.freenode.net sets mode: +ct-s | 03:29 | |
| *** changbl has joined #openstack-metering | 03:29 | |
| *** chmouel_ has joined #openstack-metering | 03:29 | |
| *** jd__ has joined #openstack-metering | 03:29 | |
| *** ryanpetrello has joined #openstack-metering | 03:29 | |
| *** fnaval_ has joined #openstack-metering | 03:29 | |
| *** fandikurnia01 has joined #openstack-metering | 03:29 | |
| *** __nikhil__ has joined #openstack-metering | 03:29 | |
| *** matsuhashi has joined #openstack-metering | 03:29 | |
| *** Shaan7 has joined #openstack-metering | 03:29 | |
| *** sdake_ has joined #openstack-metering | 03:29 | |
| *** giroro_ has joined #openstack-metering | 03:29 | |
| *** taplax has joined #openstack-metering | 03:29 | |
| *** nijaba_ has joined #openstack-metering | 03:29 | |
| *** lsmola has joined #openstack-metering | 03:29 | |
| *** flwang has joined #openstack-metering | 03:29 | |
| *** sdake has joined #openstack-metering | 03:29 | |
| *** janisg has joined #openstack-metering | 03:29 | |
| *** openstackgerrit has joined #openstack-metering | 03:29 | |
| *** sileht has joined #openstack-metering | 03:29 | |
| *** sgran has joined #openstack-metering | 03:29 | |
| *** harmw has joined #openstack-metering | 03:29 | |
| *** briancline has joined #openstack-metering | 03:29 | |
| *** asalkeld has joined #openstack-metering | 03:29 | |
| *** Joy has joined #openstack-metering | 03:29 | |
| *** aspiers has joined #openstack-metering | 03:29 | |
| *** krtaylor has joined #openstack-metering | 03:29 | |
| *** yjiang5_ has joined #openstack-metering | 03:29 | |
| *** ityaptin has joined #openstack-metering | 03:29 | |
| *** llu_away has joined #openstack-metering | 03:29 | |
| *** raghuvrao has joined #openstack-metering | 03:29 | |
| *** EmilienM has joined #openstack-metering | 03:29 | |
| *** Guest74104 has joined #openstack-metering | 03:29 | |
| *** annegentle_ has joined #openstack-metering | 03:29 | |
| *** clouded_tux has joined #openstack-metering | 03:29 | |
| *** vvechkanov has joined #openstack-metering | 03:29 | |
| *** ekarlso has joined #openstack-metering | 03:29 | |
| *** zigo_ has joined #openstack-metering | 03:29 | |
| *** tasdomas has joined #openstack-metering | 03:29 | |
| *** dhellmann has joined #openstack-metering | 03:29 | |
| *** bogdando has joined #openstack-metering | 03:29 | |
| *** dah has joined #openstack-metering | 03:29 | |
| *** apmelton has joined #openstack-metering | 03:29 | |
| *** dragondm has joined #openstack-metering | 03:29 | |
| *** Daviey has joined #openstack-metering | 03:29 | |
| *** lakspace has joined #openstack-metering | 03:29 | |
| *** jburkhart has joined #openstack-metering | 03:29 | |
| *** Ranzpirat has joined #openstack-metering | 03:29 | |
| *** fc__ has joined #openstack-metering | 03:29 | |
| *** swann has joined #openstack-metering | 03:29 | |
| *** sbadia has joined #openstack-metering | 03:29 | |
| *** hobana.freenode.net changes topic to "Project: https://launchpad.net/ceilometer | Meetings on Wed or Thu see Agenda: http://wiki.openstack.org/Meetings/MeteringAgenda | Blueprint: http://wiki.openstack.org/EfficientMetering" | 03:29 | |
| fandikurnia01 | ok doen | 04:08 |
|---|---|---|
| *** shakayumi has quit IRC | 04:24 | |
| *** boris-42 has joined #openstack-metering | 05:03 | |
| *** SergeyLukjanov has joined #openstack-metering | 05:27 | |
| *** Shaan7 has quit IRC | 05:48 | |
| *** fandikurnia01 has quit IRC | 05:49 | |
