| *** DinaBelova has joined #openstack-telemetry | 00:02 | |
| *** aignatov has joined #openstack-telemetry | 00:02 | |
| *** rbak has quit IRC | 00:03 | |
| *** david-lyle has quit IRC | 00:05 | |
| *** hfu has joined #openstack-telemetry | 00:23 | |
| *** catintheroof has joined #openstack-telemetry | 00:23 | |
| *** gordc has quit IRC | 00:35 | |
| *** thorst has joined #openstack-telemetry | 00:39 | |
| *** thorst has quit IRC | 00:42 | |
| *** dschultz has joined #openstack-telemetry | 00:52 | |
| *** hfu has quit IRC | 00:52 | |
| *** harlowja has joined #openstack-telemetry | 00:52 | |
| *** hfu has joined #openstack-telemetry | 00:54 | |
| *** dschultz has quit IRC | 00:56 | |
| *** dschultz has joined #openstack-telemetry | 00:56 | |
| *** zhurong has joined #openstack-telemetry | 01:00 | |
| *** vint_bra has joined #openstack-telemetry | 01:06 | |
| *** vint_bra has quit IRC | 01:07 | |
| *** thorst has joined #openstack-telemetry | 01:11 | |
| *** thorst has quit IRC | 01:11 | |
| *** catintheroof has quit IRC | 01:13 | |
| *** gongysh has joined #openstack-telemetry | 01:14 | |
| *** catintheroof has joined #openstack-telemetry | 01:14 | |
| *** catintheroof has quit IRC | 01:14 | |
| *** vint_bra has joined #openstack-telemetry | 01:33 | |
| *** gongysh has quit IRC | 01:40 | |
| *** gongysh has joined #openstack-telemetry | 01:42 | |
| *** gongysh has quit IRC | 01:42 | |
| *** lhx__ has joined #openstack-telemetry | 01:46 | |
| *** gongysh has joined #openstack-telemetry | 01:50 | |
| *** gongysh has quit IRC | 01:53 | |
| *** lhx_ has joined #openstack-telemetry | 01:53 | |
| *** lhx__ has quit IRC | 01:54 | |
| *** Jack_Iv has joined #openstack-telemetry | 02:11 | |
| *** thorst has joined #openstack-telemetry | 02:15 | |
| *** Jack_Iv has quit IRC | 02:15 | |
| *** llu has joined #openstack-telemetry | 02:18 | |
| *** vint_bra has quit IRC | 02:20 | |
| openstackgerrit | qin.jiang proposed openstack/ceilometer: Trivial: add white space of error message https://review.openstack.org/422341 | 02:20 |
|---|---|---|
| *** thorst has quit IRC | 02:20 | |
| *** fc___ has joined #openstack-telemetry | 02:33 | |
| *** links has joined #openstack-telemetry | 02:33 | |
| *** larainema_ has joined #openstack-telemetry | 02:35 | |
| *** darrenc_ has joined #openstack-telemetry | 02:35 | |
| *** dewanee_ has joined #openstack-telemetry | 02:36 | |
| *** larainema has quit IRC | 02:41 | |
| *** adriant has quit IRC | 02:41 | |
| *** fc__ has quit IRC | 02:41 | |
| *** swamireddy has quit IRC | 02:41 | |
| *** hyakuhei has quit IRC | 02:41 | |
| *** dewanee has quit IRC | 02:41 | |
| *** darrenc has quit IRC | 02:41 | |
| *** fc___ is now known as fc__ | 02:41 | |
| *** larainema_ is now known as larainema | 02:42 | |
| *** yarkot has quit IRC | 02:44 | |
| *** yarkot has joined #openstack-telemetry | 02:50 | |
| *** adriant has joined #openstack-telemetry | 02:51 | |
| openstackgerrit | Ha Dung proposed openstack/ceilometer: Fixes typo1 for of Ceilometer project. https://review.openstack.org/422355 | 02:53 |
| openstackgerrit | Ha Dung proposed openstack/ceilometer: Fixes typo2 for vmware folder- Ceilometer project. https://review.openstack.org/422358 | 02:57 |
| *** sam_nowitzki has joined #openstack-telemetry | 02:59 | |
| *** hyakuhei has joined #openstack-telemetry | 03:01 | |
| *** swamireddy has joined #openstack-telemetry | 03:04 | |
| *** tlian has quit IRC | 03:27 | |
| *** thorst has joined #openstack-telemetry | 03:30 | |
| *** zhurong has quit IRC | 03:34 | |
| *** thorst has quit IRC | 03:36 | |
| *** jp_ has quit IRC | 03:43 | |
| *** r-mibu has quit IRC | 04:03 | |
| *** sudipto_ has joined #openstack-telemetry | 04:07 | |
| *** sudipto has joined #openstack-telemetry | 04:07 | |
| *** Jack_Iv has joined #openstack-telemetry | 04:12 | |
| *** Jack_Iv has quit IRC | 04:16 | |
| openstackgerrit | Merged openstack/ceilometer: Trivial: add white space of error message https://review.openstack.org/422341 | 04:16 |
| *** lhx_ has quit IRC | 04:23 | |
| *** gongysh has joined #openstack-telemetry | 04:33 | |
| *** r-mibu has joined #openstack-telemetry | 04:33 | |
| *** thorst has joined #openstack-telemetry | 04:33 | |
| *** thorst has quit IRC | 04:38 | |
| *** gongysh has quit IRC | 04:42 | |
| *** xiaofandh12 has joined #openstack-telemetry | 04:42 | |
| *** donghao has quit IRC | 04:45 | |
| *** sudipto_ has quit IRC | 04:49 | |
| *** sudipto has quit IRC | 04:49 | |
