| *** vishwana_ has joined #openstack-telemetry | 00:02 | |
| *** vishwanathj has quit IRC | 00:06 | |
| *** vishwanathj has joined #openstack-telemetry | 00:14 | |
| *** vishwana_ has quit IRC | 00:16 | |
| *** diogogmt has quit IRC | 00:17 | |
| *** thorst_ has quit IRC | 00:32 | |
| *** mgarza_ has quit IRC | 00:33 | |
| *** thorst has joined #openstack-telemetry | 00:33 | |
| *** thorst has quit IRC | 00:37 | |
| *** cheneydc has joined #openstack-telemetry | 01:02 | |
| *** Liuqing has joined #openstack-telemetry | 01:09 | |
| *** jwcroppe_ has quit IRC | 01:11 | |
| *** jwcroppe has joined #openstack-telemetry | 01:12 | |
| *** diogogmt has joined #openstack-telemetry | 01:14 | |
| *** jwcroppe has quit IRC | 01:16 | |
| *** prashantD_ has quit IRC | 01:39 | |
| *** guillaume_ has quit IRC | 01:40 | |
| *** shakamunyi has joined #openstack-telemetry | 01:48 | |
| *** ljxiash has joined #openstack-telemetry | 01:53 | |
| *** ljxiash has quit IRC | 01:58 | |
| *** ljxiash has joined #openstack-telemetry | 01:59 | |
| *** ljxiash has quit IRC | 01:59 | |
| *** ljxiash has joined #openstack-telemetry | 01:59 | |
| *** shakamunyi has quit IRC | 02:10 | |
| *** shakamunyi has joined #openstack-telemetry | 02:11 | |
| *** guillaume_ has joined #openstack-telemetry | 02:15 | |
| *** thorst has joined #openstack-telemetry | 02:37 | |
| *** thorst has quit IRC | 02:42 | |
| *** jwcroppe has joined #openstack-telemetry | 02:57 | |
| *** ljxiash has quit IRC | 03:06 | |
| *** ljxiash has joined #openstack-telemetry | 03:07 | |
| *** achatterjee has joined #openstack-telemetry | 03:07 | |
| *** mragupat has joined #openstack-telemetry | 03:10 | |
| *** mragupat has quit IRC | 03:11 | |
| *** mragupat has joined #openstack-telemetry | 03:12 | |
| *** jwcroppe_ has joined #openstack-telemetry | 03:23 | |
| *** jwcroppe has quit IRC | 03:26 | |
| *** cheneydc_ has joined #openstack-telemetry | 03:44 | |
| *** yarkot has joined #openstack-telemetry | 03:45 | |
| *** guillaume_ has quit IRC | 03:47 | |
| *** vishwanathj has quit IRC | 03:53 | |
| *** Liuqing has quit IRC | 04:20 | |
| *** Liuqing has joined #openstack-telemetry | 04:21 | |
| openstackgerrit | Xia Linjuan proposed openstack/python-aodhclient: Fix aodh client fails when command with the arg --time-constraint https://review.openstack.org/272946 | 04:37 |
|---|---|---|
| *** guillaume_ has joined #openstack-telemetry | 04:47 | |
| *** nicodemus_ has quit IRC | 04:49 | |
| *** idonotknow_ has joined #openstack-telemetry | 04:50 | |
| idonotknow_ | Hi,everyone. I found my ceilometer could not find neutron_client ,and got the warning message "The resource could not be found.".I do not know why it always happens. Can anyone help me? | 04:52 |
| *** mragupat has quit IRC | 04:55 | |
| *** mragupat has joined #openstack-telemetry | 04:56 | |
| *** tongli has quit IRC | 05:00 | |
| *** yprokule has joined #openstack-telemetry | 05:28 | |
| *** vishwanathj has joined #openstack-telemetry | 05:41 | |
| *** ljxiash has quit IRC | 05:42 | |
| *** vishwanathj is now known as vishwanathj_zzz | 05:43 | |
| *** ljxiash has joined #openstack-telemetry | 05:44 | |
| *** ljxiash has quit IRC | 05:44 | |
| *** ljxiash_ has joined #openstack-telemetry | 05:44 | |
| *** efoley has joined #openstack-telemetry | 05:48 | |
| *** efoley has left #openstack-telemetry | 05:48 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Imported Translations from Zanata https://review.openstack.org/273346 | 06:08 |
| *** panda has joined #openstack-telemetry | 06:21 | |
| *** panda is now known as Guest81915 | 06:22 | |
| openstackgerrit | yuntongjin proposed openstack/ceilometer: event to sample publisher https://review.openstack.org/227106 | 06:22 |
| *** idonotknow_ has quit IRC | 06:23 | |
| *** ifat_afek has joined #openstack-telemetry | 06:34 | |
| *** thorst has joined #openstack-telemetry | 06:41 | |
| *** thorst has quit IRC | 06:48 | |
| *** _nadya_ has joined #openstack-telemetry | 06:55 | |
| *** ifat_afek has left #openstack-telemetry | 06:58 | |
| *** mragupat has quit IRC | 07:14 | |
| *** mragupat has joined #openstack-telemetry | 07:15 | |
| *** boris-42 has joined #openstack-telemetry | 07:16 | |
| *** _nadya_ has quit IRC | 07:18 | |
| *** mragupat has quit IRC | 07:19 | |
| *** Guest81915 has left #openstack-telemetry | 07:20 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 07:31 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Disable rados thread system https://review.openstack.org/272650 | 07:33 |
| sileht | jd__, I have finished repair my devstack vm, and test my thought about ceph get_xattrs: http://paste.openstack.org/show/485209/ | 07:33 |
| sileht | jd__, the rados binding around get_xattrs in awful, get_xattrs start a thread to return the iterator of xattr, and start a new thread each times we call next on it to get the next key, value... | 07:48 |
| *** guillaume_ has quit IRC | 07:56 | |
| *** belmoreira has joined #openstack-telemetry | 07:59 | |
| *** thorst has joined #openstack-telemetry | 08:03 | |
| *** thorst has quit IRC | 08:07 | |
| *** barra204 has joined #openstack-telemetry | 08:17 | |
| *** thorst has joined #openstack-telemetry | 08:17 | |
| *** shakamunyi has quit IRC | 08:19 | |
| *** permalac has joined #openstack-telemetry | 08:19 | |
| *** thorst has quit IRC | 08:22 | |
| *** cheneydc_ has quit IRC | 08:23 | |
| *** cheneydc has quit IRC | 08:23 | |
| *** thorst has joined #openstack-telemetry | 08:24 | |
| *** thorst has quit IRC | 08:27 | |
| *** thorst has joined #openstack-telemetry | 08:27 | |
| *** cdent has joined #openstack-telemetry | 08:47 | |
| openstackgerrit | Xia Linjuan proposed openstack/python-aodhclient: add alarm_history.py unit test https://review.openstack.org/273397 | 08:50 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 08:55 |
| openstackgerrit | Xia Linjuan proposed openstack/python-aodhclient: add alarm_history.py unit test https://review.openstack.org/273397 | 08:56 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 08:58 |
| *** shardy has joined #openstack-telemetry | 09:00 | |
| *** jaypipes has joined #openstack-telemetry | 09:01 | |
| *** safchain has joined #openstack-telemetry | 09:05 | |
| *** mattyw has joined #openstack-telemetry | 09:09 | |
| *** yassine__ has joined #openstack-telemetry | 09:18 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 09:20 |
| *** _nadya_ has joined #openstack-telemetry | 09:40 | |
| *** gzhai has quit IRC | 09:49 | |
| *** gzhai has joined #openstack-telemetry | 09:50 | |
