I have configured a CPU threshold alarm condition. I would like to make an API call to ceilometer to get the following information - the current CPU util value that crossed the threshold, the threshold value, the time when the threshold was crossed, the instance, etc.

Here are my settings - a) Alarm condition: cpu_util > 0.03 during 2 x 65s; alarm action: log b) Evaluation interval (in ceilometer.conf) set to 70s. c) Pipeline timers (in pipeline.yaml) set to 60s. d) I'm running Juno.

My understanding is evaluation interval (70s) should be greater than pipeline interval (60s) and alarm period (65s).

When I run ceilometer alarm-list command, I see the state as "alarm". It transitions from "insufficient data".

I have the following questions - 1. When the threshold crosses, where do I look to confirm that an alert is getting raised? 2. Do my 3 timer settings make sense? 3. What specific API call will give me the information I'm looking for? GET v2/alarms gives me the threshold value but not the actual CPU utilization value.

Thanks.