| *** SergeyLukjanov has quit IRC | 06:02 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/49680 | 06:03 |
| *** eglynn has joined #openstack-metering | 06:14 | |
| *** ryanpetrello has quit IRC | 06:14 | |
| *** neumerance has joined #openstack-metering | 06:22 | |
| *** SergeyLukjanov has joined #openstack-metering | 06:25 | |
| *** ityaptin_ has joined #openstack-metering | 06:26 | |
| *** ityaptin has quit IRC | 06:26 | |
| *** SergeyLukjanov has quit IRC | 06:37 | |
| *** fandikurnia01 has joined #openstack-metering | 06:46 | |
| *** SergeyLukjanov has joined #openstack-metering | 06:46 | |
| *** clouded_tux has quit IRC | 06:52 | |
| *** SergeyLukjanov has quit IRC | 06:53 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ceilometer: Set python-six minimum version https://review.openstack.org/50538 | 06:53 |
| *** clouded_tux has joined #openstack-metering | 06:54 | |
| *** yfujioka has joined #openstack-metering | 06:56 | |
| *** neumerance_ has joined #openstack-metering | 06:57 | |
| *** neumerance has quit IRC | 06:57 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ceilometer: Set python-six minimum version https://review.openstack.org/50538 | 07:04 |
| *** Shaan7 has joined #openstack-metering | 07:05 | |
| *** eglynn has quit IRC | 07:15 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ceilometer: Set python-six minimum version https://review.openstack.org/50538 | 07:25 |
| openstackgerrit | Stephen Gran proposed a change to openstack/ceilometer: New metric for instance state https://review.openstack.org/50546 | 07:37 |
| *** shardy has joined #openstack-metering | 07:37 | |
| *** sdake has quit IRC | 07:45 | |
| *** sdake_ has quit IRC | 07:45 | |
| *** eglynn has joined #openstack-metering | 08:02 | |
| *** boris-42 has quit IRC | 08:06 | |
| *** bogdando has quit IRC | 08:22 | |
| *** sdake has joined #openstack-metering | 08:23 | |
| *** sdake has quit IRC | 08:23 | |
| *** sdake has joined #openstack-metering | 08:23 | |
| *** bogdando has joined #openstack-metering | 08:25 | |
| openstackgerrit | Stephen Gran proposed a change to openstack/ceilometer: New metric for instance state https://review.openstack.org/50546 | 08:35 |
| *** sdake has quit IRC | 08:36 | |
| *** sdake has joined #openstack-metering | 08:43 | |
| *** sdake has quit IRC | 08:49 | |
| sgran | jd__: hi | 09:04 |
| jd__ | sgran: hi | 09:04 |
| sgran | I'm confused. Why should that be a @staticmethod ? None of the others are | 09:04 |
| sgran | in https://review.openstack.org/50546 | 09:05 |
| sgran | I'm happy to do it if you prefer it that way, but I was trying to keep to the idiom already present | 09:06 |
| *** sdake has joined #openstack-metering | 09:09 | |
| *** Bada has joined #openstack-metering | 09:15 | |
| *** Bada has quit IRC | 09:20 | |
| *** Shaan7 has quit IRC | 09:25 | |
| *** sdake has quit IRC | 09:25 | |
| *** sdake has joined #openstack-metering | 09:27 | |
| *** sdake has joined #openstack-metering | 09:27 | |
| *** sdake has quit IRC | 09:35 | |
| *** sdake has joined #openstack-metering | 09:36 | |
| *** Shaan7 has joined #openstack-metering | 09:40 | |
| *** SergeyLukjanov has joined #openstack-metering | 09:41 | |
| *** sdake has quit IRC | 09:41 | |
| *** sdake has joined #openstack-metering | 10:11 | |
| *** sdake has quit IRC | 10:11 | |
| *** sdake has joined #openstack-metering | 10:11 | |
| *** neumerance_ has quit IRC | 10:14 | |
| *** lsmola has quit IRC | 10:25 | |
| *** lsmola has joined #openstack-metering | 10:37 | |
| *** ityaptin_ has quit IRC | 10:40 | |
