| *** thorst has joined #openstack-telemetry | 00:02 | |
| *** thorst has quit IRC | 00:08 | |
| *** thorst has joined #openstack-telemetry | 00:36 | |
| *** thorst has quit IRC | 00:36 | |
| *** zhurong has joined #openstack-telemetry | 00:47 | |
| *** tovin07_ has joined #openstack-telemetry | 00:52 | |
| *** zhangguoqing has joined #openstack-telemetry | 00:57 | |
| *** dixiaoli has joined #openstack-telemetry | 01:04 | |
| *** liusheng has quit IRC | 01:04 | |
| *** liusheng has joined #openstack-telemetry | 01:09 | |
| *** dave-mccowan has joined #openstack-telemetry | 01:33 | |
| *** thorst has joined #openstack-telemetry | 01:52 | |
| *** thorst has quit IRC | 01:56 | |
| *** lhx_ has joined #openstack-telemetry | 02:02 | |
| *** zhurong has quit IRC | 02:43 | |
| *** thorst has joined #openstack-telemetry | 03:53 | |
| *** psachin has joined #openstack-telemetry | 03:56 | |
| *** thorst has quit IRC | 03:57 | |
| *** dave-mccowan has quit IRC | 04:05 | |
| *** sanchitmalhotra has joined #openstack-telemetry | 04:12 | |
| *** sanchitmalhotra has quit IRC | 04:13 | |
| *** sanchitmalhotra has joined #openstack-telemetry | 04:13 | |
| *** lhx_ has quit IRC | 04:21 | |
| *** zhurong has joined #openstack-telemetry | 04:46 | |
| *** thorst has joined #openstack-telemetry | 05:54 | |
| *** yprokule has joined #openstack-telemetry | 05:57 | |
| *** thorst has quit IRC | 05:59 | |
| *** dixiaoli has quit IRC | 06:18 | |
| *** dixiaoli has joined #openstack-telemetry | 06:18 | |
| *** rcernin has joined #openstack-telemetry | 06:47 | |
| *** harlowja has quit IRC | 07:11 | |
| *** shardy has joined #openstack-telemetry | 07:14 | |
| openstackgerrit | Deepthi V V proposed openstack/telemetry-specs master: Implements network-statistics-from-opendaylight https://review.openstack.org/475241 | 07:22 |
|---|---|---|
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Ceilometer meter support to collect network statistics. https://review.openstack.org/477883 | 07:25 |
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Gnocchi support for SDN controller meters https://review.openstack.org/478159 | 07:25 |
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Project_id for SDN controller meters https://review.openstack.org/480040 | 07:25 |
| *** yprokule_ has joined #openstack-telemetry | 07:35 | |
| *** tesseract has joined #openstack-telemetry | 07:36 | |
| *** yprokule has quit IRC | 07:37 | |
| *** yprokule_ is now known as yprokule | 07:37 | |
| *** masber has quit IRC | 07:40 | |
| *** dixiaoli has quit IRC | 07:45 | |
| *** dixiaoli has joined #openstack-telemetry | 07:46 | |
| *** zhangguoqing has quit IRC | 07:51 | |
| *** masber has joined #openstack-telemetry | 07:53 | |
| *** thorst has joined #openstack-telemetry | 07:54 | |
| *** yprokule_ has joined #openstack-telemetry | 07:58 | |
| *** thorst has quit IRC | 08:00 | |
| *** yprokule has quit IRC | 08:01 | |
| *** yprokule_ is now known as yprokule | 08:01 | |
| *** shardy has quit IRC | 08:06 | |
| *** shardy has joined #openstack-telemetry | 08:07 | |
| *** yprokule_ has joined #openstack-telemetry | 08:22 | |
| *** yprokule has quit IRC | 08:25 | |
| *** yprokule_ is now known as yprokule | 08:25 | |
| *** efoley has joined #openstack-telemetry | 08:31 | |
| *** afazekas|away is now known as afazekas | 08:34 | |
| *** yprokule_ has joined #openstack-telemetry | 08:36 | |
| *** yprokule has quit IRC | 08:39 | |
| *** yprokule_ is now known as yprokule | 08:39 | |
| *** tovin07_ has quit IRC | 08:40 | |
| *** yprokule has quit IRC | 09:12 | |
| *** zhangguoqing has joined #openstack-telemetry | 09:12 | |
| *** yprokule has joined #openstack-telemetry | 09:13 | |