| *** r-mibu has quit IRC | 09:59 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Ensure description exists for HTTP error. https://review.openstack.org/273428 | 10:05 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 10:05 |
| *** Liuqing has quit IRC | 10:13 | |
| openstackgerrit | Ryota MIBU proposed openstack/ceilometer: make even-alarm supported in default https://review.openstack.org/273432 | 10:15 |
| *** r-mibu has joined #openstack-telemetry | 10:15 | |
| *** rcernin has joined #openstack-telemetry | 10:18 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Add some resource types tests https://review.openstack.org/270419 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute int and float https://review.openstack.org/270091 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute bool https://review.openstack.org/270418 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute uuid https://review.openstack.org/270090 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Don't create Ceilometer resource types by default. https://review.openstack.org/270322 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move legacy Ceilometer resource into indexer. https://review.openstack.org/270266 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move resource type into their own sql table https://review.openstack.org/269843 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource type CRUD. https://review.openstack.org/269844 | 10:44 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute string https://review.openstack.org/269888 | 10:44 |
| sileht | jd__, do we release batching or not yet ? | 10:47 |
| jd__ | sileht: we did not | 10:47 |
| sileht | cool :) | 10:47 |
| jd__ | that's why i'm proposing we fix it :) | 10:47 |
| sileht | that will be better | 10:48 |
| jd__ | sileht: how confident are you with https://review.openstack.org/#/c/272650 ? should I backport it to 1.3.4? | 10:48 |
| sileht | jd__, for me it works well, but yesterday mnaser have reported another issue after having applied my batch (the issue is perhaps unrelated) | 10:50 |
| jd__ | what was the issue? | 10:51 |
| sileht | jd__, http://paste.openstack.org/show/485163/ | 10:54 |
| *** yassine__ has quit IRC | 10:56 | |
| sileht | jd__, the error looks weird for me it means that the class have started to be used before __init__ ends | 10:57 |
| sileht | that's clearly impossible | 10:57 |
| *** permalac has quit IRC | 10:57 | |
| sileht | So I think this is because I was monkeypatching at Runtime, so the module got monkeypatch multiple times | 10:58 |
| sileht | I have moved that at module load times, so that should not occurs anymore | 10:59 |
| sileht | jd__, perhaps we can wait mnaser to retest, except you want to release 1.3.4 today | 10:59 |
| sileht | jd__, do you have a suggestion for batching endpoint ? all that comes in my mind sucks | 11:00 |
| jd__ | sileht: yeah I'll go with 1.3.4 today and i'll do it in 1.3.5 next week or so | 11:01 |
| sileht | > /v1/batch/metrics-measure and /v1/batch/named-metrics-measure | 11:01 |
| jd__ | sileht: not really… | 11:01 |
| * jd__ thinks | 11:02 | |
| sileht | > /v1/batch/metrics-measure and /v1/batch/resource-named-metrics-measure | 11:02 |
| sileht | > /v1/batch/metrics/measure and /v1/batch/resource-named-metrics/measure | 11:02 |
| jd__ | or /v1/batch/metrics/measure and /v1/batch/resources/metrics/measure ? | 11:02 |
| sileht | I like, then we could add PATCH /v1/batch/resources | 11:03 |
| sileht | to update resource attributes | 11:04 |
| sileht | and POST /v1/batch/resources/metrcs to add metrics in batch | 11:04 |
| sileht | (not sure the last one will be helpful) | 11:04 |
| *** jaypipes has quit IRC | 11:06 | |
| openstackgerrit | Julien Danjou proposed openstack/aodh: Update alarm history only if change in alarm property https://review.openstack.org/273463 | 11:12 |
| *** jaypipes has joined #openstack-telemetry | 11:14 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 11:16 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics https://review.openstack.org/273368 | 11:19 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 11:25 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 11:26 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 11:27 |
| *** gzhai1 has joined #openstack-telemetry | 11:33 | |
| *** gzhai has quit IRC | 11:36 | |
| cdent | EmilienM: I posted on your thread with "Joe had it in for ceilometer". I hope that's sufficiently honest. | 11:38 |
| openstackgerrit | Xia Linjuan proposed openstack/ceilometer-specs: Support collect neutron port meters https://review.openstack.org/273475 | 11:42 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 11:48 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Ensure description exists for HTTP error. https://review.openstack.org/273428 | 11:48 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Fix resource representation for a metric https://review.openstack.org/273477 | 11:48 |
| *** idegtiarov has quit IRC | 11:53 | |
| EmilienM | cdent: nice writting, it is honest to me. Thanks | 11:53 |
| * cdent bows | 11:54 | |
| EmilienM | can someone help me? I'm having 403 when I run gnocchi archive-policy-create with admin credentials | 11:54 |
| EmilienM | I'm running that: http://paste.openstack.org/show/485257/ | 11:56 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 11:57 |
| *** idegtiarov has joined #openstack-telemetry | 11:57 | |
| EmilienM | jd__, sileht ^ maybe if you have 2 min | 11:57 |
| EmilienM | running mitaka master | 11:58 |
| sileht | EmilienM, can you paste your api-paste.conf, policy.json and gnocchi.conf ? | 11:58 |
| jd__ | also your credit card number | 11:58 |
| sileht | jd__, that's obvious | 11:59 |
| EmilienM | lol | 11:59 |
| EmilienM | ahahah | 11:59 |
| EmilienM | ok, wait | 11:59 |
| jd__ | it's weird to have that being denied if have the admin role and use the default policy.json I'd say | 11:59 |
| sileht | agreed | 12:00 |
| EmilienM | /etc/gnocchi/gnocchi.conf : http://ur1.ca/ogggz | 12:00 |
| EmilienM | /etc/gnocchi/policy.json : http://ur1.ca/oggh1 | 12:00 |
| EmilienM | /etc/gnocchi/api-paste.ini : http://ur1.ca/oggh2 | 12:00 |
| *** mattyw has quit IRC | 12:00 | |
| EmilienM | /etc/gnocchi/policy.json is managed by RDO packaging, not by Puppet | 12:00 |
| EmilienM | same for /etc/gnocchi/api-paste.ini | 12:00 |
| EmilienM | and gnocchi.conf is managed by puppet-gnocchi | 12:00 |
| EmilienM | so we might have a bug in there | 12:01 |
| sileht | EmilienM, read the first line of api-paste.ini ;) | 12:01 |
