Chapter 3 System Preparation
HPSS Installation Guide September 2002 159
Release 4.5, Revision 2
Totest whether an IP address is reachable (non-zero exit status indicates the ping was not
successful):
% ping -c 1 <ipAddress>
• Determinewhich networks will be used for control vs. data paths. DCE should not use all
availablenetworks on a multi-homed system unless each of those networks is guaranteed
tohave connectivity to other DCE services. If a particular network is removed (physically,
or routing is changed), that connection remains in DCE's RPC mappings. DCE will
continue to try and use that network despite the loss of connectivity. The timeouts and
retries can have an adverse affect on HPSS as well as other applications relying on DCE.
◆Isolate DCE communication to the designated control network(s) on all HPSS, DCE,
and SFS nodes in order to separate the HPSS control and data paths.
◆Asdesired, tell DCE to ignore certain network interfaces in order to separate the HPSS
control and data paths. For example, to ignore FDDI (fd0) and HIPPI (hp0), add the
following line to/etc/environment:
RPC_UNSUPPORTED_NETIFS=fd0:hp0
• Placeall HPSS, SFS, and DCE server machine IP addresses in a local host table (/etc/hosts).
ForAIX, configure the machine to use the table as backup in the event of a DNS failure. The
file/etc/netsvc.conf should be modified to look like the following:
hosts=bind,local
The netsvc.conf file is used to specify the ordering of host name resolution. In the above
ordering,DNS will be used first, if the host name is not found, then the local /etc/host file
will be used.
4. For each ethernet network interface, verify that both theen0 and et0 interfaces are not
configured at the same time (we recommend only usingen0 unless the other machines in
thenetwork are all using the 802.3 et* interface). Configure the local name service with the
uniquehostname for each network interface on all nodes and verify that each hostname is
resolvable from other nodes.
To understand and optimize the operation of HPSS, some baseline measurement, evaluation, and tuning of
the underlying network and IO subsystems is necessary. Appropriate tuning of these components is
necessaryfor HPSS to perform as expected. Any one component that is not performing at its optimal rate will
havean adverse affect on the performance of the entire system. The steps and tools listed here will help a site
make the best use of their available resources. The measurements taken should be saved for future reference.
If performance problems occur later on, these values can be compared with new measurements to determine
if the performance problems are related to changes in the subsystems. Though disk and tape configurations
rarelychange without an administrator's knowledge, networks can "mysteriously" down grade for a variety
of reasons. This is especially true for client access to the HPSS system.
• Verify that network TCP throughput has been optimized and the performance of each
network is at expected levels in both directions (especially check HPSS data networks
betweenMovers and between Mover and client nodes). Using ttcp or another network tool,
measurethe performance of all the networks that will be used in communications between
DCE,HPSS, SFS, and client machines. If multiple paths between machines are to be used,
thenall of them need to be measured as well. The transfer rates for networks differ greatly
depending upon the individual technology and network settings used. It is important to