You built (or you're about to build) a private AI assistant on your own box. Here's what nobody tells you: standing it up is the easy part. Keeping it running is the part that decides whether you still have an assistant in a month.This is the runbook for that part. It's Module 10 of Build Your Own Self-Hosted AI Assistant, and it's the chapter people come back to — so it stands on its own.What you get (7 pages, PDF):• "Healthy but silent" — the #1 failure: the service says active (running) while your assistant is quietly switched off. How to triage it in 30 seconds (ping first, always).• The box disappears — no ping, no SSH. Is it the machine, the network, or DNS? The decision tree that saves you an hour.• Empty replies / "the model seems broken" — when it's the provider, when it's your config, when it's an empty wallet wearing a "rate limit" mask.• Plus four more: auth/key failures, the boot-time race that auto-pauses your chat platform, runaway costs, and the backup you didn't test until you needed it.• Every failure: symptom → diagnosis → fix → how to prevent it. All from real production boxes, not theory.Who it's for: anyone running an LLM assistant on a home server, mini-PC, Pi, or VPS who wants the "when it breaks" answers before it breaks.This is 1 of 11 modules. If you want the whole build — stand it up, give it a personality, wire it to your files/calendar/email, schedule it, back it up — grab The Guide or The Guide + Kit. This runbook's price comes off if you upgrade.Instant download. PDF. Yours forever. Tax handled at checkout.