| *** rcernin has joined #openstack-telemetry | 00:00 | |
| *** harlowja has quit IRC | 00:04 | |
| *** thorst_ has joined #openstack-telemetry | 00:45 | |
| *** hfu has joined #openstack-telemetry | 00:47 | |
| *** hfu has quit IRC | 00:49 | |
| *** rcernin has quit IRC | 00:51 | |
| *** rcernin has joined #openstack-telemetry | 00:54 | |
| *** thorst_ has quit IRC | 00:55 | |
| *** rcernin has quit IRC | 00:55 | |
| *** rcernin has joined #openstack-telemetry | 00:55 | |
| *** leitan has joined #openstack-telemetry | 01:22 | |
| *** leitan has quit IRC | 01:27 | |
| *** thorst_ has joined #openstack-telemetry | 01:27 | |
| *** hfu has joined #openstack-telemetry | 02:05 | |
| *** dims_ has joined #openstack-telemetry | 02:14 | |
| *** dims has quit IRC | 02:14 | |
| *** thorst_ has quit IRC | 02:18 | |
| *** thorst_ has joined #openstack-telemetry | 02:20 | |
| *** dims_ has quit IRC | 02:47 | |
| *** dims has joined #openstack-telemetry | 02:50 | |
| *** Jack_Iv has joined #openstack-telemetry | 03:07 | |
| *** Jack_Iv has quit IRC | 03:11 | |
| *** leitan has joined #openstack-telemetry | 03:23 | |
| *** leitan has quit IRC | 03:28 | |
| *** yarkot has joined #openstack-telemetry | 03:31 | |
| *** thorst_ has quit IRC | 03:40 | |
| *** r-mibu has quit IRC | 05:03 | |
| *** r-mibu has joined #openstack-telemetry | 05:09 | |
| *** nadya has joined #openstack-telemetry | 05:20 | |
| *** leitan has joined #openstack-telemetry | 05:25 | |
| *** leitan has quit IRC | 05:29 | |
| *** Jack_Iv has joined #openstack-telemetry | 05:33 | |
| *** david-lyle has quit IRC | 05:46 | |
| *** thorst_ has joined #openstack-telemetry | 05:56 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 05:59 | |
| openstackgerrit | aditi sharma proposed openstack/ceilometermiddleware: Remove link to modindex https://review.openstack.org/426074 | 05:59 |
|---|---|---|
| *** Jack_Iv_ has quit IRC | 06:01 | |
| *** thorst_ has quit IRC | 06:01 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 06:01 | |
| *** Jack_Iv has quit IRC | 06:01 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:02 | |
| *** nadya has quit IRC | 06:03 | |
| *** Jack_Iv_ has quit IRC | 06:06 | |
| *** Jack_Iv has quit IRC | 06:09 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:10 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/ceilometer: Switch to decorators.idempotent_id https://review.openstack.org/426084 | 06:12 |
| openstackgerrit | Ghanshyam Mann proposed openstack/aodh: Switch to decorators.idempotent_id https://review.openstack.org/426085 | 06:12 |
| *** donghao has joined #openstack-telemetry | 06:41 | |
| *** Jack_Iv has quit IRC | 06:44 | |
| *** Jack_Iv has joined #openstack-telemetry | 06:48 | |
| *** Jack_Iv has quit IRC | 06:53 | |
| *** hfu has quit IRC | 07:09 | |
| *** adriant has quit IRC | 07:11 | |
| *** hfu has joined #openstack-telemetry | 07:17 | |
| *** rcernin has quit IRC | 07:24 | |
| *** leitan has joined #openstack-telemetry | 07:26 | |
| *** donghao has quit IRC | 07:26 | |
| *** leitan has quit IRC | 07:30 | |
| *** donghao has joined #openstack-telemetry | 07:56 | |
| *** donghao has quit IRC | 07:56 | |
| *** thorst_ has joined #openstack-telemetry | 07:57 | |
| *** nadya has joined #openstack-telemetry | 07:59 | |
| *** thorst_ has quit IRC | 08:01 | |
| *** donghao has joined #openstack-telemetry | 08:27 | |
| *** donghao has quit IRC | 08:31 | |
| *** elemoine has quit IRC | 08:31 | |
| *** shardy_afk is now known as shardy | 08:52 | |
| *** efoley has joined #openstack-telemetry | 09:22 | |
| *** leitan has joined #openstack-telemetry | 09:27 | |
