Red Hat GLOBAL FILE SYSTEM 5.0 Instrukcja Użytkownika Pobierz pdf (Strona 27)

Once you create the cluster, you can start managing it using the clusvcadm and clustat

utilities. Below are some examples of using these utilities:

• Displaying the status of a clustered service:

# clustat -s <service_name>

• Restarting a clustered service:

# clusvcadm -R <service_name>

• Stopping a clustered service:

# clusvcadm -s <service_name>

• Relocating a clustered service from one cluster node to another:

# clusvcadm -r <service_name> - m <member>

For more information on both utilities, see their man pages.

Testing Clusters

To test the created cluster, you can do the following:

• Use the clusvcadm utility to relocate a clustered service from one node to another:

# clusvcadm -r <service_name> - m <member>

• Shut down or unplug a power cord on a cluster node.

If the cluster is operating correctly, the vz and parallels-server services are relocated from the

source server to the destination one (if you are using manual fencing, you may need to use the

fence_ack_manual tool to inform the cluster of the fenced node). If some service has failed to

start, investigate the cluster logs stored in /var/log/messages by default. Please keep in mind

that the information you are looking for may be placed on different cluster nodes.

C HAPTER 5

Managing Clusters

1 2 ... 22 23 24 25 26 27 28

Brak uwag

Red Hat GLOBAL FILE SYSTEM 5.0 Instrukcja Użytkownika Strona 27