upgraded to the newer version and you want to restore the previous software version, which
was more stable. In this scenario, you can perform the software rollback either from the web
interface or the CLI. If you have a two-node controller cluster, you must have cluster backup
on both of the nodes.
To ensure that you will be able to roll back to a previous version, Ruckus Wireless
®
strongly
recommends the following before attempting to upgrade the controller software:
• Always back up the controller before attempting a software upgrade. If you are managing a
multi-node cluster, back up the entire cluster, and then verify that the backup process
completes successfully. See Creating a Cluster Backup on page 341 for the local backup
instructions. If you have a local backup and you want to roll back the controller to a previous
software version, follow the same procedure described in Restoring a Cluster Backup on
page 342.
• If you have an FTP server, back up the entire cluster and upload the backup files from all the
nodes in a cluster to a remote FTP server. See Backing Up to an FTP Server on page 345 for
remote backup instructions and Restoring from an FTP Server on page 347 for remote restore
instructions.
Recovering a Cluster from an Unsuccessful Upgrade
If an issue occurs during the upgrade process (for example, a power outage occurs or one of
the interfaces goes down), you can recover the cluster if the controller has either a local cluster
backup or a remote (FTP) configuration backup.
If the Controller Has Local Cluster Backup
Follow these steps to recover a cluster when the controller has a cluster backup stored locally.
1. Unplug the cluster interface cables of each node in the cluster to isolate each individual node.
2. On each of the nodes in the cluster, perform the following:
a) Log on to the CLI, and then execute restore local.
NOTE:
Issue the restore local command at the same time for both nodes, or temporally
disconnect the cluster interface and issue the command. After issuing the command,
connect the cluster interface back.
This command will restore the system configuration of the node from a local backup.
b) When the CLI indicates that the restore local command has been completed successfully,
plug in the cluster interface cable.
You have completed recovering the controller cluster using a local cluster backup.
If the Controller Has an FTP Backup
Follow these steps to recover a cluster when the controller has a configuration backup on a
remote FTP server.
See Backing Up to an FTP Server on page 345 for more information.
You must perform steps on each of the nodes in the cluster.
SmartCell Gateway 200/Virtual SmartZone High-Scale for Release 3.4.1 Administrator Guide
365
Performing Administrative Tasks
Upgrading the Controller
To Next Page
Loading...