| *** donghao has joined #openstack-telemetry | 09:28 | |
| *** leitan has quit IRC | 09:32 | |
| *** nadya has quit IRC | 09:32 | |
| *** donghao has quit IRC | 09:33 | |
| *** nadya has joined #openstack-telemetry | 09:38 | |
| *** hfu has quit IRC | 09:47 | |
| *** hfu has joined #openstack-telemetry | 09:47 | |
| *** hfu has quit IRC | 09:48 | |
| *** hfu has joined #openstack-telemetry | 09:48 | |
| *** hfu has quit IRC | 09:49 | |
| *** hfu has joined #openstack-telemetry | 09:49 | |
| *** hfu has quit IRC | 09:49 | |
| *** hfu has joined #openstack-telemetry | 09:50 | |
| *** hfu has quit IRC | 09:50 | |
| *** hfu has joined #openstack-telemetry | 09:51 | |
| *** hfu has quit IRC | 09:52 | |
| *** thorst_ has joined #openstack-telemetry | 09:58 | |
| *** nadya has quit IRC | 09:59 | |
| *** thorst_ has quit IRC | 10:02 | |
| *** cdent has joined #openstack-telemetry | 10:23 | |
| *** shardy has quit IRC | 10:45 | |
| *** shardy has joined #openstack-telemetry | 10:46 | |
| *** vint_bra has joined #openstack-telemetry | 11:08 | |
| *** vint_bra has quit IRC | 11:12 | |
| *** hfu has joined #openstack-telemetry | 11:13 | |
| *** Jack_Iv has joined #openstack-telemetry | 11:17 | |
| *** Jack_Iv has quit IRC | 11:18 | |
| *** Jack_Iv has joined #openstack-telemetry | 11:19 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 11:21 | |
| *** Jack_Iv_ has quit IRC | 11:22 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 11:22 | |
| *** Jack_Iv_ has quit IRC | 11:24 | |
| *** chlong has joined #openstack-telemetry | 11:24 | |
| *** Jack_Iv has quit IRC | 11:24 | |
| *** nadya has joined #openstack-telemetry | 11:26 | |
| *** leitan has joined #openstack-telemetry | 11:28 | |
| *** donghao has joined #openstack-telemetry | 11:29 | |
| *** sanchitmalhotra has quit IRC | 11:29 | |
| *** sanchitmalhotra has joined #openstack-telemetry | 11:29 | |
| *** yassine has joined #openstack-telemetry | 11:31 | |
| *** yassine is now known as Guest21909 | 11:32 | |
| *** leitan has quit IRC | 11:33 | |
| *** donghao has quit IRC | 11:33 | |
| *** Jack_Iv has joined #openstack-telemetry | 11:36 | |
| *** thorst_ has joined #openstack-telemetry | 11:59 | |
| *** Jack_Iv has quit IRC | 12:00 | |
| *** Jack_Iv has joined #openstack-telemetry | 12:00 | |
| *** leitan has joined #openstack-telemetry | 12:02 | |
| *** Jack_Iv_ has joined #openstack-telemetry | 12:02 | |
| *** thorst_ has quit IRC | 12:03 | |
| *** catintheroof has joined #openstack-telemetry | 12:04 | |
| *** khushbu has joined #openstack-telemetry | 12:05 | |
| *** Jack_Iv_ has quit IRC | 12:05 | |
| *** Jack_Iv has quit IRC | 12:06 | |
| *** shardy is now known as shardy_lunch | 12:18 | |
| *** vint_bra has joined #openstack-telemetry | 12:30 | |
| *** thorst__ has joined #openstack-telemetry | 12:43 | |
| *** cdent has quit IRC | 13:04 | |
| *** cdent has joined #openstack-telemetry | 13:04 | |
| *** catintheroof has quit IRC | 13:05 | |
| *** gordc has joined #openstack-telemetry | 13:11 | |
| *** khushbu has quit IRC | 13:13 | |
| *** nadya has quit IRC | 13:14 | |
| *** Jack_Iv has joined #openstack-telemetry | 13:14 | |
| *** catintheroof has joined #openstack-telemetry | 13:16 | |
| *** shardy_lunch is now known as shardy | 13:17 | |
| *** Jack_Iv has quit IRC | 13:19 | |
| *** pradk has joined #openstack-telemetry | 13:21 | |
| *** pradk has quit IRC | 13:31 | |
| *** khushbu_ has joined #openstack-telemetry | 13:37 | |
| *** thorst__ is now known as thorst_ | 13:42 | |
| *** rbak_ has joined #openstack-telemetry | 13:50 | |
| *** pradk has joined #openstack-telemetry | 14:05 | |
