Back up Cisco Container Platform
You can back up the Cisco Container Platform application data that pertains to the following components:
-
Application users
-
Virtualization providers
-
Tenant clusters
The logging or monitoring data from Prometheus, Grafana, and the EFK stack is not included in the backup archive.
Ensure that the backup is up-to-date before restoring. Tenant cluster tasks such as creating, deleting, upgrading, and scaling tenant clusters, or altering the number of Load Balancer Virtual IP addresses create changes in the data that will not be present in the backup. If you perform such tasks, make a fresh backup of the control plane.
Note |
Restoring a Cisco Container Platform environment from an outdated backup may lead to unexpected IP address conflicts, unmanageable tenant clusters, or an unsuccessful restore. |
Before you begin
When the target for a restore is a new cluster, you must ensure that additional free IP addresses are available to avoid conflicts with the IP addresses that are currently in use.
For more information on the requirement for additional free IP addresses, refer to the Managing Networks section of the Cisco Container Platform User Guide.
Procedure
Step 1 |
Log in to the console of the master node of the Cisco Container Platform Control Plane.
|
||
Step 2 |
Run the backup-k8s-artifacts.sh script to create a backup of the Kubernetes artifacts. backup-k8s-aritfacts.sh
Run the following command:
|
||
Step 3 |
Copy the tar file generated in Step 2 to a secure location outside of the current master node of the control plane. |
||
Step 4 |
Run the percona-backup.sh script to create a backup of the percona database that contains data related to ccp-api, ccp-networks, and ccp-appdata. This data is used when restoring the control plane data. percona-backup.sh
Run the following command:
|
||
Step 5 |
After running the backup script, note down the following information from the console:
|
||
Step 6 |
Use the scp utility to copy the ccp-percona-db-backup.tar file from Step 4 to a secure location outside of the master node control plane.
|