| *** boris-42 has joined #openstack-metering | 10:55 | |
| *** fandikurnia01 has quit IRC | 11:07 | |
| *** zul has joined #openstack-metering | 11:14 | |
| *** Bada has joined #openstack-metering | 11:30 | |
| *** Bada has quit IRC | 11:35 | |
| openstackgerrit | Christian Berendt proposed a change to openstack/ceilometer: six has to be in version 1.4.0 or higher https://review.openstack.org/50602 | 11:51 |
| eglynn | sileht, jd__: I've been playing a little with the alarm threshold validation in the api layer and a couple things surprised me | 12:01 |
| eglynn | mainly that we're validating the rule *twice* | 12:02 |
| eglynn | once on the incoming param validation | 12:02 |
| eglynn | then again on when the newly created alarm is init'd from the DB model for the response | 12:02 |
| eglynn | not necessarily a huge issue except that there's the potential to *modify* rule query as a side-effect of the validation | 12:03 |
| eglynn | seems wrong to do a possibly mutating validation step *after* the thing has been created | 12:03 |
| eglynn | (i.e. too late to modify at that stage anyway) | 12:03 |
| eglynn | seems better to separate out into (a) pure non-mutating validation and (b) modification ... | 12:04 |
| eglynn | (a) would then called as a side-effect of the WSME validate decorator | 12:04 |
| eglynn | (b) could be called explicitly inline | 12:04 |
| eglynn | sileht, jd__: any thoughts on that ^^^? | 12:05 |
| jd__ | eglynn: agreed | 12:05 |
| eglynn | (context is figuring out how to cleanly avoid the leakage of admin-ness from alarms created by admin on behalf of a non-admin) | 12:05 |
| eglynn | jd__: cool | 12:05 |
| jd__ | I think this kind of problems comes from the fact WSME does not always do the job correctly | 12:05 |
| jd__ | actually I'm working on that as we speak | 12:06 |
| sileht | eglynn, yes I have already think about this, this due of usage of __init__ of wsme BaseType | 12:06 |
| jd__ | trying to make WSME work :] | 12:06 |
| eglynn | yeah, it seems odd to have mix of implicit and explicit validation | 12:06 |
| * sileht try to find the wsme issue about this | 12:06 | |
| eglynn | cool | 12:06 |
| jd__ | basically @wsme.validate should not be used, as it does nothing anyway with pecan | 12:10 |
| jd__ | and wsexpose should do the job | 12:10 |
| jd__ | at least this is what I'm trying to make work for a couple of days now :) | 12:10 |
| sileht | eglynn, https://bugs.launchpad.net/wsme/+bug/1220906 | 12:11 |
| *** Bada has joined #openstack-metering | 12:11 | |
| eglynn | jd__: are you thinking RC2 or Icehouse for your WIP changes? | 12:11 |
| sgran | when someone has a moment, I'm hoping to get https://review.openstack.org/#/c/50351/ and https://review.openstack.org/#/c/50546/ merged | 12:11 |
| jd__ | eglynn: Icehouse | 12:12 |
| eglynn | jd__: fair nuffski | 12:12 |
| jd__ | we don't have a RC2 windows because we don't have enough bugs | 12:12 |
| sgran | I can make more :) | 12:12 |
| jd__ | s/enough/important enough/ | 12:12 |
| sileht | eglynn, we have think with jd__ that we can just add a static method (Alarm.from_db) to build Alarm API object like wsme does to bypass validation when the alarm come from the database | 12:13 |
| jd__ | yeah I think dhellmann wrote about that too somewhere | 12:13 |
| eglynn | sileht: a-ha, cool, yeah that would work I think | 12:13 |
| *** Bada has quit IRC | 12:16 | |
| *** gordc has joined #openstack-metering | 12:23 | |
