Add infrastructure management: backup scripts and documentation

2026-02-09 10:58:30 -05:00
parent fab217c259
commit 61236e864c
3 changed files with 407 additions and 0 deletions
--- a/infrastructure/scripts/backup-compose-files.sh
+++ b/infrastructure/scripts/backup-compose-files.sh
@@ -0,0 +1,78 @@
+#!/bin/bash
+# backup-compose-files.sh
+# Backs up all compose files from the cluster to local repo
+
+set -e
+
+BACKUP_DIR="/Users/timothy.bendt/developer/cloud-compose/infrastructure"
+CONTROLLER="ubuntu@192.168.2.130"
+DATE=$(date +%Y-%m-%d)
+
+echo "📦 Backing up compose files from cluster..."
+echo "Date: $DATE"
+echo ""
+
+# Backup Dokploy compose files
+echo "📁 Backing up Dokploy compose files..."
+ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "find /etc/dokploy/compose -name 'docker-compose.yml' -type f" | while read file; do
+    # Get the project name from the path
+    project=$(echo $file | grep -oP 'compose/\K[^/]+' | head -1)
+    if [ ! -z "$project" ]; then
+        echo "  - $project"
+        mkdir -p "$BACKUP_DIR/compose/$project"
+        scp -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER:"$file" "$BACKUP_DIR/compose/$project/"
+    fi
+done
+
+# Backup Traefik configuration
+echo "📁 Backing up Traefik configuration..."
+mkdir -p "$BACKUP_DIR/traefik"
+scp -r -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER:/etc/dokploy/traefik/* "$BACKUP_DIR/traefik/" 2>/dev/null || echo "  (No Traefik config or permission denied)"
+
+# Backup local compose files from home directory
+echo "📁 Backing up local compose files..."
+scp -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER:~/minio-stack.yml "$BACKUP_DIR/stacks/" 2>/dev/null || true
+
+# Create manifest
+echo "📄 Creating backup manifest..."
+cat > "$BACKUP_DIR/BACKUP_MANIFEST.md" << EOF
+# Infrastructure Backup Manifest
+
+**Backup Date:** $DATE
+**Source:** $CONTROLLER
+
+## Contents
+
+### Compose Files
+All Docker Compose files from Dokploy-managed projects:
+\`\`\`
+$(ls -1 $BACKUP_DIR/compose/)
+\`\`\`
+
+### Stack Files
+Standalone stack definitions:
+\`\`\`
+$(ls -1 $BACKUP_DIR/stacks/ 2>/dev/null || echo "None")
+\`\`\`
+
+### Traefik Configuration
+Reverse proxy configuration backed up to ./traefik/
+
+## Restoration
+To restore these configurations:
+1. Copy compose files to /etc/dokploy/compose/<project>/code/
+2. Copy stack files to ~/ on the controller
+3. Copy Traefik config to /etc/dokploy/traefik/
+4. Redeploy via Dokploy UI or docker stack deploy
+
+## Services Backed Up
+$(ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker service ls --format '- {{.Name}}'" 2>/dev/null || echo "Unable to fetch service list")
+
+---
+Generated: $(date)
+EOF
+
+echo ""
+echo "✅ Backup complete!"
+echo "📁 Location: $BACKUP_DIR/"
+echo "📄 Manifest: $BACKUP_DIR/BACKUP_MANIFEST.md"
--- a/infrastructure/scripts/backup-critical-data.sh
+++ b/infrastructure/scripts/backup-critical-data.sh
@@ -0,0 +1,139 @@
+#!/bin/bash
+# backup-critical-data.sh
+# Backs up critical databases and data from the homelab
+
+set -e
+
+BACKUP_DIR="/Users/timothy.bendt/developer/cloud-compose/backups"
+CONTROLLER="ubuntu@192.168.2.130"
+NAS="tim@192.168.2.18"
+DATE=$(date +%Y-%m-%d)
+RETENTION_DAYS=30
+
+echo "💾 Starting critical data backup..."
+echo "Date: $DATE"
+echo ""
+
+mkdir -p "$BACKUP_DIR"
+
+# Function to backup a PostgreSQL database
+backup_postgres() {
+    local service=$1
+    local db_name=$2
+    local output_file="$BACKUP_DIR/${service}-${db_name}-${DATE}.sql"
+    
+    echo "  📊 Backing up PostgreSQL: $service/$db_name..."
