*** yassine_ has quit IRC | 00:26 | |
*** yassine_ has joined #openstack-telemetry | 00:32 | |
*** hfu has joined #openstack-telemetry | 00:36 | |
*** liusheng has joined #openstack-telemetry | 00:57 | |
*** sam_nowitzki has joined #openstack-telemetry | 01:12 | |
*** catintheroof has quit IRC | 01:13 | |
*** thorst_afk has joined #openstack-telemetry | 01:17 | |
*** dave-mccowan has joined #openstack-telemetry | 01:28 | |
*** iceyao has joined #openstack-telemetry | 01:31 | |
*** thorst_afk has quit IRC | 01:41 | |
*** thorst_afk has joined #openstack-telemetry | 01:56 | |
*** lhx__ has joined #openstack-telemetry | 02:17 | |
openstackgerrit | Merged openstack/ceilometer master: set OS_AUTH_TYPE in gate https://review.openstack.org/434972 | 02:50 |
---|---|---|
*** thorst_afk has quit IRC | 02:50 | |
*** thorst_afk has joined #openstack-telemetry | 03:03 | |
*** thorst_afk has quit IRC | 03:04 | |
*** donghao has joined #openstack-telemetry | 03:09 | |
*** donghao has quit IRC | 03:13 | |
*** pradk has quit IRC | 03:32 | |
*** thorst_afk has joined #openstack-telemetry | 03:35 | |
*** thorst_afk has quit IRC | 03:35 | |
*** deepbook5broo has joined #openstack-telemetry | 03:45 | |
*** deepbook5broo has left #openstack-telemetry | 03:45 | |
*** dave-mccowan has quit IRC | 03:47 | |
*** david-lyle has quit IRC | 03:49 | |
*** david-lyle has joined #openstack-telemetry | 03:50 | |
*** sudipto_ has joined #openstack-telemetry | 03:57 | |
*** sudipto has joined #openstack-telemetry | 03:57 | |
*** links has joined #openstack-telemetry | 04:03 | |
*** sudipto has quit IRC | 04:12 | |
*** sudipto_ has quit IRC | 04:12 | |
*** gongysh has joined #openstack-telemetry | 04:26 | |
*** sudipto has joined #openstack-telemetry | 04:39 | |
*** sudipto_ has joined #openstack-telemetry | 04:39 | |
sam_nowitzki | hi, I use rabbitmqctl list_queues and see metering.sample and event.sample | 04:43 |
*** sudipto has quit IRC | 04:43 | |
*** sudipto_ has quit IRC | 04:43 | |
*** adriant has quit IRC | 04:48 | |
*** sudipto_ has joined #openstack-telemetry | 04:51 | |
*** sudipto has joined #openstack-telemetry | 04:51 | |
*** Jack_Iv has joined #openstack-telemetry | 04:52 | |
*** Jack_Iv has quit IRC | 04:56 | |
*** thorst_afk has joined #openstack-telemetry | 04:57 | |
*** thorst_afk has quit IRC | 05:02 | |
*** rcernin has joined #openstack-telemetry | 05:08 | |
*** gongysh has quit IRC | 05:10 | |
*** faizy has joined #openstack-telemetry | 05:32 | |
*** tlian has quit IRC | 05:32 | |
*** lhx__ has quit IRC | 06:09 | |
*** nadya has joined #openstack-telemetry | 06:10 | |
*** rcernin has quit IRC | 06:21 | |
*** nadya has quit IRC | 06:26 | |
*** lhx__ has joined #openstack-telemetry | 06:40 | |
*** lhx__ has quit IRC | 06:52 | |
*** lhx__ has joined #openstack-telemetry | 06:53 | |
*** thorst_afk has joined #openstack-telemetry | 06:59 | |
*** rcernin has joined #openstack-telemetry | 07:01 | |
*** thorst_afk has quit IRC | 07:03 | |
*** tesseract has joined #openstack-telemetry | 07:11 | |
*** yprokule has joined #openstack-telemetry | 07:26 | |
*** zhangguoqing has joined #openstack-telemetry | 07:35 | |
*** Jack_Iv has joined #openstack-telemetry | 07:37 | |
*** gongysh has joined #openstack-telemetry | 07:41 | |
*** jd__ has quit IRC | 