| *** Jack_Iv has joined #openstack-telemetry | 14:08 | |
| *** nadya has joined #openstack-telemetry | 14:10 | |
| *** Jack_Iv has quit IRC | 14:12 | |
| *** nadya has quit IRC | 14:23 | |
| *** nadya has joined #openstack-telemetry | 14:28 | |
| *** fguillot has joined #openstack-telemetry | 14:31 | |
| openstackgerrit | gordon chung proposed openstack/gnocchi: fix live gabbi test https://review.openstack.org/425962 | 14:32 |
| *** sanfern has joined #openstack-telemetry | 14:34 | |
| *** efoley has quit IRC | 14:37 | |
| *** efoley has joined #openstack-telemetry | 14:38 | |
| *** efoley has quit IRC | 14:39 | |
| *** tlian has joined #openstack-telemetry | 14:39 | |
| *** efoley has joined #openstack-telemetry | 14:39 | |
| *** khushbu_ has quit IRC | 14:59 | |
| *** efoley_ has joined #openstack-telemetry | 15:05 | |
| *** efoley has quit IRC | 15:09 | |
| jd__ | pradk: u had a fix for that in tripleo right? https://bugs.launchpad.net/aodh/+bug/1659867 | 15:24 |
| openstack | Launchpad bug 1659867 in Aodh "aodh-dbsync fails on missing table 'ceilometer.alarm_change'" [Undecided,New] | 15:24 |
| jd__ | pradk: should we reassign? | 15:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: string → UUID conversion for resource.id to be unique per user https://review.openstack.org/413017 | 15:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: make sure original_resource_id is never NULL https://review.openstack.org/425823 | 15:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: tests: factorize measure generation to speed up test https://review.openstack.org/426268 | 15:25 |
| pradk | jd__, well yea ths fix is really to move aodh to its own db , so i fixed from tripleo side. you can close this | 15:25 |
| jd__ | pradk: k | 15:26 |
| *** Jack_Iv has joined #openstack-telemetry | 15:27 | |
| *** Jack_Iv has quit IRC | 15:27 | |
| *** Jack_Iv has joined #openstack-telemetry | 15:28 | |
| *** eglynn has joined #openstack-telemetry | 15:29 | |
| *** Jack_Iv has quit IRC | 15:32 | |
| *** nadya has quit IRC | 15:37 | |
| *** Guest21909 has quit IRC | 15:43 | |
| gordc | shame. looks like the devstack patch is going to fail again. time to nap | 15:55 |
| *** Guest21909 has joined #openstack-telemetry | 15:58 | |
| dims | gordc : CI needs the weekend to catch up :( | 15:58 |
| gordc | dims: the weekend: the secret to merging on the last week of the cycle. :) | 16:04 |
| *** efoley__ has joined #openstack-telemetry | 16:06 | |
| *** Jack_Iv has joined #openstack-telemetry | 16:06 | |
| *** Jack_Iv has quit IRC | 16:07 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: string → UUID conversion for resource.id to be unique per user https://review.openstack.org/413017 | 16:08 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: make sure original_resource_id is never NULL https://review.openstack.org/425823 | 16:08 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: tests: do not create a resource with a custom resource type in 2.2 https://review.openstack.org/426289 | 16:08 |
| *** Jack_Iv has joined #openstack-telemetry | 16:08 | |
| *** efoley_ has quit IRC | 16:09 | |
| *** donghao has joined #openstack-telemetry | 16:11 | |
| *** efoley__ has quit IRC | 16:12 | |
| *** donghao has quit IRC | 16:13 | |
| *** Jack_Iv has quit IRC | 16:14 | |
| *** Jack_Iv has joined #openstack-telemetry | 16:15 | |
| openstackgerrit | Joseph Davis proposed openstack/ceilometer: cap python-novaclient/oslo.vmware/gabbi https://review.openstack.org/425005 | 16:30 |
| *** Jack_Iv has quit IRC | 16:38 | |
| *** Jack_Iv has joined #openstack-telemetry | 16:39 | |