| *** sudipto has joined #openstack-telemetry | 05:27 | |
| *** sudipto_ has joined #openstack-telemetry | 05:27 | |
| *** joadavis has joined #openstack-telemetry | 05:27 | |
| *** gongysh has joined #openstack-telemetry | 05:31 | |
| *** adriant has quit IRC | 05:37 | |
| *** lhx_ has joined #openstack-telemetry | 05:38 | |
| *** nadya has joined #openstack-telemetry | 05:48 | |
| *** nadya has quit IRC | 05:53 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:00 | |
| *** Jack_Iv has quit IRC | 06:00 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:01 | |
| *** Jack_Iv has quit IRC | 06:02 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:02 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 06:02 | |
| *** Jack_Iv has quit IRC | 06:06 | |
| *** thorst has joined #openstack-telemetry | 06:34 | |
| *** thorst has quit IRC | 06:39 | |
| *** Jack_Iv_ has quit IRC | 06:40 | |
| openstackgerrit | Hanxi Liu proposed openstack/ceilometer: Support extended declaring exchanges https://review.openstack.org/422411 | 06:43 |
| *** zhurong has joined #openstack-telemetry | 07:07 | |
| openstackgerrit | Shaoman Zhang proposed openstack/ceilometer: Replace disk with virtualDisk for VMware https://review.openstack.org/411261 | 07:15 |
| *** zyun has joined #openstack-telemetry | 07:25 | |
| zyun | Dears, when I run "ceilometer meter-list", I encounter one error: | 07:26 |
| sileht | / | 07:29 |
| *** masber has quit IRC | 07:30 | |
| zyun | paste.openstack.org | 07:35 |
| zyun | http://paste.openstack.org/show/595555/ | 07:36 |
| zyun | this is the error msg | 07:36 |
| zyun | Who can give me a little clue about it? It is Kilo release | 07:36 |
| *** dschultz has quit IRC | 07:38 | |
| *** dschultz has joined #openstack-telemetry | 07:38 | |
| *** dschultz has quit IRC | 07:41 | |
| zyun | Anyone can help me out ? Thanks in advance!! | 07:52 |
| *** sanchitmalhotra1 has joined #openstack-telemetry | 07:52 | |
| *** sanchitmalhotra has quit IRC | 07:54 | |
| *** sanchitmalhotra1 is now known as sanchitmalhotra | 07:54 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for serialization https://review.openstack.org/421766 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: handle timestamps from struct with numpy https://review.openstack.org/421570 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: prepare datetime for pandas.to_datetime() https://review.openstack.org/421208 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Don't call clean_ts() when unserialize https://review.openstack.org/421714 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: remove a pandas.iteritems() https://review.openstack.org/421516 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for unserialization https://review.openstack.org/421901 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: indexer: fix migration script "no_more_slash" https://review.openstack.org/421900 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: rest: returns orignal resource id https://review.openstack.org/417836 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: don't use groupby for split https://review.openstack.org/422445 | 07:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: add metricd tester for profiling https://review.openstack.org/422446 | 07:57 |
| sileht | jd__, I have fixed split(), now it's almost immediate | 07:58 |
| sileht | next on my list is a smarter merge() | 07:59 |
| *** yprokule has joined #openstack-telemetry | 08:01 | |
| lhx_ | sileht, Could I change the default configurations in pipeline.yaml? | 08:02 |
| sileht | lhx_: I have proposed a change that already does that | 08:03 |
| lhx_ | sileht, haha, seems I don't need do that in the patch, and jd__ should know that | 08:06 |
| lhx_ | https://review.openstack.org/#/c/413920/ | 08:06 |
| sileht | lhx_, I'm guessing jd haven't seen my change | 08:08 |
| lhx_ | sileht, lol, I think so | 08:09 |
| sileht | lhx_, I have added a link to my change to your review | 08:10 |
| lhx_ | sileht, cool ;) | 08:13 |
| *** tesseract has joined #openstack-telemetry | 08:13 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: Allow to configure the default publisher https://review.openstack.org/420681 | 08:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: Don't load many times the same publisher https://review.openstack.org/420655 | 08:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: pipeline: remove tests helper from runtime code https://review.openstack.org/420636 | 08:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: Change default publishers to Gnocchi https://review.openstack.org/422458 | 08:15 |
| lhx_ | sileht, jd__, are you preparing a session for Boston? | 08:17 |
| sileht | lhx_, I think jd__ have proposed one about performance of a big Gnocchi deployement | 08:18 |
| sileht | lhx_, I haven't proposed anything | 08:18 |
| lhx_ | sileht, how about gordc? | 08:19 |
| sileht | no idea | 08:19 |
| lhx_ | lhx, haha, the session commit deadline is coming. | 08:24 |
| lhx_ | sileht, sorry, wrong ping myself | 08:24 |
| sileht | lhx_, will/did you propose something ? | 08:24 |
| lhx_ | sileht, the time during I work on OpenStack is so short, nothing to propose, lol | 08:26 |
| sileht | lhx_, short, but I can see you're doing very cool and useful stuff for Telemetry project | 08:27 |
| lhx_ | sileht, acutally a little, I know you say it no seriously, :) | 08:30 |
| sileht | lhx_, i'm serious, cleaning the TODO/FIXME and helping to remove item from the coming feature list is awesome | 08:31 |
| sileht | lhx_, for example the deprecation of the collector is planned for two cycles, and it will almost being done for this cycle | 08:32 |
| *** openstackgerrit has quit IRC | 08:33 | |
| *** masber has joined #openstack-telemetry | 08:34 | |
| *** thorst has joined #openstack-telemetry | 08:35 | |
| lhx_ | lhx__, I will do it better with honor. maybe Boston is the first summit I can go to meet you, :) | 08:35 |
| lhx_ | sileht, wrong ping twice, lol | 08:35 |
| sileht | cool :) | 08:35 |
| *** zhurong has quit IRC | 08:35 | |
| *** thorst has quit IRC | 08:39 | |
| *** dschultz has joined #openstack-telemetry | 08:42 | |
| *** dschultz has quit IRC | 08:47 | |
| *** openstackgerrit has joined #openstack-telemetry | 08:53 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: nova: track flavor name https://review.openstack.org/417313 | 08:53 |
| *** hfu has quit IRC | 09:02 | |
| *** shardy_afk is now known as shardy | 09:03 | |
| *** hfu has joined #openstack-telemetry | 09:03 | |
| jd__ | lhx_: sileht: cool guys | 09:07 |
| jd__ | this probably should have been one patch :) | 09:07 |
| jd__ | though I don't see how they are related | 09:08 |
| jd__ | lhx_: yes I have proposed a talk for Boston about telemetry scalability :) | 09:09 |
| *** efoley has joined #openstack-telemetry | 09:13 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: add metricd tester for profiling https://review.openstack.org/422446 | 09:21 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: don't use groupby for split https://review.openstack.org/422445 | 09:21 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Add benchmark for split() https://review.openstack.org/422496 | 09:21 |
| *** nadya has joined #openstack-telemetry | 09:22 | |
| sileht | jd__, https://penandpants.com/2014/09/05/performance-of-pandas-series-vs-numpy-arrays/ | 09:24 |
| *** Jack_Iv has joined #openstack-telemetry | 09:24 | |
| *** yassine has joined #openstack-telemetry | 09:28 | |
| *** yassine is now known as Guest30600 | 09:28 | |
| *** Jack_Iv has quit IRC | 09:33 | |
| jd__ | sileht: haha | 09:34 |
| *** hyakuhei has quit IRC | 09:36 | |
| *** hyakuhei has joined #openstack-telemetry | 09:36 | |
| *** hyakuhei has quit IRC | 09:36 | |
| *** hyakuhei has joined #openstack-telemetry | 09:36 | |
| sileht | django drop py2 \o/ | 09:45 |
| *** zyun has quit IRC | 09:53 | |
| *** Guest30600 has quit IRC | 10:01 | |
| *** nadya has quit IRC | 10:02 | |
| *** efoley_ has joined #openstack-telemetry | 10:03 | |
| *** efoley has quit IRC | 10:05 | |
| jd__ | sileht: can you check https://review.openstack.org/#/c/421454/ so I can fix pifpaf? :) | 10:10 |
| *** Guest30600 has joined #openstack-telemetry | 10:16 | |
| *** sam_nowitzki has quit IRC | 10:23 | |
| *** masber has quit IRC | 10:26 | |
| sileht | jd__, damn I have to rerun my bench with the new split code, I haven't created enought measures: 7911 - 0 = 7911 | 10:27 |
| *** gongysh has quit IRC | 10:30 | |
| *** zhurong has joined #openstack-telemetry | 10:30 | |
| *** nadya has joined #openstack-telemetry | 10:33 | |
| *** joadavis_ has quit IRC | 10:34 | |
| *** Jack_Iv has joined #openstack-telemetry | 10:35 | |
| *** thorst has joined #openstack-telemetry | 10:35 | |
| *** Jack_Iv has quit IRC | 10:36 | |
| *** Jack_Iv has joined #openstack-telemetry | 10:36 | |