| jd__ | that was easy | 12:01 |
| * jd__ forgets EmilienM is working on master | 12:02 | |
| EmilienM | sorry guys I don't get your private jokes | 12:02 |
| sileht | while some production system still run icehouse | 12:02 |
| *** yassine__ has joined #openstack-telemetry | 12:02 | |
| sileht | EmilienM, keystoneauth is not enabled by default | 12:03 |
| sileht | EmilienM, you have to add it in api-paste.ini | 12:03 |
| EmilienM | ok, we never did that in other projects but ok | 12:03 |
| sileht | EmilienM, because this is a huge bottle neck | 12:04 |
| jd__ | you know how we are, always ahead of the next fashion trend | 12:04 |
| EmilienM | sileht: do you have an example of file please? | 12:04 |
| EmilienM | I'm going to try first | 12:04 |
| EmilienM | jd__: is it because you can run gnocchi without keystone? | 12:05 |
| EmilienM | from an user perspective, you should enable keystone middleware by default, it's a bit confusing as Gnocchi is an OpenStack project | 12:06 |
| cdent | EmilienM: jd__ is only riding the openstack bus for luxury | 12:06 |
| cdent | ;) | 12:06 |
| EmilienM | sileht, jd__ : any example of file? | 12:08 |
| sileht | EmilienM, you can edit it with the ini lib of puppet by juste changing the pipeline to "keystone_authtoken gnocchi" | 12:11 |
| EmilienM | sileht: in [pipeline:main] ? | 12:12 |
| sileht | EmilienM, yes | 12:13 |
| EmilienM | ok | 12:13 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 12:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Ensure description exists for HTTP error. https://review.openstack.org/273428 | 12:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Fix resource representation for a metric https://review.openstack.org/273477 | 12:15 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client https://review.openstack.org/273429 | 12:17 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Ensure description exists for HTTP error. https://review.openstack.org/273428 | 12:17 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Fix resource representation for a metric https://review.openstack.org/273477 | 12:17 |
| * sileht -> lunch | 12:19 | |
| EmilienM | sileht: it works thanks | 12:22 |
| openstackgerrit | Zi Lian Ji proposed openstack/python-aodhclient: Add some unit test cases https://review.openstack.org/273021 | 12:26 |
| *** leitan has joined #openstack-telemetry | 12:28 | |
| openstackgerrit | Xia Linjuan proposed openstack/ceilometer-specs: Support collect neutron port meters https://review.openstack.org/273475 | 12:32 |
| gordc | cdent: damn. you naming names? | 12:34 |
| openstackgerrit | Xia Linjuan proposed openstack/ceilometer-specs: Support collect neutron port meters https://review.openstack.org/273475 | 12:35 |
| cdent | he was already named by referencing the review where monasca was discussed | 12:35 |
| gordc | cdent: true true. *shrugs* he might get mad on the internet. | 12:36 |
| cdent | I don't think he's ever been shy about that feeling | 12:37 |
| gordc | well in person... | 12:37 |
| *** _nadya_ has quit IRC | 12:39 | |
| gordc | i'm not sure a certain company will be fond of your email. it's a nice start to my day though. so much appreciated. | 12:39 |
| *** ildikov has quit IRC | 12:41 | |
| *** ljxiash_ has quit IRC | 12:42 | |
| *** ljxiash has joined #openstack-telemetry | 12:43 | |
| *** ljxiash has quit IRC | 12:43 | |
| *** ljxiash has joined #openstack-telemetry | 12:43 | |
| *** ljxiash has quit IRC | 12:44 | |
| *** ljxiash has joined #openstack-telemetry | 12:45 | |
| *** ljxiash has quit IRC | 12:45 | |
| *** efoley has joined #openstack-telemetry | 12:48 | |
| *** mattyw has joined #openstack-telemetry | 12:51 | |
| cdent | gordc: I could just as easily have named monty | 12:56 |
| cdent | should I do that too? | 12:56 |
| *** nicodemus_ has joined #openstack-telemetry | 12:58 | |
| *** nicodemus_ has quit IRC | 12:58 | |
| gordc | cdent: i think its' pretty evident who the people that were blindly +2'ing stuff were and what company they worked for. | 12:59 |
| gordc | cdent: you should be more loyal to your company :P | 12:59 |
| *** ildikov has joined #openstack-telemetry | 13:00 | |
| cdent | wait, this is openstack, I thought we were supposed to be loyal to it | 13:00 |
| gordc | what's openstack? | 13:00 |
| jd__ | our IRC network | 13:02 |
| jd__ | I see it prepended to this channel name | 13:02 |
| cdent | You make a valid point jd__ | 13:02 |
| gordc | jd__: how did heat get away with not using it. | 13:02 |
| jd__ | maybe they got a premium account | 13:03 |
| cdent | ha! | 13:03 |
| gordc | naw. can't pay. we humble folks | 13:04 |
| * cdent wants his money back | 13:05 | |
| *** therve has joined #openstack-telemetry | 13:10 | |
| therve | Yop yop | 13:10 |
| therve | We're getting intermittent failures in Heat integration test with Ceilometer | 13:13 |
| therve | I've found one aodh error which may be in the right window: http://logs.openstack.org/53/273253/2/gate/gate-heat-dsvm-functional-convg-mysql/faab21e/logs/screen-aodh-evaluator.txt.gz#_2016-01-28_11_37_54_429 | 13:13 |
| therve | Any idea? | 13:13 |
| * jd__ frowns | 13:14 | |
| jd__ | therve: never saw that but that looks like a bug | 13:14 |
| therve | jd__, OK. It's intermittent so it's likely an annoying one | 13:15 |
| gordc | therve: yep. let's make it a bug. | 13:15 |
| jd__ | i'm checking the code | 13:15 |
| jd__ | therve: gordc: any reason this is not using Gnocchi btw? | 13:16 |
| therve | jd__, I don't know. What is it using? | 13:16 |
| jd__ | the old Ceilometer storage and the old threshold alarms AFAIU | 13:17 |
| gordc | jd__: it's heat test... we don't have people applying gnocchi to other places. (except for sileht) | 13:17 |
| gordc | and i failed to clone him many times already | 13:17 |
| * jd__ wonders how gordc tried that | 13:17 | |
| gordc | don't ask! | 13:18 |
| therve | Okay. No I don't think there is a reason, I'd happy to use it. We "just" need to custom our gate job I presume | 13:18 |
| * jd__ shakes head | 13:18 | |
| jd__ | therve: i'll propose a "fix" in the meantime anyway | 13:18 |
| therve | jd__, FWIW I think it started yesterday | 13:18 |
| jd__ | nothing has been merged for a week it seems in aodh | 13:19 |
| jd__ | though it could be Ceilometer | 13:19 |
| therve | Right | 13:21 |
| therve | jd__, https://bugs.launchpad.net/heat/+bug/1539069 FWIW if you want a bug number | 13:22 |
| openstack | Launchpad bug 1539069 in heat "CeilometerAlarmTest is failing intermittently" [Medium,New] | 13:22 |
| jd__ | thanks therve | 13:22 |
