Verifying Connectivity After Configuring Deployment

After you configure a deployment, you need to verify connectivity between the networks to ensure that traffic is optimized on either side.

This section describes ping -r and traceroute, as well as the pros and cons of using each. Finally, it summarizes a procedure for verifying connectivity.

ping

ping is a good general tool to verify reachability. However, it is not the best tool to use to verify correct deployment of WAN optimization appliances because:

1	It doesn’t verify the path that traffic takes.

It’s important to verify the path, not just reachability, because the appliance must intercept traffic on both sides of the WAN for optimization and acceleration to be effective.

2	It relies on ICMP, and some redirection methods (for example, WCCP) don’t support ICMP.

You need a tool that can verify paths by revealing all hops taken along a path. Some tools you can use to verify the paths taken are ping -r and traceroute.

ping -r [or ping -R]: ping with Record Route option

The exact syntax for ping with record route option depends on the operating system you’re using. For ease of discussion, we’ll use the notation ping -r.

Environment	Syntax	As described by OS help...
MS Windows	ping -r 9	Record route for count hops.
Linux/Silver Peak	ping -R	Record Route. Includes the RECORD_ROUTE option in the ECHO_REQUEST packet and displays the route buffer on returned packets. Note that the IP header is only large enough for nine such routes. Some hosts ignore or discard this option.

Pros

•	Most (but not guaranteed all) network devices support it, whether they are routers or not.

•	Shows the return path, too.

Cons

•	Limited to nine devices in the traffic path, including the source and destination.

•	ping -r may fail to verify connectivity with some WCCP deployments.

traceroute

Windows and Unix each have slightly different versions.

•	Both are suitable for non-WCCP deployments, but because Windows traceroute uses ICMP, it isn’t suitable for WCCP.

•	For WCCP deployments, you need to use Unix traceroute or a 3rd-party Windows traceroute that uses UDP instead of ICMP.

•	The downside of traceroute is that only router hops display.

Basic procedure

1	Verify connectivity for optimized traffic.

•	In Router mode (out-of-path deployment), Silver Peak appliances look like router hops. They’ll display in both ping -r and traceroute.

•	In Bridge mode (in-line deployment), Silver Peak appliances look like bridges. They’ll display in ping -r, but not in traceroute.

2	Verify connectivity for pass-through traffic.

As a best practice, always verify connectivity for all devices in the network. For example, if you’ve configured a route policy to cause certain traffic from certain devices to be handled as pass-through or pass-through unshaped, you should also verify connectivity for these devices.

3	Test network connectivity by using your applications. For example, do a CIFS mount or an FTP transfer.

Please send comments or suggestions regarding user documentation to techpubs@silver-peak.com.