Dockerfile

@@ -29,4 +29,8 @@ EXPOSE 8892
 HEALTHCHECK --interval=30s --timeout=3s --start-period=10s --retries=3 \
   CMD python -c "import urllib.request,sys; sys.exit(0 if urllib.request.urlopen('http://localhost:8892/healthz').status==200 else 1)" || exit 1
 
-CMD ["uvicorn", "analytics_mcp:app", "--host", "0.0.0.0", "--port", "8892", "--workers", "2"]
+# Single worker: this is a low-traffic read-only proxy for a handful of analysts, and
+# the DB connection budget = workers × MCP_POOL_MAX. One worker (× default pool 8) caps
+# the MCP at 8 backends instead of 16, which matters on a shared 100-connection DB.
+# Scale out by raising MCP_POOL_MAX, not workers, so the budget stays obvious.
+CMD ["uvicorn", "analytics_mcp:app", "--host", "0.0.0.0", "--port", "8892", "--workers", "1"]

scripts/analytics_ro_role.sql

@@ -62,3 +62,9 @@ ALTER DEFAULT PRIVILEGES FOR ROLE postgres         IN SCHEMA fuel       GRANT EX
 ALTER ROLE analytics_ro SET default_transaction_read_only = on;
 ALTER ROLE analytics_ro SET statement_timeout = '30s';
 ALTER ROLE analytics_ro SET idle_in_transaction_session_timeout = '60s';
+-- Cap idle POOLED connections too (these sit idle but NOT in a transaction, so the
+-- idle_in_transaction timeout never reaps them). On a shared 100-connection DB this
+-- returns slots the MCP isn't using. Safe with the server's dead-connection handling:
+-- a reaped pooled connection is discarded + transparently retried, not surfaced as an
+-- error. (Requires PostgreSQL 14+.)
+ALTER ROLE analytics_ro SET idle_session_timeout = '5min';