Local AI Infrastructure Notes (6/15) — Mac Mini Home Server Troubleshooting: macOS Auto-Update

Root cause of remote access failure and a power/update configuration checklist for server-mode Macs

Summary

  • Mac Mini M4 home server became unreachable — cause: macOS software update triggered an automatic restart.
  • After restart, the system sat at the login screen → Tailscale and LaunchAgents did not start → extended downtime.
  • Fix: block automatic update installation (allow download only) + enable auto-login to prevent recurrence.

Background

The Mac Mini M4 runs as a headless home server connected via Tailscale VPN, with multiple agents and automation scripts running continuously. Stability is a hard requirement in this configuration.

When macOS reboots for an automatic update, LaunchAgents do not start until a user session is established. This brings down all user-space services — Tailscale VPN, SSH, agents — and leaves no remote path to diagnose the failure.

Body

1. Diagnosis

After gaining physical access to the machine, trace the cause. macOS logs power events and software update history in separate log channels.

① Check reboot history

last reboot

An unexpected reboot entry appears in the log. If you have no record of manually restarting the machine, that timestamp marks the start of the outage.

② Check power management log

pmset -g log | grep -i "restart\|shutdown\|sleep"

Look for a RestartCountdownOperationIdleSleep assertion from SoftwareUpdateNotificationManager. This confirms that the software update subsystem triggered the restart.

SoftwareUpdateNotificationManager is the assertion macOS registers with the power management subsystem when scheduling an update-install restart. It operates independently of pmset sleep suppression settings.

③ Reboot without a preceding shutdown

A reboot entry with no prior shutdown record is the signature of a macOS-forced update restart — no user involvement.

④ System halted at login screen after restart

This is the actual service failure. The restart itself is benign, but while the system waits for a password at the login screen:

  • Tailscale is not running → remote access unavailable
  • LaunchAgents are not running → all agents and schedulers stopped
  • The server is effectively down

In macOS, user-session services (LaunchAgents, user environment variables, Tailscale, etc.) only start after login. Without auto-login, the server cannot self-recover after a restart without physical intervention.

2. Root Cause Summary

Item Detail
Direct cause macOS software update automatic restart
Root cause System waiting at login screen after restart → user services not started
Existing configuration sleep=0, autorestart=1 — power settings were correct
Gap Software update restarts bypass the pmset mechanism

pmset power settings alone cannot prevent a software-update restart. macOS software updates trigger restarts through a separate code path.

3. Fix: Block Auto-Install + Enable Auto-Login

Core principle: do not disable automatic updates entirely — allow download, block installation (restart).

Keeping downloads enabled ensures security patches are ready, while you retain control over when installation occurs.

① Block automatic update installation (keep downloads enabled)

sudo defaults write /Library/Preferences/com.apple.SoftwareUpdate AutomaticallyInstallMacOSUpdates -bool false

sudo defaults write /Library/Preferences/com.apple.SoftwareUpdate AutomaticDownload -bool true

② Enable auto-login (recovery after any restart)

sudo defaults write /Library/Preferences/com.apple.loginwindow autoLoginUser "username"

Even if a manually installed update causes a restart, auto-login ensures Tailscale and LaunchAgents start immediately. The server recovers without physical intervention.

4. Server-Mode Mac Power & Update Configuration Checklist

Setting Value Command
Disable sleep 0 sudo pmset -a sleep 0
Auto-restart after power loss on sudo pmset -a autorestart 1
Wake on LAN on sudo pmset -a womp 1
Automatic update download on sudo defaults write ...SoftwareUpdate AutomaticDownload -bool true
Automatic update installation off sudo defaults write ...SoftwareUpdate AutomaticallyInstallMacOSUpdates -bool false
Auto-login on sudo defaults write ...loginwindow autoLoginUser "username"

5. Diagnostic Commands

Commands to trace the cause when the server stops responding.

last reboot

pmset -g log

uptime

sysctl kern.boottime

Caveats

  • pmset settings alone are not sufficient. Power management and software updates are separate mechanisms. Both must be configured.

