Decision-first: One model per GPU (cloud-main + local-wake-filter for multi-model); unload-and-verify before every load; never lower the Docker Desktop VM cap; tunnel to loopback to dodge macOS Local Network Privacy; serialize loads and don’t download during inference.

Scope & freshness: Apple-Silicon Mac + minikube/Docker Desktop and a single-GPU LLM host (GB10), as of 2026-05-25. Incident patterns are durable; specific recovery commands assume kubectl/minikube/Docker Desktop.

A field runbook of failure modes seen running local LLMs next to development Kubernetes clusters. Each is a real incident pattern, not a hypothetical. (This whole doc is effectively a “what didn’t work” catalog — that’s the point.)

PostgreSQL Debugging#

When PostgreSQL breaks, it usually falls into a handful of patterns. This is a reference for diagnosing each one with specific queries and commands.

Connection Refused#

Work through these in order:

1. Is PostgreSQL running?

sudo systemctl status postgresql-16

2. Is it listening on the right address?

ss -tlnp | grep 5432

If it shows 127.0.0.1:5432 but you need remote access, set listen_addresses = '*' in postgresql.conf.

3. Does pg_hba.conf allow the connection? Check logs for no pg_hba.conf entry for host: