Upload 3 files
Browse files- Dockerfile +17 -0
- README.md +76 -6
- sync_data.sh +204 -0
Dockerfile
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM ghcr.io/open-webui/open-webui:main
|
| 2 |
+
|
| 3 |
+
# Install required tools
|
| 4 |
+
RUN apt-get update && apt-get install -y \
|
| 5 |
+
git \
|
| 6 |
+
git-lfs \
|
| 7 |
+
sqlite3 \
|
| 8 |
+
curl \
|
| 9 |
+
&& git lfs install \
|
| 10 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 11 |
+
|
| 12 |
+
COPY sync_data.sh sync_data.sh
|
| 13 |
+
|
| 14 |
+
RUN chmod -R 777 ./data && \
|
| 15 |
+
chmod -R 777 /app/backend/open_webui/static && \
|
| 16 |
+
chmod +x sync_data.sh && \
|
| 17 |
+
sed -i "1r sync_data.sh" ./start.sh
|
README.md
CHANGED
|
@@ -1,10 +1,80 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: docker
|
| 7 |
-
pinned:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: OpenWebUI
|
| 3 |
+
emoji: 🤖
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: purple
|
| 6 |
sdk: docker
|
| 7 |
+
pinned: true
|
| 8 |
+
app_port: 8080
|
| 9 |
+
suggested_hardware: cpu-basic
|
| 10 |
+
suggested_storage: medium
|
| 11 |
+
short_description: OpenWebUI with automated GitHub data sync
|
| 12 |
+
tags:
|
| 13 |
+
- openwebui
|
| 14 |
+
- chat
|
| 15 |
+
- llm
|
| 16 |
+
- ai
|
| 17 |
+
- docker
|
| 18 |
+
- data-sync
|
| 19 |
+
- backup
|
| 20 |
---
|
| 21 |
|
| 22 |
+
# OpenWebUI with Data Synchronization
|
| 23 |
+
|
| 24 |
+
A self-hosted OpenWebUI instance with automated data backup and synchronization to GitHub.
|
| 25 |
+
|
| 26 |
+
## Features
|
| 27 |
+
|
| 28 |
+
- 🤖 **Full OpenWebUI**: Chat interface for AI models
|
| 29 |
+
- 🔄 **Auto Sync**: Backs up `webui.db`, `uploads/`, `vector_db/` to GitHub
|
| 30 |
+
- 🛡️ **Safe Backups**: SQLite hot backups prevent corruption
|
| 31 |
+
- 📦 **Git LFS**: Handles large files in uploads/vector_db
|
| 32 |
+
- ⏰ **Configurable**: Custom sync intervals and timezone
|
| 33 |
+
- 🌐 **WebDAV Support**: Optional secondary backup location
|
| 34 |
+
|
| 35 |
+
## Environment Variables
|
| 36 |
+
|
| 37 |
+
Set these in your Space's **Repository Secrets**:
