Bulk Data Export

HFS implements the FHIR Bulk Data Access $export family asynchronously: kick-off, poll, manifest, download, delete.

Endpoints

Operation	Method	URL
system kick-off	GET/POST	/$export
patient kick-off	GET/POST	/Patient/$export
group kick-off	GET/POST	/Group/{id}/$export
status or manifest	GET	/export-status/{job_id}
cancel and delete	DELETE	/export-status/{job_id}
HFS-served download	GET	/export-file/{job_id}/{type}-{part}

All kick-offs require Prefer: respond-async. The default response is 202 Accepted with a Content-Location status URL.

Environment

Variable	Default	Description
HFS_BULK_EXPORT_ENABLED	true	Master switch; false returns 501 for all export endpoints
HFS_BULK_EXPORT_OUTPUT_BACKEND	local-fs	Output store: local-fs or s3
HFS_BULK_EXPORT_OUTPUT_DIR	${HFS_DATA_DIR}/exports	Local filesystem output root
HFS_BULK_EXPORT_S3_BUCKET	none	S3 bucket, required when output backend is s3
HFS_BULK_EXPORT_S3_ENDPOINT	AWS	S3-compatible endpoint URL, such as MinIO
HFS_BULK_EXPORT_S3_FORCE_PATH_STYLE	false	Path-style addressing for S3-compatible providers
HFS_BULK_EXPORT_REQUIRES_ACCESS_TOKEN	auto	Manifest posture: auto, true, or false; false is invalid with local-fs
HFS_BULK_EXPORT_FILE_URL_TTL	3600	Pre-signed download URL lifetime in seconds
HFS_BULK_EXPORT_OUTPUT_TTL	86400	Output retention after job completion in seconds
HFS_BULK_EXPORT_WORKER_CONCURRENCY	2	In-process worker pool size
HFS_BULK_EXPORT_DISABLE_LOCAL_WORKER	false	Disable in-pod workers for separate exporter deployments
HFS_BULK_EXPORT_MAX_CONCURRENT_PER_TENANT	4	Per-tenant active job cap; kick-off returns 429 if exceeded
HFS_BULK_EXPORT_BATCH_SIZE	1000	Resources per fetch_export_batch
HFS_BULK_EXPORT_LEASE_DURATION	60	Initial lease length in seconds; must exceed heartbeat interval
HFS_BULK_EXPORT_HEARTBEAT_INTERVAL	20	Worker heartbeat cadence in seconds
HFS_BULK_EXPORT_CLEANUP_INTERVAL	300	Cleanup scan interval in seconds
HFS_BULK_EXPORT_SINCE_NEWLY_ADDED	include	Group export _since toggle: include or exclude

Job-state storage reuses the same backend and connection pool that holds FHIR resources. SQLite deployments share ./data/hfs.db. PostgreSQL deployments share HFS_DATABASE_URL. There is no separate job-store configuration.

Bulk export is currently available on sqlite, postgres, sqlite-elasticsearch, and postgres-elasticsearch. Other backends return 501 until job-state implementations exist.

Single-instance Recipe

cargo run --bin hfs

This starts HFS with bulk export enabled, job state in the same SQLite database as FHIR resources, NDJSON output under ./data/exports/, and an in-process worker pool.

curl -H 'Prefer: respond-async' http://localhost:8080/Patient/\$export

Multi-instance Recipe

PostgreSQL plus S3 or MinIO:

HFS_STORAGE_BACKEND=postgres \
HFS_DATABASE_URL=postgresql://hfs:hfs@localhost/hfs \
HFS_BULK_EXPORT_OUTPUT_BACKEND=s3 \
HFS_BULK_EXPORT_S3_BUCKET=hfs-export \
HFS_BULK_EXPORT_S3_ENDPOINT=http://localhost:9000 \
HFS_BULK_EXPORT_S3_FORCE_PATH_STYLE=true \
HFS_BULK_EXPORT_REQUIRES_ACCESS_TOKEN=false \
cargo run --bin hfs --features postgres,s3

The full local stack is in docker/bulk-export/docker-compose.yml: HFS, Postgres, MinIO, and Keycloak. GitHub Actions does not use this compose file for bulk export tests. The manual conformance workflow is .github/workflows/inferno-bulk-data.yml.

Behavior Notes

• _typeFilter is parsed and applied.
• Unsupported result-control params inside _typeFilter are rejected with 400 regardless of Prefer: handling: _sort, _include, _revinclude, _count, _elements.
• _elements is implemented: subset to listed paths plus id, resourceType, and meta, with a SUBSETTED meta.tag added.
• Unsupported parameters includeAssociatedData, organizeOutputBy, and allowPartialManifests return 400 when Prefer: handling=strict is set. Without strict handling, or with lenient handling, they are ignored and a warning is logged.
• Group export _since late membership uses include by default, returning pre-_since resources for patients added after _since.
• exclude is reserved for a follow-up that requires group-membership-history tracking.
• Group export flattens nested Group/ members iteratively with a visited-set cycle guard.