*** yuntongjin has joined #openstack-telemetry | 00:17 | |
*** openstack has joined #openstack-telemetry | 00:26 | |
*** sheeprine has quit IRC | 00:30 | |
*** yuntongjin1 has joined #openstack-telemetry | 00:32 | |
*** sheeprine has joined #openstack-telemetry | 00:34 | |
*** yuntongjin has quit IRC | 00:34 | |
*** yuntongjin has joined #openstack-telemetry | 00:40 | |
*** yuntongjin1 has quit IRC | 00:42 | |
*** yuntongjin1 has joined #openstack-telemetry | 00:43 | |
*** yuntongjin has quit IRC | 00:45 | |
*** afazekas has quit IRC | 00:53 | |
*** dave-mccowan has joined #openstack-telemetry | 00:57 | |
*** afazekas has joined #openstack-telemetry | 00:59 | |
*** david-lyle has joined #openstack-telemetry | 01:06 | |
*** boris-42 has quit IRC | 01:24 | |
*** ljxiash has joined #openstack-telemetry | 01:25 | |
*** afazekas has quit IRC | 01:25 | |
*** afazekas has joined #openstack-telemetry | 01:28 | |
*** ljxiash has quit IRC | 01:35 | |
*** ljxiash has joined #openstack-telemetry | 01:36 | |
*** vishwanathj has joined #openstack-telemetry | 01:37 | |
*** ljxiash has quit IRC | 01:40 | |
*** mdih has joined #openstack-telemetry | 01:58 | |
*** mdih has quit IRC | 02:06 | |
*** ljxiash has joined #openstack-telemetry | 02:12 | |
*** mdih has joined #openstack-telemetry | 02:12 | |
*** mdih has quit IRC | 02:13 | |
*** mdih has joined #openstack-telemetry | 02:13 | |
*** liusheng has quit IRC | 02:16 | |
*** caisan has joined #openstack-telemetry | 02:16 | |
*** liusheng has joined #openstack-telemetry | 02:16 | |
*** thorst_ has quit IRC | 02:18 | |
*** mdih has quit IRC | 02:18 | |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add name support for show, update and delete https://review.openstack.org/292305 | 02:22 |
---|---|---|
openstackgerrit | liusheng proposed openstack/python-aodhclient: Add --filter to "alarm list" https://review.openstack.org/292233 | 02:26 |
*** mdih has joined #openstack-telemetry | 02:31 | |
*** liamji has joined #openstack-telemetry | 02:32 | |
openstackgerrit | liusheng proposed openstack/python-aodhclient: Add --filter to "alarm list" https://review.openstack.org/292233 | 02:49 |
*** dave-mccowan has quit IRC | 03:17 | |
*** thorst has joined #openstack-telemetry | 03:19 | |
*** thorst has quit IRC | 03:27 | |
*** yuntongjin1 has quit IRC | 03:28 | |
*** ljxiash has quit IRC | 03:38 | |
*** vishwanathj has quit IRC | 04:05 | |
*** links has joined #openstack-telemetry | 04:10 | |
*** flwang1 has quit IRC | 04:16 | |
*** thorst has joined #openstack-telemetry | 04:25 | |
*** thorst has quit IRC | 04:32 | |
*** ljxiash has joined #openstack-telemetry | 05:09 | |
*** zqfan has joined #openstack-telemetry | 05:10 | |
*** ljxiash has quit IRC | 05:10 | |
*** ljxiash has joined #openstack-telemetry | 05:10 | |
*** boris-42 has joined #openstack-telemetry | 05:13 | |
*** thorst has joined #openstack-telemetry | 05:30 | |
*** vishwanathj has joined #openstack-telemetry | 05:31 | |
*** vishwanathj has quit IRC | 05:31 | |
*** thorst has quit IRC | 05:37 | |
*** david-lyle has quit IRC | 05:45 | |
*** david-lyle has joined #openstack-telemetry | 05:46 | |
*** vishwanathj has joined #openstack-telemetry | 05:48 | |
*** vishwanathj has quit IRC | 05:51 | |
*** _nadya_ has joined #openstack-telemetry | 06:00 | |
*** flwang1 has joined #openstack-telemetry | 06:01 | |
*** _nadya_ has quit IRC | 06:01 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Imported Translations from Zanata https://review.openstack.org/292715 | 06:07 |
*** flwang1 has quit IRC | 06:09 | |
*** vishwanathj