| *** sandywalsh has joined #openstack-metering | 12:30 | |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part7 https://review.openstack.org/50613 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part4 https://review.openstack.org/47145 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part3 https://review.openstack.org/47144 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part6 https://review.openstack.org/47147 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part5 https://review.openstack.org/47146 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Update oslo in ceilometer part1 https://review.openstack.org/46476 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part1 https://review.openstack.org/46480 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Replace tests.base part2 https://review.openstack.org/47143 | 12:43 |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack/ceilometer: Update oslo in ceilometer part2 https://review.openstack.org/47142 | 12:43 |
| jd__ | dhellmann: ah I'm pleased to see you did the same remark as me on https://review.openstack.org/#/c/44652/ :) | 12:43 |
| *** bpokorny has joined #openstack-metering | 12:45 | |
| *** Bada has joined #openstack-metering | 12:53 | |
| *** julim has joined #openstack-metering | 12:54 | |
| *** BadaBoum has joined #openstack-metering | 12:56 | |
| *** eglynn is now known as eglynn-lunch | 12:57 | |
| *** Bada has quit IRC | 12:58 | |
| *** anteaya has joined #openstack-metering | 13:01 | |
| *** thomasm has joined #openstack-metering | 13:02 | |
| *** eglynn-lunch has quit IRC | 13:02 | |
| thomasm | Hey all | 13:03 |
| *** BadaBoum has quit IRC | 13:05 | |
| *** Bada has joined #openstack-metering | 13:05 | |
| *** fandikurnia01 has joined #openstack-metering | 13:07 | |
| *** jergerber has joined #openstack-metering | 13:08 | |
| *** Shaan7 has quit IRC | 13:10 | |
| *** fandikurnia01 has quit IRC | 13:14 | |
| *** ryanpetrello has joined #openstack-metering | 13:14 | |
| *** pwiltsey has joined #openstack-metering | 13:16 | |
| *** pwiltsey has quit IRC | 13:17 | |
| *** pwiltsey has joined #openstack-metering | 13:18 | |
| *** changbl has quit IRC | 13:21 | |
| dhellmann | jd__: yeah, that function's existence bothers me | 13:24 |
| gordc | dhellmann: since i have your partial attention, do you remember what bug stopped us from continuing with alembic migration scripts? | 13:27 |
| dhellmann | gordc: I'm not sure if we ever filed a bug. There were issues running the alembic migrations in unit tests, iirc. | 13:28 |
| sileht | gordc, perhaps this one https://bugs.launchpad.net/ceilometer/+bug/1217156 | 13:29 |
| *** matsuhashi has quit IRC | 13:30 | |
| *** yfujioka has quit IRC | 13:30 | |
| *** matsuhashi has joined #openstack-metering | 13:30 | |
| gordc | cool cool. just asking because we have this https://review.openstack.org/#/c/45064/ (in sql-migrate) which fixes resource table, but the patch to fix meter table was done in alembic... | 13:31 |
| *** krtaylor has quit IRC | 13:32 | |
| sileht | gordc, the issue affect only sqlite, so, but we really need to do some thing for this issue because continue to use both is very dangerous for future db migration | 13:33 |
| sileht | each time we add a new sqlachemy script, this one is played before the alembic one, this is not very good | 13:34 |
| gordc | sileht: agreed. i'm already confused. | 13:34 |
| *** matsuhashi has quit IRC | 13:34 | |
| *** fnaval_ has quit IRC | 13:40 | |
| *** dafter has joined #openstack-metering | 13:40 | |