| *** thorst has joined #openstack-telemetry | 09:23 | |
| *** thorst has quit IRC | 09:28 | |
| *** lhx_ has joined #openstack-telemetry | 10:00 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/aodh master: Imported Translations from Zanata https://review.openstack.org/481573 | 10:59 |
| *** dixiaoli has quit IRC | 11:00 | |
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Ceilometer meter support to collect network statistics. https://review.openstack.org/477883 | 11:14 |
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Gnocchi support for SDN controller meters https://review.openstack.org/478159 | 11:14 |
| openstackgerrit | Deepthi V V proposed openstack/ceilometer master: Project_id for SDN controller meters https://review.openstack.org/480040 | 11:14 |
| openstackgerrit | zhurong proposed openstack/panko master: Devstack install with pankoclient https://review.openstack.org/481550 | 11:25 |
| *** gordc has joined #openstack-telemetry | 11:45 | |
| *** thorst has joined #openstack-telemetry | 11:53 | |
| *** dave-mccowan has joined #openstack-telemetry | 11:55 | |
| *** pradk has joined #openstack-telemetry | 12:02 | |
| *** lhx__ has joined #openstack-telemetry | 12:25 | |
| *** lhx_ has quit IRC | 12:26 | |
| *** lhx__ has quit IRC | 12:32 | |
| *** zhurong has quit IRC | 12:36 | |
| *** jmlowe has quit IRC | 12:40 | |
| *** catintheroof has joined #openstack-telemetry | 13:04 | |
| *** masber has quit IRC | 13:06 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/aodh master: gnocchi: return better 'insufficient data' reason https://review.openstack.org/470778 | 13:18 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/aodh master: Expose alarm state reason to API https://review.openstack.org/470777 | 13:18 |
| *** donghao has joined #openstack-telemetry | 13:31 | |
| *** vint_bra has joined #openstack-telemetry | 13:54 | |
| *** xiaofandh12 has joined #openstack-telemetry | 14:01 | |
| *** donghao has quit IRC | 14:04 | |
| *** lhx_ has joined #openstack-telemetry | 14:06 | |
| *** vint_bra has quit IRC | 14:07 | |
| *** chlong_ has joined #openstack-telemetry | 14:23 | |
| *** psachin has quit IRC | 14:28 | |
| *** zhurong has joined #openstack-telemetry | 14:30 | |
| *** dixiaoli has joined #openstack-telemetry | 14:33 | |
| *** dixiaoli has quit IRC | 14:33 | |
| *** lhx_ has quit IRC | 14:38 | |
| *** zhangguoqing has quit IRC | 14:43 | |
| *** toddnni has quit IRC | 14:48 | |
| *** zhurong has quit IRC | 14:56 | |
| *** rcernin has quit IRC | 14:57 | |
| *** xiaofandh12 has quit IRC | 15:02 | |
| *** lhx_ has joined #openstack-telemetry | 15:03 | |
| *** chlong_ has quit IRC | 15:16 | |
| *** liusheng has quit IRC | 15:21 | |
| *** liusheng has joined #openstack-telemetry | 15:22 | |
| *** chlong_ has joined #openstack-telemetry | 15:30 | |
| *** nicodemus_ has joined #openstack-telemetry | 15:35 | |
| nicodemus_ | hello gordc, do you have a moment? I just need to ask a quick question | 15:37 |
| nicodemus_ | Is the Ocata gnocchi dispatcher able to choose which archive policy to use when pushing metrics? I remember it to be possible in older versions, specifying the archive policy in the ceilometer config file | 15:39 |
| jd_ | nicodemus_: archive policy rules in the answer | 15:40 |
| nicodemus_ | jd_, so... the answer would be "no", right? | 15:41 |
| jd_ | but… I say it's archive policy rules, not "no"! :) | 15:41 |
| jd_ | but yeah IIRC we removed the settings from Ceilometer | 15:42 |
| nicodemus_ | thing is, you map a metric to an archive policy, but in the event that two different clients want to push the same metric you have no means to separate metric A from client A versus metric A from client B | 15:42 |