| *** Jack_Iv has quit IRC | 16:39 | |
| *** Jack_Iv has joined #openstack-telemetry | 16:40 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: drop api and storage references from install-guide https://review.openstack.org/426304 | 16:49 |
| *** jmlowe1 has joined #openstack-telemetry | 16:56 | |
| jmlowe1 | ok, so I think my problem is that my omap is way too large to return all the metrics to be processed before the ceph osd times out and dies | 16:57 |
| jmlowe1 | Kraken was recently release with a change in omap behavior | 16:58 |
| jmlowe1 | "The librados omap get_keys and get_vals operations include a start key and a limit on the number of keys to return. The OSD now imposes a configurable limit on the number of keys and number of total bytes it will respond with, which means that a librados user might get fewer keys than they asked for. This is necessary to prevent careless users from requesting an unreasonable amount of data from the cluster in a single operat | 16:58 |
| jmlowe1 | will gnocchi handle this change well if there is a very large backlog? will only getting a partial list of keys from omap be a problem? | 16:59 |
| gordc | jmlowe1: hmm.. we should only be grabbing the first x key/val in omap... although maybe it's pulling in all and than grabbing first x? | 17:00 |
| gordc | let me take a quick look | 17:00 |
| gordc | regardless, you should be keeping backlog relatively low until we get new scheduling out | 17:01 |
| gordc | (get started) | 17:01 |
| jmlowe1 | I'm looking at getting a couple of nvme ssd's and putting dedicated osd's on them for gnocchi so that I won't knock over the osd's that back my cinder and glance | 17:03 |
| jmlowe1 | I figure I might have a chance at clearing my backlog as well | 17:03 |
| jmlowe1 | it's about 3.5M measurements | 17:03 |
| jmlowe1 | it looks like people have had similar trouble by putting 300M files in a single directory on cephfs which puts metadata in omap | 17:06 |
| jd__ | those are good questions for sileht | 17:07 |
| gordc | we can set a limit which we don't... that might help | 17:09 |
| gordc | it won't fix reporting though | 17:09 |
| gordc | jmlowe1: you can try something like http://paste.openstack.org/show/596756/ | 17:13 |
| gordc | but i feel like you will still have issues if you don't turn off reporting processing. | 17:14 |
| gordc | because the process requires it to list all 3.5M measures. | 17:14 |
| jmlowe1 | ok, is that in the docs somewhere? | 17:14 |
| gordc | disable reporting process? i don't think you can without hacking :/ | 17:15 |
| *** nadya has joined #openstack-telemetry | 17:15 | |
| gordc | jmlowe1: you can comment out https://github.com/openstack/gnocchi/blob/master/gnocchi/cli.py#L301 | 17:16 |
| gordc | there is also a opt to configure how often it runs `metric_reporting_delay` but i've no idea if it runs right away and then waits, or waits and then runs. | 17:17 |
| *** nadya has quit IRC | 17:17 | |
| openstackgerrit | gordon chung proposed openstack/gnocchi: ceph: set return limit https://review.openstack.org/426314 | 17:22 |
| jmlowe1 | shoved in those patches and so far so good | 17:23 |
| jmlowe1 | I may get out of this mess yet | 17:23 |
| gordc | jmlowe1: interesting... my hail mary worked?lol | 17:24 |
| gordc | jmlowe1: i missed an if in there. so it might | 17:24 |
| gordc | error | 17:24 |
| jmlowe1 | so far so good | 17:25 |
| jmlowe1 | I'm guessing just turning off metric reporting might have been enough | 17:25 |
| jd__ | good to know | 17:26 |
| jmlowe1 | I'm down about 100k objects in the gnocchi pool so far | 17:28 |