| *** fnaval_ has joined #openstack-metering | 13:40 | |
| *** eglynn-lunch has joined #openstack-metering | 13:44 | |
| *** fnaval_ has quit IRC | 13:45 | |
| *** fandikurnia01 has joined #openstack-metering | 13:47 | |
| *** Shaan7 has joined #openstack-metering | 14:00 | |
| *** fandikurnia01 has quit IRC | 14:08 | |
| *** fnaval_ has joined #openstack-metering | 14:10 | |
| *** Bada has quit IRC | 14:13 | |
| sandywalsh | so, I've started working on this: https://blueprints.launchpad.net/nova/+spec/host-metric-hook which should be a low-impact hook to getting hypervisor samples from Nova. Good application of the UDP service. | 14:20 |
| sandywalsh | eglynn-lunch: heat should benefit from it too ^ | 14:20 |
| sandywalsh | (the spec has more detail) | 14:21 |
| *** eglynn-lunch is now known as eglynn | 14:21 | |
| eglynn | sandywalsh: cool | 14:21 |
| *** SergeyLukjanov has quit IRC | 14:30 | |
| *** changbl has joined #openstack-metering | 14:34 | |
| *** julim has quit IRC | 14:35 | |
| *** julim has joined #openstack-metering | 14:40 | |
| *** sdake_ has joined #openstack-metering | 14:47 | |
| *** sdake_ has joined #openstack-metering | 14:47 | |
| *** sdake has quit IRC | 14:47 | |
| *** Shaan7 has quit IRC | 14:54 | |
| *** SergeyLukjanov has joined #openstack-metering | 15:06 | |
| *** krtaylor has joined #openstack-metering | 15:12 | |
| *** changbl has quit IRC | 15:17 | |
| *** terriyu has joined #openstack-metering | 15:27 | |
| *** SergeyLukjanov has quit IRC | 15:34 | |
| *** gordc1 has joined #openstack-metering | 15:41 | |
| *** gordc has quit IRC | 15:42 | |
| *** gordc1 has quit IRC | 15:43 | |
| *** gordc has joined #openstack-metering | 15:43 | |
| *** lsmola has quit IRC | 15:46 | |
| thomasm | Hey, eglynn: It looks like gate is passing the SQLAlchemy timestamp ordering backport: https://review.openstack.org/#/c/48039/ | 15:53 |
| thomasm | =] | 15:53 |
| thomasm | If you want to take a look | 15:53 |
| *** giroro_ has quit IRC | 16:01 | |
| *** boris-42 has quit IRC | 16:01 | |
| *** boris-42 has joined #openstack-metering | 16:02 | |
| *** Ruetobas has joined #openstack-metering | 16:03 | |
| *** anteaya has quit IRC | 16:07 | |
| *** Ruetobas has quit IRC | 16:08 | |
| *** Ruetobas has joined #openstack-metering | 16:14 | |
| *** pwiltsey has quit IRC | 16:20 | |
| *** boris-42 has quit IRC | 16:23 | |
| *** boris-42 has joined #openstack-metering | 16:23 | |
| openstackgerrit | Stephen Gran proposed a change to openstack/ceilometer: New metric for instance state https://review.openstack.org/50546 | 16:45 |
| *** chmouel_ is now known as chmouel | 16:51 | |
| *** boris-42 has quit IRC | 16:54 | |
| *** SergeyLukjanov has joined #openstack-metering | 16:54 | |
| *** boris-42 has joined #openstack-metering | 16:55 | |
| *** boris-42 has quit IRC | 16:58 | |
| *** litong has joined #openstack-metering | 17:21 | |
| *** dafter has quit IRC | 17:23 | |
| *** tvb|afk has joined #openstack-metering | 17:25 | |
| *** tvb|afk has quit IRC | 17:27 | |
| *** tvb|afk has joined #openstack-metering | 17:27 | |
| openstackgerrit | Stephen Gran proposed a change to openstack/ceilometer: New metric for instance state https://review.openstack.org/50546 | 17:37 |
| *** tvb|afk has quit IRC | 17:37 | |
| *** dafter has joined #openstack-metering | 17:38 | |
| *** dafter has quit IRC | 17:38 | |
| *** dafter has joined #openstack-metering | 17:38 | |
| *** tvb|afk has joined #openstack-metering | 17:40 | |
| *** tvb|afk has quit IRC | 17:40 | |
