| *** openstackgerrit has quit IRC | 00:04 | |
| *** openstackgerrit has joined #openstack-metering | 00:04 | |
| *** krtaylor has joined #openstack-metering | 00:07 | |
| *** matsuhashi has joined #openstack-metering | 00:27 | |
| *** shanewang1 has joined #openstack-metering | 00:35 | |
| *** shanewang1 has left #openstack-metering | 00:35 | |
| *** sandywalsh_ has joined #openstack-metering | 00:42 | |
| *** sandywalsh has quit IRC | 00:43 | |
| *** nosnos has joined #openstack-metering | 00:44 | |
| *** fnaval_ has joined #openstack-metering | 00:48 | |
| *** krtaylor has quit IRC | 00:57 | |
| *** fandikurnia01 has joined #openstack-metering | 01:17 | |
| *** asalkeld is now known as asalkeld_lunch | 01:20 | |
| *** terriyu has quit IRC | 01:43 | |
| *** shakayumi has joined #openstack-metering | 02:06 | |
| *** asalkeld_lunch is now known as asalkeld | 02:14 | |
| fandikurnia01 | chat | 02:22 |
|---|---|---|
| *** krtaylor has joined #openstack-metering | 02:26 | |
| *** shakayumi has quit IRC | 02:39 | |
| *** shakayumi has joined #openstack-metering | 02:49 | |
| *** matsuhashi has quit IRC | 02:57 | |
| *** matsuhashi has joined #openstack-metering | 03:24 | |
| *** matsuhashi has quit IRC | 03:31 | |
| *** matsuhashi has joined #openstack-metering | 03:32 | |
| *** nosnos has quit IRC | 03:34 | |
| *** evanjfraser has quit IRC | 03:34 | |
| *** Ruetobas has quit IRC | 03:34 | |
| *** Ruetobas has joined #openstack-metering | 03:34 | |
| *** nosnos has joined #openstack-metering | 03:34 | |
| *** evanjfraser has joined #openstack-metering | 03:35 | |
| *** matsuhashi has quit IRC | 03:36 | |
| *** matsuhashi has joined #openstack-metering | 03:41 | |
| *** matsuhashi has quit IRC | 03:41 | |
| *** matsuhashi has joined #openstack-metering | 03:42 | |
| *** matsuhashi has quit IRC | 03:46 | |
| *** matsuhashi has joined #openstack-metering | 03:49 | |
| *** SergeyLukjanov has joined #openstack-metering | 03:54 | |
| fandikurnia01 | hello all | 04:16 |
| fandikurnia01 | more question | 04:16 |
| fandikurnia01 | why i found error when try ceilometer meter-list | 04:16 |
| fandikurnia01 | [root@controller-2 ceilometer]# ceilometer meter-list | 04:16 |
| fandikurnia01 | No handlers could be found for logger "ceilometerclient.common.http" | 04:16 |
| fandikurnia01 | Invalid OpenStack Identity credentials. | 04:16 |
| *** yfujioka has joined #openstack-metering | 04:33 | |
| *** Shaan7 has joined #openstack-metering | 04:40 | |
| *** Shaan7 has quit IRC | 04:40 | |
| *** Shaan7 has joined #openstack-metering | 04:40 | |
| *** Shaan7 has quit IRC | 04:41 | |
| *** Shaan7 has joined #openstack-metering | 04:42 | |
| *** Shaan7 has quit IRC | 04:42 | |
| *** Shaan7 has joined #openstack-metering | 04:43 | |
| *** Shaan7 has quit IRC | 04:47 | |
| *** changbl has quit IRC | 04:47 | |
| *** Shaan7 has joined #openstack-metering | 04:48 | |
| *** Shaan7 has quit IRC | 04:48 | |
| *** Shaan7 has joined #openstack-metering | 04:48 | |
| *** changbl has joined #openstack-metering | 04:51 | |
| *** flwang has quit IRC | 04:52 | |
| *** fandikurnia01 has quit IRC | 05:00 | |
| *** fandikurnia01 has joined #openstack-metering | 05:01 | |
| *** evanjfraser has quit IRC | 05:01 | |
| *** boris-42 has joined #openstack-metering | 05:04 | |
| *** flwang has joined #openstack-metering | 05:07 | |
| *** SergeyLukjanov has quit IRC | 05:08 | |
| *** ryanpetrello has joined #openstack-metering | 05:17 | |
| *** fandikurnia01 has quit IRC | 05:20 | |
| *** fandikurnia01 has joined #openstack-metering | 05:24 | |
| *** changbl has quit IRC | 05:38 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/49680 | 06:03 |
| *** yfujioka has quit IRC | 06:06 | |
| *** fandikurnia01 has quit IRC | 06:57 | |
| *** fandikurnia01 has joined #openstack-metering | 06:57 | |
| *** eglynn has quit IRC | 07:07 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Avoid leaking admin-ness into combination alarms https://review.openstack.org/50989 | 07:10 |
| *** dafter has joined #openstack-metering | 07:29 | |
| *** shardy_afk is now known as shardy | 07:31 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ceilometer: Ceilometer has no such project-list subcommand https://review.openstack.org/51133 | 07:32 |
| *** shakayumi has quit IRC | 07:39 | |
| *** eglynn has joined #openstack-metering | 08:03 | |
| jd__ | the gate is broken because of https://bugs.launchpad.net/ceilometer/+bug/1238529 | 08:26 |
| jd__ | I'm working on it and I'll need fast review | 08:26 |
| sileht | jd__, \o/ | 08:34 |
| sileht | jd__, on a friday !!! | 08:34 |
| jd__ | worst thing is that it blocks the release of rc2 | 08:34 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: Add support for keystoneclient 0.4.0 https://review.openstack.org/51151 | 08:56 |
| jd__ | please review this ^^^^^ | 08:57 |
| sileht | jd__, I think I can fast approval this one no ? | 09:02 |
| sileht | Someone else is here to unblock gate ? asalkeld, dhellmann eglynn ? | 09:04 |
| eglynn | sileht: looking now | 09:04 |
| sileht | eglynn, thx | 09:04 |
| eglynn | sileht: looks good, approved | 09:05 |
| sileht | eglynn, thanks a lot ! | 09:05 |
| jd__ | thanks guys | 09:07 |
| *** Shaan7 has quit IRC | 09:24 | |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: Add support for keystoneclient 0.4.0 https://review.openstack.org/51151 | 09:30 |
| jd__ | we need to merge everything for rc2 asap now | 09:42 |
| jd__ | eglynn: sileht: I've to go for lunch, if you guys can check the patch list and do reverify if needed, that'd be awesome | 09:42 |
| jd__ | be back soon | 09:42 |
| *** vvechkanov has joined #openstack-metering | 10:00 | |
| vvechkanov | Hello all. I again have a little question. I can't found in the code where type of meters is processed? What it is used for? Or it is just a information for a user? | 10:02 |
| vvechkanov | I am sorry for very often questions, but I'm reserching ceilometer for my company and nobody exept you can help me with it. I'm also start searching bugs to fix for ceilometer, so you can be sure that your help will not be wasted) | 10:04 |
| *** boris-42 has quit IRC | 10:55 | |
| *** SergeyLukjanov has joined #openstack-metering | 11:35 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 11:37 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 11:37 | |
| *** fandikurnia01 has quit IRC | 11:42 | |
| *** thomasm has joined #openstack-metering | 11:55 | |
| *** thomasm has quit IRC | 11:55 | |
| *** thomasm has joined #openstack-metering | 11:56 | |
| thomasm | Hey all! | 11:56 |
| *** boris-42 has joined #openstack-metering | 11:59 | |
| *** fandikurnia01 has joined #openstack-metering | 12:27 | |
| *** matsuhashi has quit IRC | 12:33 | |
| *** matsuhashi has joined #openstack-metering | 12:34 | |
| *** aspiers has quit IRC | 12:34 | |
| *** nosnos has quit IRC | 12:35 | |
| *** nosnos has joined #openstack-metering | 12:35 | |
| *** aspiers has joined #openstack-metering | 12:38 | |
