Failback

9. Make sure that you are logged in to the actual, failed node:

[VTL_Plus]vtladmin# uname -a

SunOS VTLPLUSN1 n.nn Generic_nnnnnn-nni86pc i386 i86pc

The system should display the expected node name.

10.If you are not logged in to the correct system, you have accidently logged into the management IP address (which always connects to the active node) rather than the service (“monitoring”) address. Close the ssh session, and ssh to the other IP address for the failed node.

11.Change to the directory that holds the VTL executables:

[VTL_Plus]vtladmin# cd /usr/local/vtl/bin

12.Before proceeding further, make sure that no I/O is being sent to the failed node. Make sure that all backup jobs have completed and that failover has completed successfully. Stop I/O, if necessary.

If host I/O is not stopped, data may be lost.

150 VTL User Guide • May 2008

G • 96267

Page 164
Image 164
Sun Microsystems Virtual Tape Library manual Make sure that you are logged in to the actual, failed node