07:47 | |
*** nadya has joined #openstack-telemetry | 07:52 | |
*** nadya has quit IRC | 07:53 | |
*** nadya has joined #openstack-telemetry | 07:53 | |
*** zhangguoqing has quit IRC | 07:57 | |
*** shardy_afk is now known as shardy | 07:57 | |
*** Jack_Iv has quit IRC | 07:58 | |
*** Jack_Iv has joined #openstack-telemetry | 07:59 | |
*** Jack_Iv has quit IRC | 08:00 | |
*** Jack_Iv has joined #openstack-telemetry | 08:00 | |
*** pcaruana has joined #openstack-telemetry | 08:22 | |
*** thorst_afk has joined #openstack-telemetry | 08:59 | |
*** thorst_afk has quit IRC | 09:04 | |
*** lhx__ has quit IRC | 09:11 | |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer master: Remove useless metric name https://review.openstack.org/435327 | 09:28 |
*** cdent has joined #openstack-telemetry | 09:50 | |
*** nadya has quit IRC | 09:59 | |
*** jd__ has joined #openstack-telemetry | 10:08 | |
*** hfu has quit IRC | 10:09 | |
*** iceyao has quit IRC | 10:11 | |
*** faizy has quit IRC | 10:12 | |
*** faizy has joined #openstack-telemetry | 10:13 | |
*** gongysh has quit IRC | 10:13 | |
*** sam_nowitzki has quit IRC | 10:29 | |
*** lhx_ has joined #openstack-telemetry | 10:32 | |
*** nadya has joined #openstack-telemetry | 10:33 | |
*** nadya has quit IRC | 10:38 | |
*** nadya has joined #openstack-telemetry | 10:50 | |
*** thorst_afk has joined #openstack-telemetry | 11:01 | |
*** thorst_afk has quit IRC | 11:05 | |
*** sudipto_ has quit IRC | 11:08 | |
*** sudipto has quit IRC | 11:08 | |
*** Jack_Iv has quit IRC | 11:18 | |
*** Jack_Iv has joined #openstack-telemetry | 11:20 | |
*** hfu has joined #openstack-telemetry | 11:35 | |
*** donghao has joined #openstack-telemetry | 12:10 | |
*** dave-mccowan has joined #openstack-telemetry | 12:13 | |
*** donghao has quit IRC | 12:16 | |
*** Jack_Iv has quit IRC | 12:19 | |
*** lhx_ has quit IRC | 12:20 | |
*** Jack_Iv has joined #openstack-telemetry | 12:23 | |
*** Jack_Iv has quit IRC | 12:23 | |
*** shardy is now known as shardy_lunch | 12:23 | |
*** catintheroof has joined #openstack-telemetry | 12:23 | |
*** vint_bra has joined #openstack-telemetry | 12:24 | |
*** thorst_afk has joined #openstack-telemetry | 12:43 | |
*** faizy has quit IRC | 12:45 | |
*** iceyao has joined #openstack-telemetry | 13:00 | |
*** iceyao has quit IRC | 13:05 | |
*** fguillot has joined #openstack-telemetry | 13:13 | |
*** lhx_ has joined #openstack-telemetry | 13:15 | |
*** sudipto has joined #openstack-telemetry | 13:16 | |
*** sudipto_ has joined #openstack-telemetry | 13:16 | |
*** iceyao has joined #openstack-telemetry | 13:22 | |
*** nadya has quit IRC | 13:26 | |
frickler | hmm, has python-gnocchiclient been dropped from pypi? | 13:29 |
frickler | ah, no, its without the "python-", nevermind :) | 13:31 |
*** shardy_lunch is now known as shardy | 13:32 | |
frickler | https://launchpad.net/python-gnocchiclient has it wrong, so its not only me *phew* | 13:33 |
*** links has quit IRC | 13:36 | |
frickler | also, it seems that one needs gnocchiclient>=3.0.0 in order to talk to gnocchi-api=3.1.0, if that is mentioned in the release notes, I didn't spot it | 13:36 |