| *** tvb|afk has joined #openstack-metering | 17:40 | |
| *** dafter has quit IRC | 17:42 | |
| *** thomasm has quit IRC | 17:43 | |
| *** boris-42 has joined #openstack-metering | 17:58 | |
| *** thomasm has joined #openstack-metering | 18:03 | |
| litong | @eglynn, eoghan, ping. | 18:03 |
| thomasm | back | 18:03 |
| eglynn | litong: 'sup? | 18:03 |
| *** thomasm has quit IRC | 18:04 | |
| litong | @eglynn, looking at the alarm/service.py and trying to get a handle on how alarm component works. | 18:04 |
| eglynn | k | 18:05 |
| litong | noticed that class AlarmService extensions_namespace is "celiometer.alarm.evaluator" | 18:05 |
| *** thomasm has joined #openstack-metering | 18:05 | |
| eglynn | yep | 18:05 |
| litong | I am a bit puzzled by that. why the name space for alarm service being set to evaluator? | 18:05 |
| eglynn | because the evaluators are loaded as stevedore extensions | 18:06 |
| litong | I thought notifier and evaluator should be in different namespace. | 18:06 |
| eglynn | (there are two separate evaluators, one for threshold-oriented alarms and another for combination alarms) | 18:06 |
| litong | @eglynn, is there an overall doc somewhere that I can read to see how they work at a higher level? | 18:08 |
| eglynn | meh, that's not good: https://github.com/openstack/ceilometer/blob/master/setup.cfg#L106 | 18:08 |
| eglynn | litong: not really, as yet | 18:08 |
| litong | both notifier and evaluator loaded based on extensions, right? one is ceilometer.alarm.notifier, the other is ceilometer.alarm.evaluator? | 18:08 |
| litong | I have both services enabled when set up devstack. but I have trouble to see how they actually get invoked and working. | 18:09 |
| litong | how do I see it in action in devstack? | 18:09 |
| litong | from rabbitmq management console, I can see the queues for alarm, but I am not seeing anything going on on the queue. such as this one | 18:10 |
| litong | ceilometer.alarm.alarm_notifier | 18:11 |
| eglynn | litong: create an alarm with a loging notification action, see it fire, watch the alarm-notifier logging output | 18:11 |
| litong | @eglynn, ok, good to know there is a way, how do I do that? | 18:12 |
| eglynn | litong: example alarm creation command line right at the end of this WIP quickstart guide http://openstack.redhat.com/CeilometerQuickStart | 18:12 |
| litong | ceilometer client? | 18:13 |
| litong | ok, using the ceilometer client. all right. let me try that. Thanks @eglynn. | 18:13 |
| eglynn | litong: sure, python-ceilometerclient | 18:13 |
| eglynn | sileht: are we missing the line ... | 18:15 |
| eglynn | combination = ceilometer.alarm.evaluator.combination:CombinationEvaluator | 18:15 |
| eglynn | from: https://github.com/openstack/ceilometer/blob/master/setup.cfg#L106 | 18:15 |
| eglynn | sileht: or how is combo evaluator loaded? | 18:15 |
| *** Shaan7 has joined #openstack-metering | 18:23 | |
| *** Shaan7 has quit IRC | 18:23 | |
| *** Shaan7 has joined #openstack-metering | 18:24 | |
| *** Shaan7 has quit IRC | 18:24 | |
| *** Shaan7 has joined #openstack-metering | 18:24 | |
| *** Shaan7 has quit IRC | 18:26 | |
| *** Shaan7 has joined #openstack-metering | 18:27 | |
| *** Shaan7 has quit IRC | 18:30 | |
| *** Shaan7 has joined #openstack-metering | 18:30 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Avoid leaking admin-ness into threshold-oriented alarms https://review.openstack.org/50708 | 18:36 |
| *** Shaan7 has quit IRC | 18:44 | |
| *** Shaan7 has joined #openstack-metering | 18:45 | |