| openstackgerrit | Julien Danjou proposed openstack/aodh: threshold: fix statistics empty case https://review.openstack.org/273531 | 13:26 |
| jd__ | this code is… not beautiful | 13:27 |
| therve | I wonder if it won't just push the issue further | 13:30 |
| therve | We should have a data point here AFAIU | 13:30 |
| jd__ | therve: maybe | 13:30 |
| jd__ | but everything is async, so are you sure it's there already? | 13:30 |
| jd__ | if ceilo is slow… | 13:31 |
| therve | As long as it gets there at some point :) | 13:31 |
| jd__ | hehe | 13:31 |
| *** Liuqing has joined #openstack-telemetry | 13:32 | |
| *** _nadya_ has joined #openstack-telemetry | 13:32 | |
| jd__ | gordc: do we have a concrete plan to deprecate Ceilometer sample storage and all finally? doc update? something in the code? | 13:32 |
| gordc | jd__: that's what this was for: https://review.openstack.org/#/c/273049/2/doc/source/install/dbreco.rst | 13:35 |
| jd__ | gordc: that looks soft | 13:35 |
| jd__ | but ok :) | 13:35 |
| gordc | tbh, i'm not sure we can fully deprecate. people actually use it. | 13:35 |
| gordc | and gnocchi doesn't cover everything | 13:35 |
| jd__ | we can do whatever we want because we are the developers you know | 13:35 |
| gordc | so the best we can do is 'hey, if you want to keep it around and do random sh*t later, sure.' | 13:36 |
| jd__ | yeah, i'd just hope we would move it out of our way | 13:36 |
| gordc | jd__: "we can do whatever we want because we are the developers you know"... you watch an intel commercial recently? | 13:37 |
| jd__ | no, should I? | 13:37 |
| gordc | well it's old... it's just a phrase i never heard befoer.lol | 13:37 |
| gordc | https://www.youtube.com/watch?v=zMlWbTqwkdU | 13:37 |
| jd__ | lol | 13:38 |
| gordc | jd__: so when you say deprecate, you mean to delete it next cycle? | 13:41 |
| *** ljxiash has joined #openstack-telemetry | 13:42 | |
| jd__ | delete or move out | 13:42 |
| _nadya_ | gordc, jd__ did you discuss this at ops session? | 13:42 |
| jd__ | the one where people were asking what Gnocchi was where when we were already working and pushing for it for a year? I don't remember, I think we did somehow | 13:43 |
| gordc | jd__: i don't mind moving out... we do need to think about how it will be registered in keystone | 13:44 |
| jd__ | my point being I'm a developer and my users don't use that so I am not going to maintain it, nor is sileht, nor is (AFAIK) any other active Telemetry developer | 13:44 |
| *** Liuqing has quit IRC | 13:44 | |
| jd__ | so unless a developer really wants to do it, well, it'll move out | 13:44 |
| gordc | _nadya_: no. jd__ is just riffing | 13:44 |
| jd__ | haha | 13:44 |
| jd__ | gordc: hm I think I'd keep the same endpoint but I'd move the event part in another endpoint, wdyt? | 13:45 |
| gordc | jd__: well it's being worked on still so i assume someone is using it. | 13:45 |
| gordc | jd__: just for reference, we use it (but in transition) | 13:45 |
| _nadya_ | gordc: a lot of companies using it | 13:45 |
| gordc | jd__: we'd need a new user? | 13:45 |
| gordc | user as in ceilometer (for polling stuff) and something else for storage? | 13:46 |
| jd__ | gordc: keystone user? not sure, why? | 13:46 |
| * cdent wonders if any one else got "we can do whatever we want because we are the developers you know" memo? | 13:46 | |
| _nadya_ | I guess the main reason to deprecate/remove it is not to adopt it to the new API. But we don't have new API | 13:46 |
| jd__ | yeah, when is Ceilometer API v3 being released? | 13:46 |
| * jd__ flies out | 13:47 | |
| gordc | _nadya_: what api are you looking for? | 13:47 |
| *** Liuqing has joined #openstack-telemetry | 13:47 | |
| * cdent shoots jd__ with his magic arrow | 13:47 | |
| * jd__ crashes loudly | 13:47 | |
| _nadya_ | gordc: I don't know :) Actually, I don't have any issues with the current one. But everybody want a new one. The main use case is billing and autoscaling | 13:49 |
| gordc | _nadya_: i don't think the current api works well for those use cases. it's much too verbose. | 13:50 |
| _nadya_ | gordc: I remember our discussion about API. Probably, API-less Ceilometer-core is a good option indeed. But, in my opinion, Telemetry needs some common API anyway. Because we need to communicate with OpenStack somehow. With Heat at least | 13:52 |
| *** Liuqing has quit IRC | 13:52 | |
| gordc | _nadya_: yeah. i was thinking api-less ceilometer as well. but we need to move dbs first. | 13:53 |
| jd__ | Aodh and Gnocchi APIs can count as Telemetry API, no? | 13:53 |
| * cdent was just going to say what jd__ said | 13:54 | |
| gordc | yes | 13:54 |
| cdent | v3 api already exists, it is just _way_ different | 13:54 |
| cdent | and -events :) | 13:54 |
| _nadya_ | jd__: probably. What is the main usecase for Gnocchi? | 13:54 |
| jd__ | did you really write that? :( | 13:55 |
| gordc | cdent: i need to figure out what to do with events. have a meeting later to discuss whether we can build some extension to leverage it. | 13:55 |
| _nadya_ | jd__: yep. I know what is Gnocchi :) But in Ceilometer we had problems with usecases, that's why I'm asking | 13:56 |
| gordc | _nadya_: https://review.openstack.org/#/c/273049/2/doc/source/install/dbreco.rst | 13:56 |
| gordc | _nadya_: i'm basically blacklisting every statistic api in ceiloemter right now | 13:56 |
| gordc | _nadya_: don't run ceilometer statistics is my recommendation... if you want that, you probably want gnochi | 13:57 |
| jd__ | _nadya_: well it solves the usecases people wanted Ceilometer API for: alarming, billing, provisionning | 13:57 |
| cdent | jd__: it doesn't solve those use cases, it provides the data that _can_ solve those use cases, two different things | 13:58 |
| *** pradk has quit IRC | 13:58 | |
| _nadya_ | gordc: hmm... need to think about this more... so, in Mitaka it will be deprecated? | 13:59 |
| _nadya_ | cdent: yep, exactly | 13:59 |
| jd__ | cdent: I stand corrected indeed, thanks :) | 14:00 |
| gordc | _nadya_: no. if we do deprecate it'll be next cycle earliest. | 14:00 |
| gordc | _nadya_: so gnocchi (for most part) can't really handle auditing... so there may be a reason to keep around current api... | 14:01 |
| gordc | i'm not sure if auditing is a use case for anyoen but yeah. | 14:01 |
| jd__ | even if it is the old API is not designed for that | 14:02 |
| *** nicodemus_ has joined #openstack-telemetry | 14:05 | |
| nicodemus_ | morning jd__, do you have a moment? | 14:05 |
| jd__ | nicodemus_: morning! sure, what's going on? | 14:06 |
| nicodemus_ | I did some tests yesterday, and found a couple of errors in the logs of gnocchi-metricd | 14:06 |
| gordc | jd__: right. but it can be redoned easier for auditing case than gnocchi case is what i'm saying. | 14:09 |
| nicodemus_ | http://paste.openstack.org/show/485273/ when possible, could you look at them? I couldn't process any new metric unfortunately | 14:10 |
| jd__ | nicodemus_: the first one should be fixed by sileht new version of the Rados patch from what I got | 14:11 |
| jd__ | nicodemus_: about the second one, which version of tooz you got? it seems too old | 14:11 |
| jd__ | you need tooz>=1.22 as specified in the requirements | 14:12 |
| *** julim has quit IRC | 14:13 | |
| gordc | jd__: i asked before. i think its 1.29 | 14:15 |
| *** gzhai2 has joined #openstack-telemetry | 14:18 | |
| *** gzhai1 has quit IRC | 14:21 | |
| cdent | gordc for silly-ometer secretary | 14:22 |
| gordc | cdent: works for me. | 14:23 |
| gordc | i get a bowl of candy to hand out | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Add some resource types tests https://review.openstack.org/270419 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute number https://review.openstack.org/270091 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute bool https://review.openstack.org/270418 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute uuid https://review.openstack.org/270090 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Don't create Ceilometer resource types by default. https://review.openstack.org/270322 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move legacy Ceilometer resource into indexer. https://review.openstack.org/270266 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource type CRUD. https://review.openstack.org/269844 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute string https://review.openstack.org/269888 | 14:23 |
| *** achatterjee has quit IRC | 14:24 | |
| *** diogogmt has quit IRC | 14:28 | |
| *** efoley_ has joined #openstack-telemetry | 14:28 | |
| *** llu-laptop has joined #openstack-telemetry | 14:28 | |
| llu-laptop | jd__: ping | 14:29 |
| *** permalac has joined #openstack-telemetry | 14:29 | |
| llu-laptop | jd__: about your comment on https://review.openstack.org/#/c/273008 | 14:31 |
| *** diogogmt has joined #openstack-telemetry | 14:32 | |
| *** efoley has quit IRC | 14:32 | |
| llu-laptop | jd__: could you give me some hints about hacking esourceSchema? I didn't see anything in voluptuous documentation of similar functionality | 14:33 |
| nicodemus_ | jd__: I pulled tooz from git on Jan 21, so should be 1.29 (the latest). | 14:34 |
| *** liamji has joined #openstack-telemetry | 14:36 | |
| nicodemus_ | and about the first one, that error shows up with and without the monkeypatch for single threading (I pulled gnocchi master yesterday) | 14:36 |
| *** efoley has joined #openstack-telemetry | 14:40 | |
| *** pradk has joined #openstack-telemetry | 14:40 | |
| pradk | jd__, hmm i see you released gnocchi 1.3.4, but the master tag remains unchanged 1.3.1.dev230 | 14:45 |
| jd__ | pradk: supposedly a bot is supposed to send a commit to update it but it didn't happen | 14:46 |
| jd__ | llu-laptop: schemas are just callable, so build your own callable (partially based on voluptuous.Schema and your own hack :) | 14:46 |
| jd__ | nicodemus_: tooz master? :/ | 14:47 |
| *** pradk has quit IRC | 14:47 | |
| *** diogogmt has quit IRC | 14:49 | |
| nicodemus_ | yes... was that wrong? | 14:49 |
| jd__ | no, but I find it odd | 14:49 |
| jd__ | about that Rados error, that sounds to be a problem with metricd not being able to connect to Ceph, but it's really strange | 14:50 |
| *** pradk has joined #openstack-telemetry | 14:50 | |
| nicodemus_ | I'm setting up a new environment (separate from production) for me to have more control | 14:51 |
| jd__ | >>> tooz.drivers.redis.RedisLock.__call__ | 14:51 |
| jd__ | <unbound method RedisLock.__call__> | 14:51 |
| jd__ | RedisLock is callable | 14:51 |
| nicodemus_ | regarding tooz, would you recommend to use stable/kilo instead of master? | 14:51 |
| jd__ | so I wonder if you don't have another tooz version installed being used? | 14:51 |
| jd__ | stable/kilo? no, it's too old | 14:52 |
| jd__ | pick the latest release or master | 14:52 |
| nicodemus_ | k, will do | 14:52 |
| nicodemus_ | the funny thing is that I didn't change tooz version since last week, yet this error showed up after pulling gnocchi and redeploying yesterday | 14:59 |
| jd__ | which is not a surprise since we change Gnocchi | 15:00 |
| jd__ | we just leveraged "new" features from tooz that were merged in 1.22 | 15:00 |
| gordc | quick meeting folks? | 15:00 |
| jd__ | unless we have a bug in tooz :) | 15:00 |
| gordc | _nadya_: sileht: ildikov: meeting? | 15:02 |
| _nadya_ | :) | 15:02 |
| ildikov | gordc: tnx :) | 15:02 |
| *** julim has joined #openstack-telemetry | 15:20 | |
| *** diogogmt has joined #openstack-telemetry | 15:21 | |
| *** diogogmt has quit IRC | 15:23 | |
| *** mattyw has quit IRC | 15:24 | |
| *** rbak has joined #openstack-telemetry | 15:30 | |
| *** guillaume_ has joined #openstack-telemetry | 15:30 | |
| openstackgerrit | Merged openstack/ceilometer-specs: Improve Nova Instance metering https://review.openstack.org/209799 | 15:34 |
| *** mragupat has joined #openstack-telemetry | 15:37 | |
| *** yprokule has quit IRC | 15:43 | |
| *** ljxiash_ has joined #openstack-telemetry | 15:44 | |
| *** _nadya_ has quit IRC | 15:45 | |
| *** ljxiash has quit IRC | 15:45 | |
| *** mgarza_ has joined #openstack-telemetry | 15:45 | |
| *** llu-laptop has quit IRC | 15:47 | |
| *** efoley has quit IRC | 15:54 | |
| *** ddieterly has joined #openstack-telemetry | 16:00 | |
| *** diogogmt has joined #openstack-telemetry | 16:02 | |
| sileht | dropping mysql will save me a lot of times | 16:04 |
| EmilienM | sileht: wait, dropping mysql from where? | 16:04 |
| *** mattyw has joined #openstack-telemetry | 16:04 | |
| sileht | everywhere ! we should use postgres | 16:05 |
| sileht | jd__, did you already got mysql error '(1060, u"Duplicate column name 'tablename'")', when the column the column doesn't exists ? | 16:06 |
| *** _nadya_ has joined #openstack-telemetry | 16:06 | |
| jd__ | sileht: that does not ring a bell | 16:06 |
| sileht | jd__, well the migration test pass with postgresql, but not with mysql with this error | 16:07 |
| jd__ | sileht: you got more details? | 16:08 |
| sileht | jd__, only the failed query: ALTER TABLE resource_type ADD COLUMN tablename VARCHAR(18)' | 16:09 |
