| *** donghao has quit IRC | 00:02 | |
| *** ljxiash_ has joined #openstack-telemetry | 00:04 | |
| *** ljxiash has quit IRC | 00:08 | |
| *** dave-mccowan has quit IRC | 00:08 | |
| *** thorst has joined #openstack-telemetry | 00:12 | |
| *** thorst has quit IRC | 00:15 | |
| *** thorst has joined #openstack-telemetry | 00:16 | |
| *** thorst has quit IRC | 00:25 | |
| *** catintheroof has quit IRC | 00:25 | |
| *** Guest76526 is now known as ioni | 00:31 | |
| *** ioni is now known as Guest97258 | 00:31 | |
| *** ljxiash_ has quit IRC | 00:46 | |
| *** hfu has joined #openstack-telemetry | 00:52 | |
| *** diogogmt has quit IRC | 00:55 | |
| *** rcernin has quit IRC | 01:01 | |
| *** leitan has joined #openstack-telemetry | 01:06 | |
| *** ljxiash has joined #openstack-telemetry | 01:08 | |
| *** ljxiash has quit IRC | 01:08 | |
| *** thorst has joined #openstack-telemetry | 01:09 | |
| *** thorst has quit IRC | 01:09 | |
| *** thorst has joined #openstack-telemetry | 01:10 | |
| *** zhurong has joined #openstack-telemetry | 01:12 | |
| *** thorst has quit IRC | 01:14 | |
| *** thorst has joined #openstack-telemetry | 01:23 | |
| *** thorst has quit IRC | 01:24 | |
| *** zhurong has quit IRC | 01:25 | |
| *** zhurong has joined #openstack-telemetry | 01:25 | |
| *** leitan has quit IRC | 01:27 | |
| openstackgerrit | yaoyuan proposed openstack/ceilometer: filtered out the phynical nics when query vm nics with VMware API https://review.openstack.org/406807 | 01:31 |
|---|---|---|
| *** Guest97258 is now known as ioni | 01:32 | |
| *** ioni is now known as Guest62050 | 01:32 | |
| openstackgerrit | yaoyuan proposed openstack/ceilometer: filtered out the phynical nics when query vm nics with VMware API https://review.openstack.org/406807 | 01:39 |
| *** Jack_Iv has joined #openstack-telemetry | 01:40 | |
| *** Jack_Iv has quit IRC | 01:45 | |
| *** sam_nowitzki has joined #openstack-telemetry | 01:50 | |
| *** lhx_ has joined #openstack-telemetry | 01:51 | |
| sam_nowitzki | Hi, I'm using Gnocchi for storing Ceilometer metric, but some metrics Ceilometer collects but Gnocchi show no data, though I config it in gnocchi.resources.yaml. For instance: network.outgoing.bytes.rate | 01:57 |
| sam_nowitzki | my gnocchi_resources.yaml | 01:57 |
| sam_nowitzki | http://pastebin.com/icjJE7NK | 01:57 |
| *** thorst has joined #openstack-telemetry | 02:05 | |
| *** thorst has quit IRC | 02:05 | |
| *** thorst has joined #openstack-telemetry | 02:06 | |
| *** thorst has quit IRC | 02:15 | |
| *** hfu has quit IRC | 02:15 | |
| *** Guest62050 is now known as ioni | 02:32 | |
| *** ioni is now known as Guest1782 | 02:33 | |
| *** Ephur has quit IRC | 02:44 | |
| *** hfu has joined #openstack-telemetry | 02:44 | |
| *** thorst has joined #openstack-telemetry | 03:12 | |
| *** thorst has quit IRC | 03:20 | |
| *** hfu has quit IRC | 03:31 | |
| *** Guest1782 is now known as ioni | 03:33 | |
| *** ioni is now known as Guest37341 | 03:34 | |
| *** hfu has joined #openstack-telemetry | 03:36 | |
| *** hfu has quit IRC | 03:40 | |
| *** hfu has joined #openstack-telemetry | 03:40 | |
| *** hfu has quit IRC | 03:44 | |
| *** hfu has joined #openstack-telemetry | 03:45 | |
| *** hfu has quit IRC | 03:47 | |
| *** hfu has joined #openstack-telemetry | 03:50 | |
| *** links has joined #openstack-telemetry | 03:52 | |
| *** hfu has quit IRC | 03:54 | |
| *** masber has joined #openstack-telemetry | 04:07 | |
| *** hfu has joined #openstack-telemetry | 04:08 | |
| *** hfu has quit IRC | 04:15 | |
| *** thorst has joined #openstack-telemetry | 04:18 | |
| *** hfu has joined #openstack-telemetry | 04:20 | |
| *** thorst has quit IRC | 04:25 | |
| sorrison | Any gnocchi devs around? Have a question about datetime formats | 04:25 |
| *** Jack_Iv has joined #openstack-telemetry | 04:25 | |
| *** afazekas has quit IRC | 04:26 | |
| *** hfu has quit IRC | 04:27 | |
| sorrison | Basically in test_storage.py the tests pass in different types of datetime objects. Not sure why and if the storage driver should have to concern itself with datatime format | 04:27 |
| sorrison | sometimes they pass in pure datetime objects. And other times they pass in utils.dt_to_unix_ns(2014, 1, 1, 12, 0, 1) which is a mumpy.datetime64 object | 04:28 |
| *** Jack_Iv has quit IRC | 04:29 | |
| *** Guest37341 is now known as ioni | 04:34 | |
| *** ioni is now known as Guest93349 | 04:35 | |
| *** afazekas has joined #openstack-telemetry | 04:58 | |
| *** zhangguoqing has joined #openstack-telemetry | 05:07 | |
| *** anteaya has quit IRC | 05:16 | |
| *** thorst has joined #openstack-telemetry | 05:24 | |
| *** anteaya has joined #openstack-telemetry | 05:29 | |
| *** thorst has quit IRC | 05:30 | |
| *** sanchitmalhotra has joined #openstack-telemetry | 05:31 | |
| *** Guest93349 is now known as ioni | 05:35 | |
| *** ioni is now known as Guest85472 | 05:35 | |
| *** hfu has joined #openstack-telemetry | 05:37 | |
| *** hfu has quit IRC | 05:40 | |
| *** hfu has joined #openstack-telemetry | 05:42 | |
| *** sanchitmalhotra1 has joined #openstack-telemetry | 05:43 | |
| *** sanchitmalhotra has quit IRC | 05:44 | |
| *** sanchitmalhotra1 is now known as sanchitmalhotra | 05:44 | |
| *** hfu has quit IRC | 05:47 | |
| *** hfu has joined #openstack-telemetry | 05:47 | |
| *** adriant has quit IRC | 05:53 | |
| *** nadya has joined #openstack-telemetry | 05:55 | |
| *** Jack_Iv has joined #openstack-telemetry | 05:56 | |
| *** hfu has quit IRC | 06:06 | |
| *** nadya has quit IRC | 06:07 | |
| *** hfu has joined #openstack-telemetry | 06:15 | |
| *** Jack_Iv has quit IRC | 06:16 | |
| *** hfu has quit IRC | 06:27 | |
| *** thorst has joined #openstack-telemetry | 06:29 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:30 | |