| jd_ | client or Ceilometer? | 15:42 |
| jd_ | or client that are Ceilometers? | 15:43 |
| nicodemus_ | client was not a good choice of word... let's say customer | 15:43 |
| nicodemus_ | customer A and customer B want to POST the same metric, but I'd like to have that metric from one customer to use a different archive policy | 15:44 |
| jd_ | wah | 15:44 |
| jd_ | err I think we never though somebody would do that | 15:44 |
| jd_ | +t | 15:44 |
| jd_ | let me check if we did remove it | 15:44 |
| jd_ | I don't recall exactly | 15:45 |
| nicodemus_ | hahah we always have a weird requirement | 15:45 |
| jd_ | no we did not removed it | 15:45 |
| jd_ | we are planning to deprecate it | 15:45 |
| gordc | i don't understand how the same metric would have different policy. | 15:45 |
| jd_ | maybe we should not | 15:45 |
| jd_ | gordc: different needs? | 15:45 |
| jd_ | gordc: from 2 different ceilometers | 15:46 |
| *** yprokule has quit IRC | 15:46 | |
| gordc | but they are two different metrics... but have the same attributes. | 15:46 |
| nicodemus_ | in this specific case, one customer wishes to have monitoring and the other wishes to have just an hourly rating | 15:46 |
| nicodemus_ | and we look to remove unnecessary processing from gnocchi | 15:47 |
| jd_ | gordc: I don't understand what you don't understand | 15:47 |
| gordc | hahah | 15:47 |
| nicodemus_ | that means not processing every 10 minutes for the customer that doesn't need monitoring | 15:47 |
| jd_ | nicodemus_: so yeah you can do the same thing with ocata | 15:48 |
| nicodemus_ | jd_, that's good news | 15:48 |
| gordc | yeah i don't understand. if userA pushes cpu_util of vmA and userB pushes cpu_util of vmA. isn't that just same metric, but an archive policy containing two granulariteis? | 15:48 |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer master: Revert "deprecate archive policy of Gnocchi dispatcher" https://review.openstack.org/482191 | 15:49 |
| gordc | granularities* | 15:49 |
| jd_ | gordc: this is not what nicodemus_ says | 15:49 |
| jd_ | gordc: ceilometerA pushes cpu_util of vmA for userA and ceilometerB pushes cpu_util of vmB for userB | 15:50 |
| nicodemus_ | ^ that's right | 15:50 |
| jd_ | but userA wants archive policy low and userB wants high | 15:50 |
| gordc | oh. that's just two separate metrics in my mind. | 15:51 |
| *** gongysh has joined #openstack-telemetry | 15:51 | |
| jd_ | gordc: it is, but you can't do that without having the option for each ceilometer to force their archive policy at metric creation time | 15:53 |
| jd_ | archive policy rule won't cover that use case | 15:53 |
| gordc | word. | 15:53 |
| openstackgerrit | Chason Chan proposed openstack/python-ceilometerclient master: Docs: switch to openstackdocstheme https://review.openstack.org/476757 | 15:54 |
| gordc | jd_: your revert only supports scenario where both users are coming from completely separate ceilometer? | 15:56 |
| jd_ | gordc: yes…? what's the trap in your question? | 16:03 |
| *** lhx_ has quit IRC | 16:03 | |
| gordc | jd_: how many people have multiple different ceilometer's running? i imagine it's one ceilometer per region... | 16:05 |
| jd_ | gordc: at least nicodemus_ ? | 16:05 |
| *** chlong_ has quit IRC | 16:06 | |
| gordc | *shrugs* must be fun figuring out how to direct messages. | 16:08 |
| *** gongysh has quit IRC | 16:15 | |
| openstackgerrit | Chason Chan proposed openstack/python-ceilometerclient master: import content from cli-reference in openstack-manuals https://review.openstack.org/482204 | 16:18 |
| *** chlong_ has joined #openstack-telemetry | 16:20 | |