has joined #openstack-telemetry | 06:34 | |
*** vishwanathj is now known as vishwanathj_zzz | 06:35 | |
*** thorst has joined #openstack-telemetry | 06:35 | |
*** thorst has quit IRC | 06:42 | |
*** yprokule has joined #openstack-telemetry | 07:05 | |
*** flwang1 has joined #openstack-telemetry | 07:13 | |
*** tesseract has joined #openstack-telemetry | 07:16 | |
*** tesseract is now known as Guest64967 | 07:17 | |
*** _nadya_ has joined #openstack-telemetry | 07:19 | |
*** _nadya_ has quit IRC | 07:21 | |
*** Guest64967 has quit IRC | 07:21 | |
*** ljxiash has quit IRC | 07:30 | |
*** flwang1 has quit IRC | 07:30 | |
*** ljxiash has joined #openstack-telemetry | 07:31 | |
*** ljxiash has quit IRC | 07:32 | |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Add some resource types tests https://review.openstack.org/270419 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute number https://review.openstack.org/270091 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute bool https://review.openstack.org/270418 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute uuid https://review.openstack.org/270090 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Don't create Ceilometer resource types by default. https://review.openstack.org/270322 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move legacy Ceilometer resource into indexer. https://review.openstack.org/270266 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute string https://review.openstack.org/269888 | 07:35 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Retry to delete resource types tables on deadlock https://review.openstack.org/292776 | 07:35 |
*** ljxiash has joined #openstack-telemetry | 07:36 | |
*** tesseract- has joined #openstack-telemetry | 07:36 | |
*** thorst has joined #openstack-telemetry | 07:40 | |
*** chlong has quit IRC | 07:42 | |
*** thorst has quit IRC | 07:48 | |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/aodh: devstack: allow uwsgi deployments https://review.openstack.org/292245 | 07:49 |
openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: devstack: allow uwsgi deployments https://review.openstack.org/292077 | 07:50 |
*** jwcroppe has quit IRC | 07:57 | |
*** belmoreira has joined #openstack-telemetry | 07:58 | |
*** jwcroppe has joined #openstack-telemetry | 07:58 | |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 08:02 |
*** jwcroppe has quit IRC | 08:02 | |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 08:13 |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 08:18 |
*** permalac has joined #openstack-telemetry | 08:20 | |
*** kbyrne has joined #openstack-telemetry | 08:22 | |
*** cdent has joined #openstack-telemetry | 08:26 | |
*** ig0r_ has quit IRC | 08:39 | |
*** hparekh_ has quit IRC | 08:42 | |
*** hparekh has joined #openstack-telemetry | 08:44 | |
*** thorst has joined #openstack-telemetry | 08:46 | |
*** thorst has quit IRC | 08:52 | |
*** safchain has joined #openstack-telemetry | 08:54 | |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 08:55 |
*** ljxiash has quit IRC | 08:58 | |
*** ljxiash has joined #openstack-telemetry | 09:05 | |
*** nijaba has quit IRC | 09:05 | |
openstackgerrit | Zhenyu Zheng proposed openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 09:10 |