| *** hfu has joined #openstack-telemetry | 06:32 | |
| *** rcernin has joined #openstack-telemetry | 06:34 | |
| *** thorst has quit IRC | 06:34 | |
| *** Guest85472 is now known as ioni | 06:36 | |
| *** hfu has quit IRC | 06:36 | |
| *** ioni is now known as Guest72550 | 06:36 | |
| *** sanchitmalhotra has quit IRC | 06:39 | |
| *** hfu has joined #openstack-telemetry | 06:46 | |
| *** hfu has quit IRC | 06:48 | |
| *** hfu has joined #openstack-telemetry | 06:53 | |
| *** Jack_Iv has quit IRC | 06:56 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:57 | |
| *** hfu has quit IRC | 06:57 | |
| *** hfu has joined #openstack-telemetry | 06:57 | |
| *** Jack_Iv has quit IRC | 07:03 | |
| openstackgerrit | Shengping Zhang proposed openstack/ceilometer: Allow rss collected as memory usage when polling cirros https://review.openstack.org/407354 | 07:10 |
| *** Jack_Iv has joined #openstack-telemetry | 07:11 | |
| *** rcernin has quit IRC | 07:12 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 07:17 | |
| *** g3ek has quit IRC | 07:20 | |
| *** Jack_Iv has quit IRC | 07:21 | |
| *** g3ek has joined #openstack-telemetry | 07:21 | |
| *** Guest66666 has quit IRC | 07:27 | |
| *** Guest66666 has joined #openstack-telemetry | 07:29 | |
| *** thorst has joined #openstack-telemetry | 07:32 | |
| *** yprokule has joined #openstack-telemetry | 07:33 | |
| *** rcernin has joined #openstack-telemetry | 07:34 | |
| *** Guest72550 is now known as ioni | 07:37 | |
| *** ioni is now known as Guest17332 | 07:37 | |
| *** thorst has quit IRC | 07:40 | |
| *** pcaruana has joined #openstack-telemetry | 07:42 | |
| *** hfu has quit IRC | 07:48 | |
| *** hfu has joined #openstack-telemetry | 07:49 | |
| *** hfu has quit IRC | 07:53 | |
| *** hfu has joined #openstack-telemetry | 07:53 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: fix oslo.db 4.15.0 breakage https://review.openstack.org/406885 | 07:59 |
| *** nadya has joined #openstack-telemetry | 08:05 | |
| openstackgerrit | Shengping Zhang proposed openstack/ceilometer: Allow rss collected as memory usage when polling cirros https://review.openstack.org/407354 | 08:15 |
| *** sanchitmalhotra has joined #openstack-telemetry | 08:16 | |
| *** ljxiash has joined #openstack-telemetry | 08:20 | |
| *** ljxiash has quit IRC | 08:24 | |
| *** Guest17332 is now known as ioni | 08:38 | |
| *** ioni is now known as Guest24621 | 08:38 | |
| *** thorst has joined #openstack-telemetry | 08:38 | |
| *** thorst has quit IRC | 08:44 | |
| *** zhangguoqing has quit IRC | 08:47 | |
| *** shardy has joined #openstack-telemetry | 09:11 | |
| *** larainema has quit IRC | 09:14 | |
| *** Jack_Iv_ has quit IRC | 09:14 | |
| *** larainema_ has joined #openstack-telemetry | 09:15 | |
| *** thorst has joined #openstack-telemetry | 09:24 | |
| *** thorst has quit IRC | 09:31 | |
| *** yassine has joined #openstack-telemetry | 09:34 | |
| *** yassine is now known as Guest17450 | 09:35 | |
| *** Guest24621 is now known as ioni | 09:38 | |
| *** ioni is now known as Guest6730 | 09:39 | |
| *** shardy is now known as shardy_mtg | 09:45 | |
| *** zhurong has quit IRC | 10:03 | |
| *** sam_nowitzki has quit IRC | 10:23 | |
| *** thorst has joined #openstack-telemetry | 10:28 | |
| *** cdent has joined #openstack-telemetry | 10:29 | |
| *** egonzalez90 has joined #openstack-telemetry | 10:31 | |
| *** egonzalez90 has left #openstack-telemetry | 10:32 | |
| *** thorst has quit IRC | 10:36 | |
| *** Guest6730 is now known as ioni | 10:39 | |
| *** egonzalez90 has joined #openstack-telemetry | 10:39 | |
| *** ioni is now known as Guest39022 | 10:40 | |
| *** Ephur has joined #openstack-telemetry | 10:47 | |
| *** Ephur has quit IRC | 10:52 | |
| *** hfu has quit IRC | 10:58 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 11:15 | |
| *** Jack_Iv_ has quit IRC | 11:19 | |
| *** shardy_mtg has quit IRC | 11:30 | |
| *** thorst has joined #openstack-telemetry | 11:34 | |
| *** Guest39022 is now known as ioni | 11:40 | |
| *** vint_bra has joined #openstack-telemetry | 11:41 | |
| *** thorst has quit IRC | 11:41 | |
| *** ioni is now known as Guest13109 | 11:41 | |
| lhx_ | sileht, is the measurements up-to-date? http://docs.openstack.org/admin-guide/telemetry-measurements.html | 11:44 |
| *** egonzalez90 has left #openstack-telemetry | 11:44 | |
| lhx_ | some placed marks 'Meters added in the Kilo release', so the measures are all Kilo release? | 11:46 |
| *** hfu has joined #openstack-telemetry | 12:11 | |
| *** nadya has quit IRC | 12:14 | |
| *** thorst has joined #openstack-telemetry | 12:30 | |
| *** nadya has joined #openstack-telemetry | 12:38 | |
| *** Guest13109 is now known as ioni | 12:41 | |
| *** ioni is now known as Guest12284 | 12:41 | |
| *** larainema_ has quit IRC | 12:46 | |
| *** larainema_ has joined #openstack-telemetry | 12:46 | |
| *** Kevin_Zheng has quit IRC | 12:46 | |
| *** Kevin_Zheng has joined #openstack-telemetry | 12:47 | |
| *** csatari has joined #openstack-telemetry | 12:48 | |
| csatari | Hi I'm looking for some help with Panko and devstack. Can someone help me to figure out if I need to add other projects to devstack then Panko? | 12:49 |
| *** nadya has quit IRC | 12:49 | |
| *** nadya has joined #openstack-telemetry | 12:51 | |
| *** lhx_ has quit IRC | 13:05 | |
| *** hfu has quit IRC | 13:07 | |
| *** dave-mccowan has joined #openstack-telemetry | 13:19 | |
| *** shardy has joined #openstack-telemetry | 13:22 | |
| *** pradk has joined #openstack-telemetry | 13:22 | |
| *** hfu has joined #openstack-telemetry | 13:23 | |
| *** hfu has quit IRC | 13:29 | |
| *** hfu has joined #openstack-telemetry | 13:29 | |
| *** lhx_ has joined #openstack-telemetry | 13:31 | |
| *** hfu has quit IRC | 13:38 | |
