Add service health checks with graceful degradation #32

New Issue

kert · 2026-03-12T17:42:28Z

kert commented

2026-03-12 17:42:28 +00:00

stack health / GET /health endpoint that checks:

DuckDB file exists and is readable
Bib SQLite accessible
Nessie catalog reachable (if configured)
Polaris catalog reachable (if configured)
Trino connection valid (if configured)
Databricks workspace reachable (if configured)

Each check independent. Report partial health when some services down.
CLI and API commands should degrade gracefully (e.g., skip lakehouse if Nessie is down).

stack health / GET /health endpoint that checks: - DuckDB file exists and is readable - Bib SQLite accessible - Nessie catalog reachable (if configured) - Polaris catalog reachable (if configured) - Trino connection valid (if configured) - Databricks workspace reachable (if configured) Each check independent. Report partial health when some services down. CLI and API commands should degrade gracefully (e.g., skip lakehouse if Nessie is down).

kert added this to the P8: Quality — 100% coverage, error handling, graceful degradation milestone 2026-03-12 17:42:28 +00:00

kert added the cli api quality labels 2026-03-12 17:42:28 +00:00

kert closed this issue

2026-03-12 22:29:49 +00:00

kert referenced this issue from a commit

2026-03-12 23:29:17 +00:00

add P8 quality: 99% coverage, health checks, snapshots, CI pipeline

Sign in to join this conversation.