  • Disabling automatic updates entirely is a security risk. Allowing downloads while blocking installation is the right balance. Apply updates manually on a regular schedule.

  • Auto-login is a security trade-off. Acceptable in physically controlled environments; not recommended for servers with external exposure.

Closing

Infrastructure failures in server operation more often originate from OS-level settings than from code bugs. This was the second infrastructure incident in a short span, following an OAuth token expiry.

The shared lesson is the same: the goal is not a system that never fails, but one that self-recovers after failure. A single auto-login setting is all it takes to make the server come back on its own after any restart.

Series overview: Series index

๋Œ“๊ธ€

๋Œ“๊ธ€ ์“ฐ๊ธฐ

์ด ๋ธ”๋กœ๊ทธ์˜ ์ธ๊ธฐ ๊ฒŒ์‹œ๋ฌผ

Agent Memory Engine (2/10) — Building an AI Agent Memory System with SQLite Alone

← 1/10 SQLite Memory Engine f… ๐Ÿ“š Series Index 3/10 memcore Completed → A memory engine that runs without daemons, without dependencies, anywhere ํ•ต์‹ฌ ์š”์•ฝ This post covers how to design an AI agent memory engine (memcore) on a single-file SQLite backend. - Structure: 25 modules, ~6,464 lines, 22 DB tables - Principles: no daemon, single-file portability, no required external dependencies, host-agnostic - Core techniques: 4-layer hybrid search (topic · vector · FTS5 · LIKE), U-tag dialectic-based confidence evolution, ingestion-layer bias rejection rules What This Post Covers One approach to building AI agent memory without a server or cloud vector DB. Specifically: (1) the limitations of text-file-based memory, (2) vector DB alternatives compared and selection criteria, (3) how to layer search tiers, (4) how to evolve confidence scores over time, and (5) how to block biased memory accumulation at the ingestion stage. Design Principles: Why Single-File SQLite Text-file-based ...
Read more »

"ML Foundations (9/9) — PyTorch vs TensorFlow, and the Road to Local LLMs"

์ด๋ฏธ์ง€
← 8/9 Deep Learning Architec… ๐Ÿ“š Series Index (series end) The final part. So far we've focused on models. Now we focus on the tools that actually run them . We'll lay out the philosophical difference between PyTorch and TensorFlow, and trace how Hugging Face transformers , llama.cpp , MLX , and Ollama built the bridge to running large language models on your own machine. By the end you should have the full mental model of "download a pretrained LLM and serve it locally." 0. Learning Objectives Explain the eager-vs-graph difference between PyTorch and TensorFlow. Explain, in graph terms, how autograd automates the backward pass. Use Hugging Face transformers ' AutoModel , AutoTokenizer , and pipeline abstractions. Describe llama.cpp's GGUF quantization, INT4 inference, and CPU-first flow. Describe Apple MLX's use of unified memory and how it differs from PyTorch. Run a local LLM with Ollama and call it through an OpenAI-compatible API. ...
Read more »

"RAG Core Study (14/26) — Evaluation Sets with RAGAS & DeepEval"

์ด๋ฏธ์ง€
Series overview: Series index ๐Ÿ“š Series Index Without an evaluation set, every RAG improvement is a story, not evidence. RAG systems fail in more than one place: retrieval, context selection, answer generation, and grounding. That means "looks better" is not a metric. Part 14 explains how to build a golden dataset , how RAGAS and DeepEval fit into the workflow, and why teams must separate retrieval quality from answer quality if they want tuning decisions to hold up. 0. Prerequisites Part 13 reranking — system quality now depends on multi-stage ranking. Part 1 grounding — a correct-looking answer is not enough. Part 12 Hybrid — multiple retrieval settings need comparison on fixed data. 1. Learning Objectives Build a small but useful golden dataset for RAG. Distinguish retrieval metrics from answer metrics . Use RAGAS and DeepEval in the right roles. Avoid the common traps in judge-model-based evaluation. 2. ํ•ต์‹ฌ ์š”์•ฝ An evaluation set for RAG is ...
Read more »

"ML Foundations (8/9) — Deep Learning Architectures: CNN, RNN, Attention"