| jd__ | is tablename a reserver keyword? | 16:09 |
| sileht | jd__, I have tried to rename it to tbname, same error | 16:09 |
| * sileht imagines the mysql errors handling: if err: raise Exception(random.sample(possible_errors, 1)[0]) | 16:10 | |
| *** r-mibu has left #openstack-telemetry | 16:24 | |
| sileht | the column doesn't exists only on the first ran test | 16:28 |
| datravis | Hi, I've submitted this bug report: https://bugs.launchpad.net/ceilometer/+bug/1539163 If I believe I have a change-set that will address it, should I go ahead and submit it? | 16:28 |
| openstack | Launchpad bug 1539163 in Ceilometer "Aggregation transformer sometimes produces incorrect values" [Undecided,New] | 16:28 |
| *** _nadya_ has quit IRC | 16:28 | |
| *** ddieterly has quit IRC | 16:31 | |
| *** belmoreira has quit IRC | 16:31 | |
| jd__ | datravis: sure | 16:32 |
| datravis | jd_: will do, thanks | 16:34 |
| *** liamji has quit IRC | 16:37 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Add some resource types tests https://review.openstack.org/270419 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute number https://review.openstack.org/270091 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute bool https://review.openstack.org/270418 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute uuid https://review.openstack.org/270090 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Don't create Ceilometer resource types by default. https://review.openstack.org/270322 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move legacy Ceilometer resource into indexer. https://review.openstack.org/270266 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Move resource type into their own sql table https://review.openstack.org/269843 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource type CRUD. https://review.openstack.org/269844 | 16:38 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute string https://review.openstack.org/269888 | 16:38 |
| *** datravis has quit IRC | 16:47 | |
| gordc | pradk: https://github.com/openstack/ceilometer/blob/juno-eol/ceilometer/objectstore/swift_middleware.py | 16:52 |
| pradk | gordc, cool | 16:53 |
| pradk | thx | 16:53 |
| gordc | np. :) | 16:53 |
| gordc | i think it's pretty broken and slows the swift apparently | 16:53 |
| gordc | which is why we changed to ceilometermiddleware which is more lightweight | 16:53 |
| pradk | gordc, so did we just move the code to ceilometermiddleware in kilo? | 16:54 |
| gordc | in theory | 16:54 |
| gordc | no. it's different | 16:54 |
| pradk | ah ok | 16:54 |
| gordc | ceilometermiddleware builds a notification with cadf payload adn that's it | 16:54 |
| gordc | legacy version did taht and sent like 3 or four samples via a pipeline which may or may not transform them.. | 16:55 |
| *** datravis has joined #openstack-telemetry | 16:55 | |
| gordc | basically it was middleware with pipeline... so it was publishing a bunch of samples | 16:55 |
| pradk | gordc, understood, thx | 16:56 |
| *** yassine__ has quit IRC | 16:57 | |
| *** thorst has quit IRC | 17:05 | |
| *** thorst has joined #openstack-telemetry | 17:05 | |
| *** jaypipes has quit IRC | 17:06 | |
| *** diogogmt has quit IRC | 17:06 | |
| *** cdent has quit IRC | 17:07 | |
| gordc | yay i merged a project-config patch and it didn't break everything (yet) | 17:08 |
| *** thorst has quit IRC | 17:10 | |
| openstackgerrit | Dan Travis proposed openstack/ceilometer: Adds timestamp option to Aggregation transformer https://review.openstack.org/273672 | 17:15 |
| openstackgerrit | gordon chung proposed openstack/ceilometer: fix locking in ceilometer https://review.openstack.org/272728 | 17:15 |
| *** diogogmt has joined #openstack-telemetry | 17:24 | |
| *** permalac has quit IRC | 17:25 | |
| *** efoley_ has quit IRC | 17:32 | |
| *** thorst has joined #openstack-telemetry | 17:32 | |
| *** thorst_ has joined #openstack-telemetry | 17:34 | |
| *** thorst has quit IRC | 17:36 | |
| *** safchain has quit IRC | 17:39 | |
| *** thorst_ has quit IRC | 17:39 | |
| *** thorst has joined #openstack-telemetry | 17:49 | |
| EmilienM | sileht, jd__: is it better now? http://logs.openstack.org/98/272298/11/check/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/77a3fde/logs/etc/gnocchi/api-paste.ini.txt.gz | 17:50 |
| jd__ | EmilienM: no, that looks identical | 17:51 |
| jd__ | EmilienM: what did you change? | 17:51 |
| EmilienM | no of course I did wrong | 17:52 |
| EmilienM | I need to change [pipeline:main] | 17:52 |
| EmilienM | not [pipeline:gnocchi+auth] | 17:53 |
| EmilienM | I put the pipeline in the wrong place with puppet -- fixing it | 17:53 |
| EmilienM | jd__: quick questions, I want to Aodh using Gnocchi, do you have an example of config? | 17:54 |
| *** liusheng has quit IRC | 17:54 | |
| jd__ | ah right :) | 17:54 |
| jd__ | hum good question | 17:54 |
| jd__ | i have it in devstack! | 17:54 |
| *** liusheng has joined #openstack-telemetry | 17:54 | |
| EmilienM | that works for me | 17:55 |
| EmilienM | jd__: you have a CI job I might want to look at? | 17:55 |
| EmilienM | so I'll copy coping | 17:55 |
| EmilienM | config* | 17:55 |
| jd__ | EmilienM: https://github.com/openstack/aodh/blob/master/devstack/plugin.sh#L218 see how complicated it is | 17:55 |
| EmilienM | yeah? | 17:56 |
| EmilienM | lol that's cool | 17:56 |
| EmilienM | jd__: which component of aodh will use it? | 17:56 |
| EmilienM | api? | 17:56 |
| jd__ | the evaluator mainly | 17:56 |
| jd__ | I'm not sure if the other uses it | 17:56 |
| EmilienM | I need to know for the puppet class | 17:57 |
| EmilienM | if I setup the param for this servic only or for all services | 17:57 |
| jd__ | hum don't bother | 17:57 |
| jd__ | it's *always* all services in our projects | 17:57 |
| jd__ | we don't have a conff file per service or whatever | 17:57 |
| jd__ | KISS :) | 17:57 |
| * jd__ is sad when he thinks about Glance for that | 17:58 | |
| EmilienM | jd__: I see in the code, API is using the param, isn't? | 17:58 |
| EmilienM | in aodh/api/controllers/v2/alarm_rules/gnocchi.py | 17:58 |
| EmilienM | maybe I'm wrong | 17:58 |
| jd__ | I think it is to see if Gnocchi is enabled | 17:58 |
| jd__ | that reminds me something ye | 17:59 |
| jd__ | p | 17:59 |
| EmilienM | I'll enable it only for evaluator for now, maybe move it later | 17:59 |
| EmilienM | we use a single config file no worries | 17:59 |
| EmilienM | except for glance | 17:59 |
| jd__ | that sounds a bad idea but your call | 17:59 |
