JOBS

Job failure rate

A large share of the jobs in the application fail.

Outcome: Reliability
Scope: Judged on the whole run
Usual fix: A change in the job code
CI gate: No budget flag: gate it with --fail-on-introduced against a baseline

An example finding

Where you see it

Dashboard

The JOBS tag marks it in the action region of the dashboard, among the widgets that rank a fix. How to read it in the findings guide.

Agent (MCP)

An assistant can read this entry with the list_detectors tool and its background with get_finding_documentation. The call and the response below come from a real run of the MCP server.

Call

{
  "tool": "list_detectors",
  "arguments": {
    "type": "jobFailureRate"
  }
}

Response

{
  "count": 1,
  "detectors": [
    {
      "type": "jobFailureRate",
      "name": "job failure rate",
      "tag": "JOBS",
      "tagSharedWith": [],
      "family": "reliability",
      "pitch": "A large share of the jobs in the application fail.",
      "region": "action",
      "scope": "app",
      "fixEffort": "code",
      "thresholdSummary": "at least 10% of jobs failing",
      "detectors": [
        {
          "detector": "jobFailureRate",
          "version": 1,
          "thresholds": {
            "infoRate": 0.1,
            "warnRate": 0.3,
            "critRate": 0.5
          }
        }
      ],
      "pages": {
        "catalog": "docs/detectors/jobFailureRate.html",
        "guide": "docs/user-guide/understanding-findings.html#jobs"
      },
      "tuning": {
        "kind": "page",
        "anchor": "#bottleneck-job-failure-rate",
        "path": "docs/tuning-reference/bottleneck-job-failure-rate.html#bottleneck-job-failure-rate"
      },
      "gate": null
    }
  ]
}

get_finding_documentation call

{
  "tool": "get_finding_documentation",
  "arguments": {
    "type": "jobFailureRate"
  }
}

Its response, with each Markdown body replaced by its length

{
  "type": "jobFailureRate",
  "name": "Job Failure Rate",
  "detectionDoc": {
    "tag": "JOBS",
    "title": "Job failure rate",
    "content": "<176 characters of Markdown>"
  },
  "tuningDoc": {
    "anchor": "#bottleneck-job-failure-rate",
    "title": "Job Failure Rate",
    "content": "<6265 characters of Markdown>"
  }
}

CI (CLI)

List this entry from the command line. The output below is a real run of the command. Budgets and exit codes are in the CI guide.

sparkforensics-analyze detectors --type jobFailureRate --format md

# Detectors (1)

## Reliability

| Tag | Type | Scope | Fix | Gate | What it finds |
| --- | --- | --- | --- | --- | --- |
| JOBS | `jobFailureRate` | app | code |  | A large share of the jobs in the application fail. |

Report only this finding type for a run with --type:

sparkforensics-analyze <event-log> --type jobFailureRate --impact critical,warning,info

It has no budget flag. Fail a build on a new one by comparing against a baseline with --fail-on-introduced.

Airflow

The operator has no budget parameter for it. Use fail_on_introduced with a baseline.

check_spark_job = SparkForensicsOperator(
    task_id="check_spark_job",
    log_source=...,
    backend=SubprocessAnalyzeHook(),
    baseline_log_source=...,
    fail_on_introduced="critical",
)

The operator works per run, and its summary counts findings per impact band. See the operator for its parameters.

How it is measured

Thresholds

At least 10% of jobs failing.

jobFailureRate version 1

Threshold	Default
`infoRate`	0.1
`warnRate`	0.3
`critRate`	0.5

Each rule's impact bands and fixed fallbacks are in the threshold tables.

Override a default per run with --thresholds; see Tuning detector thresholds.

Docs

This finding has its own tuning page: Spark tuning reference.
Findings guide: JOBS in Understanding findings.