*** ljxiash has quit IRC | 09:10 | |
*** nijaba has joined #openstack-telemetry | 09:11 | |
*** nijaba has quit IRC | 09:11 | |
*** nijaba has joined #openstack-telemetry | 09:11 | |
*** ljxiash has joined #openstack-telemetry | 09:35 | |
*** yassine__ has joined #openstack-telemetry | 09:36 | |
*** _nadya_ has joined #openstack-telemetry | 09:37 | |
*** ljxiash has quit IRC | 09:40 | |
openstackgerrit | Merged openstack/ceilometer: change dns and trove notifications to declarative https://review.openstack.org/292136 | 09:42 |
*** thorst has joined #openstack-telemetry | 09:50 | |
*** openstackgerrit has quit IRC | 09:53 | |
*** openstackgerrit_ is now known as openstackgerrit | 09:53 | |
*** openstackgerrit has quit IRC | 09:53 | |
*** openstackgerrit_ has joined #openstack-telemetry | 09:54 | |
*** openstackgerrit_ is now known as openstackgerrit | 09:54 | |
*** openstackgerrit_ has joined #openstack-telemetry | 09:55 | |
*** openstackgerrit has quit IRC | 09:55 | |
*** openstackgerrit has joined #openstack-telemetry | 09:56 | |
*** thorst has quit IRC | 09:57 | |
*** lsmola has joined #openstack-telemetry | 10:05 | |
*** fwdit has joined #openstack-telemetry | 10:21 | |
*** caisan has quit IRC | 10:37 | |
*** ljxiash has joined #openstack-telemetry | 10:51 | |
*** thorst has joined #openstack-telemetry | 10:55 | |
*** ljxiash has quit IRC | 10:56 | |
*** pece has joined #openstack-telemetry | 10:59 | |
*** flwang1 has joined #openstack-telemetry | 10:59 | |
*** thorst has quit IRC | 11:02 | |
*** mdih has quit IRC | 11:12 | |
*** openstackgerrit has quit IRC | 11:18 | |
*** openstackgerrit has joined #openstack-telemetry | 11:19 | |
*** liamji has quit IRC | 11:27 | |
openstackgerrit | zhaozhilong proposed openstack/gnocchi: Add a decorate for noauth function https://review.openstack.org/293381 | 11:31 |
*** thorst has joined #openstack-telemetry | 11:35 | |
*** cdent has quit IRC | 11:43 | |
*** gordc has joined #openstack-telemetry | 11:44 | |
*** dave-mccowan has joined #openstack-telemetry | 11:52 | |
*** pece has quit IRC | 12:01 | |
*** pece has joined #openstack-telemetry | 12:09 | |
*** cdent has joined #openstack-telemetry | 12:10 | |
*** pece has quit IRC | 12:10 | |
jd__ | sileht: https://bugs.launchpad.net/gnocchi/+bug/1557724 if you did not notice | 12:12 |
openstack | Launchpad bug 1557724 in Gnocchi "gnocchi-metricd causes slow queries on ceph pool" [Medium,Triaged] | 12:12 |
*** pece has joined #openstack-telemetry | 12:13 | |
*** pcaruana has quit IRC | 12:28 | |
openstackgerrit | litong01 proposed openstack/ceilometer: Add http publisher https://review.openstack.org/290624 | 12:28 |
*** boris-42 has quit IRC | 12:34 | |
*** fwdit has quit IRC | 12:39 | |
*** fwdit has joined #openstack-telemetry | 12:39 | |
*** pradk has joined #openstack-telemetry | 12:51 | |
*** rcernin has joined #openstack-telemetry | 13:03 | |
openstackgerrit | Merged openstack/python-aodhclient: Don't support "alarm search" and mandatory --type in "alarm list" https://review.openstack.org/283958 | 13:07 |
openstackgerrit | Merged openstack/python-aodhclient: Add --query to alarm list https://review.openstack.org/291541 | 13:07 |
*** jwcroppe has joined #openstack-telemetry | 13:08 | |
*** peristeri has joined #openstack-telemetry | 13:08 | |
*** kro_sinus_hell is now known as krot_sinus_purga | 13:15 | |
*** krot_sinus_purga is now known as kro_sniffles | 13:15 | |
*** ljxiash has joined #openstack-telemetry | 13:22 | |
*** ig0r_ has joined #openstack-telemetry | 13:24 | |
*** _nadya_ has quit IRC | 13:25 | |
*** ljxiash has quit IRC | 13:26 | |
*** _nadya_ has joined #openstack-telemetry | 13:29 | |
openstackgerrit | zhangguoqing proposed openstack/ceilometer: Use assertIn for test https://review.openstack.org/293428 | 13:32 |
cdent | I'm surprised you didn't burn me out gordc | 13:32 |
