HP ProCurve 6120G/XG, Руководство по настройке

Страница: 200 / 662

Configuring IP Addressing
Configuring a Single Source IP Address
Configuring a Single Source IP Address
Overview
This feature applies to the following software applications:
• RADIUS
• SNTP
• System Logging applications
• TACACS
• Telnet
• TFTP
The above IP-based software applications use a client-server communication
model, that is, the client’s source IP address is used for unique client identifi­
cation. The source IP address is determined by the system and is usually the
IP address of the outgoing interface in the routing table. However, routing
switches may have multiple routing interfaces due to load balancing or routing
redundancy, and outgoing packets can potentially be sent by different paths
at different times. This results in different source IP addresses, which creates
a client identification problem on the server site. For example, there is no way
to designate a fixed IP address for outgoing packets for RADIUS or TACACS,
so it is necessary to configure in the RADIUS or TACACS database all possible
IP addresses that are configured on the switch as valid clients. When using
system logging, it can be difficult to interpret the logging and accounting data
on the server site as the same client can be logged with different IP addresses.
To decrease the amount of administrative work involved, a configuration
model is provided that allows the selection of an IP address to use as the
source address for all outgoing traffic generated by a specified software
application on the switch. This allows unique identification of the software
application on the server site regardless of which local interface has been used
to reach the destination server.
Specifying the Source IP Address
The CLI command
ip source-interface
is used to specify the source IP address
for an application. Different source IP addresses can be used for different
software applications, but only one source IP address can be specified for
each application.
8-20