ai-stack: vendor the local-ai full AI stack as a new service

Vendor the functional source of github.com/outis1one/local-ai into ./ai-stack (22 files) and add services/ai-stack.sh, which copies the source to ~/docker/ai-stack and hands off to the app's VRAM-aware installer (local-ai-setup.sh). The stack bundles Ollama, Open WebUI, RAG + MCP servers, ChromaDB, SearXNG, Kiwix, Gitea, InvokeAI, ComfyUI and Portainer. Cloud LLM providers (Groq/DeepInfra/OpenAI/OpenRouter) are optionally wired into Open WebUI via the plural OPENAI_API_BASE_URLS list, with the local RAG connection kept as the first entry so RAG keeps working. Open WebUI ships built-in auth, so Caddy is configured without Authelia. Excludes the upstream's two bundled copies of this very project (ubuntu-post-install.sh, ubuntu-post-install-main.zip) — stale and circular. Coexists with the existing ai-gpu service. Also fix the install-function names for ai-gpu and ai-stack: the dispatcher calls install_<raw-name>, so the function must be install_ai-gpu / install_ai-stack (hyphen), matching the working mail-archiver / wg-easy services. ai-gpu was previously uninstallable. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
2026-06-26 12:41:44 +00:00
parent 084922afaa
commit c6576178b4
25 changed files with 7524 additions and 3 deletions
@@ -0,0 +1,67 @@
+#!/bin/bash
+# =============================================================================
+# Ollama Model Auto-Updater
+# Pulls latest version of every installed model. Ollama compares digests
+# server-side — no download happens if the model is already current.
+# Safe to run as a systemd timer.
+# =============================================================================
+
+OLLAMA_URL="${OLLAMA_HOST:-http://localhost:11434}"
+LOG="$HOME/docker/ai-stack/logs/ollama-update.log"
+mkdir -p "$(dirname "$LOG")"
+
+G='\033[0;32m'; Y='\033[1;33m'; C='\033[0;36m'; R='\033[0;31m'; N='\033[0m'
+ok()  { echo -e "${G}[OK]${N}  $1" | tee -a "$LOG"; }
+inf() { echo -e "${C}[..]${N}  $1" | tee -a "$LOG"; }
+wrn() { echo -e "${Y}[!!]${N}  $1" | tee -a "$LOG"; }
+err() { echo -e "${R}[XX]${N}  $1" | tee -a "$LOG"; }
+
+echo "" | tee -a "$LOG"
+echo "=== Ollama Model Update: $(date) ===" | tee -a "$LOG"
+echo "" | tee -a "$LOG"
+
+# Check Ollama is reachable
+if ! curl -sf "$OLLAMA_URL/api/tags" > /dev/null; then
+    err "Ollama not reachable at $OLLAMA_URL — is the container running?"
+    exit 1
+fi
+
+# Get installed model names from the API
+mapfile -t MODELS < <(
+    curl -sf "$OLLAMA_URL/api/tags" | \
+    grep -oP '"name"\s*:\s*"\K[^"]+' | \
+    sort -u
+)
+
+if [[ ${#MODELS[@]} -eq 0 ]]; then
+    wrn "No models found — nothing to update"
+    exit 0
+fi
+
+inf "Found ${#MODELS[@]} model(s): ${MODELS[*]}"
+echo "" | tee -a "$LOG"
+
+UPDATED=0
+FAILED=0
+
+for model in "${MODELS[@]}"; do
+    inf "Pulling $model..."
+    output=$(ollama pull "$model" 2>&1)
+    exit_code=$?
+
+    if [[ $exit_code -ne 0 ]]; then
+        err "$model: pull failed"
+        echo "$output" >> "$LOG"
+        FAILED=$((FAILED + 1))
+    elif echo "$output" | grep -q "up to date"; then
+        ok "$model: already up to date"
+    else
+        ok "$model: updated"
+        UPDATED=$((UPDATED + 1))
+    fi
+done
+
+echo "" | tee -a "$LOG"
+inf "Done — $UPDATED updated, $FAILED failed"
+echo "=== Complete: $(date) ===" | tee -a "$LOG"
+echo "" | tee -a "$LOG"