*** faizy has joined #openstack-telemetry | 13:41 | |
*** gordc has joined #openstack-telemetry | 13:45 | |
*** pradk has joined #openstack-telemetry | 13:53 | |
*** thorst_afk is now known as thorst_ | 13:56 | |
*** vint_bra has quit IRC | 13:59 | |
*** vint_bra has joined #openstack-telemetry | 14:05 | |
*** donghao has joined #openstack-telemetry | 14:10 | |
*** chlong has joined #openstack-telemetry | 14:15 | |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 14:26 |
*** nadya has joined #openstack-telemetry | 14:27 | |
catintheroof | gordc: sileht guys, hi, is there any insight about if ceilometer supports to push RBD disk metrics ? here's the reference: https://bugs.launchpad.net/mos/+bug/1539253 | 14:30 |
openstack | Launchpad bug 1539253 in Mirantis OpenStack 9.x "Ceilometer: Disk.usage metric is not available for ceph volume. " [Medium,Fix released] - Assigned to Ildar Svetlov (isvetlov) | 14:30 |
sileht | catintheroof, yes you need to implement it in KVM | 14:30 |
catintheroof | sileht: do you have any reference on how to do it ? | 14:31 |
sileht | catintheroof, no idea | 14:31 |
sileht | catintheroof, the only I know is Ceilometer rely on libvirt that relies on KVM and KVM doesn't provide this information for RBD disk | 14:31 |
*** nicodemus_ has joined #openstack-telemetry | 14:31 | |
catintheroof | sileht: oh, that's bad | 14:32 |
gordc | i know less than sileht | 14:32 |
gordc | lol | 14:32 |
*** hfu has quit IRC | 14:32 | |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 14:33 |
*** pradk has quit IRC | 14:34 | |
*** yassine_ has quit IRC | 14:35 | |
*** yassine has joined #openstack-telemetry | 14:35 | |
*** rcherrueau has joined #openstack-telemetry | 14:35 | |
*** yassine is now known as Guest70824 | 14:35 | |
catintheroof | sileht: does this means https://bugs.launchpad.net/ceilometer/+bug/1457440 that on ubuntu's 16.04 libvirt version this might be fixed ? | 14:39 |
openstack | Launchpad bug 1457440 in Ceilometer "Compute agent virDomainGetBlockInfo error for instances with RBD backend" [Medium,Triaged] | 14:39 |
sileht | catintheroof, maybe I haven't tested recently | 14:41 |
sileht | catintheroof, I have just tested: http://paste.openstack.org/show/599417/ | 14:43 |
sileht | catintheroof, so yes the version of KVM and libvirt in ubuntu 16.04 have this implemented | 14:43 |
*** iceyao has quit IRC | 14:43 | |
catintheroof | sileht: niceeee! thanks for testing that one out | 14:45 |
gordc | sileht: red hat still doesn't know you don't have fedora? :P | 14:45 |
sileht | gordc, shut up ! | 14:45 |
sileht | gordc, I use both :) | 14:46 |
sileht | (but one more than the other) | 14:46 |
catintheroof | hahahaahah | 14:46 |
gordc | sileht: lol | 14:47 |
gordc | understood | 14:47 |
gordc | don't worry, i'll raise fedora bugs through you. then they think you are qa'ing | 14:48 |
*** tlian has joined #openstack-telemetry | 14:53 | |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 14:54 |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 14:59 |
*** pradk has joined #openstack-telemetry | 15:03 | |
lhx_ | gordc, sileht, have you ever beringei? | 15:05 |
lhx_ | do you guys know beringei? | 15:05 |
gordc | lhx_: no, sorry. | 15:06 |
sileht | lhx_, I have just cross it, but it's a 'inmemory' implementation so it's fast, it doesn't store the data... | 15:06 |
lhx_ | gordc, Facebook made it open source recently as time-series storage | 15:07 |
