PI Scenario fails, resend does not work, test message does

[former_member569931]

Hello Guys,

so, as stated in title, I have this scenario on my hands.

Little information:

It's async SAP -> Third Party Service scenario. It's based on IC and basically just converts an IDOC to HTTP POST.

This scenario is in production environment and processes about 350 messages a day successfully. About one message in 1000 fails. This is a small number, however, this functionality is critical, so we still try to solve it.

This all runs on PO 8.0.

Now to my question:

The failed messages returns a:

com.sap.aii.adapter.http.api.HttpAdapterException: STATUS_CODE_NOT_OK-Internal Server Error

When I hit resend, I get back the exact same error and the sending fails.

But here is the problem: If I just take the IDOC from first step (before mapping), and i run it through NWA->SOA->Monitoring->Send Test Message using the same exact scenario that failed in the first place, and failed via resending, the message will process and get delivered successfuly.

Do you guys have any idea why this might be the case, or give some pointer as to what to look for?

[dhivakar]

Hi Jakub,

Generally this issue occurs due to the content type parameter in the HTTP channel. But in your case it occurs only for certain messages.

What is the frequency of the message flow in this HTTP receiver adapter?

[former_member569931]

Hi Dhivakar,

Thank you for your response.

I'm not sure if this is what you are asking, but I think the messages come through every few minutes.

If this is not what you were asking, please let me know how I can find the required information.

I also add the header information of both a failed and a successful message:

Best Regards,

Jakub

[dhivakar]

Thanks for sharing the information Jakub!

This is not due to any of your PI configuration. This is mostly due to the huge frequency of messages that hits the target server at the same time which results in certain failures. but it will be okay once you resend the messages.

Hope this is something with the target systems capacity to receive the huge number of messages simultaneously.

Best Regards!