Orchestrator Coverage Design

Context

Both orchestrator files are core to the pipeline system but have coverage gaps:

• definition_orchestrator.py — 68% (validation edge cases, skip logic, error paths)
• orchestrator.py — 60% (resume, retry/backoff, fallback, error handlers)

Approach

Sequential per-file: definition_orchestrator first (simpler), then orchestrator (retry/backoff needs careful mocking). Pure test additions — no source changes needed.

definition_orchestrator.py (68% → 100%)

Validation tests

Test	Lines
No nodes defined	75
Missing node id	83
Duplicate node id	86
Missing node type	90
Handler validation exception	103-105
Unknown next node reference	113

Execution tests

Test	Lines
_should_skip with skip_if_errors	195-200
_should_skip with skip_if_condition .has_errors	362-370
Ingest node propagates incident_id	219-222
Required node exception → pipeline failed	240-251
Outer exception handler	265-274
_execute_node exception re-raises with stage record	328-334

orchestrator.py (60% → 100%)

Resume tests

Test	Lines
resume_pipeline run not found	216-219
resume_pipeline wrong status	221-222
resume_pipeline success	224-225

Execution tests

Test	Lines
Skip completed stages on resume	274-284
ANALYZE with fallback_used continues	337-342
Generic exception handler	389-405

Retry tests (mock time.sleep)

Test	Lines
StageExecutionError retry + backoff	499-521
Generic exception retry + backoff	523-544
Unreachable safety net	547-550

Target

100% branch coverage on both files.