Identifying Source Routes - Compaq ServerNet ServerNet II Installation Manual

H-2 Compaq ServerNet II PCI Adapter and Switch Installation Guide
subsequent file name increments the name by 1. The file name restarts at results001.txt when all
test result files are removed.
NOTE: For example, if files results001.txt and results003.txt already exist, the next file name will be
results002.txt. After that, the next file name is results004.txt.

Identifying Source Routes

It is useful to be able to identify the path messages use to traverse the SAN. The "source route,"
a sequence of numbers, is one way to identify such a path. Each number identifies a port of a
node in a SAN. Each port can be on a different node, depending upon the SAN topology.
From the perspective of data communications, transactions occur between a local host computer
and a remote node. The local host computer initiates the communications and the remote node
replies to the initiation. From the narrow view of a single transaction, each Switch, also called a
node, has two ends: the end closest to the local host computer, and the end closest to the remote
node.
A source route has real meaning only when it is associated with a starting node on the SAN,
usually the local host computer. The source route is an identifier with the following usage:
The first number in the source route refers to the port on the starting node. The first
I
number may begin with either an "X" or a "Y", indicating the ServerNet II fabric.
The second number refers to the end port on the neighbor node.
I
The third, and subsequent numbers, refer to the remote-node exit ports on all the
I
neighbor's neighboring nodes.
It is possible to specify an invalid source route, that is, a source route that does not exist on the
SAN in question. When the source route is valid, it can also be used to refer to a node in the
SAN, sometimes referred to as the "target node."