| *** Shaan7 has quit IRC | 18:45 | |
| *** Shaan7 has joined #openstack-metering | 18:47 | |
| *** Shaan7 has quit IRC | 18:48 | |
| *** Shaan7 has joined #openstack-metering | 18:48 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Ensure combination alarm evaluator can be loaded https://review.openstack.org/50710 | 18:51 |
| sgran | does anyone have a minute to double check me? | 18:54 |
| sgran | the end goal of what I'm trying to do is have ceilometer collect a metric, and poke heat if it meets an alarm condition | 18:54 |
| sgran | the metric I'm collecting is 'instance_state' as returned by libvirt - an enum from 0 to 8, and some string states | 18:55 |
| sgran | I'm returning (I think) the numeric value as the value of the metric, and the string state in extra data. If someone has a moment, a set of eyeballs from someone who is more sure of this than I am would be appreciated | 18:56 |
| sgran | this is https://review.openstack.org/#/c/50546/ | 18:56 |
| sgran | I haven't yet made the alarm condition for heat, so that will be my next task :) | 18:56 |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Ensure combination alarms can be evaluated https://review.openstack.org/50714 | 19:07 |
| *** Shaan7 has quit IRC | 19:11 | |
| *** Shaan7 has joined #openstack-metering | 19:11 | |
| eglynn | jd__: looks like we're gonna need an RC2 | 19:12 |
| eglynn | (IMHO at least) | 19:12 |
| eglynn | unless I've completely misunderstood what's going on, the combination alarm feature is broken without https://review.openstack.org/50710 & https://review.openstack.org/50714 | 19:13 |
| eglynn | jd__: adding the to-RC2-or-not-to-RC2 question as an agenda topic for discussion at our team meeting shortly | 19:20 |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Avoid leaking admin-ness into threshold-oriented alarms https://review.openstack.org/50708 | 19:28 |
| *** lsmola has joined #openstack-metering | 19:32 | |
| *** sdake has joined #openstack-metering | 19:37 | |
| litong | @eglynn, eoghan, what triggers an evaluator? | 19:43 |
| eglynn | litong: do you mean what caused the evaluator to trigger an alarm? | 19:45 |
| eglynn | litong: or what causes the evaluator to do an evaluation? | 19:45 |
| litong | @eglynn, eoghan, I mean what makes an evaluator's evaluate being invoked? | 19:45 |
| litong | @eglynn, eoghan, the later. | 19:45 |
| eglynn | litong: it's a periodic task | 19:45 |
| litong | @eglynn, periodically. ok, I see. | 19:46 |
| litong | no wonder there is an interval. | 19:46 |
| eglynn | litong: see https://github.com/openstack/ceilometer/blob/master/ceilometer/alarm/service.py#L129 | 19:46 |
| eglynn | litong: and https://github.com/openstack/ceilometer/blob/master/ceilometer/alarm/service.py#L183 | 19:47 |
| litong | @eglynn, eoghan, yeah, making sense now. That is where I was lost, I sort of figured out how the notifers work. but did not get how evaulators work. | 19:48 |
| eglynn | litong: cool :) | 19:48 |
| litong | @eglynn, I think I am quite clear now. I thought evaluators work differently (like in a pipeline which filter through all the meters) | 19:48 |
| litong | so that only meters triggers the evaluator (not a periodic task). | 19:49 |
| litong | for example, threshold evaulator will remember how many some kind of meters it sees , when it reaches certain limit, it post something to the alarm queue. | 19:50 |
| eglynn | litong: the intention of not doing in-line evaluation like that was to allow for "widely dimensioned" alarms (e.g. alarms based on metrics aggregated across a group of instances) | 19:52 |
