(Rocky) Telemetry alarm yields Insuff Data [closed]

asked 2020-03-12 16:17:53 -0600

PQUIGS gravatar image

updated 2020-03-13 09:47:34 -0600

I am attempting a manual install/config of OpenStack Rocky Telemetry services (Ceilometer, Aodh, Gnocchi) following the Installation Guides, filling in the blanks with results from Google searches. I have a heat template that deploys 1 VM along with scaling policies & alarms. Eventually, I want to expand this to multiple VMs front-ended by a load balancer (Octavia). <btw -="" i="" can="" do="" already="" do="" that="" in="" my="" devstack="" environment="">

My alarms are stuck in "Insufficient Data" state:

openstack alarm list
+--------------------------------------+--------------------------------------------+-----------------------------------+-------------------+----------+---------+
| alarm_id                             | type                                       | name                              | state             | severity | enabled |
+--------------------------------------+--------------------------------------------+-----------------------------------+-------------------+----------+---------+
| 96a08963-ff0e-4277-a9aa-388313068040 | gnocchi_aggregation_by_resources_threshold | PQASG-cpu_alarm_low-hwqqbjfhhyhh  | insufficient data | low      | True    |
| 878ed5ca-dd21-46d6-a392-9f59fc885a39 | gnocchi_aggregation_by_resources_threshold | PQASG-cpu_alarm_high-6chchkxf7hxp | insufficient data | low      | True    |
+--------------------------------------+--------------------------------------------+-----------------------------------+-------------------+----------+---------+

I have looked at the log files for Ceilometer, Gnocchi, and Aodh. But, I can't spot the error. Any help is very much appreciated!!

Here are some messages from those log files: In aodh-evaluator.log:

2020-03-12 19:37:22.192 3129 INFO aodh.evaluator [-] initiating evaluation cycle on 2 alarms
2020-03-12 19:37:22.192 3129 DEBUG aodh.evaluator [-] evaluating alarm 96a08963-ff0e-4277-a9aa-388313068040 _evaluate_alarm /usr/lib/python2.7/dist-packages/aodh/evaluator/__init__.py:263
2020-03-12 19:37:22.192 3129 DEBUG aodh.evaluator.threshold [-] query stats from 2020-03-12 19:31:22.192591 to 2020-03-12 19:37:22.192591 _bound_duration /usr/lib/python2.7/dist-packages/aodh/evaluator/threshold.py:71
2020-03-12 19:37:22.611 3129 WARNING aodh.evaluator.threshold [-] Expecting 5 datapoints but only get 0: InsufficientDataError: aggregation mean does not exist for at least one metric of the query
2020-03-12 19:37:22.614 3129 INFO aodh.evaluator [-] alarm 96a08963-ff0e-4277-a9aa-388313068040 transitioning to insufficient data because aggregation mean does not exist for at least one metric of the query

In gnocchi-metricd.log

<messages, but none related to this>

In ceilometer--agent-compute.log:

2020-03-12 18:23:52.540 3168 INFO ceilometer.polling.manager [-] Polling pollster cpu_util in the context of all_pollsters
2020-03-12 18:23:52.540 3168 DEBUG ceilometer.compute.pollsters [-] 79ae4228-cc45-4e03-ab6e-17be528a149f/cpu_util volume: Unavailable _stats_to_sample /usr/lib/python2.7/dist-packages/ceilometer/compute/pollsters/__init__.py:113
2020-03-12 18:23:52.540 3168 WARNING ceilometer.compute.pollsters [-] cpu_util statistic in not available for instance 79ae4228-cc45-4e03-ab6e-17be528a149f: NoVolumeException

In ceilometer-agent-notification.log:

2020-03-12 20:39:21.866 3138 DEBUG ceilometer.publisher.gnocchi [-] Resource instance-0000001a-cc55881f-498e-4694-9ee4-ace67ea3e2b1-tapf544b079-e3 updated _update_resource /usr/lib/python2.7/dist-packages/ceilometer/publisher/gnocchi.py:448
2020-03-12 20:39:21.870 3138 DEBUG ceilometer.publisher.gnocchi [-] 1 measures posted against 1 metrics through 1 resources batch_measures /usr/lib/python2.7/dist-packages/ceilometer/publisher/gnocchi.py:440
2020-03-12 20:39:21.873 3138 DEBUG ceilometer.publisher.gnocchi [-] Resource cc55881f-498e-4694-9ee4-ace67ea3e2b1 updated _update_resource /usr/lib/python2.7/dist-packages/ceilometer/publisher/gnocchi.py:448
2020-03-12 20:39:21.873 3138 DEBUG ceilometer.pipeline.sample [-] Pipeline cpu_sink: Transform sample <name: cpu, volume: 223347566871, resource_id: cc55881f-498e-4694-9ee4-ace67ea3e2b1, timestamp: 2020-03-12T20:39:21.035988> from 0 transformer _publish_samples /usr/lib/python2.7/dist-packages/ceilometer/pipeline/sample.py:149
2020-03-12 20:39:21.874 3138 DEBUG ceilometer.transformer.conversions [-] handling sample <name: cpu, volume: 223347566871, resource_id: cc55881f-498e-4694-9ee4-ace67ea3e2b1, timestamp: 2020-03-12T20:39:21.035988> handle_sample /usr/lib/python2.7/dist-packages/ceilometer/transformer/conversions.py:186
2020-03-12 20:39 ...
(more)
edit retag flag offensive reopen merge delete

Closed for the following reason the question is answered, right answer was accepted by PQUIGS
close date 2020-04-02 10:58:02.195890