lhx_ | sileht, yeah, so fast | 15:08 |
lhx_ | with high performance | 15:08 |
sileht | they say 'scalable' but it's not 'scalable verticaly or horizontaly' | 15:08 |
sileht | so I don't got the direction of the scalability ;à | 15:09 |
gordc | lhx_: i hacked gnocchi to write to ramdisk once. it was 'so fast' also :P | 15:09 |
sileht | I personnaly use it with the 'incoming data' into a tmpfs :) it's fast also | 15:10 |
sileht | I don't think 'Beringei' and 'Gnocchi' goals are the same | 15:10 |
gordc | we need to do our benchmarks on that. :) | 15:10 |
sileht | I got how they do the scaling part, they do sharding with the client... | 15:12 |
lhx_ | gordc, i guess you can compare them in some aspects, but you can't do quantitative comparison | 15:13 |
gordc | sileht: like ceph? | 15:13 |
gordc | lhx_: was there something unique it does that we can do as well? idea 'sharing' | 15:13 |
sileht | gordc, I don't found the detail | 15:16 |
gordc | sileht: ack. maybe i'll look at it later | 15:17 |
lhx_ | not very clear, just know it has extremely fast query performance in-memory storage and extremely efficient streaming compression algorithm | 15:18 |
*** donghao has quit IRC | 15:20 | |
lhx_ | the percent of compress data would be higher than 90% | 15:20 |
lhx_ | seems better than gnocchi | 15:21 |
*** donghao has joined #openstack-telemetry | 15:21 | |
lhx_ | can also integrate with Grafana | 15:22 |
lhx_ | s/integrate/be integrated | 15:22 |
*** faizy has quit IRC | 15:22 | |
gordc | lhx_: i imagine you need some context to compression percentage. if i compress 1000 '1' and 1 '0', the compression will be amazing also. :) | 15:24 |
sileht | I think I will test it to see how it works more in detail | 15:24 |
lhx_ | gordc, lol, you are right, we should do quantitative comparison | 15:27 |
lhx_ | sileht, sounds great! | 15:29 |
*** faizy has joined #openstack-telemetry | 15:32 | |
*** nadya has quit IRC | 15:34 | |
*** donghao has quit IRC | 15:49 | |
*** rbak has joined #openstack-telemetry | 16:08 | |
*** tlian2 has joined #openstack-telemetry | 16:10 | |
*** tlian has quit IRC | 16:12 | |
*** tlian2 has quit IRC | 16:15 | |
*** lhx_ has quit IRC | 16:19 | |
*** pcaruana has quit IRC | 16:21 | |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 16:31 |
*** Jack_Iv has joined #openstack-telemetry | 16:33 | |
*** rcernin has quit IRC | 16:35 | |
*** Jack_Iv has quit IRC | 16:38 | |
*** nadya has joined #openstack-telemetry | 16:39 | |
*** sudipto_ has quit IRC | 16:39 | |
*** sudipto has quit IRC | 16:39 | |
*** tesseract has quit IRC | 16:41 | |
*** Jack_Iv has joined #openstack-telemetry | 16:43 | |
*** rbak_ has joined #openstack-telemetry | 16:50 | |
*** rbak has quit IRC | 16:50 | |
*** faizy has quit IRC | 16:54 | |
*** donghao has joined #openstack-telemetry | 17:04 | |
*** donghao has quit IRC | 17:11 | |
openstackgerrit | Julien Danjou proposed openstack/aodh master: Add keystoneauth1 in requirements https://review.openstack.org/435538 | 17:12 |
*** Jack_Iv has quit IRC | 17:28 | |
catintheroof | gordc: what is the situation where agent-compute pushes to notification.critical ? | 17:28 |
