Using Fiddler as a server to server tamper

[Shish]

Hey all,

I have server to server XML based methods which I use to engage the communication between 2 servers. 

I cannot use SoapUI since we are not using WSDL services, hence I would like to simulate server calls and responses by using Fiddler.

Is there a way of doing so? and if there is, could some one please aid me doing so?

Thanks

[Eric Lawrence]

Hi, Shish--

Can you elaborate a bit about what you mean when you say "Simulate"? You can use Fiddler to issue requests using its COMPOSER tab (either by resending existing requests exactly or after modifications, or by generating wholly new requests). You can get Fiddler to respond to requests without hitting a server by using Fiddler's AUTORESPONDER tab (either by returning previously captured responses exactly or after modifications, or by returning files off the local disk).

Regards,
Eric Lawrence
Telerik

[Shish]

Hi Eric,

Thanks for your reply.

When I said simulate, I meant generating requests, like the ones you mentioned in the composer tab.

But my issues are a bit complex than regular requests. I have a client, and 2 different servers communicating between them.

My client cannot capture the calls between the servers as it only communicate with 1 server and with limited method requests.

To make things simpler, the client communicates with server #1, and server #1 sends requests to server #2 (which he then responses to server #1). In that fashion, the client communicates only with server #1.

I'd like to initiate requests from my client to server #1 but to manipulate the request from server #1 to server #2, and manipulate responses from server #2 to server #1. I only know how to generate requests and responses that are independent, but I don't really know how to do it that in an end to end scenario like I just described.

Thanks

[Eric Lawrence]

Hello,

From the sound of it, server 1 is a client of server 2. So, you want to change

 Server1<---->Server2

to be:

 Server1<---->Fiddler<---->Server2

You do that by configuring the proxy settings on Server1 to point at a Fiddler instance. You then use rules in Fiddler (e.g. Rules > Customize Rules) to rewrite requests from Server1 to Server2, or rewrite responses from Server2 to Server1.

Regards,
Eric Lawrence
Telerik

[Shish]

Hi,

Thanks for the answer.

Indeed, Server #1 is a client of server #2.Is there a guide to configure the proxy in the way you mentioned?

Also, we have a lot of other requests to server #2 as not only  server #1 is a client, meaning we have a lot of traffic going to/from server #2 besides server#1. Isn't there a risk of setting such proxy, which can get the traffic collapse entirely?

[Eric Lawrence]

Hello, Shish--

Typically, what you'd do in this case is configure the proxy settings on "Server #1" to point to a Fiddler instance (running either on "Server #1 or elsewhere). Then, the traffic from Server #1 would flow through the Fiddler instance.

As to how to configure the proxy on Server #1-- that entirely depends on what technologies you're using that need to have their proxy settings adjusted. Instructions for most platforms can be found in the documentation, e.g. http://docs.telerik.com/fiddler/Configure-Fiddler/Tasks/ConfigureDotNETApp

Regards,
Eric Lawrence
Telerik

[Shish]

Hi,

I forgot to mention that Server#1 isn't configurable by me, as it is an external server.

So what should I do?

Also - regarding the consequences of a proxy in the middle of heavy traffic - is there a risk of placing a proxy in the middle of heavily requested server?

[Eric Lawrence]

Hello, Shish--

As you seem to have guessed, there's no way for you to capture traffic between Server #1 and Server #2 unless you have the ability to configure either of those servers to use Fiddler.

Regards,
Eric Lawrence
Telerik