Managing Systems and Workgroups: A Guide for HP-UX System Administrators

Setting Up and Administering an HP-UX NFS Diskless Cluster
Administering Your NFS Diskless Cluster
Chapter 10928
Administering Your NFS Diskless Cluster
If you have chosen “shared” for the cluster policies and you manage all
printers/plotters and file systems as cluster-wide resources, your HP-UX
cluster will look and act much like a single, multiuser computer. For the
end-user there is little difference between HP-UX running on a
standalone system and any member (server or client) of such a cluster.
However, administering a cluster, even with shared policies and
cluster-wide resources, involves significant differences from managing a
standalone system. “What Is an NFS Diskless Cluster?” on page 899
explains the characteristics that make a cluster different from a
standalone system (a computer that is not part of a cluster). This section
shows how these characteristics affect system administration tasks in
practice. Refer to the NFS Diskless Concepts and Administration White
Paper (supplied in /usr/share/doc/NFSD_Concepts_Admin.ps on most
10.x systems) for detailed information on cluster administration, such as
single point administration and “DUX” versus NFS diskless
administration differences.
In the day-to-day administration of a cluster, it is important to
understand where (on which cluster node) to perform a given task.
Table 10-3, “Where to Perform Tasks,” on page 928 summarizes where to
perform selected tasks. Table 10-4, “Tasks Required by Specific Events,”
on page 931 summarizes which tasks to perform given a specific event.
Table 10-3 Where to Perform Tasks
Task Where to Perform the Task
Configure a cluster (define,
modify, install, and remove
clients)
Server
Back up a cluster (full backup,
incremental backup, or full
archival backup after first
boot)
a
Server
Back up private client file
system
Client to which the disk containing
the file system is attached