| sileht | jd__, for merge(), I have tested 3 solutions with numpy, but all returns the perf :( | 10:37 |
| *** nadya has quit IRC | 10:38 | |
| *** thorst has quit IRC | 10:40 | |
| *** masber has joined #openstack-telemetry | 10:41 | |
| *** nadya has joined #openstack-telemetry | 10:45 | |
| sileht | jd__, the split() change: 15372 measures in 10 minutes :) | 10:57 |
| sileht | instead of ~4000 yesterday | 10:57 |
| *** Jack_Iv has quit IRC | 11:02 | |
| *** masber has quit IRC | 11:06 | |
| openstackgerrit | qin.jiang proposed openstack/ceilometer: Extract the same code of string definition https://review.openstack.org/422555 | 11:11 |
| *** Deep_Thought has joined #openstack-telemetry | 11:16 | |
| *** sudipto_ has quit IRC | 11:39 | |
| *** sudipto has quit IRC | 11:39 | |
| *** efoley__ has joined #openstack-telemetry | 12:01 | |
| *** Jack_Iv has joined #openstack-telemetry | 12:03 | |
| *** efoley_ has quit IRC | 12:05 | |
| *** zhurong has quit IRC | 12:05 | |
| *** Jack_Iv has quit IRC | 12:08 | |
| jd__ | sileht: damn that's how much better? | 12:09 |
| sileht | jd__, I'm replaying my bench to verify my numbers | 12:09 |
| jd__ | sileht on the way to achieve negative perf | 12:10 |
| *** masber has joined #openstack-telemetry | 12:20 | |
| sileht | jd__, I have restarted because I have found that gnocchi have noisy flatmate on the server | 12:24 |
| sileht | jd__, I also change the methodology and will just compare how many time a defined number of metrics take to be processed. | 12:29 |
| sileht | instead of grab metricd logs | 12:30 |
| Deep_Thought | Hello, does gnocchi provide some way to consume data using websockets ? | 12:32 |
| *** shardy is now known as shardy_lunch | 12:34 | |
| jd__ | Deep_Thought: not really, what would you have in mind? | 12:36 |
| *** thorst has joined #openstack-telemetry | 12:37 | |
| openstackgerrit | Julien Danjou proposed openstack/aodh: Move policy.json out of etc https://review.openstack.org/422220 | 12:38 |
| openstackgerrit | qin.jiang proposed openstack/ceilometer: Fix same simple message error https://review.openstack.org/422590 | 12:39 |
| Deep_Thought | jd__: a way to open a stream (using websocket) to e.g /v1/metric/XXX and get data as soon as they come | 12:41 |
| jd__ | Deep_Thought: that's not supported yet but I think it'd be a good idea | 12:41 |
| Deep_Thought | ok | 12:42 |
| openstackgerrit | qin.jiang proposed openstack/ceilometer: Fix same type of simple message error https://review.openstack.org/422590 | 12:43 |
| *** catintheroof has joined #openstack-telemetry | 12:43 | |
| jd__ | Deep_Thought: you want to build a shiny interface? :) | 12:47 |
| *** dave-mccowan has joined #openstack-telemetry | 12:48 | |
| Deep_Thought | jd__: Not me personnaly but the front-end team would love to do so :D | 12:48 |
| jd__ | Deep_Thought: hehe | 12:57 |
| sileht | jd__, the most impressive improvement is the split() method | 12:58 |
| jd__ | nice | 12:58 |
| sileht | jd__, when I compare profiles of _add_measures(), before split() was representing 14% of work, now it's 3% | 13:01 |
| *** nadya has quit IRC | 13:04 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for serialization https://review.openstack.org/421766 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Add benchmark for split() https://review.openstack.org/422496 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: handle timestamps from struct with numpy https://review.openstack.org/421570 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: prepare datetime for pandas.to_datetime() https://review.openstack.org/421208 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Don't call clean_ts() when unserialize https://review.openstack.org/421714 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: remove a pandas.iteritems() https://review.openstack.org/421516 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: add metricd tester for profiling https://review.openstack.org/422446 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: don't use groupby for split https://review.openstack.org/422445 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for unserialization https://review.openstack.org/421901 | 13:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: combine_first optimisation https://review.openstack.org/422606 | 13:07 |
| *** links has quit IRC | 13:11 | |
| *** tlian has joined #openstack-telemetry | 13:14 | |