| jd__ | especially because you can't know if we will change that usage later in another component | 18:00 |
| jd__ | you're just putting more work on your shoulders | 18:00 |
| jd__ | or is there a good reason? | 18:00 |
| EmilienM | ok | 18:00 |
| *** leitan has quit IRC | 18:00 | |
| *** nicodemus_ has quit IRC | 18:00 | |
| EmilienM | I'll put it in init.pp, which means this param will be for all nodes running aodh | 18:00 |
| jd__ | perfect | 18:00 |
| jd__ | that's how all our projects should be configured AFAIK :) | 18:00 |
| jd__ | simple! | 18:00 |
| *** thorst has quit IRC | 18:03 | |
| *** thorst has joined #openstack-telemetry | 18:03 | |
| *** julim has quit IRC | 18:04 | |
| EmilienM | jd__: so alarms will be stored in gnocchi? | 18:06 |
| jd__ | EmilienM: oh no | 18:06 |
| jd__ | it's stored in SQL or whatever you configured | 18:06 |
| *** julim has joined #openstack-telemetry | 18:06 | |
| EmilienM | ok | 18:06 |
| jd__ | Gnocchi is used to evaluate the alarms for threshold etc | 18:06 |
| EmilienM | oh yeah it makes sense | 18:06 |
| jd__ | is cpu.util > 90? etc :) | 18:06 |
| EmilienM | ok | 18:07 |
| *** thorst has quit IRC | 18:08 | |
| EmilienM | I think I got the scenario working | 18:08 |
| EmilienM | jd__: I'll let you review once it works | 18:08 |
| EmilienM | review configuration files and make sure we test the right things | 18:08 |
| EmilienM | jd__: the patch will be https://review.openstack.org/272298 if you want to monitor | 18:10 |
| EmilienM | jd__: thx for your helpp | 18:11 |
| *** boris-42 has quit IRC | 18:13 | |
| *** pradk has quit IRC | 18:20 | |
| *** jdandrea has quit IRC | 18:27 | |
| *** jdandrea has joined #openstack-telemetry | 18:28 | |
| *** pradk has joined #openstack-telemetry | 18:34 | |
| *** nicodemus_ has joined #openstack-telemetry | 18:55 | |
| jd__ | EmilienM: you're welcome | 19:08 |
| openstackgerrit | gordon chung proposed openstack/ceilometer: rolling upgrades https://review.openstack.org/273081 | 19:09 |
| *** mgarza_ has quit IRC | 19:09 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: rewriting history https://review.openstack.org/273049 | 19:10 |
| *** mgarza_ has joined #openstack-telemetry | 19:10 | |
| EmilienM | ok, now tempest is failing http://paste.openstack.org/show/ynZzB68HXOHPchUeETUx/ | 19:10 |
| EmilienM | I guess I'm missing something in tempest.conf | 19:10 |
| EmilienM | gordc: ^ | 19:12 |
| EmilienM | TelemetryNotificationAPITestJSON is failing because i'm using Gnocchi | 19:12 |
| gordc | errr... none of those tests will work with ceilometer+gnocchi | 19:12 |
| EmilienM | http://www.nooooooooooooooo.com/ | 19:13 |
| EmilienM | so I did all of that for nothing | 19:13 |
| gordc | i don't know what you did? | 19:13 |
| gordc | i told you we don't have tempest. | 19:13 |
| gordc | not for gnocchi | 19:13 |
| EmilienM | gordc: modifying our telemetry scenario to use gnocchi | 19:14 |
| gordc | those tests won't make any sense in gnocchi api. sorry. | 19:14 |
| EmilienM | but tempest.api.telemetry.test_alarming_api_negative.TelemetryAlarmingNegativeTest.test_get_non_existent_alarm works | 19:14 |
| EmilienM | only telemetry.test_telemetry_notification_api fails | 19:15 |
| gordc | works or skipped? | 19:15 |
| EmilienM | works | 19:15 |
| EmilienM | https://jenkins07.openstack.org/job/gate-puppet-openstack-integration-scenario001-tempest-dsvm-centos7/178/consoleFull | 19:15 |
| gordc | ceilometer alarms api redirects to aodh | 19:15 |
| gordc | because it's a straight move | 19:15 |
| EmilienM | I use gnocchi+aodh | 19:15 |
| EmilienM | and it does not fail in tempest | 19:15 |
| gordc | ceilometer metering db is not related to gnocchi at all | 19:15 |
| gordc | tempest.api.telemetry.test_alarming_api_negative.TelemetryAlarmingNegativeTest.test_get_non_existent_alarm uses gnocchi? | 19:16 |
| EmilienM | gordc: I don't know | 19:16 |
| EmilienM | I guess not? | 19:16 |
| gordc | i think that really only hits ceilometer api... which points to aodh. and that's not it. | 19:16 |
| gordc | sorry. 'that is it' ignore 'not' | 19:16 |
| EmilienM | ok | 19:16 |
| EmilienM | no problem, I'm already happy to have deployed gnocchi + ceilo + aodh with puppet :-) we'll just disable the config for ceilometer and keep it for aodh only | 19:17 |
| EmilienM | I mean, we'll disable the dispatcher | 19:17 |
| gordc | if you have a test that actually tests for an alarm to trigger, than it's aodh+gnocchi | 19:17 |
| gordc | EmilienM: if you disable dispatcher, you disable gnocchi? or you write from another source? | 19:18 |
| EmilienM | test_alarming_api do not fail with gnocchi so I guess it works | 19:18 |
| EmilienM | gordc: I'll use "database" dispatcher for ceilometer | 19:18 |
| EmilienM | the goal is to test test_telemetry_notification_api in tempest and it fails when using meter_dispatchers = gnocchi in ceilometer.conf | 19:18 |
| openstackgerrit | gordon chung proposed openstack/ceilometer: cleanup devstack plugin for gnocchi https://review.openstack.org/270818 | 19:19 |
| gordc | test_alarming_api probably doesn't call gnocchi... just ability to do crud events | 19:19 |
| gordc | as you wish. i'm pretty sure none of the non-alarm, non-event tests will pass for gnocchi | 19:20 |
| EmilienM | I just want *something* in Puppet CI that says "Gnocchi is configured by Puppet and it's working for at least one service" | 19:20 |
| EmilienM | I know you don't plan to patch tempest, on my side I'm just truing to deploy your software and test it the same way we test other services | 19:21 |
| EmilienM | so i'll skip ceilometer <-> gnocchi because it makes tempest failing now. | 19:21 |
| gordc | EmilienM: you can look at this https://review.openstack.org/#/q/status:open+project:openstack/ceilometer+branch:master+topic:tempest2 | 19:22 |
| gordc | it might speed up your ceilometer+gnocchi tempest support. | 19:22 |
| EmilienM | wow nice | 19:22 |
| gordc | i said 'might' | 19:22 |
| gordc | the idea is we will have more control here so we don't need to wait on bugging random qa folks... or just jordan. | 19:23 |
| EmilienM | ok | 19:23 |
| gordc | EmilienM: we need this too https://review.openstack.org/#/c/255161/ | 19:23 |
| EmilienM | gordc: do I still need to configure archive policy & archive policy rule if I only use Gnocchi / Aodh ? | 19:25 |
| EmilienM | I remember I needed to create them when running ceilometer with gnocchi | 19:25 |
| gordc | archive policy only relates to gnocchi | 19:26 |
| *** thorst has joined #openstack-telemetry | 19:26 | |