| gordc | that is definitely going to be problematic... can't really hack a fix for that | 17:28 |
| jd__ | the reporting? | 17:28 |
| gordc | jd__: yeah | 17:28 |
| jmlowe1 | it will remove stuff from omap as it processes no? | 17:29 |
| jd__ | gordc: is there a "paging" api in Ceph? | 17:29 |
| gordc | jd__: well if we do the bucketise scheduling stuff, it'll help | 17:29 |
| jd__ | jmlowe1: yes | 17:29 |
| gordc | jd__: yeah, there is i believe. | 17:29 |
| jmlowe1 | so just a disable reporting option that you can turn on when you get in trouble might be enough | 17:30 |
| stevelle | current docs suggest setting gnocchi as an event_dispatcher? http://docs.openstack.org/developer/ceilometer/install/dbreco.html#moving-from-ceilometer-to-gnocchi | 17:30 |
| jmlowe1 | in kraken there is an offset | 17:30 |
| jd__ | stevelle: yes | 17:31 |
| jd__ | jmlowe1: that's what we'd need then | 17:31 |
| jmlowe1 | give me all keys starting with x | 17:31 |
| *** shardy has quit IRC | 17:31 | |
| jd__ | gordc: that'd be helpful in your patch with the limit | 17:31 |
| *** Jack_Iv has quit IRC | 17:31 | |
| gordc | jd__: i don' think that api exists. (also is kraken released? | 17:31 |
| jmlowe1 | you have no idea how long I've been waiting for this, like 6 years http://ceph.com/releases/v11-2-0-kraken-released/ | 17:32 |
| jd__ | gordc: jmlowe1 just said it does | 17:32 |
| *** Jack_Iv has joined #openstack-telemetry | 17:33 | |
| gordc | jd__: hmm... been there for a while. https://github.com/ceph/ceph/blame/master/src/pybind/rados/rados.pyx#L3238 | 17:35 |
| gordc | i have no idea if it's an offset or a marker... | 17:35 |
| gordc | seems like marker | 17:35 |
| gordc | which means it could disappear... | 17:36 |
| jmlowe1 | release notes lead me to say marker | 17:36 |
| gordc | jmlowe1: yeah, i'm guessing that too... it expects string | 17:36 |
| jmlowe1 | I don't want to even contemplate how to handle that | 17:36 |
| jmlowe1 | on the other hand, does it matter if you get all of them? | 17:37 |
| jmlowe1 | nm, order probably counts | 17:37 |
| *** nadya has joined #openstack-telemetry | 17:38 | |
| jd__ | the problem is that getting all of them makes it crash no? | 17:39 |
| gordc | jmlowe1: in theory the reporting number isn't completely accurate anyways. | 17:39 |
| jd__ | if the key does not exist it just returns everything? | 17:39 |
| gordc | jd__: right. but if it's a marker and it disappears, i think it could end up in some massive loop trying to page to the end | 17:39 |
| gordc | it returns everthing if you don't give marker and don't give limit | 17:40 |
| gordc | i would think if the marker doesnt exist, it would start from beginning | 17:40 |
| jmlowe1 | the problem is that there is a heartbeat and if it times out the osd dies, grabbing all of the keys exceeds the heartbeat time out and failover begins, you get flapping osd's | 17:41 |
| jd__ | hehe | 17:41 |
| gordc | jd__: i think we need to bucketise our incoming measures regardless. right now all our incoming measures hit the same osd+filestore. same for swift i believe according to notmynames ML | 17:43 |
| jd__ | yup | 17:43 |
| gordc | or well i guess ceph hits different osds but same filestore... something for gnocchi version next | 17:44 |
| *** nadya has quit IRC | 17:44 | |
| jd__ | yeah | 17:45 |
| openstackgerrit | gordon chung proposed openstack/gnocchi: ceph: set return limit https://review.openstack.org/426314 | 17:50 |
| *** nicodemus_ has joined #openstack-telemetry | 18:00 | |
| *** harlowja has joined #openstack-telemetry | 18:02 | |
