get_run_summary

Reference for the get_run_summary tool of the SparkForensics MCP server.

App/stage/job/sql counts, duration, and how the run ended: no findings. failedJobs/totalJobs count only jobs with an end record, and failureReason is the first line of Spark's own recorded reason when a job failed, the same line the dashboard verdict quotes (failureReasonStageId is the stage it came from, or null when there is no reason or it came from a job's exception). runShape carries the dashboard's run-shape figures: wallClockMs, efficiencyPct (the share of the run with a stage running, the Scorecard's Efficiency), unusedCoreTimePct (the share of executor core time that ran no task, against the run's allocation, cores times the time each executor was alive; minEfficiencyPct checks 100 minus this), etlPhasesMs (extract/transform/load summed stage time, so a phase can exceed the run) and peakBusyCores (busy cores at the peak of Core Usage by Locality, averaged over one chart bucket, so it can be fractional). Each is null where the dashboard shows "Not measured" or "Unavailable".

Parameters: source, runId and redact, as in diagnose_run.

Example call:

{ "name": "get_run_summary", "arguments": { "source": { "path": "app-20260101.zstd" } } }

Example response:

{
  "runId": "3f9c2b7e-...",
  "app": { "id": "application_0000000000000_0001", "name": "t", "sparkVersion": null },
  "stageCount": 0,
  "jobCount": 0,
  "sqlExecutionCount": 0,
  "executorCount": { "added": 0, "removed": 0 },
  "durationMs": 100,
  "runComplete": true,
  "failedJobs": 0,
  "totalJobs": 0,
  "failureReason": null,
  "failureReasonStageId": null,
  "runShape": { "wallClockMs": 100, "efficiencyPct": null, "unusedCoreTimePct": null, "etlPhasesMs": null, "peakBusyCores": null }
}