|
| 38 |
+
|
| 39 |
+
### Required
|
| 40 |
+
|
| 41 |
+
| Variable | Description | Example |
|
| 42 |
+
|----------|-------------|---------|
|
| 43 |
+
| `G_NAME` | GitHub repo | `username/openwebui-db` |
|
| 44 |
+
| `G_TOKEN` | GitHub PAT | `github_pat_xxx...` |
|
| 45 |
+
|
| 46 |
+
### Optional
|
| 47 |
+
|
| 48 |
+
| Variable | Description | Default |
|
| 49 |
+
|----------|-------------|---------|
|
| 50 |
+
| `SYNC_INTERVAL` | Sync frequency (seconds) | `3600` |
|
| 51 |
+
| `SYNC_TIMEZONE` | Timezone for logs | `Asia/Kolkata` |
|
| 52 |
+
| `GIT_USER_NAME` | Commit author | `AutoSync Bot` |
|
| 53 |
+
| `GIT_USER_EMAIL` | Commit email | `autosync@bot.com` |
|
| 54 |
+
| `WEBDAV_URL` | WebDAV server URL | - |
|
| 55 |
+
| `WEBDAV_USERNAME` | WebDAV user | - |
|
| 56 |
+
| `WEBDAV_PASSWORD` | WebDAV password | - |
|
| 57 |
+
|
| 58 |
+
## GitHub Token Setup
|
| 59 |
+
|
| 60 |
+
1. Go to [GitHub Token Settings](https://github.com/settings/tokens)
|
| 61 |
+
2. Create a **Fine-grained token**
|
| 62 |
+
3. Select your backup repository
|
| 63 |
+
4. Grant **Contents**: Read and Write
|
| 64 |
+
|
| 65 |
+
## Setup
|
| 66 |
+
|
| 67 |
+
1. Create a private GitHub repository for backups
|
| 68 |
+
2. Generate a GitHub Personal Access Token
|
| 69 |
+
3. Add `G_NAME` and `G_TOKEN` to Space secrets
|
| 70 |
+
4. Deploy and enjoy persistent storage!
|
| 71 |
+
|
| 72 |
+
## How It Works
|
| 73 |
+
|
| 74 |
+
1. **Startup**: Clones your GitHub repo and restores data
|
| 75 |
+
2. **Runtime**: Syncs every `SYNC_INTERVAL` seconds
|
| 76 |
+
3. **Backup**: Creates SQLite hot backup, syncs uploads/vector_db
|
| 77 |
+
4. **Push**: Commits and pushes changes to GitHub
|
| 78 |
+
5. **WebDAV**: Optionally uploads timestamped backup to WebDAV
|
| 79 |
+
|
| 80 |
+
Check out the [configuration reference](https://huggingface.co/docs/hub/spaces-config-reference) for more details.
|
sync_data.sh
ADDED
|
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/bin/bash
|
| 2 |
+
# OpenWebUI Data Sync Script
|
| 3 |
+
# Syncs webui.db, uploads/, and vector_db/ to GitHub with optional WebDAV backup
|
| 4 |
+
|
| 5 |
+
set -e
|
| 6 |
+
|
| 7 |
+
# ============================================
|
| 8 |
+
# CONFIGURATION
|
| 9 |
+
# ============================================
|
| 10 |
+
|
| 11 |
+
# Set timezone
|
| 12 |
+
export TZ="${SYNC_TIMEZONE:-Asia/Kolkata}"
|
| 13 |
+
|
| 14 |
+
# Absolute paths (critical for reliability)
|
| 15 |
+
APP_DIR="/app/backend"
|
| 16 |
+
DATA_DIR="${APP_DIR}/data"
|
| 17 |
+
GITHUB_DATA_DIR="${DATA_DIR}/github_data"
|
| 18 |
+
|
| 19 |
+
# Configurable variables with defaults
|
| 20 |
+
SYNC_INTERVAL=${SYNC_INTERVAL:-3600}
|
| 21 |
+
GIT_USER_NAME=${GIT_USER_NAME:-"AutoSync Bot"}
|
| 22 |
+
GIT_USER_EMAIL=${GIT_USER_EMAIL:-"autosync@bot.com"}
|
| 23 |
+
|
| 24 |
+
# ============================================
|
| 25 |
+
# VALIDATION
|
| 26 |
+
# ============================================
|
| 27 |
+
|
| 28 |
+
echo "=========================================="
|
| 29 |
+
echo "OpenWebUI Data Sync Starting..."
|
| 30 |
+
echo "Timezone: $TZ"
|
| 31 |
+
echo "Current time: $(date '+%Y-%m-%d %H:%M:%S')"
|
| 32 |
+
echo "=========================================="
|
| 33 |
+
|
| 34 |
+
if [ -z "$G_NAME" ] || [ -z "$G_TOKEN" ]; then
|
| 35 |
+
echo "[ERROR] Missing required environment variables G_NAME or G_TOKEN"
|
| 36 |
+
echo "[WARN] Sync disabled. OpenWebUI will start without data sync."
|
| 37 |
+
exit 0 # Don't exit 1, let OpenWebUI start anyway
|
| 38 |
+
fi
|
| 39 |
+
|
| 40 |
+
echo "[INFO] GitHub Repo: $G_NAME"
|
| 41 |
+
echo "[INFO] Sync Interval: ${SYNC_INTERVAL}s"
|
| 42 |
+
|
| 43 |
+
# ============================================
|
| 44 |
+
# INITIAL CLONE & RESTORE
|
| 45 |
+
# ============================================
|
| 46 |
+
|
| 47 |
+
REPO_URL="https://${G_TOKEN}@github.com/${G_NAME}.git"
|
| 48 |
+
mkdir -p "$GITHUB_DATA_DIR"
|
| 49 |
+
|
| 50 |
+
echo "[INFO] Cloning repository..."