| *** matsuhashi has quit IRC | 12:38 | |
| *** dafter has quit IRC | 12:40 | |
| *** nosnos has quit IRC | 12:40 | |
| *** dafter has joined #openstack-metering | 12:41 | |
| *** fandikurnia01 has quit IRC | 12:45 | |
| *** dafter has quit IRC | 12:46 | |
| *** fandikurnia01 has joined #openstack-metering | 12:46 | |
| *** bpokorny1 has quit IRC | 12:55 | |
| openstackgerrit | Stephen Gran proposed a change to openstack/ceilometer: Change alarm_history.detail to text type https://review.openstack.org/50351 | 12:57 |
| jd__ | doesn't look like rc2 will be released today | 13:03 |
| *** fandikurnia01 has quit IRC | 13:04 | |
| thomasm | :( | 13:11 |
| *** bpokorny has joined #openstack-metering | 13:14 | |
| *** julim has joined #openstack-metering | 13:16 | |
| *** sandywalsh_ has quit IRC | 13:16 | |
| *** bpokorny has quit IRC | 13:17 | |
| *** Shaan7 has joined #openstack-metering | 13:17 | |
| *** bpokorny has joined #openstack-metering | 13:18 | |
| *** gordc has joined #openstack-metering | 13:21 | |
| *** sandywalsh has joined #openstack-metering | 13:28 | |
| *** Shaan7 has quit IRC | 13:29 | |
| *** Shaan7 has joined #openstack-metering | 13:29 | |
| *** dafter has joined #openstack-metering | 13:38 | |
| *** dafter has quit IRC | 13:38 | |
| *** dafter has joined #openstack-metering | 13:38 | |
| *** aspiers has quit IRC | 13:49 | |
| *** aspiers has joined #openstack-metering | 13:53 | |
| *** fnaval_ has quit IRC | 14:02 | |
| *** fnaval_ has joined #openstack-metering | 14:02 | |
| *** changbl has joined #openstack-metering | 14:03 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 14:04 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 14:04 | |
| *** fnaval_ has quit IRC | 14:07 | |
| *** fnaval_ has joined #openstack-metering | 14:29 | |
| *** ruhe has joined #openstack-metering | 14:35 | |
| * eglynn is super-braindead today ... | 14:40 | |
| eglynn | OK real dumb question chaps | 14:41 |
| eglynn | is there anything beyond letting api.host defaul to 0.0.0.0 required to allow the ceilo API service be invoked from outside the localhost? | 14:41 |
| eglynn | here's the scenario ... | 14:41 |
| eglynn | ceilo API service (plus nova-api, glance-api, keystone etc.) all running on one rhel VM | 14:41 |
| eglynn | nova-compute & ceilo-compute running on a second VM | 14:41 |
| eglynn | from VM2, I can call out to all services ports on VM1 | 14:42 |
| eglynn | *except* 8777, which fails with EHOSTUNREACH | 14:42 |
| sileht | eglynn, strange it should work by default (the default is 0.0.0.0 I think) | 14:43 |
| sileht | eglynn, ceilometer-api should show in log the binded IP | 14:44 |
| eglynn | sileht: yep, that's the default all right ... netstat confirms listening on 0.0.0.0 | 14:44 |
| eglynn | ... in api.log, "serving on 0.0.0.0:8777, view at http://127.0.0.1:8777" | 14:46 |
| eglynn | .... BTW that 0.0.0.0 and 127.0.0.1 are both hard-coded in the log message when api.host is 0.0.0.0: https://github.com/openstack/ceilometer/blob/master/ceilometer/api/app.py#L124 | 14:49 |
| sileht | eglynn, the 'view at ...' is not very useful | 14:50 |
| eglynn | yeap | 14:50 |
| sileht | eglynn, perhaps a ipv6 issue | 14:56 |
| sileht | eglynn, do you use the hostname resolvable in ipv6 or somehing like taht ? | 14:56 |
| eglynn | sileht: well just using IPv4 sytyle addressing for everything else | 14:57 |
| *** boris-42 has quit IRC | 15:03 | |
| * gordc is joining braindead group. | 15:10 | |
| gordc | do we not require admin role for api access? | 15:10 |