| *** hfu has quit IRC | 18:02 | |
| *** cdent has quit IRC | 18:02 | |
| *** chlong has quit IRC | 18:03 | |
| *** nadya has joined #openstack-telemetry | 18:11 | |
| *** donghao has joined #openstack-telemetry | 18:14 | |
| *** donghao has quit IRC | 18:18 | |
| openstackgerrit | Merged openstack/ceilometer: cap python-novaclient/oslo.vmware/gabbi https://review.openstack.org/425005 | 18:22 |
| *** khushbu has joined #openstack-telemetry | 18:28 | |
| *** khushbu has quit IRC | 18:38 | |
| *** gordc has quit IRC | 18:50 | |
| *** Jack_Iv has quit IRC | 19:03 | |
| *** Jack_Iv has joined #openstack-telemetry | 19:04 | |
| nicodemus_ | hello | 19:04 |
| nicodemus_ | one question... if a Nova instance has X metadata key/values, is ceilometer able to gather those metadata values? | 19:06 |
| *** gordc has joined #openstack-telemetry | 19:07 | |
| nicodemus_ | I'm wondering if I could create a resource_type containing for example 10 different possible metadata values, and have ceilometer pushing to gnocchi each resource with that metadata (if it exists) | 19:07 |
| *** Jack_Iv_ has joined #openstack-telemetry | 19:11 | |
| *** Jack_Iv has quit IRC | 19:14 | |
| *** donghao has joined #openstack-telemetry | 19:15 | |
| *** Jack_Iv_ has quit IRC | 19:18 | |
| *** Gleb has joined #openstack-telemetry | 19:19 | |
| *** Jack_Iv has joined #openstack-telemetry | 19:19 | |
| *** donghao has quit IRC | 19:19 | |
| Gleb | Hello! I am in the process of configuring the HTTP dispatcher in Ceilometer and I have a question. In ceilometer.conf, I have the option to set two different targets under [http_dispatcher]: event_target and target. What is the difference between these two? | 19:20 |
| *** nadya has quit IRC | 19:21 | |
| *** Jack_Iv has quit IRC | 19:21 | |
| *** Jack_Iv has joined #openstack-telemetry | 19:21 | |
| *** nadya has joined #openstack-telemetry | 19:22 | |
| jmlowe1 | guessing (I'm curious to know if I'm right) event_target is for events (this thing happened at this time like successful auth or instance launch) and target is for measurements (think samples) | 19:22 |
| *** Jack_Iv has quit IRC | 19:24 | |
| Gleb | Right now I set both targets to different ports, and I only receive http requests on the address set in "target". I seem to obtain events at the address set by "target". | 19:25 |
| Gleb | The requests only get sent when something happens (e.g. instance launch), and have a event_type field. | 19:26 |
| Gleb | However there are also periodical requests (every 10 mins) that describe the state of all the resources. | 19:27 |
| gordc | nicodemus_: we can now... but not before ocata | 19:35 |
| gordc | or maybe you can. it's just not as easy | 19:35 |
| nicodemus_ | gordc, so... I should look at current master and backport some code if I wanted to make it work with a Newton deploy, right? | 19:36 |
| gordc | nicodemus_: we added this https://review.openstack.org/#/c/413236/ | 19:36 |
| gordc | i believe in theory you could add metadata before this but you need to know exact value and add it to event_definitions | 19:37 |
| gordc | and then you'd need to create equivalent resource type in gnocchi to capture the data | 19:38 |
| gordc | and probbaly update teh gnocchi mappings https://github.com/openstack/ceilometer/blob/master/ceilometer/dispatcher/data/gnocchi_resources.yaml | 19:38 |
| gordc | (i've never done it recently so i'm just yelling things out as they come into my head) | 19:39 |
| nicodemus_ | gordc, I see. I imagined that I would need to modify gnocchi_resources.yaml | 19:39 |
| gordc | nicodemus_: yep | 19:39 |