cdent | that would have been irony at its finest | 13:32 |
gordc | cdent: i don't understand irony | 13:32 |
gordc | although i should've now that you mention it | 13:32 |
cdent | gordc: watch "reality bites", learn all about irony | 13:33 |
cdent | https://www.youtube.com/watch?v=7GM--22zOlw | 13:35 |
gordc | i can't remember if i've seen this. | 13:35 |
gordc | the title sounds like some 'jersey shore' type show | 13:36 |
gordc | i don't know the uk equivalent. | 13:36 |
cdent | here its geordi shore | 13:37 |
cdent | same things baically | 13:37 |
openstackgerrit | zhangguoqing proposed openstack/aodh: Use assertIn for test https://review.openstack.org/293439 | 13:40 |
*** shardy has joined #openstack-telemetry | 13:42 | |
*** zqfan has quit IRC | 13:52 | |
*** rbak has joined #openstack-telemetry | 13:56 | |
openstackgerrit | gordon chung proposed openstack/ceilometer: core status cleanup https://review.openstack.org/293448 | 14:03 |
gordc | p!hist | 14:03 |
gordc | p!hist | 14:03 |
gordc | cdent: i'm assuming what i just typed isn't private... | 14:04 |
cdent | I can see you saying p!hist | 14:04 |
cdent | but now the responses | 14:04 |
cdent | not | 14:04 |
gordc | cool cool | 14:04 |
*** mgagne has quit IRC | 14:07 | |
*** mgagne has joined #openstack-telemetry | 14:07 | |
*** alextricity25 has quit IRC | 14:07 | |
*** alextricity25 has joined #openstack-telemetry | 14:08 | |
*** KrishR has joined #openstack-telemetry | 14:13 | |
openstackgerrit | zhangguoqing proposed openstack/ceilometer: Use assertIn and assertNotIn https://review.openstack.org/293428 | 14:18 |
openstackgerrit | zhangguoqing proposed openstack/aodh: Use assertIn and assertNotIn for test https://review.openstack.org/293439 | 14:21 |
openstackgerrit | gordon chung proposed openstack/aodh: cleanup core list https://review.openstack.org/293463 | 14:25 |
*** nicodemus_ has joined #openstack-telemetry | 14:29 | |
*** _nadya_ has quit IRC | 14:41 | |
*** _nadya_ has joined #openstack-telemetry | 14:45 | |
nicodemus_ | hello gordc | 15:06 |
nicodemus_ | do you have a minute? | 15:06 |
*** david-lyle has quit IRC | 15:15 | |
*** david-lyle has joined #openstack-telemetry | 15:15 | |
*** purplerbot has quit IRC | 15:16 | |
*** purplerbot has joined #openstack-telemetry | 15:17 | |
*** david-lyle_ has joined #openstack-telemetry | 15:34 | |
*** david-lyle has quit IRC | 15:34 | |
*** vishwanathj_zzz is now known as vishwanathj | 15:38 | |
*** jroll has quit IRC | 15:41 | |
*** jroll has joined #openstack-telemetry | 15:41 | |
*** jroll has quit IRC | 15:41 | |
*** jroll has joined #openstack-telemetry | 15:41 | |
*** yprokule has quit IRC | 15:47 | |
*** cdent has quit IRC | 16:05 | |
*** belmoreira has quit IRC | 16:11 | |
*** david-lyle_ is now known as david-lyle | 16:20 | |
*** fwdit has quit IRC | 16:30 | |
*** cdent has joined #openstack-telemetry | 16:30 | |
*** ljxiash has joined #openstack-telemetry | 16:36 | |
*** links has quit IRC | 16:38 | |
gordc | nicodemus_: sorry, was moving stuff. whatsup? | 16:46 |
*** flwang1 has quit IRC | 16:49 | |
nicodemus_ | no problem gordc, wanted to ask you a quick question if you have a moment | 16:49 |
gordc | sure. | 16:51 |
gordc | nicodemus_: going to wait out this rain before going to lunch anyways | 16:51 |
nicodemus_ | if the gnocchi backlog is large, does it make sense for metricd to take longer to process measures? | 16:54 |
gordc | nicodemus_: i think it more depends no the granularity of your policies. but even then, it should cap out since we time split the measures. | 16:57 |