*** rcherrueau has quit IRC | 17:28 | |
*** Jack_Iv has joined #openstack-telemetry | 17:28 | |
gordc | catintheroof: ceilometer will never push to *.critical (not that i know of). we listen to additional topics such as *.error in case the other services do | 17:42 |
*** Jack_Iv has quit IRC | 17:45 | |
*** Jack_Iv has joined #openstack-telemetry | 17:45 | |
*** cdent has quit IRC | 17:47 | |
*** Jack_Iv has quit IRC | 17:48 | |
*** Jack_Iv has joined #openstack-telemetry | 17:49 | |
*** Jack_Iv_ has joined #openstack-telemetry | 17:51 | |
*** Jack_Iv has quit IRC | 17:51 | |
catintheroof | gordc: sorry, was my terminal that showed me wrong values | 17:53 |
*** Jack_Iv has joined #openstack-telemetry | 17:54 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer stable/mitaka: Updated from global requirements https://review.openstack.org/328529 | 17:55 |
*** Jack_Iv has quit IRC | 17:56 | |
*** Jack_Iv_ has quit IRC | 17:56 | |
*** Jack_Iv has joined #openstack-telemetry | 17:57 | |
*** Jack_Iv has quit IRC | 17:57 | |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 17:57 |
*** nadya has quit IRC | 17:57 | |
*** Jack_Iv has joined #openstack-telemetry | 18:01 | |
*** Jack_Iv has quit IRC | 18:09 | |
*** tlian has joined #openstack-telemetry | 18:18 | |
nicodemus_ | hello | 18:22 |
nicodemus_ | Suppose I want ceilometer-agent to push an extra key in the samples. This extra key needs to be accepted by gnocchi, but I don't need to have it on the resource_type definition (since this value will go to generic) | 18:24 |
nicodemus_ | How could I make gnocchi to accept it if it's not on the resource_type definition? I'm imagining adding an "if" somewhere in indexer/sqlalchemy.py, but don't know exactly where... | 18:26 |
nicodemus_ | my guess is that here: https://github.com/openstack/gnocchi/blob/master/gnocchi/indexer/sqlalchemy.py#L747 the create_resource method gets the resource definition, is that so? | 18:30 |
gordc | nicodemus_: i don't think generic stores any attributes so i'm not sure what you're asking is possible | 18:45 |
gordc | you can ask on ML. i dont know the answer myself and not sure anyone else is online anymore. | 18:46 |
openstackgerrit | gordon chung proposed openstack/ceilometer master: add missing instance meta https://review.openstack.org/435109 | 18:54 |
*** tomoiaga has joined #openstack-telemetry | 18:55 | |
*** Pneumatic_ has joined #openstack-telemetry | 18:58 | |
-openstackstatus- NOTICE: Restarting gerrit due to performance problems | 19:04 | |
*** fguillot has quit IRC | 19:12 | |
*** yprokule has quit IRC | 19:18 | |
*** rbak_ has quit IRC | 19:21 | |
*** Pneumatic_ has quit IRC | 19:30 | |
harlowja | gordc jd__ the polling rate of different pollsters is all configurable right? | 19:32 |
harlowja | https://github.com/openstack/ceilometer/blob/master/etc/ceilometer/polling.yaml though sets everything to 10 minutes? | 19:33 |
harlowja | i guess most people change that :-P | 19:33 |
gordc | harlowja: yeah, it's the default. | 19:35 |
gordc | i think we might have actually asked whether people changed that in a prior survey... let me check | 19:35 |
harlowja | kk | 19:36 |
gordc | but yeah 10 minutes is long.lol | 19:36 |
harlowja | for certain pollsters it may be, for certain others it might be ok :-P | 19:39 |
*** rbak has joined #openstack-telemetry | 19:40 | |
gordc | harlowja: we asked what ideal poll interval was in survey back in Oct2015, ~65% said 5-10 minutes, ~27% said 1-5min | 19:47 |