| *** fguillot has joined #openstack-telemetry | 13:41 | |
| *** Guest12284 is now known as ioni | 13:42 | |
| *** ioni is now known as Guest19475 | 13:42 | |
| *** catintheroof has joined #openstack-telemetry | 13:43 | |
| *** swamireddy has quit IRC | 13:43 | |
| *** hfu has joined #openstack-telemetry | 13:53 | |
| *** hfu has quit IRC | 13:58 | |
| *** swamireddy has joined #openstack-telemetry | 13:58 | |
| *** nadya has quit IRC | 13:59 | |
| *** nadya has joined #openstack-telemetry | 14:03 | |
| *** nadya has quit IRC | 14:03 | |
| *** nadya has joined #openstack-telemetry | 14:03 | |
| *** nadya has quit IRC | 14:04 | |
| *** gordc has joined #openstack-telemetry | 14:04 | |
| *** nadya has joined #openstack-telemetry | 14:06 | |
| *** hfu has joined #openstack-telemetry | 14:08 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: storage: split s3 driver https://review.openstack.org/402154 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: storage: split swift driver https://review.openstack.org/402153 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: storage: split ceph driver https://review.openstack.org/402033 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: storage: split file driver https://review.openstack.org/402032 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Introduce new storage groups for storage https://review.openstack.org/402320 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: storage: remove temporary incoming setup https://review.openstack.org/402321 | 14:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: metricd: move metricd options in metricd group https://review.openstack.org/402341 | 14:16 |
| *** swamireddy has quit IRC | 14:17 | |
| *** donghao has joined #openstack-telemetry | 14:24 | |
| *** hfu has quit IRC | 14:28 | |
| *** hfu has joined #openstack-telemetry | 14:29 | |
| *** swamireddy has joined #openstack-telemetry | 14:32 | |
| *** hfu has quit IRC | 14:35 | |
| *** hfu has joined #openstack-telemetry | 14:38 | |
| *** hfu has quit IRC | 14:40 | |
| *** links has quit IRC | 14:40 | |
| *** nadya has quit IRC | 14:41 | |
| *** Guest19475 is now known as ioni | 14:42 | |
| *** nadya has joined #openstack-telemetry | 14:43 | |
| *** ioni is now known as Guest47493 | 14:43 | |
| *** nadya has quit IRC | 14:46 | |
| *** nadya has joined #openstack-telemetry | 14:46 | |
| *** Ephur has joined #openstack-telemetry | 14:49 | |
| *** Ephur has quit IRC | 14:53 | |
| *** rbak has joined #openstack-telemetry | 14:56 | |
| lhx_ | gordc, should the measurements be up-to-date? | 15:03 |
| lhx_ | http://docs.openstack.org/admin-guide/telemetry-measurements.html | 15:03 |
| *** hfu has joined #openstack-telemetry | 15:04 | |
| gordc | lhx_: they aren't up to date. | 15:06 |
| gordc | you can update them if you have time | 15:06 |
| gordc | i think you need to delete a lot of things | 15:07 |
| lhx_ | gordc, some places mark 'Meters added in the Kilo release', so the measures are in Kilo release? | 15:07 |
| lhx_ | or measures were just added when kilo release? and we should update with the latest measures? | 15:09 |
| *** larainema_ is now known as larainema | 15:11 | |
| gordc | lhx_: it means they were added in Kilo (2nd option) | 15:13 |
| gordc | lhx_: but we dropped a lot of them | 15:13 |
| gordc | ie. no more instance meter | 15:13 |
| gordc | the magnetodb stuff is gone becuase that project doesn't exist anymore | 15:13 |
| gordc | the sahara (data processing) stuff is gone as well | 15:14 |
| gordc | i would just try to grep meter in code, if you can't find it, it's gone | 15:14 |
| lhx_ | gordc, some measures like IPMI based meters, how can I get the latest? | 15:16 |
| gordc | lhx_: https://github.com/openstack/ceilometer/tree/master/ceilometer/ipmi | 15:17 |
| gordc | sileht: you know why we require testresourcs now? | 15:19 |
| gordc | it an oslo.db thing? | 15:19 |
| sileht | gordc, because this is a extra dep of oslo.db | 15:19 |
| gordc | i see. so everything that uses it needs it now? | 15:19 |
| sileht | gordc, yes | 15:19 |
| gordc | kk, adding to aodh then | 15:20 |
| gordc | throwing error for me | 15:20 |
| sileht | gordc,when we depends on oslo.db>=4.14.0 we could install oslo.db with somthing like oslo.db[fixture,mysql] | 15:20 |
| sileht | gordc, and remove pymysql/pscyco2/testresources from our requirements | 15:21 |
| gordc | yeah, that seems cleaner. i'll let you do it :P | 15:21 |
| sileht | lol | 15:21 |
| *** jwcroppe has joined #openstack-telemetry | 15:21 | |
| sileht | raise oslo.db to 4.14 instead 4.8 will break newton compat | 15:22 |
| sileht | I dunno if we want that now | 15:22 |
| gordc | probably not. i guess we wait until next cycle | 15:22 |
| openstackgerrit | gordon chung proposed openstack/aodh: add testresources req https://review.openstack.org/407564 | 15:23 |
| *** dotplus has joined #openstack-telemetry | 15:25 | |
| dotplus | is openstack/gnocchi on-topic for this channel? or should I be elsewhere? | 15:30 |
| gordc | dotplus: yep. go for it. no promises on good advice though | 15:31 |
| openstackgerrit | gordon chung proposed openstack/aodh: add testresources req https://review.openstack.org/407564 | 15:34 |
| dotplus | I'm wondering whether there any public examples of using gnocchi? In particular, I'd love to see a demo or even just graphs based on gnocchi-stored TS data that show ... well, anything. Something that is actually useful. I'm considering a side project to help our OS eng folk get a better understanding of how to improve various performance metrics and I'd love to use an OS project if it makes sense instead more obvious choices like ... | 15:35 |
| dotplus | ... something built on influxdb or whatever. | 15:35 |
