Healthcheck
filedge healthcheck probes the Audit DB and destination connector without creating tables or writing rows. Use it for Kubernetes liveness/readiness probes, uptime monitors, or a quick operator check before a scheduled Run.
The command exits 0 when both dependencies are reachable. It exits 1 when any check fails.
filedge run performs the same preflight before scanning files. If a dependency is unreachable, the Run exits before audit tables, destination tables, or audit rows are created:
JSON Output
Use --json for probes and monitors:
The command writes one JSON object to stdout:
{
"healthy": true,
"checks": [
{"name": "audit_db", "ok": true, "error": null, "latency_ms": 1.234},
{"name": "destination", "ok": true, "error": null, "latency_ms": 2.345}
]
}
Check names are stable:
| Name | Probe |
|---|---|
audit_db |
SELECT 1 against the Audit DB URL |
destination |
Connector-specific read-only round trip, such as SELECT 1 |
Kubernetes Example
readinessProbe:
exec:
command:
- filedge
- healthcheck
- --config
- /config/pipeline.yaml
- --audit-db-url
- "$(FILEDGE_AUDIT_DB_URL)"
- --json
initialDelaySeconds: 5
periodSeconds: 30
livenessProbe:
exec:
command:
- filedge
- healthcheck
- --config
- /config/pipeline.yaml
- --audit-db-url
- "$(FILEDGE_AUDIT_DB_URL)"
initialDelaySeconds: 10
periodSeconds: 60