Hope i get some help here. We are seeing this problem only in one of our environments, and stopped us from moving ahead.
We have a customer, who is sending the data thru HTTP Post.
I see the contentStream object in the server log for both DEV and TST01 and the service works fine. (I used the tracePipeline)
But the same code we moved into TST02 environment and we are not seeing contentStream object, but rather we see the broken data as string in the server log.
We spent all day yesterday to see the diff in servers and found nothing.
I really need some help here… Please shed some light on whats going wrong.
I’d suggest first putting a savePipeline step in the beginning of your service and then comparing the saved pipeline from your Development server to the one from your Test server.
You may also want to get your network team involved and ask them to sniff the packets coming in from the customer. It could be that the customer is simply sending the wrong thing to your test server (ex. the content-type may be set incorrectly.)
Atlast I could find a solution for this issue, but not sure if it is the right one.
It is working as expected when I unregister the content type “application/x-www-form-urlencoded”. Please let me know it will lead to any other issues??
Rajesh,
application/x-www-form-urlencoded is the default content type for HTTP POST requests. Unregistering the WM default/built-in content handler for application/x-www-form-urlencoded means that you’ll be receiving HTTP Post requests as raw content Streams as opposed to the name value pairs (form parameters) parsed by the HTTP handler.
~tS