gordc | do you know which metric is taking a long time? or is it all of them taking longer? | 16:58 |
nicodemus_ | I have now a 100k backlog and, on average, all of them take longer compared to a 50k backlog | 17:00 |
gordc | hmm.. this is for ceph right? | 17:02 |
nicodemus_ | some take 300 secs, I'm guessing because of that worker waiting to acquire a rw lock on the measure object. But those are the least, the average for 100k backlog is around 10 secs | 17:02 |
nicodemus_ | yes, for a 22 osd, 4-node ceph cluster | 17:02 |
gordc | i see. if you enable the extremely verbose 'retrieve, write, expire' logs. can you check to see if anything sticks out as slow? | 17:04 |
*** _nadya_ has quit IRC | 17:07 | |
nicodemus_ | gordc, let me check | 17:10 |
nicodemus_ | gordc, there is one very slow retrieve | 17:16 |
nicodemus_ | http://paste.openstack.org/show/490761/ | 17:17 |
nicodemus_ | 4 lines before the end, one retrieve took 153 secs | 17:17 |
stevelle | the retrieve and store each are outliers there | 17:19 |
gordc | i see. i imagine it's probably that locking issue that you mentioned | 17:20 |
gordc | it doesn't seem to be consistent with other granularities of similar/larger size or related to the aggregation type | 17:21 |
nicodemus_ | here is another case: http://paste.openstack.org/show/490763/ | 17:22 |
nicodemus_ | where the delay is for a different object (granularity & aggregation method) | 17:23 |
nicodemus_ | these are extreme cases, not all metrics take that much time... but on average, I think the larger the backlog the longer it takes to process measures | 17:25 |
stevelle | one of those outliers will heavily skew not only the average, but the flow rate | 17:25 |
gordc | stevelle: are you folks still in planning/configuration stage for setting up testing? (i'm assuming you're working with llu to test some performance stuff) | 17:27 |
stevelle | gordc: yes, we were not going to do ceph for our first tests though | 17:27 |
gordc | stevelle: cool cool. that's fine. i was hoping to do some testing myself but lab is getting delayed constantly. | 17:28 |
gordc | nicodemus_: is this with 2 * 24 workers? | 17:28 |
stevelle | labs are always 2 months or more behind the plan. :(( | 17:28 |
nicodemus_ | gordc, yes. Plus those 2 are the 2 dedicated servers (instance metricd are slower) | 17:29 |
gordc | stevelle: yeah, unfortuantely. i'm at "or more" stage. | 17:30 |
gordc | nicodemus_: so you have 4 separate metricd services? | 17:31 |
nicodemus_ | gordc, in an effort not to get behing in measure processing, I have now 2 HW servers w/24 CPU cores and 2 instances w/16 vCPU | 17:32 |
nicodemus_ | all of them have 24 workers each | 17:32 |
nicodemus_ | those logs are with the 2 servers only | 17:32 |
gordc | nicodemus_: didn't it perform 'better' when you disabled the two instance metricd services yesterday? | 17:33 |
nicodemus_ | another thing I see is that metricd cannot scale horizontally beyond a certain point... adding more metricd workers causes overall measure processing to slow down | 17:33 |
nicodemus_ | yeah, that's why I left it with the 2 servers now | 17:33 |
stevelle | I don't know a lot yet but I would suspect that this is a ceph saturation, not metricd | 17:33 |
nicodemus_ | ceph does complain about slow queries: https://bugs.launchpad.net/gnocchi/+bug/1557724 | 17:34 |
openstack | Launchpad bug 1557724 in Gnocchi "gnocchi-metricd causes slow queries on ceph pool" [Medium,Triaged] | 17:34 |
gordc | stevelle: yeah, that's what i was thinking -- we're just exhaustnig the number of connections. | 17:35 |