| gordc | dotplus: nebeliu has a few demos where they use cloudkitty+gnocchi for their rating system. they've done a few demos. | 15:37 |
| gordc | they have their own ui though. | 15:37 |
| jd__ | sileht: gordc: well we can keep Newton and 3.0 together and move on as far as I'm concern | 15:37 |
| gordc | i have it in our environment with grafana, but i don't have any public demos. | 15:38 |
| gordc | jd__: oh. i was thinking more about aodh | 15:38 |
| jd__ | gordc: hm so the answer is even more obvious now? we never had any compat' between different cycles of other openstack projects | 15:39 |
| jd__ | dotplus: did you check the website http://gnocchi.xyz/ as a first step? | 15:39 |
| gordc | jd__: i won't lie, i don't understand how this whole requirements stuff works in openstack with this false upper-constraints bs. | 15:39 |
| gordc | i'll just nod and agree | 15:40 |
| jd__ | gordc: works for me :p | 15:40 |
| sileht | gordc, it's simple, when a project want to stay in the past it blocks the whole openstack | 15:42 |
| *** Guest47493 is now known as ioni | 15:43 | |
| *** ioni is now known as Guest29621 | 15:44 | |
| gordc | sileht: lol | 15:44 |
| gordc | i just read that thread yesterday. | 15:45 |
| sileht | the fact is oslo.messaging can't work with the old version and monasca can't with the new one... | 15:46 |
| * lhx_ thinking about the issue what you are talking | 15:48 | |
| gordc | sileht: sucks how mutually exclusive projects must share same requirements | 15:49 |
| sileht | gordc, ceilometer already requires kafka>=1.0.0 but don't follow the requirements stuffs :) | 15:50 |
| *** nadya has quit IRC | 15:50 | |
| gordc | lhx_: you want to fix requirements? you will need a lot of time. | 15:51 |
| lhx_ | gordc, sorry, it isn't in my plan | 15:53 |
| gordc | :) | 15:54 |
| dotplus | gordc: ok I have the nubeliu demo queued. what sort of graphs do you have in grafana? traditional system-level stuff like load averages etc.? or OS-specific stuff like instance creation timing, instance/network/volume count over time? | 15:54 |
| *** Jack_Iv_ has joined #openstack-telemetry | 15:54 | |
| gordc | dotplus: if using just opnestack services, it's whatever measurements ceilometer generates. | 15:56 |
| lhx_ | gordc, these days I have done some preparing to use ceilometer in hardware, it costs much of my times | 15:57 |
| gordc | dotplus: my environment has a few more metrics... gnocchi just stores time-series and you can configure grafana to graph whatever you want. | 15:58 |
| openstackgerrit | gordon chung proposed openstack/aodh: Replace retrying with tenacity https://review.openstack.org/367600 | 15:58 |
| gordc | sileht: i blocked oslo.db in this patch too. https://review.openstack.org/#/c/407564/ | 15:59 |
| sileht | gordc, sure | 16:00 |
| openstackgerrit | gordon chung proposed openstack/aodh: Replace retrying with tenacity https://review.openstack.org/367600 | 16:01 |
| lhx_ | will this be remove? | 16:03 |
| lhx_ | https://github.com/openstack/ceilometer/blob/master/setup.cfg#L115 | 16:03 |
| gordc | lhx_: yes https://review.openstack.org/#/c/400973/ | 16:07 |
| lhx_ | gordc, all instance on compute and central node? | 16:08 |
| lhx_ | I see you just remove the central node instance | 16:09 |
| gordc | lhx_: the deuce... there's two?lol | 16:11 |
| gordc | lhx_: you should -1 my patch then. i'll remove it from both. | 16:11 |
| lhx_ | gordc, maybe I can propose one, lol | 16:12 |
| gordc | lhx_: sure. i think another patch i'm depending on polls from central. so maybe nothing is needed | 16:13 |
| gordc | my patch only shows one and master only shows one | 16:13 |
| gordc | brb | 16:13 |
| *** shardy has quit IRC | 16:14 | |
| *** catinthe_ has joined #openstack-telemetry | 16:14 | |
| *** yprokule has quit IRC | 16:16 | |
| *** catintheroof has quit IRC | 16:16 | |
| dotplus | gordc: ok. so you would say gnocchi/grafana might well be suitable for storing/visualizing metrics in general? For example, my immediate use case is Jenkins job build steps - I want to store how long various build steps take (and possibly even profiling each command within a buildstep that is a shell command) and how that changes over time. | 16:16 |
| *** rcernin has quit IRC | 16:17 | |
| *** catinthe_ has quit IRC | 16:18 | |
| dotplus | jd__: sorry, yes. I did look over the gnocchi docs briefly. Also, some of the material on your blog | 16:19 |
| *** pcaruana has quit IRC | 16:19 | |
| *** cdent has quit IRC | 16:22 | |
| *** cdent has joined #openstack-telemetry | 16:26 | |
| dotplus | jd__: perhaps you could comment on why you started gnocchi (as opposed to using an existing collector/tsdb)? to ease integration with ceilometer and OS more generally? for project org reasons such as foundation control instead of external? for some technical reason(s) why the alternatives were not suitable? All of the above or something else? | 16:32 |
| dotplus | (I think any one or combination of the above could be valid, I don't want to come across as combative here - I'm seeking to understand the thinking to help decide direction for *our* team's effort) | 16:33 |
| *** fguillot has quit IRC | 16:33 | |
| *** jmlowe has joined #openstack-telemetry | 16:35 | |
| jd__ | dotplus: there might be some context written on my slides https://julien.danjou.info/talks/storing-metrics-at-scale-with-gnocchi.pdf | 16:35 |
| dotplus | aha, I'll look there. thanks | 16:36 |
| jd__ | dotplus: no, that's good questions :) don't worry | 16:36 |
| jmlowe | gordc: Can you give me some hints on what to look at for this? http://paste.openstack.org/show/591563/ | 16:36 |
| *** Jack_Iv_ has quit IRC | 16:42 | |
| gordc | dotplus: yep. that's a good use case. basically it's designed to measure anything over time. our system we're using it to partially visualise load, and partially to do some pattern recognition stuff. | 16:43 |
