Backup and restore ensure the security and durability of the applications and data, helping enterprises avoid data loss due to system failures, accidents, and human errors. The IEC platform provides a powerful, multi-tiered solution for backup and restore. You can flexibly choose backup and restore strategies based on namespaces, applications, or label selectors to protect applications and data based on the importance and recovery needs of the application. In addition, when upgrading your Kubernetes cluster or having your existing Kubernetes services migrate to IEC, all applications and data can be transferred seamlessly to the new environment.
The IEC platform can fully back up applications and persistent data in Kubernetes. When backing up a persistent volume that does not support cloud snapshots, the system will back up data in the file level: Application resources and persistent volume data are packed, encrypted, and stored in a remote backup location. This mechanism ensures the consistency and security of applications and data.
When performing a backup operation, the system automatically identifies all Kubernetes resources related to the application, including Deployment, StatefulSet, ConfigMap, Secret, etc., and the persistent volumes used by the application. Whether these volumes are created from local storage or other cloud storage devices, the system will add these resources to the backup process.
Kubernetes API resources will be saved in a backup file. This contains all the configuration and status information needed to run the application, which realizing fast application recovery when restored.
For persistent volumes, each file in the volume will be compressed and encrypted, and then transferred to the remote backup storage location. This ensures can prevent data damage or loss that even when the cloud snapshot function cannot be used. The encryption prevents unauthorized data access. Since encryption and compression are completed before transmission, ensures that the data cannot be accessed without authorization even if it is stored remotely.
In IEC, resources that can be backed up and cannot be backed up are listed a below:
| Can be backed up | Cannot be backed up |
|
|
The IEC platform provides incremental backup by default, which only backs up data that has changed since the last backup, effectively reducing storage space usage and the time required for backup. This is suitable for frequent backups and scenarios where long-term data history needs to be retained.
Note:
- Only administrators can perform backup and restore.
- To protect etcd data, use etcd's backup tool (such as etcdctl) to perform a standalone backup.
- To back up Kubelet-related configurations, use other system-level backup tools.
To back up your data of applications, consider your backup strategy and plans before starting backup. For more information, see Backup Strategies.
Backup Strategies
To appropriate back up your data of applications, before starting backup, consider the following:
- Backup destinations
- Backup frequency and schedules
- Pre-backup/post-backup actions
Backup Destinations
IEC supports to back up your data to cloud storage and local storage system (external storage to the cluster):
- Cloud storage: AWS S3-compatible cloud storage solution, suitable for enterprise environments that require scalability and geographical redundancy.
- Local (external) storage: EonStor GS/GSx storage systems can work as object storage. With this function enabled, you can back up your data to these on-premises appliances.
Backup Frequency and Schedules
When it comes to backup frequencies, configure flexible schedules to meet the data protection needs of different applications:
- Multiple times in a day: Suitable for applications with extremely high business continuity requirements, ensuring real-time data protection.
- Daily: Backup on a daily basis help balance backup time and storage space.
- Weekly or monthly: Suitable for long-term backup plans to regularly protect data to avoid long-term storage occupation.
According to the backup schedule that you configured, backup tasks can be performed regularly to ensure that important data is protected at all times. Or, when there are specific needs, manually perform real-time backup. This is useful especially before system maintenance and upgrade.
Note:
There can only be one task at a time. For example, if a backup task “Task A” is running at 10:00, a restore task or another backup task will be postponed until Task A is finished.
Pre-backup/Post-backup Actions
To ensure that backup operations proceed smoothly according to application requirements, you can pre-backup/post-backup rules.
- Pre-backup rules: Before performing a backup, you can set specific operations, such as pausing application running, locking the database, or stopping non-essential services, thereby ensuring that the data is backed up in a consistent state and avoiding inconsistent data being written to the backup.
- Post-backup rules: After the backup is completed, the application automatically resumes its operation according to your settings, or restart stopped services to ensure that the system quickly returns to normal and reduces the impact on the business.
For more information about pre-backup/post-backup rules, see Backup Rules.