harlowja | for all the pollsters? | 19:47 |
harlowja | do people usually set the globally ? | 19:47 |
gordc | we didn't ask specifically per pollster | 19:47 |
harlowja | k | 19:47 |
gordc | also, this was only 18 replies | 19:47 |
harlowja | ya, fair nuff | 19:47 |
harlowja | thx :) | 19:47 |
gordc | np | 19:48 |
*** donghao has joined #openstack-telemetry | 19:56 | |
*** jfluhmann has joined #openstack-telemetry | 20:00 | |
*** donghao has quit IRC | 20:00 | |
jd__ | harlowja: yes | 20:10 |
harlowja | ya, i thought so, just wondered what people were setting them to | 20:10 |
jd__ | we set it to 10 min a while back to not overflow mongodb | 20:10 |
jd__ | i imagine we could set it to 60 now that we default to gnocchi | 20:10 |
harlowja | kk | 20:10 |
*** lionel has quit IRC | 20:11 | |
*** lionel has joined #openstack-telemetry | 20:12 | |
nicodemus_ | excuse me gordc, I got sidetracked. Let me rephrase the question: when/where are a resource's attributes checked when the dispatcher tries to POST something to gnocchi? | 20:19 |
nicodemus_ | Imagine a new resource that needs to be created, and the dispatcher has X amount of attributes mapped to that resource in the gnocchi_resources.yaml file. The dispatcher tries to POST to gnocchi API, how does gnocchi verify that said POST contains an extra key? | 20:20 |
sileht | nicodemus_, it should return 400 for invalid input | 20:23 |
nicodemus_ | sileht, it does indeed | 20:23 |
nicodemus_ | but... I'm trying to avoid returning that 400 error | 20:23 |
sileht | nicodemus_, we use volouptuous lib to check input, and we write in the code we want this behavior | 20:24 |
sileht | nicodemus_, do you want to ignore the extra key ? | 20:24 |
nicodemus_ | yes | 20:25 |
nicodemus_ | I need to get this extra key and use it for something else, with gnocchi not complaining about it | 20:25 |
nicodemus_ | is the input check done here? https://github.com/openstack/gnocchi/blob/e11e7b4f438e474f78355a02bf17b54068370931/gnocchi/rest/__init__.py#L109 | 20:26 |
*** tomoiaga has quit IRC | 20:31 | |
*** jfluhmann has quit IRC | 20:31 | |
sileht | nicodemus_, you can hack here, but we don't want to do that upstream | 20:49 |
nicodemus_ | sileht, obviously not... I'm trying to create a monster | 20:49 |
*** g3ek has quit IRC | 21:04 | |
*** nicodemus_ has quit IRC | 21:15 | |
*** g3ek has joined #openstack-telemetry | 21:18 | |
*** gordc has quit IRC | 21:49 | |
*** thorst_ has quit IRC | 21:53 | |
*** vint_bra has quit IRC | 22:00 | |
*** Guest70824 has quit IRC | 22:03 | |
*** Guest70824 has joined #openstack-telemetry | 22:03 | |
*** rbak has quit IRC | 22:14 | |
*** dave-mccowan has quit IRC | 22:17 | |
*** jd__ has quit IRC | 22:59 | |
*** rbak has joined #openstack-telemetry | 23:02 | |
*** catintheroof has quit IRC | 23:21 | |
*** rbak has quit IRC | 23:21 | |
*** catintheroof has joined #openstack-telemetry | 23:22 | |
*** catintheroof has quit IRC | 23:26 | |
*** thorst has joined #openstack-telemetry | 23:27 | |
*** thorst has quit IRC | 23:32 | |
*** Guest70824 has quit IRC | 23:45 | |
*** chlong has quit IRC | 23:46 | |
*** Guest70824 has joined #openstack-telemetry | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!