Virtual TapeServer 6.03.42 Operations and Administration Guide

Administrative Tasks | 147
The following example of the /etc/fstab file shows Vault10 configured as a GFS partition:
LABEL=/VAULT00 /VAULT00 ext3 defaults 0 0
/dev/pool/pool_gfs /VAULT10 gfs defaults 1 2
Ensure the mount points to an existing pool in /dev/pool, which is where the GFS pool is
defined when configured. GFS mount points start at Vault10.
6. Change the owner of Vault10 to the super user on the VTS server:
[root@VTS001 root]# chown –R bill.bill /VAULT10
Shutting down GFS on a single server
If you need to reboot one server in the cluster, use the procedure below. If you do not follow
these steps, data may be damaged or lost. PERFORM THESE STEPS BEFORE
REBOOTING.
1. Verify that shutting down will not cause cluster instability.
[root@VTS001 root]# gulm_tool nodelist localhost
Here is an example of the output from this command:
Name: VTS001
ip = 192.168.80.108
state = Logged in
mode = Slave
missed beats = 0
last beat = 1202843621267030
delay avg = 10000357
max delay = 10010365
Name: VTS002
ip = 192.168.80.106
state = Logged in
mode = Slave
missed beats = 0
last beat = 1202843624251337
delay avg = 0
max delay = 0
Name: VTS003
ip = 192.168.80.104
state = Logged in
mode = Master
missed beats = 0
last beat = 1202843616024219
delay avg = 10000715
max delay = 10010717
If the server is hostname is listed, ensure that the remaining systems constitute more
than 50% of the list.
2. Unmount GFS /Vaults.
[root@VTS001 root]# umount /VAULT10
Repeat this step until all vaults are unmounted.