+    
+    # Find the container
+    container=$(ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker ps -q -f name=$service" | head -1)
+    
+    if [ ! -z "$container" ]; then
+        ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker exec $container pg_dump -U postgres $db_name" > "$output_file"
+        echo "    ✅ Saved to: $output_file"
+    else
+        echo "    ❌ Container not found: $service"
+    fi
+}
+
+# Function to backup a MariaDB database
+backup_mariadb() {
+    local service=$1
+    local db_name=$2
+    local output_file="$BACKUP_DIR/${service}-${db_name}-${DATE}.sql"
+    
+    echo "  📊 Backing up MariaDB: $service/$db_name..."
+    
+    container=$(ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker ps -q -f name=$service" | head -1)
+    
+    if [ ! -z "$container" ]; then
+        ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker exec $container mariadb-dump -u root -p'3QU5eA&U^Y&3DQm6' $db_name" > "$output_file" 2>/dev/null || \
+        echo "    ⚠️  Could not backup (check credentials)"
+    else
+        echo "    ❌ Container not found: $service"
+    fi
+}
+
+# Function to backup volume data
+backup_volume() {
+    local service=$1
+    local volume=$2
+    local output_file="$BACKUP_DIR/${service}-data-${DATE}.tar.gz"
+    
+    echo "  💿 Backing up volume: $volume..."
+    
+    ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker run --rm -v $volume:/data -v /tmp:/backup alpine tar czf /backup/${service}-backup.tar.gz -C /data ."
+    scp -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER:/tmp/${service}-backup.tar.gz "$output_file"
+    ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "rm -f /tmp/${service}-backup.tar.gz"
+    
+    echo "    ✅ Saved to: $output_file"
+}
+
+echo "📦 Backing up databases..."
+
+# Dokploy database
+backup_postgres "dokploy-postgres" "dokploy"
+
+# Immich database
+backup_postgres "immich3-compose-ubyhe9-immich-database" "immich"
+
+# BewCloud database  
+backup_postgres "bewcloud-postgres-in40hh" "bewcloud"
+
+# Pancake database
+backup_mariadb "bendtstudio-pancake-bzgfpc" "pancake"
+
+echo ""
+echo "📦 Backing up application data..."
+
+# Memos data
+backup_volume "bewcloud-memos-ssogxn-memos" "bewcloud-memos-ssogxn_memos_data"
+
+# Gitea data
+echo "  💿 Backing up Gitea repositories..."
+container=$(ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker ps -q -f name=gitea" | head -1)
+if [ ! -z "$container" ]; then
+    ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker exec $container tar czf /tmp/gitea-backup.tar.gz -C /data ."
+    scp -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER:/tmp/gitea-backup.tar.gz "$BACKUP_DIR/gitea-data-${DATE}.tar.gz"
+    ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "rm -f /tmp/gitea-backup.tar.gz"
+    echo "    ✅ Saved to: $BACKUP_DIR/gitea-data-${DATE}.tar.gz"
+fi
+
+echo ""
+echo "🧹 Cleaning up old backups (older than $RETENTION_DAYS days)..."
+find "$BACKUP_DIR" -type f -mtime +$RETENTION_DAYS -delete
+echo "  ✅ Cleanup complete"
+
+echo ""
+echo "📄 Creating backup report..."
+cat > "$BACKUP_DIR/BACKUP_REPORT-${DATE}.md" << EOF
+# Backup Report - $DATE
+
+## Backup Location
+$BACKUP_DIR
+
+## Files Backed Up
+\`\`\`
+$(ls -lh $BACKUP_DIR/*-${DATE}* 2>/dev/null || echo "No files found")
+\`\`\`
+
+## Services Status
+$(ssh -o ConnectTimeout=10 -i ~/.ssh/id_ed25519 $CONTROLLER "docker service ls --format 'table {{.Name}}\t{{.Replicas}}\t{{.Image}}'" 2>/dev/null || echo "Unable to fetch status")
+
+## Next Steps
+1. Verify backup files are valid
+2. Copy backups to offsite storage (MinIO, external drive, etc.)
+3. Test restore procedure periodically
+
+## Retention Policy
+Backups older than $RETENTION_DAYS days are automatically deleted.
+
+---
+Generated: $(date)
+EOF
+
+echo ""
+echo "✅ Backup complete!"
+echo "📁 Location: $BACKUP_DIR/"
+echo "📄 Report: $BACKUP_DIR/BACKUP_REPORT-${DATE}.md"
+echo ""
+echo "⚠️  Remember to copy backups to offsite storage!"