| gordc | jmlowe: weird. you know what version you have running? | 16:44 |
| gordc | jd__: ^ not sure if you've seen this? seems it's not doing split properly. | 16:44 |
| *** Guest29621 is now known as ioni | 16:44 | |
| jmlowe | was 3.0.1 now 3.0.2 as of an hour ago | 16:44 |
| *** ioni is now known as Guest99990 | 16:45 | |
| jmlowe | pulled from cheeseshop | 16:45 |
| jd__ | gordc: what makes you say that? | 16:46 |
| gordc | NoneType' object has no attribute 'serialize' | 16:46 |
| *** nadya has joined #openstack-telemetry | 16:46 | |
| jmlowe | this ceph object strikes me as odd gnocchi_b53d78b8-6312-4eda-a28f-3b764f5e2421_none_v3, is that right? | 16:47 |
| jd__ | gordc: right… that is weird | 16:48 |
| jd__ | jmlowe: no that's right | 16:48 |
| jd__ | jmlowe: it's the unaggregated timeserie containing the current back window of points | 16:48 |
| jd__ | "none" is for unaggregated, obviously (sigh :p) | 16:49 |
| jmlowe | ok, is there one of those objects I should dump and examine? | 16:49 |
| gordc | we are very good at naming. | 16:49 |
| jd__ | gordc: yeah ts.split() returns None, wtf? | 16:49 |
| gordc | jmlowe: this is thrown after you upgraded to 3.0.2? | 16:50 |
| jd__ | right knowing it's a problem in 3.0.2 and not in 3.0.1 might help :p | 16:51 |
| gordc | did we require upgrade between 3.0.1 to 3.0.2? i feel like we might have | 16:51 |
| jmlowe | it was doing it on 3.0.1, saw there was some message packing changes in 3.0.2 so I went for it | 16:51 |
| gordc | oh ok. | 16:51 |
| jmlowe | what's the worst that could happen | 16:51 |
| jd__ | ok so the problem is in the rewrite code | 16:51 |
| jd__ | jmlowe: it's just not treating any new measure for that metric currently, until we fix the bug :p | 16:52 |
| jmlowe | that's usually the start of how I get into trouble | 16:52 |
| jd__ | (if it's a bug, though it looks like one) | 16:52 |
| jmlowe | is there anything I can get for you that would make it easier to diagnose and fix? | 16:53 |
| jd__ | jmlowe: can you send me a copy of the files? | 16:55 |
| gordc | just looking through code now. there a few scenarios which could trigger it it seems | 16:55 |
| jd__ | gordc: I'm trying to find those scenarios yeah, I'm not seeing it clearly yet :/ | 16:55 |
| jmlowe | all of the ones listed in the paste? | 16:55 |
| jd__ | jmlowe: yep | 16:55 |
| jmlowe | ok, I'll get cracking | 16:55 |
| jd__ | if we can find the precise case you're in it should be easy to fix | 16:55 |
| jd__ | likely the new measures are very far away of the latest measure or something like that | 16:56 |
| *** nadya has quit IRC | 16:56 | |
| jd__ | something that might be not tested in our test I imagine? | 16:56 |
| jd__ | (wild guess) | 16:56 |
| gordc | jmlowe: did you change what aggregations are calculated by any chance? (add a new aggregation?) | 16:58 |
| jd__ | gordc: this is very weird because the code says basically that there's a key like e.g. 1477440000.0 that is old and should be compressed, but then when it tries to get it from Ceph, it gets no data, so it gets None | 16:59 |
| jd__ | gordc: yeah that's a good question ^^ | 16:59 |
| jmlowe | I don't think I did | 16:59 |
| gordc | jd__: yeah, those seem like the two obvious places that could return None | 16:59 |
| jmlowe | so the history is ceilometer liberty -> mitaka, gnocchi 1.x->2.x->3.x | 17:00 |
| *** nadya has joined #openstack-telemetry | 17:01 | |
| jmlowe | the move from gnocchi 2.x to 3.x had some measures that had to be thrown away, the api got started before metricd finished conversion if I remember correctly | 17:01 |
| jd__ | wah, you mean the upgrade code actually worked, somehow? lol | 17:01 |
| jmlowe | more or less | 17:01 |
| gordc | jd__: apparently not.lol | 17:01 |
| jd__ | I'm impressed :) | 17:01 |
| jmlowe | sometimes less | 17:01 |
| jd__ | :D | 17:01 |
| jd__ | sorry to hear that | 17:02 |
| jd__ | it's too bad you did not report the bug(s) though | 17:02 |
| jmlowe | eh, it's not critical data | 17:02 |
| jd__ | jmlowe: can you give us the details of the archive policy used by metric b53d78b8-6312-4eda-a28f-3b764f5e2421 ? | 17:02 |
| jmlowe | I don't think I realized there was a bug, there was at least some degree of operator error | 17:02 |
| jmlowe | I have those files tar'ed up, how to you want them, ~4k | 17:03 |
| *** Ephur has joined #openstack-telemetry | 17:03 | |
| *** efoley has joined #openstack-telemetry | 17:03 | |
| gordc | jd__: there's std_300.0 missing | 17:03 |
| *** catintheroof has joined #openstack-telemetry | 17:03 | |
| *** catintheroof has quit IRC | 17:03 | |
| gordc | or not in paste | 17:04 |
| jmlowe | http://paste.openstack.org/show/591568/ | 17:04 |
| *** catintheroof has joined #openstack-telemetry | 17:04 | |
| jd__ | gordc: here you are eye of lynx | 17:04 |
| jd__ | jmlowe: so where is std_300.0? did you hide it somewhere? monster! | 17:05 |
| jd__ | ;) | 17:05 |
| jd__ | FREE STD_300.0 | 17:05 |
| jmlowe | I don't tolerate deviation | 17:05 |
| jd__ | haha | 17:05 |
| jd__ | even if it's a standard one? c'mon | 17:05 |
| jmlowe | rados get -p gnocchi gnocchi_b53d78b8-6312-4eda-a28f-3b764f5e2421_1478520000.0_std_300.0_v3 gnocchi_b53d78b8-6312-4eda-a28f-3b764f5e2421_1478520000.0_std_300.0_v3 | 17:06 |
| jmlowe | error getting gnocchi/gnocchi_b53d78b8-6312-4eda-a28f-3b764f5e2421_1478520000.0_std_300.0_v3: (2) No such file or directory | 17:06 |
| jmlowe | so it's really missing? | 17:06 |
| jd__ | yeah so that's the problem | 17:07 |
| jd__ | we need to log a message if that happens | 17:07 |
| jd__ | jmlowe: I imagine there was a problem at some stage? which one I could not say, upgrade maybe? | 17:07 |
| jd__ | sudden shutdown, data corruption, a random file eraser? who knows | 17:07 |
