Skip to main content
Proton v5

Indicators for a period

GET 

/v1/metrics

The operational indicators of the runs finished in the period (by their end), in total and grouped by groupBy. The default period is the last 30 days, and the maximum is 366.

  • Rates: success and failure over the total runs (a stopped run counts in the total).
  • Failure: business error, technical error, environment failure, not dispatched and runner lost.
  • Duration: average, p95 and maximum, without the runs that were not dispatched.
  • Reprocessings: the batch's automatic re-run and the re-run from the exception queue.
  • MTTR: from a dataset's failure to its next success, in the period or later. open counts the failures still without a success. A stopped run neither opens nor closes.
  • Exceptions: opened and resolved in the period, pending now and the average time to resolve. Null without the exception queue or without its permission.
  • Runners: availability and downtime of the shared runners, from the down and back events. Personal and cloud runners are left out. eventsSince says since when there are events. Null without the runners' permission.
  • Items processed: reserved (null) for the work queue.

Request​

Responses​

The indicators.