|
| 51 |
+
if git clone "$REPO_URL" "$GITHUB_DATA_DIR" 2>/dev/null; then
|
| 52 |
+
echo "[SUCCESS] Repository cloned"
|
| 53 |
+
|
| 54 |
+
# Configure Git LFS
|
| 55 |
+
cd "$GITHUB_DATA_DIR"
|
| 56 |
+
git lfs install --local 2>/dev/null || true
|
| 57 |
+
git lfs pull 2>/dev/null || true
|
| 58 |
+
cd "$APP_DIR"
|
| 59 |
+
else
|
| 60 |
+
echo "[WARN] Clone failed - initializing empty sync directory"
|
| 61 |
+
mkdir -p "$GITHUB_DATA_DIR"
|
| 62 |
+
cd "$GITHUB_DATA_DIR"
|
| 63 |
+
git init
|
| 64 |
+
git remote add origin "$REPO_URL" 2>/dev/null || true
|
| 65 |
+
cd "$APP_DIR"
|
| 66 |
+
fi
|
| 67 |
+
|
| 68 |
+
# Restore data from repository
|
| 69 |
+
echo "[INFO] Restoring data from repository..."
|
| 70 |
+
|
| 71 |
+
if [ -f "$GITHUB_DATA_DIR/webui.db" ]; then
|
| 72 |
+
cp "$GITHUB_DATA_DIR/webui.db" "$DATA_DIR/webui.db"
|
| 73 |
+
echo "[SUCCESS] Restored webui.db"
|
| 74 |
+
else
|
| 75 |
+
echo "[INFO] No webui.db found in repo, will create on first sync"
|
| 76 |
+
fi
|
| 77 |
+
|
| 78 |
+
if [ -d "$GITHUB_DATA_DIR/uploads" ]; then
|
| 79 |
+
mkdir -p "$DATA_DIR/uploads"
|
| 80 |
+
cp -r "$GITHUB_DATA_DIR/uploads/"* "$DATA_DIR/uploads/" 2>/dev/null || true
|
| 81 |
+
echo "[SUCCESS] Restored uploads directory"
|
| 82 |
+
fi
|
| 83 |
+
|
| 84 |
+
if [ -d "$GITHUB_DATA_DIR/vector_db" ]; then
|
| 85 |
+
mkdir -p "$DATA_DIR/vector_db"
|
| 86 |
+
cp -r "$GITHUB_DATA_DIR/vector_db/"* "$DATA_DIR/vector_db/" 2>/dev/null || true
|
| 87 |
+
echo "[SUCCESS] Restored vector_db directory"
|
| 88 |
+
fi
|
| 89 |
+
|
| 90 |
+
# ============================================
|
| 91 |
+
# SYNC FUNCTION
|
| 92 |
+
# ============================================
|
| 93 |
+
|
| 94 |
+
sync_data() {
|
| 95 |
+
while true; do
|
| 96 |
+
echo "[INFO] Next sync in ${SYNC_INTERVAL} seconds..."
|
| 97 |
+
sleep "$SYNC_INTERVAL"
|
| 98 |
+
|
| 99 |
+
CURRENT_TIME=$(date '+%Y-%m-%d %H:%M:%S')
|
| 100 |
+
echo ""
|
| 101 |
+
echo "=========================================="
|
| 102 |
+
echo "[SYNC] Starting sync at $CURRENT_TIME"
|
| 103 |
+
echo "=========================================="
|
| 104 |
+
|
| 105 |
+
# Navigate to GitHub data directory
|
| 106 |
+
cd "$GITHUB_DATA_DIR"
|
| 107 |
+
|
| 108 |
+
# Configure Git
|
| 109 |
+
git config user.name "$GIT_USER_NAME"
|
| 110 |
+
git config user.email "$GIT_USER_EMAIL"
|
| 111 |
+
|
| 112 |
+
# Ensure on correct branch
|
| 113 |
+
git checkout main 2>/dev/null || git checkout master 2>/dev/null || git checkout -b main
|
| 114 |
+
|
| 115 |
+
# Pull latest changes
|
| 116 |
+
git pull origin HEAD --rebase 2>/dev/null || echo "[WARN] Pull failed (might be empty repo)"
|
| 117 |
+
|
| 118 |
+
# ----------------------------------------
|
| 119 |
+
# 1. Safe Database Backup
|
| 120 |
+
# ----------------------------------------
|
| 121 |
+
if [ -f "$DATA_DIR/webui.db" ]; then
|
| 122 |
+
echo "[INFO] Creating safe backup of webui.db..."