| gordc | i hope it's not because it's deleting too aggressively... although it doens't make sense the other aggreagtes are there. | 17:07 |
| jd__ | that should not be /fatal/ though, we could just ignore it and log an warning | 17:08 |
| *** lhx_ has quit IRC | 17:08 | |
| gordc | jmlowe: you can switch this to 'return' https://github.com/openstack/gnocchi/blob/master/gnocchi/storage/_carbonara.py#L241 that may be the quick hack | 17:09 |
| gordc | assuming it throws aggregationdoesnotexist and not some granularitydoesnotexist exception | 17:10 |
| jmlowe | so just to verify s/pass/return/ in line 241 | 17:11 |
| gordc | i think? lol that would've been my patch. jd__ ? | 17:11 |
| gordc | confidence inspiring. | 17:12 |
| jmlowe | http://paste.openstack.org/show/591569/ | 17:13 |
| jd__ | gordc: jmlowe: yes this is what I'm doing | 17:14 |
| jd__ | well you need an if split is None: return just before L249 to be exact | 17:14 |
| jmlowe | same, unless of course I should have whacked the pyc | 17:14 |
| jd__ | wait a sec I'm pushing a patch | 17:14 |
| *** nadya has quit IRC | 17:15 | |
| jd__ | jmlowe: gordc https://review.openstack.org/#/c/407614/ | 17:16 |
| gordc | jmlowe: hmm... maybe add a try/except around that and print out key_as_str, aggregation, archive_policy_def.granularity) | 17:17 |
| gordc | or that. | 17:17 |
| jmlowe | ok, firing up that patch | 17:18 |
| jd__ | I hope I did not make a formatting mistake in the LOG.warning lol | 17:19 |
| jd__ | that would be embarassing | 17:19 |
| jd__ | :P | 17:19 |
| gordc | jd__: no test. cannot be verified.lol | 17:20 |
| gordc | or jmlowe is the test | 17:20 |
| jd__ | jmlowe will be the test for now, then it should be easy to add a test | 17:21 |
| jd__ | just create a metric and delete a file :p | 17:21 |
| jd__ | bbl | 17:24 |
| jmlowe | I've got to go, be back in 2 hours or so | 17:25 |
| jmlowe | looks promising though http://paste.openstack.org/show/591571/ | 17:25 |
| jmlowe | like I might converge on zero eventually | 17:25 |
| gordc | jmlowe: :) cool cool. thanks for raising and verifying | 17:26 |
| jmlowe | np | 17:26 |
| jmlowe | thanks to both of you | 17:26 |
| *** rcernin has joined #openstack-telemetry | 17:26 | |
| jmlowe | and there we go, down to 1 metric 1 measure | 17:27 |
| *** jmlowe has quit IRC | 17:28 | |
| gordc | yay | 17:28 |
| *** diogogmt has joined #openstack-telemetry | 17:32 | |
| *** Guest99990 is now known as ioni | 17:45 | |
| *** ioni is now known as Guest97896 | 17:46 | |
| *** fguillot has joined #openstack-telemetry | 17:46 | |
| *** leitan has joined #openstack-telemetry | 17:47 | |
| *** Guest17450 has quit IRC | 18:01 | |
| *** jwcroppe has quit IRC | 18:03 | |
| jd__ | cool! | 18:12 |
| jd__ | dotplus: see, that's why you should use Gnocchi, an awesome support team fixing bugs :p | 18:12 |
| jd__ | so what's wrong with our integration tests? gordc, sileht ?http://logs.openstack.org/85/406885/5/check/gate-telemetry-dsvm-integration-gnocchi-ubuntu-xenial/a728f36/console.html | 18:14 |
| *** leitan has quit IRC | 18:16 | |
| *** rbak has quit IRC | 18:20 | |
| *** efoley has quit IRC | 18:23 | |
| gordc | jd__: https://review.openstack.org/#/c/407622/1 i think that | 18:26 |
| jd__ | ok, that's a good change at least | 18:27 |
| gordc | but no one in heat channel will confirm so i'm just going to run with it. | 18:27 |
| jd__ | I should yell at therve a bit for a change | 18:27 |
| gordc | go for it :) | 18:27 |
| gordc | disclaimer: 100% of my patches to project-config are guesses.lol | 18:28 |
| dotplus | jd__: very good, decision made. | 18:29 |
| jd__ | gordc: I feel you bro | 18:29 |
| *** jwcroppe has joined #openstack-telemetry | 18:32 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: carbonara: log a message and do not fail if a file is missing https://review.openstack.org/407614 | 18:38 |
| *** Guest97896 is now known as ioni | 18:46 | |
| *** ioni is now known as Guest42461 | 18:46 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 19:01 | |
| *** Jack_Iv_ has quit IRC | 19:03 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 19:04 | |
| *** Jack_Iv_ has quit IRC | 19:10 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 19:11 | |
| *** catinthe_ has joined #openstack-telemetry | 19:14 | |
| *** bapalm_ has quit IRC | 19:14 | |
| *** nadya has joined #openstack-telemetry | 19:15 | |
| *** Jack_Iv_ has quit IRC | 19:16 | |
| *** catintheroof has quit IRC | 19:16 | |
| *** bapalm has joined #openstack-telemetry | 19:18 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 19:20 | |
| *** Jack_Iv_ has quit IRC | 19:20 | |
| *** nadya has quit IRC | 19:22 | |
| *** AlexeyAbashkin has joined #openstack-telemetry | 19:22 | |
| *** AlexeyAbashkin has quit IRC | 19:23 | |
| *** nadya has joined #openstack-telemetry | 19:25 | |
| *** nadya has quit IRC | 19:26 | |
| *** rbak has joined #openstack-telemetry | 19:40 | |
| *** Guest42461 is now known as ioni | 19:47 | |
| *** ioni is now known as Guest6222 | 19:47 | |
| *** jmlowe has joined #openstack-telemetry | 19:50 | |
| *** jwcroppe has quit IRC | 19:54 | |
| *** rwsu has quit IRC | 20:02 | |
| *** catintheroof has joined #openstack-telemetry | 20:02 | |
| *** catinthe_ has quit IRC | 20:04 | |
| *** Marx314 has joined #openstack-telemetry | 20:05 | |
| *** jmlowe has quit IRC | 20:09 | |
| *** hfu has quit IRC | 20:12 | |
| *** rwsu has joined #openstack-telemetry | 20:22 | |
| *** adriant has joined #openstack-telemetry | 20:25 | |
| *** jwcroppe has joined #openstack-telemetry | 20:25 | |
| *** leitan has joined #openstack-telemetry | 20:34 | |
| *** leitan has quit IRC | 20:37 | |
| *** leitan_ has joined #openstack-telemetry | 20:37 | |
| *** fguillot has quit IRC | 20:47 | |