| eglynn | sileht, gordc: meh! told ye I was being dumb today, no iptables ACCEPT rule for 8777 ... d'oh! | 15:17 |
| sileht | eglynn, :) | 15:17 |
| *** sandywalsh has quit IRC | 15:17 | |
| gordc | eglynn: problem solved, time for a break. :) | 15:18 |
| *** sandywalsh has joined #openstack-metering | 15:24 | |
| *** ruhe has quit IRC | 15:46 | |
| thomasm | jd__, ping =] | 15:50 |
| jd__ | thomasm: ? | 15:52 |
| thomasm | jd__, So CM used to use EnabledExtensionManager for the notification plugins? | 15:52 |
| thomasm | And have a disabled config? | 15:52 |
| *** sandywalsh has quit IRC | 15:53 | |
| thomasm | In the collector | 15:53 |
| *** sandywalsh has joined #openstack-metering | 15:53 | |
| *** sandywalsh has quit IRC | 15:54 | |
| thomasm | jd__, Regarding the comment here: https://review.openstack.org/#/c/49244/3/ceilometer/collector/service.py | 15:54 |
| thomasm | line 64 | 15:54 |
| jd__ | we don't have a disabled config | 15:55 |
| jd__ | we should probably use only ExtensionManager | 15:55 |
| jd__ | we don't want a disabled config | 15:55 |
| thomasm | Why? | 15:56 |
| jd__ | because we don't want to add useless config options :) | 15:56 |
| jd__ | ah we do use ExtensionManager, your patch change that sorry misunderstood | 15:56 |
| *** sandywalsh has joined #openstack-metering | 15:57 | |
| *** sandywalsh has quit IRC | 15:57 | |
| thomasm | Yeah, I changed it to use EnabledExtensionManager, to allow us to turn notification plugins on and off. If you negate a meter in the pipeline.yaml, it still loads the extension and the samples from it just get dropped on the floor. This is intended to make it easier to just put a lid on an entire plugin and resulting stream. | 15:58 |
| *** sandywalsh has joined #openstack-metering | 15:58 | |
| thomasm | Also, to decouple the consumer configuration from the plugin dependency. If we don't load plugins, we still want to be able to store events if we want. If we aren't storing events, no consumer gets set up. | 15:59 |
| thomasm | s/if we aren't storing events/if we aren't storing events also/ | 15:59 |
| *** Ruetobas has quit IRC | 16:01 | |
| jd__ | thomasm: I think I can agree with that decoupling | 16:02 |
| jd__ | I just don't think the disabled/enabled load is worth it; relying on what's in the pipeline should be enough | 16:03 |
| *** Ruetobas has joined #openstack-metering | 16:03 | |
| jd__ | i.e. I'm not against the spirit of your change, just the implementation :) | 16:03 |
| thomasm | It'd be nice to be able to turn off sampling in one fell swoop (easier config), and it'd be nice to be able to just disable a misbehaving plugin. Also, more flexibility over what events are stored in the absence of the pipeline. | 16:06 |
| jd__ | I'm not against turning off | 16:07 |
| jd__ | as said, I'm ok with this feature | 16:07 |
| thomasm | If I were to disable all or most meters in pipeline.yaml, I still HAVE to load all of the plugins and turn all notifications into samples. | 16:07 |
| jd__ | but the way to do it is not IMHO "don't load" | 16:07 |
| thomasm | What way would you do it? | 16:08 |
| jd__ | thomasm: yeah, that's not normal I agree | 16:08 |
| jd__ | thomasm: the collector needs to get smarter wrt what the pipeline wants? | 16:08 |
| *** Ruetobas has quit IRC | 16:08 | |
| jd__ | sorry, I don't have everything in mind so I can't reply more precisely right now | 16:08 |