| EmilienM | ok | 19:27 |
| EmilienM | gordc: can you do a quick last round on https://review.openstack.org/272298 please? | 19:28 |
| gordc | EmilienM: yes, i am a puppet expert | 19:30 |
| EmilienM | gordc: no, but some things might be obvious to say "it's wrong" | 19:31 |
| EmilienM | the commit message explains my motivation | 19:31 |
| gordc | i was being honest | 19:33 |
| gordc | i'm the next ptl remember? | 19:34 |
| EmilienM | lol | 19:34 |
| gordc | sorry team liaison | 19:35 |
| gordc | it has same acronym too! | 19:35 |
| EmilienM | gordc: that's not a bad name, indeed | 19:36 |
| *** _nadya_ has joined #openstack-telemetry | 19:36 | |
| EmilienM | gordc: you should do politics | 19:36 |
| gordc | everyone can be happy. we change the lead part. and enterprise still see ptl | 19:36 |
| gordc | EmilienM: i do. every damn day. *sigh* | 19:36 |
| gordc | i deal with you, deal with cdent... all these damn companies. | 19:37 |
| *** guillaume_ has quit IRC | 19:45 | |
| *** guillaume_ has joined #openstack-telemetry | 20:00 | |
| jd__ | i'm eager to check this mail about ceilometer/pbr and then laugh at global-req again | 20:01 |
| jd__ | Phucking Team Liaison | 20:03 |
| jd__ | that'd be nice for PTL :) | 20:03 |
| gordc | you can make pins... | 20:09 |
| gordc | i'm changing my linkedin. change starts here. | 20:09 |
| openstackgerrit | Merged openstack/python-ceilometerclient: improve readme contents https://review.openstack.org/235208 | 20:09 |
| *** vishwanathj_zzz is now known as vishwanathj | 20:19 | |
| *** thorst has quit IRC | 20:46 | |
| *** thorst has joined #openstack-telemetry | 20:47 | |
| *** thorst has quit IRC | 20:55 | |
| *** _nadya_ has quit IRC | 20:57 | |
| *** rcernin has quit IRC | 21:02 | |
| *** nicodemus_ has quit IRC | 21:06 | |
| *** nicodemus_ has joined #openstack-telemetry | 21:07 | |
| *** leitan has joined #openstack-telemetry | 21:10 | |
| *** nicodemus_ has quit IRC | 21:11 | |
| *** jwcroppe_ has quit IRC | 21:16 | |
| *** jwcroppe has joined #openstack-telemetry | 21:17 | |
| *** jwcroppe_ has joined #openstack-telemetry | 21:21 | |
| *** jwcroppe has quit IRC | 21:21 | |
| *** rcernin has joined #openstack-telemetry | 21:28 | |
| *** nicodemus_ has joined #openstack-telemetry | 21:31 | |
| *** AndChat|224721 has joined #openstack-telemetry | 21:38 | |
| *** shardy has quit IRC | 21:38 | |
| *** nicodemus_ has quit IRC | 21:41 | |
| *** datravis has quit IRC | 21:43 | |
| *** nicodemus_ has joined #openstack-telemetry | 21:45 | |
| *** AndChat|224721 has quit IRC | 21:48 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: test coordination https://review.openstack.org/273774 | 21:56 |
| *** datravis has joined #openstack-telemetry | 21:59 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: do not configure worker specific items in init https://review.openstack.org/273792 | 22:23 |
| stevelle | anyone able to help me figure out what is wrong with my gnocchi.conf file here? http://paste.openstack.org/show/9j9d2fcFrqJu0QuStEiu/ | 22:34 |
| *** nicodemus_ has quit IRC | 22:35 | |
| *** nicodemus_ has joined #openstack-telemetry | 22:35 | |
| *** julim has quit IRC | 22:50 | |
| gordc | stevelle: sorry, never seen that before. what's your apache template look like? or gnocchi.conf? | 22:50 |
| stevelle | gordc: /var/www/cgi-bin/gnocchi/gnocchi-api looks like this: http://paste.openstack.org/show/QLF2nX0IFXJ8DXnIQk1S/ | 22:51 |
| stevelle | that seems to be where the problem is. | 22:51 |
| stevelle | and this seems like something that changed in the last 2 weeks maybe? | 22:51 |
| *** julim has joined #openstack-telemetry | 22:52 | |
| stevelle | gordc: I figured out that it wasn't the gnocchi.conf itself by running under werkzeug | 22:53 |
| gordc | i don't get it. what's all the execfile stuff at top? https://github.com/openstack/gnocchi/blob/master/gnocchi/rest/app.wsgi | 22:55 |
| stevelle | gordc: running under a virtualenv | 22:55 |
| gordc | stevelle: what does your api-paste.ini file look like? | 22:59 |
| gordc | the only thing about api i remember changing is we disabled keystone by default. | 22:59 |
| stevelle | gordc: I might have an older template then if that changed. http://paste.openstack.org/show/wtwYUxWOQdi9GFsa3wnd/ | 23:00 |
| stevelle | only thing I modded there was adding healthcheck | 23:00 |
| gordc | stevelle: this is the patch https://review.openstack.org/#/c/259367/10/doc/source/configuration.rst | 23:02 |
| stevelle | gordc: yeah, that doesn't seem like it should be the source of the issue for me | 23:03 |
| gordc | stevelle: yeah, it looks ok on my end too.. i'm not sure. never really seen that error before. | 23:05 |
| gordc | only other thing is a bad keystonemiddleware release since that's where the bug is failing. | 23:06 |
| *** guillaume_ has quit IRC | 23:07 | |
| stevelle | keystonemiddleware==4.1.0 | 23:07 |
| stevelle | my working environment from 2 weeks ago is keystonemiddleware==2.4.1 | 23:08 |
| gordc | https://github.com/openstack/requirements/blob/master/global-requirements.txt#L73 | 23:10 |
| gordc | 4.1.0 is broken. | 23:10 |
| *** julim has quit IRC | 23:11 | |
| gordc | stevelle: not really sure what is broken in it | 23:11 |
| *** nicodemus_ has quit IRC | 23:14 | |
| *** nicodemus_ has joined #openstack-telemetry | 23:14 | |
| gordc | actually i remember vague details on that bug and i think it was blocked because of performance and not any errors | 23:15 |
| stevelle | not seeing anything in the 4.2.0 release notes | 23:16 |
| stevelle | I found the perf. issue, but that is mitigated by using memcached, which I am | 23:16 |
| stevelle | keystoneauth1==2.2.0 which is the latest | 23:17 |
| *** guillaume_ has joined #openstack-telemetry | 23:18 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: rewriting history https://review.openstack.org/273049 | 23:21 |
| *** prashantD has joined #openstack-telemetry | 23:22 | |
| *** AndChat|224721 has joined #openstack-telemetry | 23:29 | |
| *** AndChat-224721 has joined #openstack-telemetry | 23:30 | |
| *** nicodemus_ has quit IRC | 23:32 | |
| *** chlong has quit IRC | 23:32 | |
| *** AndChat|224721 has quit IRC | 23:34 | |
| *** gordc has quit IRC | 23:38 | |
| *** mattyw has quit IRC | 23:39 | |
| *** nicodemus_ has joined #openstack-telemetry | 23:41 | |
| *** AndChat-224721 has quit IRC | 23:41 | |
| *** zqfan has joined #openstack-telemetry | 23:52 | |
| *** leitan has quit IRC | 23:52 | |
| *** mgarza_ has quit IRC | 23:56 | |
| *** guillaume_ has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!