Preparing for switch configuration

Strategy for initial configuration

Strategies for initial configuration of transparent proxying are as follows:

Set up the appliances offline.
Setting up the appliances offline before passing all your network traffic to the L4 or L7 switch will have the least impact on your users. If you deploy transparent proxying during planned network downtime, you might be able to work directly on the network without affecting your users.

The procedure described in the online Help assumes that you will configure the appliance offline, test the configuration, then pass your network traffic to the switch.
Set up transparent proxying on an installed appliance.
You can set up transparent proxying on an installed appliance if a spare network interface is available. A separate network interface card enables the existing NetCache service to continue during the conversion process.

If you choose to use this method, ensure that for HTTP transparent proxying you do the following:
- Leave the HTTP Proxy Ports setting as it is so that users can still proxy Web requests through the existing NetCache configuration.
- Add an additional HTTP Proxy Ports setting of 80 for HTTP transparent proxying.

Avoiding traffic loops

If you have more than one network interface and you are deploying transparent proxying with an L4 or L7 switch, make sure that the port through which outgoing NetCache traffic is sent is not configured as a transparent port. If outgoing NetCache traffic is sent back through an incoming cache port on the switch, the switch passes the traffic back to the appliance, creating a traffic loop. If a traffic loop occurs, no requests are serviced. For HTTP, MMS, and RTSP transparent proxying, this event is recorded in the Messages log file. For NNTP transparent proxying, the NNTP trace log file can help you trace the path that the request has taken. The NNTP trace log is available only through the CLI.