| *** Guest6222 is now known as ioni | 20:48 | |
| *** diogogmt has quit IRC | 20:48 | |
| *** ioni is now known as Guest27417 | 20:48 | |
| *** david-lyle has quit IRC | 20:55 | |
| *** catintheroof has quit IRC | 21:06 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 21:08 | |
| *** dave-mccowan has quit IRC | 21:10 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: add polling.yaml docs https://review.openstack.org/407711 | 21:13 |
| *** gordc has quit IRC | 21:16 | |
| *** leitan_ has quit IRC | 21:25 | |
| *** Jack_Iv_ has quit IRC | 21:29 | |
| *** chlong has joined #openstack-telemetry | 21:43 | |
| *** gordc has joined #openstack-telemetry | 21:47 | |
| *** Guest27417 is now known as ioni | 21:48 | |
| *** ioni is now known as Guest17320 | 21:49 | |
| sorrison | jd__ you around? Got a quick question about gnocchi and datetime formats | 21:53 |
| sorrison | or any other gnocchi devs? | 21:54 |
| gordc | sorrison: whatsup? | 21:54 |
| gordc | this might end up with me telling you i don't know and to ask on mailing list.lol | 21:55 |
| sorrison | In the gnocchi storage tests some tests pass in datetimes others pass in numpy.datetime64 | 21:55 |
| sorrison | Just seems strange that the storage driver needs to worry about converting datetimes | 21:55 |
| sorrison | writing a storage driver for influxdb and got it down to one failing test and wondering how to handle | 21:56 |
| sorrison | http://logs.openstack.org/60/390260/20/check/gate-gnocchi-tox-db-py27-mysql-ubuntu-xenial/6dcfcb9/console.html | 21:56 |
| gordc | you have some examples? iirc, we're storing neither technically. | 21:56 |
| sorrison | In gnocchi/tests/test_storage.py | 21:57 |
| gordc | kk, taking look at code now | 21:58 |
| sorrison | the date conversion/handling to the right format happens at a higher level I think it's just the tests that test the storage driver itself that seem to pass in different formats | 21:58 |
| sorrison | eg https://github.com/openstack/gnocchi/blob/master/gnocchi/tests/test_storage.py#L397 | 21:59 |
| gordc | yeah, it should be converting it to numpy.datetime64 | 21:59 |
| sorrison | https://github.com/openstack/gnocchi/blob/master/gnocchi/tests/test_storage.py#L572 | 21:59 |
| gordc | the api that is | 21:59 |
| sorrison | I'm just not sure what format I should be handling | 21:59 |
| *** Ephur has quit IRC | 22:01 | |
| gordc | the actual timestamps passed for storage should be nanosecond epoch time | 22:01 |
| gordc | the code you're failing at is i think the range timestamps | 22:02 |
| sorrison | the tests for get_measures pass in datetimes and the tests for get_cross_metric_measures pass in numpy.datetime64 | 22:03 |
| *** leitan has joined #openstack-telemetry | 22:03 | |
| gordc | hmm. so i don't think it matters in carbanara driver because it ends up getting normalised by pandas.Timestamp (which handles both) | 22:06 |
| gordc | for get_cross_metric_measures specifically | 22:06 |
| *** chlong has quit IRC | 22:06 | |
| gordc | seems like it's passing in datetime.datetime.utcfromtimestamp | 22:07 |
| *** chlong has joined #openstack-telemetry | 22:07 | |
| *** thorst has quit IRC | 22:09 | |
| *** thorst has joined #openstack-telemetry | 22:09 | |
| gordc | i'm pretty sure datetime is not the format api is passing so we can probably change tests... i just need to figure out what datetime.datetime.utcfromtimestamp | 22:12 |
| gordc | is | 22:12 |
| *** chlong has quit IRC | 22:12 | |
| gordc | oh. it's a datetime. nm then | 22:13 |
| gordc | datetime it is | 22:13 |
| *** jmlowe has joined #openstack-telemetry | 22:14 | |
| sorrison | ok so you think all the test_storage tests should pass in datetimes? | 22:15 |
| *** thorst has quit IRC | 22:18 | |
| gordc | sorrison: yeah, it seems like that's what is being passed in by api. | 22:18 |
| gordc | they all validate using utils.to_datetime which eventually just return https://github.com/openstack/gnocchi/blob/411626f871984f0f0776f6a4fd37bb3a47c05a62/gnocchi/utils.py#L121-L127 | 22:19 |
| sorrison | ok cool, thanks gordc I'll have a play changing that and see how it goes | 22:19 |
| gordc | i'll push a patch to switch to datetime. won't merge since our gate is sort of blocked right now | 22:20 |
| gordc | or you can push a patch. | 22:20 |
| gordc | if you have it already | 22:20 |
| sorrison | na I haven't done anything | 22:20 |
| gordc | kk. i'll do it now and give it a test. seems fine after i changed one | 22:20 |
| *** vint_bra has quit IRC | 22:22 | |
| *** david-lyle has joined #openstack-telemetry | 22:24 | |
| sorrison | Great thanks, I'll rebase my change once its in gerrit | 22:24 |
| *** mfranc213_ has joined #openstack-telemetry | 22:29 | |
| *** dave-mccowan has joined #openstack-telemetry | 22:32 | |
| mfranc213_ | hello. i am trying to fine where ceilometer sets the volume for a cpu_util request for libvirt. i understand that the rate_of_change transformation actually produces the % based on current and previous cpu_times, but i cannot see where s.volume is set in this case. would someone be able to assist? | 22:32 |
| openstackgerrit | gordon chung proposed openstack/gnocchi: use datetime when defining series range https://review.openstack.org/407737 | 22:32 |
| mfranc213_ | s/fine/find :) | 22:32 |
| *** dave-mcc_ has joined #openstack-telemetry | 22:33 | |
| gordc | sorrison: let me know if that works for you ^ | 22:34 |
| gordc | mfranc213_: it's done here https://github.com/openstack/ceilometer/blob/master/ceilometer/transformer/conversions.py#L208-L211 | 22:35 |
| *** dave-mccowan has quit IRC | 22:37 | |
| mfranc213_ | gordc: thank you. i guess my question is: where is s.volume set when it's used, for example, here: https://github.com/openstack/ceilometer/blob/master/ceilometer/transformer/conversions.py#L204 | 22:37 |