|
| 123 |
+
if command -v sqlite3 &> /dev/null; then
|
| 124 |
+
sqlite3 "$DATA_DIR/webui.db" ".backup '$GITHUB_DATA_DIR/webui.db'"
|
| 125 |
+
echo "[SUCCESS] SQLite hot backup created"
|
| 126 |
+
else
|
| 127 |
+
cp "$DATA_DIR/webui.db" "$GITHUB_DATA_DIR/webui.db"
|
| 128 |
+
echo "[WARN] sqlite3 not found, used cp (potentially unsafe)"
|
| 129 |
+
fi
|
| 130 |
+
else
|
| 131 |
+
echo "[INFO] No webui.db found yet"
|
| 132 |
+
fi
|
| 133 |
+
|
| 134 |
+
# ----------------------------------------
|
| 135 |
+
# 2. Sync Uploads Directory
|
| 136 |
+
# ----------------------------------------
|
| 137 |
+
if [ -d "$DATA_DIR/uploads" ]; then
|
| 138 |
+
echo "[INFO] Syncing uploads directory..."
|
| 139 |
+
mkdir -p "$GITHUB_DATA_DIR/uploads"
|
| 140 |
+
cp -r "$DATA_DIR/uploads/"* "$GITHUB_DATA_DIR/uploads/" 2>/dev/null || true
|
| 141 |
+
fi
|
| 142 |
+
|
| 143 |
+
# ----------------------------------------
|
| 144 |
+
# 3. Sync Vector DB
|
| 145 |
+
# ----------------------------------------
|
| 146 |
+
if [ -d "$DATA_DIR/vector_db" ]; then
|
| 147 |
+
echo "[INFO] Syncing vector_db directory..."
|
| 148 |
+
mkdir -p "$GITHUB_DATA_DIR/vector_db"
|
| 149 |
+
cp -r "$DATA_DIR/vector_db/"* "$GITHUB_DATA_DIR/vector_db/" 2>/dev/null || true
|
| 150 |
+
fi
|
| 151 |
+
|
| 152 |
+
# ----------------------------------------
|
| 153 |
+
# 4. Git Commit & Push
|
| 154 |
+
# ----------------------------------------
|
| 155 |
+
if [[ -n $(git status -s) ]]; then
|
| 156 |
+
echo "[INFO] Changes detected, committing..."
|
| 157 |
+
git add .
|
| 158 |
+
git commit -m "Auto sync: $CURRENT_TIME"
|
| 159 |
+
|
| 160 |
+
if git push origin HEAD; then
|
| 161 |
+
echo "[SUCCESS] GitHub push successful"
|
| 162 |
+
else
|
| 163 |
+
echo "[WARN] Push failed, retrying..."
|
| 164 |
+
sleep 5
|
| 165 |
+
git pull origin HEAD --rebase 2>/dev/null || true
|
| 166 |
+
git push origin HEAD || echo "[ERROR] Retry failed"
|
| 167 |
+
fi
|
| 168 |
+
else
|
| 169 |
+
echo "[INFO] No changes detected"
|
| 170 |
+
fi
|
| 171 |
+
|
| 172 |
+
# Return to app directory
|
| 173 |
+
cd "$APP_DIR"
|
| 174 |
+
|
| 175 |
+
# ----------------------------------------
|
| 176 |
+
# 5. WebDAV Sync (Optional)
|
| 177 |
+
# ----------------------------------------
|
| 178 |
+
if [ -n "$WEBDAV_URL" ] && [ -n "$WEBDAV_USERNAME" ] && [ -n "$WEBDAV_PASSWORD" ]; then
|
| 179 |
+
echo "[INFO] Starting WebDAV sync..."
|
| 180 |
+
FILENAME="webui_$(date +'%Y%m%d_%H%M%S').db"
|
| 181 |
+
|
| 182 |
+
if [ -f "$DATA_DIR/webui.db" ]; then
|
| 183 |
+
if curl -s -T "$DATA_DIR/webui.db" \
|
| 184 |
+
--user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" \
|
| 185 |
+
"$WEBDAV_URL/$FILENAME"; then
|
| 186 |
+
echo "[SUCCESS] WebDAV upload successful: $FILENAME"
|
| 187 |
+
else
|
| 188 |
+
echo "[ERROR] WebDAV upload failed"
|
| 189 |
+
fi
|
| 190 |
+
fi
|
| 191 |
+
fi
|
| 192 |
+
|
| 193 |
+
echo "[SYNC] Completed at $(date '+%Y-%m-%d %H:%M:%S')"
|
| 194 |
+
done
|
| 195 |
+
}
|
| 196 |
+
|
| 197 |
+
# ============================================
|
| 198 |
+
# START BACKGROUND SYNC
|
| 199 |
+
# ============================================
|
| 200 |
+
|
| 201 |
+
echo "[INFO] Starting background sync process..."
|
| 202 |
+
sync_data &
|
| 203 |
+
SYNC_PID=$!
|
| 204 |
+
echo "[INFO] Sync process started with PID: $SYNC_PID"
|