| nicodemus_ | gordc, in our case it is one ceilometer per domain | 16:23 |
| nicodemus_ | with our nasty domain patch | 16:23 |
| *** shardy is now known as shardy_afk | 16:37 | |
| *** harlowja has joined #openstack-telemetry | 16:37 | |
| *** chlong_ has quit IRC | 16:38 | |
| *** tesseract has quit IRC | 17:01 | |
| *** efoley has quit IRC | 17:30 | |
| *** Kevin_Zheng has quit IRC | 17:33 | |
| *** jmlowe has joined #openstack-telemetry | 17:58 | |
| *** shardy_afk is now known as shardy | 18:59 | |
| *** jmlowe has quit IRC | 19:14 | |
| *** nicodemus_ has quit IRC | 20:29 | |
| *** thorst has quit IRC | 20:42 | |
| *** thorst has joined #openstack-telemetry | 20:45 | |
| *** thorst has quit IRC | 20:49 | |
| *** catintheroof has quit IRC | 20:53 | |
| *** thorst has joined #openstack-telemetry | 21:00 | |
| flwang | gordc: jd_: questions about gnocchi | 21:37 |
| flwang | gordc: jd_: does gnocchi support a pluggable transformer for metrics? | 21:38 |
| flwang | the background is we're experiencing a performance issue of ceilometer + mongodb | 21:38 |
| flwang | so we're planning to gnocchi to improve the API performance when collecting data from ceilometer/gnocchi | 21:39 |
| gordc | what does pluggable transformer mean? | 21:39 |
| flwang | but we're wondering if we can do it more aggressive | 21:39 |
| flwang | gordc: 1 sec, let me show you something | 21:39 |
| flwang | gordc: see https://github.com/openstack/distil/blob/master/distil/transformer/conversion.py | 21:40 |
| flwang | distil is our billing system | 21:40 |
| flwang | now we're collecting data from ceilometer and then wash/transform them into the format we can use for billing | 21:40 |
| flwang | so that means Distil needs to call Ceilometer/Gnocchi API to get the metrics, transform them, save in local Distil Database | 21:41 |
| flwang | instead of getting the data and saving locally, i'm thinking if we can transform the data at the first place, by Gnocchi | 21:42 |
| flwang | especially given now Gnocchi is a general metrics as a service | 21:42 |
| flwang | does that make sense? gordc, jd_ ^ | 21:43 |
| gordc | flwang: are you just computing the lifecycle of a resource? | 21:43 |
| flwang | kind of, another transformer class https://github.com/openstack/distil/blob/master/distil/transformer/arithmetic.py | 21:44 |
| gordc | err... i don't really know what context your using these transformer... the arithmetic stuff seems like what gnocchi does already with aggregations. | 21:46 |
| *** gongysh has joined #openstack-telemetry | 21:47 | |
| gordc | is raw_data the output of running ceilometer sample-list -m meter_name? | 21:48 |
| flwang | gordc: yep | 21:51 |
| *** vishwanathj has joined #openstack-telemetry | 21:52 | |
| gordc | flwang: i'll be honest, i have no idea how start/end related to the raw_data. gnocchi simplest description is roughly equivalent to "ceilometer statistics -m meter --resource_id blah" | 21:57 |
| gordc | it does capture lifecycle as well but probably not for the NetworkServiceTransformer you have | 21:58 |
| *** vint_bra has joined #openstack-telemetry | 21:58 | |
| *** thorst has quit IRC | 22:02 | |
| *** shardy has quit IRC | 22:22 | |
| *** deep-book-gk_ has joined #openstack-telemetry | 22:32 | |
| *** deep-book-gk_ has quit IRC | 22:35 | |
| *** catintheroof has joined #openstack-telemetry | 22:43 | |
| *** gordc has quit IRC | 22:53 | |
| *** thorst has joined #openstack-telemetry | 23:17 | |
| *** thorst has quit IRC | 23:21 | |
| *** vint_bra has quit IRC | 23:25 | |
| *** thorst has joined #openstack-telemetry | 23:57 | |
| *** catintheroof has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!