| *** thorst_ has joined #openstack-telemetry | 22:38 | |
| *** thorst_ has quit IRC | 22:38 | |
| gordc | mfranc213_: when it's used? | 22:39 |
| sorrison | so far so good gordc | 22:39 |
| mfranc213_ | gordc: sorry that i'm not being clear. L204 shows that s.volume is already set. where was it set? | 22:40 |
| gordc | basically it calculates rate of change and then if you follow _convert() part, you should see it starts doing stuff there. | 22:40 |
| gordc | line 204 is just extracting the volume from current sample (which it takes from cpu sample) | 22:41 |
| *** david-lyle has quit IRC | 22:41 | |
| mfranc213_ | gordc: the rate of change based on difference fetches of cpu_time. where does that fetch happen? | 22:41 |
| gordc | mfranc213_: oh. there's a compute agent which polls hypervisor for cputime | 22:41 |
| mfranc213_ | gordc: i looked at sample.Sample and elsewhere (and everywhere) but not in the right place :( | 22:41 |
| mfranc213_ | gordc: i looked in the agent code also. | 22:42 |
| gordc | sorrison: good to hear | 22:42 |
| mfranc213_ | gordc: i'm probably just being obtuse. | 22:42 |
| gordc | mfranc213_: no comment :P j/k | 22:42 |
| gordc | it starts here: https://github.com/openstack/ceilometer/blob/master/ceilometer/compute/pollsters/cpu.py#L36-L49 | 22:43 |
| mfranc213_ | gordc: looking ... | 22:43 |
| gordc | that will run on compute agent and generate cputime samples. | 22:43 |
| *** jmlowe has quit IRC | 22:43 | |
| *** harlowja has quit IRC | 22:43 | |
| gordc | that will push onto a queue and get picked up by notificaiton agent which does the transformation into cpu_util | 22:43 |
| *** harlowja has joined #openstack-telemetry | 22:43 | |
| sorrison | gordc wrt gate brokenness, are you talking about the oslo.db 4.15.0 breakage? | 22:44 |
| mfranc213_ | gordc: ah yes, but that's CPUPollster, and setup.cfg sends the caller to CPUUtilPollster, whose get_samples calles inspect_cpu_util, which is not implemented for libvirt. | 22:44 |
| mfranc213_ | gordc: you have having at least 2 conversations simultaneously. this is very impressive! :) | 22:45 |
| gordc | sorrison: that... and heat also switched to devstack plugins recently so we need to upgrade our gate to explicitly enable heat https://review.openstack.org/#/c/407622/ | 22:45 |
| gordc | then we can merge the oslo.db stuff | 22:45 |
| sorrison | ah ok I see, lovely little tangle | 22:46 |
| gordc | mfranc213_: cpuutilpollster will not generate anything if on libvirt since it doesn't provide that functionality | 22:46 |
| openstackgerrit | Sam Morrison proposed openstack/gnocchi: WIP Resurrect influxdb backend driver https://review.openstack.org/390260 | 22:47 |
| gordc | sorrison: i'm hoping that's all the issues for blocked gate. | 22:47 |
| mfranc213_ | gordc: yes, so how does a call to cpu_util end up producing the desired result in the case of libvirt? i understand what you said above about the notification agent but don't see the tie from the cpu_util request to the result. i think this will be my last question since i've hit my allotment. | 22:48 |
| *** Guest17320 is now known as ioni | 22:49 | |
| *** ioni is now known as Guest1890 | 22:50 | |
| *** cdent has quit IRC | 22:53 | |
| gordc | mfranc213_: no worries. in the libvirt path, there is no call to cpu_util. it generates cputime | 22:54 |
| gordc | if you specify cpu_util transformation in pieline.yaml (which it does by default), it will end up deriving cpu_util samples as well | 22:55 |
| mfranc213_ | gordc: thank you very much. i'm going to go away and think about what you said, and go exploring again. i really appreciate your help. | 22:56 |
| gordc | np | 22:58 |
| *** rcernin has quit IRC | 23:02 | |
| sorrison | gordc you mind if I rebase your change ontop of "fix oslo.db 4.15.0 breakage" change? | 23:03 |
| gordc | sorrison: sure. go ahead. | 23:08 |
| *** dave-mcc_ has quit IRC | 23:08 | |
| *** leitan_ has joined #openstack-telemetry | 23:16 | |
| *** leitan has quit IRC | 23:18 | |
| sorrison | hmm gerrit won't let me do that for some reason | 23:21 |
| openstackgerrit | Sam Morrison proposed openstack/gnocchi: use datetime when defining series range https://review.openstack.org/407737 | 23:21 |
| sorrison | web interface lets me | 23:21 |
| openstackgerrit | Sam Morrison proposed openstack/gnocchi: WIP Resurrect influxdb backend driver https://review.openstack.org/390260 | 23:22 |
| openstackgerrit | gordon chung proposed openstack/aodh: Replace retrying with tenacity https://review.openstack.org/367600 | 23:23 |
| *** jwcroppe has quit IRC | 23:26 | |
| gordc | sorrison: usually it's other way around. | 23:26 |
| gordc | heading off. ttyl | 23:26 |
| *** gordc has quit IRC | 23:26 | |
| *** jwcroppe has joined #openstack-telemetry | 23:26 | |
| *** Jack_Iv has joined #openstack-telemetry | 23:29 | |
| *** jwcroppe has quit IRC | 23:31 | |
| *** Jack_Iv has quit IRC | 23:34 | |
| *** masber has quit IRC | 23:37 | |
| *** thorst_ has joined #openstack-telemetry | 23:39 | |
| *** jmlowe has joined #openstack-telemetry | 23:41 | |
| *** rbak has quit IRC | 23:45 | |
| jmlowe | gordc: the ceilometer compute agent generates the thing I'm interested in storing in gnocchi, hypervisor cpu load, for some reason I have to report this number to the NSF | 23:45 |
| *** rbak has joined #openstack-telemetry | 23:45 | |
| *** rbak has quit IRC | 23:46 | |
| *** thorst_ has quit IRC | 23:48 | |
| *** Guest1890 is now known as ioni | 23:50 | |
| *** ioni is now known as Guest81706 | 23:51 | |
| *** lhx_ has joined #openstack-telemetry | 23:53 | |
| *** jmlowe has quit IRC | 23:56 | |
| *** catintheroof has joined #openstack-telemetry | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!