์ด๋ฏธ์ง€
← 7/9 Deep Learning Training ๐Ÿ“š Series Index 9/9 PyTorch vs TensorFlow → What did we keep adding on top of an MLP? Vision went CNN, sequences went RNN/LSTM, and eventually both converged on Attention and the Transformer. Each architecture is easier to remember if you read it as what it gave up to gain something else . This part is the one-line summary of the decisive inflection points: LeNet → AlexNet → ResNet → LSTM → Attention → Transformer. 0. Learning Objectives Explain why CNN's convolution, pooling, and weight sharing match images so well. Trace what got unblocked from LeNet → AlexNet → VGG → ResNet as depth grew. State BPTT for RNNs and the vanishing/exploding gradient problem. Write the LSTM gate equations (forget, input, output) with intuition. Write the attention formula in query/key/value form and the scaled dot-product variant. State the precise reasons Transformers replaced RNNs (parallelism + long-range dependencies). 1. ํ•ต์‹ฌ ์š”์•ฝ CNN : weight sharin...
Read more »

"ML Foundations (7/9) — Deep Learning Training: Optimizers, Regularization, Initialization"

← 6/9 Neural Networks ๐Ÿ“š Series Index 8/9 Deep Learning Architec… → Building an MLP, as we did in Part 6, is not the same as getting it to train well . The right optimizer, regularization, initialization, and learning rate — when these line up, deep networks converge. When they don't, the network refuses to learn at all. This part is the catalog of those crafts, with formulas, paper citations, and code in one place. 0. Learning Objectives Compare and write the update rules for SGD, Momentum, Nesterov, and Adam. Explain how Dropout, BatchNorm, and LayerNorm work and where they belong in a model. Derive the variance formulas for Xavier and He initialization and match them to activations. Implement step, cosine, and warmup learning-rate schedules in PyTorch. Explain why gradient clipping is effectively required for RNNs and Transformers. Diagnose the most common training failures (NaN, plateau, overfitting) and apply first-line fixes. 1. ํ•ต์‹ฌ ์š”์•ฝ SGD : \(w \leftarro...
Read more »

OpenClaw to Hermes Migration (2/13) — What to Preserve, Partially Port, or Discard

← 1/13 Current Structure Inve… ๐Ÿ“š Series Index 3/13 What Moves to Hermes → A code review record classifying operational scripts into three tiers: preserve, partial-preserve, and discard. What This Post Covers An asset classification framework (preserve / partial-preserve / discard) for migrating a legacy agent system to a successor A concrete case applying the Strangler Fig + Subset Migration strategy to a memory pipeline Classification results evaluated across LOC, external dependencies, and replaceability axes A method for slimming down a bloated single file (1,310 LOC) using the migration as a natural entry point Problem Definition: Port Everything or Cut Deep? The OpenClaw inventory comprises multiple agents, multiple scheduler/daemon processes, and multiple operational scripts. When moving to the successor system, Hermes, the choice is not binary. A full port carries accumulated technical debt along with it; a minimal port discards hard-won domain logic. Code-revie...
Read more »

AI Agents I Built (5/7) — Building an Automated Blogger API Publishing System

← 4/7 My Life Organizer Agent ๐Ÿ“š Series Index 6/7 Dual → Designing a Markdown-Based Auto-Publish Pipeline with Blogger API v3 + OAuth 2.0 Key Summary Blogger API v3 + OAuth 2.0 enables a pipeline that converts Markdown files into publishable HTML posts blogger_publish.py handles the full flow — frontmatter parsing → body cleanup → HTML conversion → CSS injection → API post — in a single script For bulk publishing, quota management (delay, retry, TOC suppression) is the critical design concern Platform Selection Criteria API access is mandatory for AI agent-driven content publishing. A comparison of platforms: Platform API Cost AdSense Decision WordPress.com REST API Paid plan required Paid only Cost overhead Ghost REST API Self-hosted or paid Manual setup Ops overhead Blogger v3 API Free Native integration Adopted Blogger is free, exposes a REST API, and integrates natively with AdSense. For personal blog automation, this combination i...
Read more »