| *** vint_bra has joined #openstack-telemetry | 13:17 | |
| *** lhx_ has quit IRC | 13:17 | |
| *** shardy_lunch is now known as shardy | 13:20 | |
| *** glb has joined #openstack-telemetry | 13:23 | |
| *** gordc has joined #openstack-telemetry | 13:30 | |
| glb | hi! i'm trying to test the ceilometer v7.0.1, and i've encoutenred an issue with cotyledon | 13:32 |
| glb | here is the stack trace | 13:32 |
| glb | Traceback (most recent call last): | 13:32 |
| glb | File "/usr/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap | 13:32 |
| glb | self.run() | 13:32 |
| glb | File "/usr/lib/python2.7/multiprocessing/process.py", line 114, in run | 13:32 |
| glb | self._target(*self._args, **self._kwargs) | 13:32 |
| *** glb has quit IRC | 13:32 | |
| *** glb has joined #openstack-telemetry | 13:33 | |
| glb | has anyone seen this before ? | 13:33 |
| glb | sorry for the flood, I tought I would make a single message | 13:35 |
| sileht | glb, please use paste.openstack.org | 13:37 |
| glb | ok thx for the tips :) | 13:38 |
| glb | http://paste.openstack.org/show/595617/ | 13:38 |
| *** sudipto has joined #openstack-telemetry | 13:43 | |
| *** sudipto_ has joined #openstack-telemetry | 13:43 | |
| *** pradk has joined #openstack-telemetry | 13:44 | |
| *** lhx_ has joined #openstack-telemetry | 13:45 | |
| *** sudipto has quit IRC | 13:45 | |
| *** sudipto_ has quit IRC | 13:45 | |
| *** dschultz has joined #openstack-telemetry | 13:49 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: combine_first optimisation https://review.openstack.org/422606 | 13:50 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: add merge() benchmark https://review.openstack.org/422637 | 13:50 |
| gordc | glb: that doesn't look related to cotyledon... seems like you have bad config | 13:54 |
| gordc | jd__: sileht: should we be recording baseline for all the benchmarks? so we actually know if something negatively changed it. | 13:55 |
| jd__ | gordc: it's really hard to do so as it really depends on the CPU you running of and we have basically no stable system to do that | 13:56 |
| jd__ | gordc: a good rule of thumb is to test any change that goes into carbonara | 13:56 |
| jd__ | with the embeded benchmark | 13:56 |
| sileht | yeah, I'm adding other test in the embeding benchmark, it's easier to compare numbers | 13:56 |
| glb | oh, thx for help gordc, do you know what part of the config it can related to that ? | 13:57 |
| *** nadya has joined #openstack-telemetry | 13:57 | |
| gordc | agreed. i just think currently running benchmark means nothing because it's just numbers with no context of past. | 13:58 |
| *** Deep_Thought has quit IRC | 13:59 | |
| gordc | i don't know what you mean by embedded benchmark but ok.lol | 13:59 |
| jd__ | gordc: the benchmark() method in carbonara | 13:59 |
| openstackgerrit | Julien Danjou proposed openstack/aodh: Move policy.json out of etc https://review.openstack.org/422220 | 13:59 |
| sileht | gordc, you can run 'python gnocchi/carbonara.py' before and after a change to see the diff | 14:00 |
| gordc | jd__: right. so if i run it, it gives me a bunch of numbers. | 14:00 |
| jd__ | we have a lot of job failing recently that's weird | 14:01 |
| gordc | jd__: http://logs.openstack.org/90/422590/2/check/gate-telemetry-dsvm-integration-ceilometer-ubuntu-xenial/a614e0f/console.html#_2017-01-19_13_36_28_869134 | 14:01 |
| gordc | something with heat tempest | 14:01 |
| jd__ | should we rant on #openstack-heat? | 14:02 |
| gordc | jd__: i leave it to you. you are pro :P | 14:02 |
| gordc | did they add openstack- to their channel? i thought they were too cool for it?lol | 14:03 |
| *** dschultz has quit IRC | 14:03 | |
| gordc | sileht: yeah, my concern is that it requires whoever is commiting/reviewing to be aware they should run that benchmark. | 14:04 |
| jd__ | gordc: dunnow | 14:04 |
| gordc | i was thinking just capturing a value and if we notice gone down a certain percentage, we should probably look back in history to see what's up. | 14:04 |
| *** fguillot has joined #openstack-telemetry | 14:05 | |
| jd__ | gordc: I'll let you try and see how much false positive you got :) | 14:05 |
| jd__ | gordc: look at the time range for a test job to run | 14:06 |
| jd__ | gordc: the same test job can took 3 minutes one day and 15 minutes the next day | 14:06 |
| jd__ | gordc: without any code change | 14:06 |
| jd__ | gordc: so good look doing that :-) | 14:06 |