| nicodemus_ | The use case would be to create a resource_type with X metadata keys that are required:false, so if a certain instance has the metadata fine, if not, no problem | 19:40 |
| nicodemus_ | I'll take a look to see how complex it would be to backport it | 19:40 |
| nicodemus_ | thanks gordc ! | 19:40 |
| gordc | nicodemus_: kk, you can specify metadata in meters.yaml file prior to ocata. but you need to specify each attribute specifically... and it might not necessarily be there either... but yeah, you'll need to debug that. | 19:42 |
| *** eglynn has quit IRC | 19:42 | |
| gordc | Gleb: the periodic 10min stuff is from polling agents | 19:43 |
| *** rbak_ has quit IRC | 19:43 | |
| *** rbak_ has joined #openstack-telemetry | 19:43 | |
| gordc | i don't actually use the http dispatcher but what jmlowe1 said was right (or should be) | 19:43 |
| Gleb | Ok. Do you have any suggestions about how to enable the event dispatcher? I only get http requests from the meter dispatcher. | 19:47 |
| gordc | Gleb: what data are you looking for? | 19:47 |
| *** Jack_Iv has joined #openstack-telemetry | 19:48 | |
| Gleb | I just noticed that ceilometer event-list returns an empty list, so I guess I need to set that up first. I'm trying to monitor the creation and deletion of resources (instances, volumes, images, etc) | 19:48 |
| gordc | you should be able to just define meter_dispatchers=http and event_dispatchers=http and then set the appropriate in http_dispatchers section | 19:48 |
| *** nadya has quit IRC | 19:49 | |
| Gleb | It seems that it's just a matter of setting up events in ceilometer to get the event dispatcher working. Thanks for the help! Will check back in if I have any other questions. | 19:50 |
| gordc | you should probably be a aware that we pushed all the ceilometer event-* storage to a new service named Panko | 19:50 |
| Gleb | Yes I know, but I'm stuck on an old release (kilo), so I think that predates Panko. | 19:51 |
| gordc | Gleb: yeah :( although you in theory could still use it since it's just an endpoint (although i've never tested that so don't quote me) | 19:52 |
| gordc | probably best to delete this entire history. | 19:53 |
| *** david-lyle has joined #openstack-telemetry | 20:01 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: update verify to use gnocchi https://review.openstack.org/426389 | 20:23 |
| *** nicodemus_ has quit IRC | 20:38 | |
| *** pradk has quit IRC | 20:47 | |
| *** nicodemus_ has joined #openstack-telemetry | 20:50 | |
| *** fguillot has quit IRC | 20:53 | |
| *** donghao has joined #openstack-telemetry | 21:16 | |
| *** leitan has quit IRC | 21:19 | |
| *** nicodemus_ has quit IRC | 21:19 | |
| *** donghao has quit IRC | 21:20 | |
| *** pradk has joined #openstack-telemetry | 21:41 | |
| *** catintheroof has quit IRC | 21:44 | |
| *** catintheroof has joined #openstack-telemetry | 21:45 | |
| *** catintheroof has quit IRC | 21:49 | |
| *** vint_bra has quit IRC | 21:54 | |
| *** Guest21909 has quit IRC | 22:00 | |
| *** pradk has quit IRC | 22:02 | |
| *** gordc has quit IRC | 22:12 | |
| *** vishwanathj has joined #openstack-telemetry | 22:13 | |
| *** vishwanathj has quit IRC | 22:14 | |
| *** vishwanathj has joined #openstack-telemetry | 22:15 | |
| *** thorst_ has quit IRC | 22:16 | |
| *** vishwanathj has quit IRC | 22:17 | |
| *** Jack_Iv has quit IRC | 22:20 | |
| *** rbak_ has quit IRC | 23:19 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/ceilometer: WIP: Use Tempest stable library interfaces https://review.openstack.org/426427 | 23:22 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!