| thomasm | Nah, I'm just trying to figure out an alternate implementation. What I'm also a bit lost on is, what's wrong with the 'don't load' approach? | 16:09 |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 16:10 | |
| *** _SergeyLukjanov has quit IRC | 16:10 | |
| thomasm | It seems easy to understand, not prone to error. It gives the deployer full control at both ends when they want it and a sensible default when they don't know or don't care. | 16:10 |
| thomasm | I'm missing the pitfall, I guess. :\ | 16:11 |
| *** terriyu has joined #openstack-metering | 16:12 | |
| *** Ruetobas has joined #openstack-metering | 16:13 | |
| thomasm | And I don't think the pipeline can define the plugins it consumes from, since it would have to produce a sample to know the meter name. | 16:15 |
| thomasm | Anywho, I'll reply so we can have a more visible discussion with more alternatives and feedback. :P | 16:17 |
| *** eglynn has quit IRC | 16:18 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Add support for keystoneclient 0.4.0 https://review.openstack.org/51151 | 16:38 |
| *** gordc has quit IRC | 17:06 | |
| sandywalsh | jd__: is there code in the ceilometer client for publishing UDP? | 17:09 |
| sandywalsh | jd__: (and I'm assuming we can use the ceilometer client as a programming library and not just a cmdline tool, yes?) | 17:09 |
| *** eglynn has joined #openstack-metering | 17:09 | |
| sandywalsh | (answered part 2, yes) | 17:11 |
| sandywalsh | (researching part 1 ...) | 17:11 |
| sandywalsh | (answered part 1 ... nope) | 17:14 |
| sandywalsh | hmm, I'm wondering if we should use keystone auth to get service catalog for UDP endpoint or make that a config option and not require keystone at all | 17:16 |
| sandywalsh | dhellmann: how about if I added a python-ceilometerudpclient to github/openstack? Just a tiny library we could use for publishing UDP data to ceilometer. Or should I make it part of python-ceilometerclient? | 17:18 |
| sandywalsh | likely the latter I suspect | 17:18 |
| dhellmann | sandywalsh: adding it to our existing client library seems like the right approach | 17:21 |
| dhellmann | sandywalsh: we have a udp collector, where is it getting data? are you sure there's not a publisher in there somewhere? | 17:21 |
| sandywalsh | dhellmann: yeah, it's going to be a wart on the side, but at least the packaging will be consistent | 17:21 |
| dhellmann | ceilometer/publisher/udp.py? | 17:22 |
| sandywalsh | well, we don't want to have to install all of ceilometer just to get the udp publisher | 17:22 |
| dhellmann | maybe we need to refactor | 17:22 |
| sandywalsh | yeah, that's the plan | 17:22 |
| dhellmann | makes sense | 17:23 |
| sandywalsh | I just put this up in nova https://review.openstack.org/#/c/51249 | 17:23 |
| sandywalsh | which will allow us to tap into all the metrics that nova already polls for | 17:23 |
| sandywalsh | on a per instance and host basis | 17:23 |
| sandywalsh | the idea is to put the udp publisher in there | 17:23 |
| sandywalsh | dhellmann: ^ | 17:24 |
| *** gordc has joined #openstack-metering | 17:34 | |
| *** SergeyLukjanov has joined #openstack-metering | 17:43 | |
| *** boris-42 has joined #openstack-metering | 17:45 | |
| *** shakayumi has joined #openstack-metering | 17:52 | |
| *** alexb|f19 has joined #openstack-metering | 17:54 | |
| *** nati_ueno has joined #openstack-metering | 17:59 | |
| * dhellmann finishes phone call | 18:02 | |
| dhellmann | sandywalsh: every time someone calls importutils in a loop I die a little inside ;-) | 18:03 |