| gordc | jd__: fair enough. i'm not advocating performance testing on vms (i still scratch my head on rally stuff) that said. these numbers will be nothing in a few weeks | 14:08 |
| jd__ | nothing? | 14:08 |
| gordc | like consider our original benchmark values when we initially did v3... i have no idea what they are | 14:08 |
| jd__ | yeah it's always relative I imagine | 14:08 |
| jd__ | gordc: though I'm not against your idea, don't get me wrong, I just don't see any practical way to implement it | 14:08 |
| jd__ | gordc: but if you have let me know | 14:09 |
| jd__ | I'd love to be able to send code and having a job saying "👍 it's faster" | 14:09 |
| gordc | i'm just going to save a copy of one run after all of silehts patches... if i die, then it dies with it. | 14:10 |
| gordc | s/it/me | 14:10 |
| gordc | glb: what value do you have here: https://github.com/openstack/ceilometer/blob/3d395e976f63433c28fd02a9a342dc7778d3a923/ceilometer/cmd/polling.py#L64 | 14:10 |
| gordc | assuming you changed it | 14:10 |
| glb | yes I launch with option --polling-namespaces central | 14:13 |
| gordc | hmm.. that should be fine. `ceilometer-polling --polling-namespaces central` | 14:15 |
| *** kelv has quit IRC | 14:16 | |
| gordc | glb: i don't know. did you modify entry_points? | 14:17 |
| *** rbak has joined #openstack-telemetry | 14:53 | |
| glb | no I don't | 14:55 |
| glb | I launch ceilometer v6.0.0 with the same config, same cli options but it failed with v7.0.1 | 14:56 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for serialization https://review.openstack.org/421766 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Add benchmark for split() https://review.openstack.org/422496 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: handle timestamps from struct with numpy https://review.openstack.org/421570 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: combine_first optimisation https://review.openstack.org/422606 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: add merge() benchmark https://review.openstack.org/422637 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: prepare datetime for pandas.to_datetime() https://review.openstack.org/421208 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Don't use clean_ts() https://review.openstack.org/421714 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: remove a pandas.iteritems() https://review.openstack.org/421516 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: add metricd tester for profiling https://review.openstack.org/422446 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: don't use groupby for split https://review.openstack.org/422445 | 14:57 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for unserialization https://review.openstack.org/421901 | 14:57 |
| jd__ | gordc: looks like heat is not aware of that failure | 15:08 |
| gordc | jd__: :/ how'd it merge in their gate but fail in ours? | 15:11 |
| jd__ | gordc: they don't run the job we run | 15:11 |
| jd__ | gordc: it's from https://review.openstack.org/#/c/348076 | 15:11 |
| gordc | oh yeah, just read heat channel | 15:12 |
| jd__ | ah sorry did not notice you were there | 15:12 |
| gordc | i wasn't reading it... was busy trying to learn numpy | 15:13 |
| *** nadya has quit IRC | 15:15 | |
| *** eglynn has joined #openstack-telemetry | 15:23 | |
| *** sheel has quit IRC | 15:27 | |
| lhx_ | gordc, could you have a look? | 15:48 |
| lhx_ | https://review.openstack.org/#/c/412508/ | 15:48 |
| *** vint_bra has quit IRC | 15:56 | |
| *** vint_bra has joined #openstack-telemetry | 15:56 | |
| glb | I see that in ceilometer log: No module named libvirt, so ceilometer need libvirt module ? but for what ? | 15:56 |
| gordc | lhx_: i don't think it's necessary it doesn't break anything if you set it even if collector not enabled. you probably still need to set it anyways since it still loads dispatchers | 16:00 |
| *** shardy has quit IRC | 16:00 | |
| gordc | glb: for compute agent... if it's asking for libvirt in central agent, seems there's a bug | 16:00 |
| *** shardy has joined #openstack-telemetry | 16:01 | |
| lhx_ | gordc, I just think there will be some unused configurations in conf when collector not enabled | 16:04 |
| *** nadya has joined #openstack-telemetry | 16:05 | |
| lhx_ | gordc, do I just need add a warn? | 16:06 |
| *** david-lyle has joined #openstack-telemetry | 16:06 | |
| *** jmlowe has joined #openstack-telemetry | 16:07 | |
