Backup & Snapshot

Learn how to protect your VMs with snapshots for quick backup and recovery.

What are Snapshots? link

Snapshots capture the complete state of a VM at a specific point in time:

• Full VM State: Memory, disk, and configuration
• Instant Creation: Takes seconds to create
• Quick Restore: Restore VM in seconds
• Multiple Snapshots: Keep several backup points

Use Cases link

Before Risky Changes:

  Create Snapshot → Make Changes → Success? Keep | Failure? Restore
  

Regular Backups:

• Daily snapshots of production VMs
• Before system updates
• Before application deployments

Testing & Development:

• Save clean state before testing
• Restore to clean state between tests
• Experiment safely

Disaster Recovery:

• Quick recovery from failures
• Rollback from bad updates
• Restore from accidental deletions

Creating a Snapshot link

Step 1: Navigate to VM link

1. Go to Virtual Machines page
2. Click on the VM you want to snapshot
3. Click the Snapshots tab

Step 2: Create Snapshot link

Click the Create Snapshot button:

A dialog will appear:

Step 3: Enter Snapshot Details link

Snapshot Name:

• Use descriptive names
• Include date/time or purpose
• Examples:
  • before-upgrade-2025-12-16
  • clean-install
  • before-database-migration
  • daily-backup-20251216

Description (Optional):

  Before upgrading to PostgreSQL 15
Installed packages: postgresql-14, nginx, nodejs
  

Step 4: Create link

Click Create to start the snapshot process:

What happens:

1. VM state is paused briefly
2. Memory contents are saved
3. Disk state is captured
4. VM resumes automatically

Time: Usually 5-15 seconds depending on VM size

Step 5: Snapshot Created link

The new snapshot appears in the list:

You’ll see:

• Snapshot name
• Creation date/time
• Size (disk + memory)
• Actions (Restore, Delete)

Restoring from Snapshot link

Warning: Restoring replaces current VM state with snapshot!

Before Restoring link

Important considerations:

• Current VM data will be lost
• VM will revert to snapshot time
• Create new snapshot of current state if needed
• Stop VM before restoring (recommended)

Restore Process link

1. Go to VM Snapshots tab
2. Find the snapshot you want to restore
3. Click Restore button

4. Confirm the restoration:

Confirmation message:

  ⚠️  Warning: This will restore VM to snapshot state.
Current data will be lost. This cannot be undone.

Snapshot: before-upgrade-2025-12-16
Created: 2025-12-16 10:30:00

Type VM name to confirm: my-vm
  

5. Type VM name and click Confirm Restore

Restoration Progress link

The system will:

1. Stop the VM (if running)
2. Replace disk with snapshot
3. Restore memory state
4. Restart the VM

Time: Usually 10-30 seconds

Verify Restoration link

After restoration:

1. Check VM is in “Running” state
2. Access console and verify data
3. Test that everything works as expected
4. Check timestamp — should match snapshot time

Example verification:

  # Check system uptime (should show recent boot)
uptime

# Check file timestamps
ls -la /var/log/

# Verify applications are running
ps aux | grep nginx
  

Managing Snapshots link

Renaming a Snapshot link

1. Click ⋮ menu next to snapshot
2. Select Rename
3. Enter new name
4. Click Save

Deleting a Snapshot link

Caution: Deleted snapshots cannot be recovered!

1. Click ⋮ menu next to snapshot
2. Select Delete
3. Confirm deletion

What happens:

• Snapshot is permanently removed
• Disk space is freed
• Cannot be restored after deletion
• VM is not affected

Snapshot Types link

Full Snapshots link

Captures complete VM state:

• ✅ All disk data
• ✅ Memory contents
• ✅ Configuration
• ✅ Independent restore point

Size: Matches VM disk + memory size

Use when: Creating major backup points

Incremental Snapshots link

Captures only changes since last snapshot:

• ✅ Changes since parent snapshot
• ✅ Smaller size
• ✅ Faster creation
• ⚠️ Requires parent snapshot

Size: Only changed data

Use when: Frequent backups of same VM

Best Practices link

Snapshot Naming link

Good names:

  before-update-2025-12-16
after-install-postgres
clean-os-install
production-daily-20251216-0300
pre-migration-backup
  

