HPA collects several metrics used to monitor the scheduled tasks (ST) executions.
Some of them are similar to Calypso status and views, some of them are different.
Captured by the DataServer agent
- The metric name pattern is
Hpa|ScheduledTask|ST id @ ST external reference. - One example
Hpa|ScheduledTask|61002@HEX_TEST_LOADTRADES
Standard metrics are captured:
| Metric | Value | Semantics |
|---|---|---|
:Average Response Time (ms) |
duration of a finished execution | GAUGE |
|start:Status |
1 per execution started | DELTA |
|success:Status |
1 per execution succeeded | DELTA |
|failed:Status |
1 per execution failed (counted once per execution) | DELTA |
|stopped:Status |
1 per execution stopped | DELTA |
|not triggered:Status |
1 per execution not triggered | DELTA |
- the number of executions is the count of
:Average Response Time (ms)(metric type COUNT in a DPS) - see Metric semantics
Those metrics are captured after the ST execution and the DataServer notification relative to the ST execution.
See Metric semantics for what GAUGE/DELTA mean and how each is aggregated.
Captured by the ST agent
Depending on the monitoring profile used to monitor a ST, numerous metrics can be captured.
One of them can be specifically mentioned:
- The metric name is
Hpa|Infrastructure|Cpu:aliveProcessTime - The agent name is containing the ST identifier
- The ST duration in then monitored in live