| gordc | lhx_: it still will be used... because setting publisher database:// or whatever it is will still require dispatchers to be set | 16:08 |
| gordc | or maybe it doesn't... i guess it doesn't really hurt... if anything i would just drop it completely. | 16:10 |
| gordc | you already have a patch with warning https://review.openstack.org/#/c/413920/ | 16:11 |
| *** chopmann has joined #openstack-telemetry | 16:14 | |
| *** chopmann has quit IRC | 16:15 | |
| lhx_ | gordc, haha, just abandon it, | 16:17 |
| glb | godrc, no bug it's because I used both compute and central namespace in a test | 16:18 |
| *** kelv has joined #openstack-telemetry | 16:18 | |
| lhx_ | gordc, another one | 16:19 |
| lhx_ | https://review.openstack.org/#/c/422411/ | 16:19 |
| *** thorst is now known as thorst_afk | 16:19 | |
| gordc | glb: i see. we did some stuff around libvirt recently so wasnt' sure if it was a regression | 16:19 |
| lhx_ | gordc, I'm not sure if what I changed catches the bug? | 16:20 |
| gordc | lhx_: sort of. we just need to make control_exchanges configurable. | 16:20 |
| gordc | lhx_: if we reuse http_control_exchanges but rename it to something more general we can use it globally in ceilometer | 16:21 |
| gordc | so we don't need all those independent config options | 16:21 |
| gordc | (i might be wrong, i haven't tested any of ^) | 16:22 |
| lhx_ | gordc, yeah, I think so, but I'm confused about original advantages of http_control_exchange? | 16:22 |
| openstackgerrit | gordon chung proposed openstack/ceilometer: test https://review.openstack.org/422732 | 16:24 |
| gordc | http_control_exchanges originally designed to listen to oslo.middleware.notifier (something like that) | 16:26 |
| gordc | it can work on any services but i don't know if it's actually still active | 16:26 |
| gordc | it was suppose to meter api request... but i can't even find that middleware anymore. | 16:28 |
| gordc | that might be a dead metric. | 16:28 |
| lhx_ | gordc, this is what I need to make sure, lol | 16:28 |
| *** fc__ has left #openstack-telemetry | 16:28 | |
| gordc | lhx_: oh. nevermind it's in oslo.messaging https://github.com/openstack/oslo.messaging/blob/master/oslo_messaging/notify/middleware.py | 16:29 |
| gordc | lhx_: i don't think anyone uses it. | 16:29 |
| lhx_ | gordc, if still active, I will choose to use http_control_exchanges to replace all and make exchange configurable | 16:29 |
| *** vint_bra has quit IRC | 16:29 | |
| gordc | regardless i think you can rename http_control_exchanges and use it for notification agent. | 16:30 |
| lhx_ | gordc, should middleware.py be rename? | 16:32 |
| gordc | i think you can leave it. i guess you can declare the config option somewhere else. | 16:33 |
| lhx_ | gordc, declare in this? | 16:34 |
| lhx_ | https://github.com/openstack/ceilometer/blob/83d805b51e7ec96118917f1873e9e072ba8150ab/ceilometer/exchange_control.py | 16:34 |
| gordc | lhx_: maybe https://github.com/openstack/ceilometer/blob/83d805b51e7ec96118917f1873e9e072ba8150ab/ceilometer/notification.py | 16:34 |
| gordc | lhx_: since we're going to deprecate everything in exchange_control.py and hopefully delete it later | 16:35 |
| lhx_ | gordc, do it need a patch for deprecating it? | 16:36 |
| gordc | yeah. mark them all deprecated | 16:36 |
| gordc | you can do it in same patch | 16:36 |
| lhx_ | gordc, ok, thanks :) | 16:37 |
| *** links has joined #openstack-telemetry | 16:37 | |
| *** thorst_afk is now known as thorst_ | 16:37 | |
| * gordc thought struct was already | 16:38 | |
| gordc | C code. | 16:38 |
| lhx_ | gordc, what's your mean? :) | 16:40 |
| *** nicodemus_ has joined #openstack-telemetry | 16:44 | |
| lhx_ | good night, guys! | 16:46 |
| *** lhx_ has quit IRC | 16:46 | |
| *** david-lyle has quit IRC | 16:55 | |
| *** vint_bra has joined #openstack-telemetry | 16:55 | |
| *** yprokule has quit IRC | 16:55 | |
| *** Marx314 has quit IRC | 17:06 | |
| *** Marx314 has joined #openstack-telemetry | 17:06 | |
| *** links has quit IRC | 17:11 | |
| *** liusheng has quit IRC | 17:18 | |
| *** liusheng has joined #openstack-telemetry | 17:18 | |
| *** Jack_Iv has joined #openstack-telemetry | 17:22 | |
| *** dschultz has joined #openstack-telemetry | 17:24 | |
| *** nadya has quit IRC | 17:26 | |
| *** Jack_Iv has quit IRC | 17:27 | |
| *** nicodemus_ has quit IRC | 17:28 | |
| *** david-lyle has joined #openstack-telemetry | 17:28 | |