stevelle | I'm not sure if it's connections or something else, but I don't know much about ceph really | 17:36 |
stevelle | what is the block_size being used here? | 17:36 |
gordc | yeah, sileht is the ceph expert among us but i don't believe he's online currently. | 17:36 |
gordc | nicodemus_: did you set a aggregation_workers_number value? | 17:37 |
nicodemus_ | gordc, yes, all metricd hosts have aggregation_workers_number = 2 | 17:40 |
nicodemus_ | as far as network goes, I can say there's no link saturation (installed a cacti a week ago to check that) | 17:41 |
gordc | oh. i'd maybe bump that up to the number of aggregates you're calculating. but that will probably lead to more parallel connections to ceph. | 17:41 |
gordc | maybe try dropping metricd workers and increasing aggregate workers (disclaimer: i'm just guessing right now) | 17:43 |
gordc | biab. going to grab some lunch. | 17:43 |
nicodemus_ | it's worth the try :) | 17:44 |
stevelle | nicodemus_: did you do similar tests with file storage earlier in Mitaka? | 17:44 |
*** _nadya_ has joined #openstack-telemetry | 17:45 | |
nicodemus_ | stevelle, only for a brief time | 17:45 |
nicodemus_ | not nearly as extensive as with ceph | 17:45 |
*** vishwanathj is now known as vishwanathj_chow | 17:45 | |
stevelle | nicodemus_: that's reasonable. did you publish any data from that publicly? | 17:45 |
stevelle | I'd like to include a review of that in our build spec | 17:46 |
nicodemus_ | I'd have to look, one test I remember doing was using file as backend, and setting file_basepath to a rbd volume | 17:48 |
*** rcernin has quit IRC | 17:53 | |
stevelle | nicodemus_: back to your current test, if the resource contention is as I suspect the size of the backlog won't make matters worse, beyond the factor of a coefficient, once you have all of the metricd workers in your pool processing blocks but the outliers are already 100x to 1000x which means that coefficient could be large | 17:54 |
*** safchain has quit IRC | 17:55 | |
nicodemus_ | stevelle, I see. I'm still looking for the 'file' log | 17:57 |
*** _nadya_ has quit IRC | 18:02 | |
nicodemus_ | One thing I remember, is that using 'file' backend there was a lock directory in the basepath | 18:02 |
nicodemus_ | and there were separate locks for different measure_ objects | 18:02 |
stevelle | right | 18:06 |
*** _nadya_ has joined #openstack-telemetry | 18:13 | |
*** tesseract- has quit IRC | 18:19 | |
*** gordc_ has joined #openstack-telemetry | 18:24 | |
*** cdent has quit IRC | 18:40 | |
*** _nadya_ has quit IRC | 18:45 | |
*** gordc_ has quit IRC | 18:53 | |
*** pece has quit IRC | 18:56 | |
*** vishwanathj_chow is now known as vishwanathj | 19:01 | |
*** harlowja has joined #openstack-telemetry | 19:04 | |
*** cdent has joined #openstack-telemetry | 19:13 | |
*** pece has joined #openstack-telemetry | 19:25 | |
*** kro_sniffles is now known as kro_mnckn_doc | 19:27 | |
openstackgerrit | gordon chung proposed openstack/ceilometer: add rc1 release notes https://review.openstack.org/293671 | 19:32 |
*** _nadya_ has joined #openstack-telemetry | 19:41 | |
*** harlowja has quit IRC | 19:46 | |
*** ig0r_ has quit IRC | 19:55 | |
*** _nadya_ has quit IRC | 20:09 | |
*** _nadya_ has joined #openstack-telemetry | 20:14 | |
*** _nadya_ has quit IRC | 20:15 | |
*** cdent has quit IRC | 20:15 | |
*** marcin12345 has joined #openstack-telemetry | 20:16 | |
marcin12345 | Has anyone have Ceilometer configured with Gnocchi that uses Swift as its storage? I cant make it work. | 20:17 |
gordc | marcin12345: what's the issue? | 20:22 |
gordc | it works for me on single node (although swiftclient has an annoying bug which makes it seem like it's exploding) | 20:23 |
*** diogogmt has joined #openstack-telemetry | 20:33 | |
nicodemus_ | marcin12345, this config worked for us: http://paste.openstack.org/show/490795/ | 20:45 |
nicodemus_ | stevelle, I can't find for the life of me the 'file' storage backend log | 20:57 |
stevelle | nicodemus_: thanks for looking | 20:57 |
gordc | stevelle: just curious, you helping to build the 1000node intel+rax cluster? | 20:58 |
nicodemus_ | if you need, next week I'll deploy a small gnocchi lab to play with, I could configure fie and gather some performance numbers | 20:59 |
stevelle | gordc: that's already built, but it's using the OSA deployment tooling | 20:59 |
*** safchain has joined #openstack-telemetry | 20:59 | |
gordc | stevelle: random qustion, you wouldn't know what RAID level is used on the racks would you? | 21:00 |
stevelle | gordc: if you like, I'll find out | 21:00 |
stevelle | nicodemus_: we can just figure it out again :) | 21:00 |
*** thorst has quit IRC | 21:01 | |
gordc | stevelle: if it's easy. | 21:01 |
gordc | nicodemus_: do you know what RAID level you have? | 21:01 |
stevelle | gordc: I'm pretty sure there is a doc that was passed around somewhere around here | 21:01 |
gordc | stevelle: yeah. i vaguely remember seeing RAID 10 as what RAX uses but it's possible i'm pulling that out of nowhere. | 21:02 |
*** thorst has joined #openstack-telemetry | 21:03 | |
nicodemus_ | gordc, I don't think I follow... you mean the RAID in our servers? | 21:03 |
gordc | nicodemus_: right. (unrelated to anything you've been asking) | 21:04 |
nicodemus_ | gordc, we use two SATADOM configured on RAID 1 | 21:05 |
gordc | nicodemus_: ok, cool cool. | 21:06 |
*** thorst has quit IRC | 21:07 | |
stevelle | gordc: found the answer, and unsurprisingly it's complicated :) | 21:18 |
gordc | stevelle: yay | 21:18 |
gordc | stevelle: i think i'm just going to roll some dice. | 21:19 |
*** kro_mnckn_doc has quit IRC | 21:19 | |
stevelle | obv swift doesn't need fancy RAID | 21:19 |
*** anteaya has quit IRC | 21:19 | |
stevelle | but the simple answer is that RAID10 is a good place to start for non-cinder/swift | 21:20 |
gordc | cool cool. that seems reasonable. might play around with that or RAID50. | 21:21 |
gordc | thanks for checking | 21:21 |
*** nicodemus_ has quit IRC | 21:22 | |
*** shardy has quit IRC | 21:28 | |
*** pradk has quit IRC | 21:32 | |
*** lsmola has quit IRC | 21:32 | |
*** lsmola has joined #openstack-telemetry | 21:45 | |
*** peristeri has quit IRC | 21:48 | |
*** permalac has quit IRC | 21:50 | |
*** gordc has quit IRC | 22:18 | |
*** r-mibu has quit IRC | 22:20 | |
*** r-mibu has joined #openstack-telemetry | 22:20 | |
*** KrishR has quit IRC | 22:29 | |
*** chlong has joined #openstack-telemetry | 22:32 | |
*** safchain has quit IRC | 22:36 | |
*** rbak has quit IRC | 22:54 | |
*** jwcroppe has quit IRC | 22:59 | |
*** jwcroppe has joined #openstack-telemetry | 22:59 | |
*** jwcroppe has quit IRC | 23:04 | |
*** vishwanathj has quit IRC | 23:06 | |
*** yassine__ has quit IRC | 23:08 | |
*** krotscheck has joined #openstack-telemetry | 23:11 | |
*** anteaya has joined #openstack-telemetry | 23:21 | |
*** diogogmt has quit IRC | 23:38 | |
*** chlong has quit IRC | 23:41 | |
*** pece has quit IRC | 23:44 | |
*** jwcroppe has joined #openstack-telemetry | 23:44 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!