Get Batch Diagnostics
Get comprehensive diagnostics for a batch.
Combines batch status, task progress, collection info, performance metrics, and actionable insights into a single response for easy frontend rendering.
Use Cases:
- Monitor batch processing in real-time
- Debug failed batches
- View performance breakdown after completion
- Get actionable next steps
Response includes:
- Overall batch status and progress
- Per-tier task details with Ray job links
- Collection document counts
- Performance insights and bottlenecks (if completed)
- Error details (if failed)
- Recommended next actions
Example:
GET /v1/analytics/performance/batches/{batch_id}/diagnostics
Perfect for:
- Real-time progress tracking UI
- Batch monitoring dashboards
- Debugging failed extractions
- Performance optimization
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
Response
Successful Response
Comprehensive batch diagnostics response.
Combines batch status, task progress, collection info, and performance insights into a single response for easy frontend rendering.
Batch ID
Batch name
Overall batch status
Source bucket ID
Current tier being processed
Total number of tiers
Overall progress percentage (0-100)
When batch was created
When batch was submitted
When processing started
When processing completed
Total duration in seconds
Estimated completion time
Diagnostic info for each tier
Status of target collections
Performance metrics summary (available after completion)
Performance insights and recommendations
Whether batch has any failures
Number of failed tiers
Derived health signal: 'ok', 'scaling' (PROCESSING with no progress yet but still inside the GPU cold-start / cluster-provisioning window — a scale-from-zero node and image pull can take ~20 min, so this is expected, not stuck), 'stuck' (PROCESSING but no progress and the active tier's job has been PENDING past the cold-start window — its driver/worker may be unschedulable), or 'degraded' (has partial failures).
Human-readable reason the batch appears stuck/blocked, or — for health='scaling' — why it is legitimately waiting (e.g. 'cluster_cold_start: GPU workers scaling from zero'). None when health is 'ok' or 'degraded'.
How long the batch has been making no progress while PROCESSING, in seconds. None when not stuck.
Total objects in batch
Recommended next steps for user