| *** shardy has quit IRC | 17:29 | |
| *** nicodemus_ has joined #openstack-telemetry | 17:40 | |
| *** jmlowe has quit IRC | 18:09 | |
| *** Jack_Iv has joined #openstack-telemetry | 18:17 | |
| *** Jack_Iv has quit IRC | 18:17 | |
| *** david-lyle has quit IRC | 18:18 | |
| *** catinthe_ has joined #openstack-telemetry | 18:18 | |
| *** catintheroof has quit IRC | 18:18 | |
| *** eglynn has quit IRC | 18:22 | |
| openstackgerrit | gordon chung proposed openstack/gnocchi: serialise: simplify array format https://review.openstack.org/422800 | 18:22 |
| *** eglynn has joined #openstack-telemetry | 18:23 | |
| *** eglynn is now known as eglynn-afk | 18:23 | |
| *** Jack_Iv has joined #openstack-telemetry | 18:28 | |
| *** Jack_Iv has quit IRC | 18:30 | |
| *** Jack_Iv has joined #openstack-telemetry | 18:31 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 18:31 | |
| *** Jack_Iv has quit IRC | 18:35 | |
| *** rbak has quit IRC | 18:46 | |
| *** Jack_Iv_ has quit IRC | 18:57 | |
| *** Jack_Iv has joined #openstack-telemetry | 18:58 | |
| *** Jack_Iv has quit IRC | 19:01 | |
| *** Jack_Iv has joined #openstack-telemetry | 19:03 | |
| *** nadya has joined #openstack-telemetry | 19:03 | |
| *** rbak has joined #openstack-telemetry | 19:03 | |
| *** chopmann has joined #openstack-telemetry | 19:07 | |
| *** Jack_Iv has quit IRC | 19:07 | |
| *** tesseract has quit IRC | 19:29 | |
| *** openstackgerrit has quit IRC | 19:33 | |
| *** Guest30600 has quit IRC | 19:47 | |
| *** david-lyle has joined #openstack-telemetry | 19:48 | |
| *** nadya has quit IRC | 20:03 | |
| *** dave-mccowan has quit IRC | 20:08 | |
| *** dave-mccowan has joined #openstack-telemetry | 20:09 | |
| *** efoley__ has quit IRC | 20:13 | |
| *** Jack_Iv has joined #openstack-telemetry | 20:14 | |
| *** darrenc_ is now known as darrenc | 20:15 | |
| *** Jack_Iv has quit IRC | 20:17 | |
| *** jmlowe has joined #openstack-telemetry | 20:29 | |
| *** openstackgerrit has joined #openstack-telemetry | 20:30 | |
| openstackgerrit | gordon chung proposed openstack/gnocchi: serialise: simplify array format https://review.openstack.org/422800 | 20:30 |
| *** pablo|500| has quit IRC | 20:57 | |
| *** jmlowe has quit IRC | 21:03 | |
| *** jmlowe has joined #openstack-telemetry | 21:07 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: Add benchmark for split() https://review.openstack.org/422496 | 21:20 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: add merge() benchmark https://review.openstack.org/422637 | 21:20 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: don't use groupby for split https://review.openstack.org/422445 | 21:20 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: use numpy for unserialization https://review.openstack.org/421901 | 21:20 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: carbonara: add bench for resample https://review.openstack.org/422877 | 21:20 |
| *** david-lyle has quit IRC | 21:35 | |
| *** nicodemus_ has quit IRC | 21:35 | |
| *** david-lyle has joined #openstack-telemetry | 21:39 | |
| *** david-lyle has quit IRC | 21:39 | |
| *** david-lyle has joined #openstack-telemetry | 21:39 | |
| *** eglynn-afk has quit IRC | 21:41 | |
| *** vint_bra has quit IRC | 21:43 | |
| *** vint_bra has joined #openstack-telemetry | 21:43 | |
| *** thorst_ has quit IRC | 21:45 | |
| *** jmlowe has quit IRC | 21:49 | |
| *** jmlowe has joined #openstack-telemetry | 21:51 | |
| *** fguillot has quit IRC | 21:56 | |
| *** vint_bra has quit IRC | 22:00 | |
| *** jmlowe has quit IRC | 22:03 | |
| *** dave-mccowan has quit IRC | 22:08 | |
| *** jmlowe has joined #openstack-telemetry | 22:10 | |
| *** thorst_ has joined #openstack-telemetry | 22:11 | |
| *** adriant has joined #openstack-telemetry | 22:12 | |
| *** thorst_ has quit IRC | 22:15 | |
| *** jmlowe has quit IRC | 22:25 | |
| *** jmlowe has joined #openstack-telemetry | 22:39 | |
| *** thorst_ has joined #openstack-telemetry | 22:40 | |
| *** thorst_ has quit IRC | 22:43 | |
| *** jmlowe has quit IRC | 22:51 | |
| *** david-lyle has quit IRC | 23:03 | |
| *** thorst_ has joined #openstack-telemetry | 23:44 | |
| *** chopmann has quit IRC | 23:46 | |
| *** chopmann has joined #openstack-telemetry | 23:46 | |
| *** thorst_ has quit IRC | 23:48 | |
| *** pradk has quit IRC | 23:49 | |
| *** rbak has quit IRC | 23:56 | |
| *** pradk has joined #openstack-telemetry | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!