Occassional "NO_MAPPINGPROGRAM_FOUND" on synchronous SOAP to RFC scenario

[Former Member]

Hello experts!

We have various SOAP web services created and working fine in production. However, we seem to randomly get this message about NO_MAPPINGPROGRAM_FOUND, it sounds as if could be a temporary cache issue. Here is the exact text from Adapter Log:

 Received XI System Error. ErrorCode: NO_MAPPINGPROGRAM_FOUND ErrorText: null ErrorStack: Mapping program is not available in runtime cache: Object ID 3A41778C9D643875AFD70C9570510B34 Software Component 3EF6DC10FD5411DFA931FD2E0A114C39 

and then...

 The message was successfully transmitted to endpoint com.sap.engine.interfaces.messaging.api.exception.MessagingException: XIServer:NO_MAPPINGPROGRAM_FOUND: using connection SOAP_http://sap.com/xi/XI/System.

finally...

 The message status was set to FAIL.

Shortly after (without any manual cache refreshes or intervention) this error goes away and the transaction works again. This has happened to a few of the web services and I don't see a pattern. It is happening like less than 1% of the time.

Any ideas on how to eliminate this glitch are greatly appreciated. We are on PI 7.11 SP03.

-Aaron

[Former Member · Accepted solution]

Hi Aaron,

You should check dump error in SAP system via tcode ST22 whenever this error occurs, this is the common error of the RFC adpater when the BAPI is dump.

Rgds,

Thien Duong

[Former Member]

Aha! The problem wasn't random, it was because of the occassional short dumps due to data lengths or validations in ABAP programs. The short dumps in ERP correspond to the alerts and failures raised in PI.

Thank you very much.

-Aaron

[suman_saha]

Hi Guys,

     I am getting the same issue in our Live system (RFC to SOAP scenario). It has thrown an error saying

"MAPPING">NO_MAPPINGPROGRAM_FOUND twice.

I have checked ST22 in ECC system and its saying ABAP Runtime Error COMPUTE_FLOAT_ZERODIVIDE . The interface is running properly for last 3-4 years and many successful messages processed even on the same day through that interface. When the XML is tested in our test system its working fine i.e. no problem with data also. What may be the root cause of the issue?

Thanks,

Suman

[Former Member]

Hi Suman,

Everytime I see this, it always has something to do with the data...Sometimes even missing config...

I would say that your best chance of isolating this issue is to somehow try to get (temporary) debug access in Prod (I know this is hard but it's sometimes necessary) and take the XML that errored out through the debugger. So you re-produce the error where it happens. That's probably your best bet at getting to the root.

Regards, Trevor

[Former Member]

So in your case it sounds like the ABAP has failed with a divide by zero (bad input data).

PI has sent the web service call to SAP, but never got a valid response from SAP for any entry showing in that ST22.

I think the No Mapping Program Found might mean that we haven't mapped an error message type on the synchronous call.

Verify the exact data parameters sent to ABAP by looking at the PI payload and match up the exact time as the ST22 messages. You can test these parameters in the ABAP program yourself or ask your ABAP team to verify their program with those parameters.

Then you can either ask them to improve their error handling or put better input validation on the web front end or in PI.

If you really don't think it is a data issue, monitor the SAP system and maybe these transactions are running too slowly sometimes, causing the ABAP to fail.

Bottom line - I think if you see ST22 errors which line up with the mapping errors in PI, the failure is occuring in the RFC call to ABAP.

Good luck!

[rajasekhar_reddy14]

Hi,

This error comes maily when your SAP System not gives reponse , so this reason could be

1)Data what you have sent is not correct.

or

2) Function Module is giivng dump might be some issues.

Regards,

Raj

[baskar_gopalakrishnan2]

Just few cents....

a) Please check your WSDL endpoint.. When you transfer the interfaces from one environment to other say then the host name or ip address will be changed. Are you recreating WSDL for every environment? or dealing with the same wsdl and manually edit the endpoint location. If the endpoint location in the WSDL is invalid, you get this error

b) Are u having web dispatcher and they wronglly point to another valid URL where URL is valid but software component does not exist over there? Then you might getting this issue.

Basically the request is delegated based on the endpoint URL available in WSDL. So try to see the endpoint location is valid one.

[Former Member]

Hi Baskar, thanks for the reply!

a) I think the endpoints are fine because the interfaces are working ok until we randomly get this message. No changes are actively being made during the times of the errors.

b) We only have 1 instance of PI so there are no other valid URLs that would even work part of the time, if I'm understanding your idea correctly.

The main point is this problem is transient. All web interfaces are tested initially, but at some later time we get one of the best synch messages error failing due to the Cache maybe or some other reason.

Thanks again but I will keep the thread open for more ideas.