| eglynn | bbiab | 19:52 |
| litong | @eglynn, yes, makes sense. did not think about the aggregated. | 19:52 |
| *** sdake_ has quit IRC | 19:58 | |
| *** sdake has quit IRC | 19:58 | |
| *** sdake has joined #openstack-metering | 20:00 | |
| *** sdake has quit IRC | 20:00 | |
| *** sdake has joined #openstack-metering | 20:00 | |
| *** JerryM has joined #openstack-metering | 20:02 | |
| *** sdake has quit IRC | 20:05 | |
| *** sdake has joined #openstack-metering | 20:09 | |
| sandywalsh | litong: I like your line of thinking ;) | 20:13 |
| litong | @sandywalsh, thanks. | 20:14 |
| litong | @sandywalsh, doing that, we do not need the alarm_evaluator server, but it can only handle the data through metering recording flow. it can resolve the aggregated data if these data come in as samples as well. | 20:16 |
| litong | @sandywalsh, but current design works as well. | 20:17 |
| *** raghuvrao has quit IRC | 20:17 | |
| *** raghuvrao has joined #openstack-metering | 20:18 | |
| *** raghuvrao has joined #openstack-metering | 20:18 | |
| *** SergeyLukjanov has quit IRC | 20:21 | |
| sandywalsh | litong: exactly, the evaluator service should disappear in that case. The event trigger stuff that dragondm is working on will handle the temporal cases. | 20:22 |
| litong | @sandysalsh, are you saying we are in the process of getting rid of the evaluator services? | 20:23 |
| *** SergeyLukjanov has joined #openstack-metering | 20:23 | |
| sandywalsh | litong: no, but it's a suggestion. This is still a WIP (addressing the wide dimension use case next) https://wiki.openstack.org/wiki/Ceilometer/AlarmImprovements | 20:24 |
| sandywalsh | litong: look forward to your feedback (will open it up to the ML once I get the existing feedback integrated) | 20:24 |
| litong | @sandywalsh, oh, ok. will do that. thanks. | 20:25 |
| *** SergeyLukjanov has quit IRC | 20:25 | |
| *** litong has quit IRC | 20:33 | |
| *** boris-42 has quit IRC | 20:45 | |
| *** boris-42 has joined #openstack-metering | 20:47 | |
| *** sandywalsh has quit IRC | 20:55 | |
| *** sandywalsh has joined #openstack-metering | 20:57 | |
| *** evanjfraser has joined #openstack-metering | 21:00 | |
| *** shardy is now known as shardy_afk | 21:09 | |
| *** egallen has joined #openstack-metering | 21:11 | |
| *** tvb|afk has quit IRC | 22:05 | |
| *** dafter has joined #openstack-metering | 22:06 | |
| *** dafter has quit IRC | 22:10 | |
| *** gordc has quit IRC | 22:12 | |
| *** eglynn has quit IRC | 22:13 | |
| *** jergerber has quit IRC | 22:13 | |
| *** bpokorny has quit IRC | 22:13 | |
| *** ryanpetrello has quit IRC | 22:23 | |
| *** thomasm has quit IRC | 22:37 | |
| *** sdake_ has joined #openstack-metering | 22:40 | |
| *** sdake_ has joined #openstack-metering | 22:40 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Ensure combination alarm evaluator can be loaded https://review.openstack.org/50710 | 22:45 |
| *** fnaval_ has quit IRC | 22:47 | |
| *** julim has quit IRC | 22:48 | |
| *** evanjfraser has quit IRC | 22:57 | |
| *** egallen has quit IRC | 22:58 | |
| *** dafter has joined #openstack-metering | 23:16 | |
| *** gordc has joined #openstack-metering | 23:18 | |
| *** dafter has quit IRC | 23:21 | |
| *** gordc has quit IRC | 23:33 | |
| *** JerryM has left #openstack-metering | 23:38 | |
| *** yfujioka has joined #openstack-metering | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!