| dhellmann | sandywalsh: but I've added myself as a reviewer and will give it a closer look | 18:04 |
| sandywalsh | dhellmann: yeah, I was going to use stevedore, but it's not part of nova (yet) | 18:07 |
| sandywalsh | that said, I like being able to fully specify a class and know exactly what I'm getting (explicit vs implicit) | 18:11 |
| sandywalsh | well, according to russell it seems stevedore is ok in nova after all :) | 18:13 |
| *** dafter has quit IRC | 18:35 | |
| *** evanjfraser has joined #openstack-metering | 18:37 | |
| *** evanjfraser has quit IRC | 18:42 | |
| *** dafter has joined #openstack-metering | 18:44 | |
| *** alexb|f19 has quit IRC | 18:53 | |
| *** shakayumi has quit IRC | 18:53 | |
| *** Bada has joined #openstack-metering | 19:00 | |
| dhellmann | sandywalsh: I just saw his comment :-) | 19:01 |
| sandywalsh | yeah, too funny | 19:01 |
| *** alexb|f19 has joined #openstack-metering | 19:21 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Avoid leaking admin-ness into threshold-oriented alarms https://review.openstack.org/50708 | 20:24 |
| *** sandywalsh has quit IRC | 20:26 | |
| *** Bada has quit IRC | 20:41 | |
| *** dafter has quit IRC | 20:41 | |
| *** briancline has quit IRC | 20:44 | |
| *** tvb|afk has joined #openstack-metering | 20:44 | |
| *** ruhe has joined #openstack-metering | 20:45 | |
| *** ruhe has quit IRC | 20:46 | |
| *** thomasm has quit IRC | 20:46 | |
| *** julim has quit IRC | 20:59 | |
| *** fnaval_ has quit IRC | 21:03 | |
| *** alexb|f19 has quit IRC | 21:04 | |
| *** fnaval_ has joined #openstack-metering | 21:05 | |
| *** SergeyLukjanov has quit IRC | 21:19 | |
| *** SergeyLukjanov has joined #openstack-metering | 21:20 | |
| *** gordc has quit IRC | 21:25 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 21:32 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 21:32 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 21:33 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 21:33 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 21:33 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 21:33 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 21:34 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 21:34 | |
| *** eglynn has quit IRC | 21:42 | |
| *** elextro_ has joined #openstack-metering | 22:03 | |
| *** SergeyLukjanov has quit IRC | 22:04 | |
| *** SergeyLukjanov has joined #openstack-metering | 22:05 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 22:06 | |
| *** _SergeyLukjanov is now known as SergeyLukjanov | 22:06 | |
| *** tvb|afk has quit IRC | 22:14 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Change alarm_history.detail to text type https://review.openstack.org/50351 | 22:17 |
| openstackgerrit | A change was merged to openstack/ceilometer: Avoid leaking admin-ness into combination alarms https://review.openstack.org/50989 | 22:18 |
| *** flwang has quit IRC | 22:19 | |
| *** SergeyLukjanov has quit IRC | 22:22 | |
| *** flwang has joined #openstack-metering | 22:24 | |
| *** bpokorny has quit IRC | 22:29 | |
| *** flwang has quit IRC | 22:31 | |
| *** changbl has quit IRC | 22:35 | |
| *** fnaval_ has quit IRC | 22:37 | |
| *** boris-42 has quit IRC | 23:01 | |
| *** flwang has joined #openstack-metering | 23:03 | |
| *** nati_uen_ has joined #openstack-metering | 23:45 | |
| *** nati_ueno has quit IRC | 23:46 | |
| *** alexb|f19 has joined #openstack-metering | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!