Celery

Why Celery Here?

Diagnostics lab workflows often require work that should not block the HTTP request:

• result enrichment (flags/interpretation)
• report generation
• sending notifications
• periodic QC summaries

Celery demonstrates reliable background processing with a separate worker process.

Configuration

Settings are read from env (see .env.example):

• CELERY_BROKER_URL
• CELERY_RESULT_BACKEND
• CELERY_ALWAYS_EAGER (useful for tests/local experiments)

Implementation:

• diagnostics_lab/celery_app.py

Tasks

Implemented tasks:

• diagnostics_lab.tasks.process_result(result_id)
• computes TestResult.interpretation

• updates LabOrder.status (roll-up)

• diagnostics_lab.tasks.daily_qc_summary()

• periodic example task configured via Celery Beat

Running

With Docker Compose:

• worker: worker service
• beat: beat service

Optional monitoring:

• docker compose --profile flower up --build