Bad names:

  snapshot1
backup
test
20251216
  

Snapshot Frequency link

Production VMs:

• Daily snapshots at off-peak hours
• Before any changes
• Keep last 7 daily snapshots
• Monthly long-term snapshots

Development VMs:

• Before major changes
• After successful configurations
• Clean state snapshots
• Keep 2-3 recent snapshots

Test VMs:

• Before each test cycle
• Clean baseline state
• Delete after testing complete

Snapshot Retention link

Recommended retention policy:

Delete old snapshots:

• Free up disk space
• Reduce clutter
• Focus on important backups
• Automate cleanup if possible

Storage Management link

Monitor snapshot storage:

1. Go to Snapshots page
2. Check total size
3. Review storage usage

Optimize storage:

• Delete unnecessary snapshots
• Use incremental snapshots
• Compress old snapshots
• Archive to external storage

Disaster Recovery link

Recovery Plan link

Scenario: VM crashed and won’t boot

Recovery steps:

1. Assess damage:

   • Try restarting VM
   • Check error messages
   • Identify last known good state

2. Find latest snapshot:

   • Go to VM Snapshots tab
   • Identify most recent working snapshot
   • Note what data will be lost

3. Restore snapshot:

   • Stop failed VM
   • Click Restore on chosen snapshot
   • Confirm restoration
   • Wait for completion

4. Verify recovery:

   • Check VM starts successfully
   • Test critical services
   • Verify data integrity
   • Document what was lost

5. Prevent recurrence:

   • Identify failure cause
   • Implement fixes
   • Create new snapshot of fixed state

Testing Restores link

Monthly practice:

1. Choose non-critical VM
2. Create test snapshot
3. Make some changes
4. Restore from snapshot
5. Verify restoration worked
6. Delete test snapshot

Why test?:

• Verify backups are valid
• Practice recovery process
• Build confidence
• Find issues before emergencies

Troubleshooting link

Issue: Snapshot Creation Fails link

Problem: Error message when creating snapshot

Solutions:

1. Check available disk space on host
2. Ensure VM is in stable state
3. Try stopping VM first, then snapshot
4. Reduce snapshot frequency
5. Contact administrator if disk full

Issue: Restore Takes Too Long link

Problem: Restoration stuck or very slow

Solutions:

1. Wait - large VMs take longer (can be minutes)
2. Check network connection to server
3. Refresh browser after 5 minutes
4. Check VM status directly
5. Contact administrator if >10 minutes

Issue: Can’t Delete Snapshot link

Problem: Delete button greyed out

Solutions:

1. Check if snapshot is in use
2. Stop dependent VMs
3. Wait for other operations to complete
4. Refresh the page
5. Check permissions

Issue: Snapshot Missing link

Problem: Expected snapshot not in list

Solutions:

1. Refresh the browser page
2. Check you’re looking at correct VM
3. Check All Snapshots page
4. Verify snapshot wasn’t auto-deleted
5. Check with team if someone deleted it

Advanced Tips link

Pre-Snapshot Checklist link

Before creating important snapshots:

  # In VM console/SSH

# 1. Stop services gracefully
systemctl stop nginx
systemctl stop postgresql

# 2. Sync filesystem
sync

# 3. Clear cache (optional)
sync; echo 3 > /proc/sys/vm/drop_caches

# 4. Create marker file
echo "Snapshot created at $(date)" > /root/snapshot-$(date +%Y%m%d).txt
  

Then create the snapshot.

Why?:

• Ensures consistent state
• Prevents corruption
• Makes restore cleaner

Snapshot Metadata link

Add useful metadata in description:

  Created: 2025-12-16 15:30:00
Purpose: Before PostgreSQL 15 upgrade
Installed: PostgreSQL 14.5, Nginx 1.24, Node.js 20
Services Running: web-api, background-worker
IP Address: 192.168.1.100
Last Updated: 2025-12-15
  

Helps identify correct restore point later!

Next Steps link

• Monitoring - Monitor VM performance
• Manage VM - VM lifecycle operations
• Create VM - Create new VMs