Compute Instance Backups

Overview

Having backups of your instances is critical to disaster recovery. There are a few ways to go about backing up your data, this article will cover these processes here.

Instance Snapshots

📘

NOTE: Scheduling Snapshots of OpenStack Flex VMs is currently not available as an option. However, you can use an alternative script/API/cronjob method found here

An Instance Snapshot captures the state of a virtual machine and creates a bootable image that can be used to launch a new instance later. Instance snapshots are useful for preserving a VM configuration before making changes, creating templates, or performing quick recovery of an entire system.

Creating an Instance Snapshot

NOTE: Shutting down, or limiting activity on the VM ahead of snapshotting is ideal when possible.

1. Log into the Skyline UI
2. Navigate to Compute > Instances in the left-hand navigation pane
3. Identify the server you want to snapshot > click More from the righthand side > Backups & Snapshots > Create Snapshot

Key characteristics:

• Captures the VM's root disk state
• Creates a Glance image that can be used to launch new instances
• Useful for rollback, cloning, and migration scenarios
• Typically crash-consistent rather than application-consistent
• Not recommended as the sole backup mechanism for databases or transactional applications

Common use cases:

• Pre-maintenance checkpoints
• Golden image creation
• VM cloning
• Short-term recovery

---

Volume Snapshots

A Volume Snapshot captures the point-in-time state of a Volume. Unlike instance snapshots, volume snapshots focus on the attached storage rather than the entire virtual machine.

Creating a Volume Snapshot

NOTE: Shutting down, or limiting activity on the volume ahead of snapshotting is ideal when possible.

1. Log into the Skyline UI
2. Navigate to Storage > Volumes in the left-hand navigation pane
3. Identify the volume you want to snapshot > click More from the righthand side > Data Protection > Create Snapshot

Key characteristics:

• Captures a specific Cinder volume at a point in time
• Can be used to create new volumes
• Generally faster and more storage-efficient than full copies
• Useful for protecting application data stored on persistent volumes
• Typically crash-consistent unless application-specific steps are taken beforehand

Common use cases:

• Application data protection
• Volume rollback
• Test and development cloning
• Short-term recovery points

---

Volume Backups

A Volume Backup creates a backup of a Volume and stores it in a configured backup repository. Unlike snapshots, backups are designed for longer-term retention and disaster recovery.

Creating a Volume Backup

NOTE: Shutting down, or limiting activity on the volume ahead of backup is ideal when possible.

1. Log into the Skyline UI
2. Navigate to Storage > Volumes in the left-hand navigation pane
3. Identify the volume you want to snapshot > click More from the righthand side > Data Protection > Create Backup

Key characteristics:

• Copies volume data to a backup target
• Supports long-term retention
• Can be restored to new volumes
• Often supports incremental backups after the initial full backup
• Better suited for disaster recovery than snapshots

Common use cases:

• Long-term data retention
• Disaster recovery planning
• Compliance requirements
• Protection against volume or storage failures

---

Snapshot vs. Backup

IMPORTANT: Neither instance snapshots nor volume snapshots should be considered a substitute for application-aware database backups. Because snapshots are generally crash-consistent, databases may require recovery procedures and cannot guarantee transactional consistency unless additional preparation steps are taken before the snapshot is created.

---

Third Party Backup Solutions

While the following are not Rackspace products and so support would be provided from these other communities - it is worth mentioning them. Since OpenStack Flex is based on more vanilla OpenStack environments, here are some third party backup solutions available that can be used to create file level backups.

A common best practice is to use application-aware backups (database dumps, Veeam agents, Bacula jobs, etc.) and store the resulting backup data in Object Storage Containers (Swift), as an addition to OpenStack snapshots.