cancel
Showing results for 
Search instead for 
Did you mean: 

JDBC Receiver Adapter: Guarnteed delivery

Former Member
0 Kudos

Hi All,

I have a JDBC receiver scenario. I was investigating the Guarnteed ( Exactly Once ) delivery option of the JDBC receiver adapter. The Following two scenarios needs to be validated

The Technical scenario is SAP R/3 --- XI -


Database

Scenario1 : The message has been triggered by SAP r/3 to XI. The Target Database then crashed. The message errors out in the Adapter engine and was not sent to the database. Would the messsage be resent again, till successful delivery. Do we need to set the Advance Mode, parameter "Number of Re-tries of database Transaction on SQL error". Or should this happen automatically from the Adapter without any setting in the adapter configuration.

Scenario2: There is a setting as Exactly Once Handling .

i have gone through the documentation on SAP help

[http://help.sap.com/erp2005_ehp_03/helpdata/EN/22/b4d13b633f7748b4d34f3191529946/frameset.htm]

But this is not very illustrative

it has 2 modes

PERSISTENCE ( values local and DB )

CONFLICT RESOLUTION ( Error or Redo )

Is this only after a post-commit system crash? Can anybody provide some more documentation, or blogs for the same.

In all we would expect that, a message Triggered from R/3 should always be delivered to the database.

please let me know know your opinions on the same.

Thanks in advance,

Best Regards

Abhishek

Accepted Solutions (0)

Answers (1)

Answers (1)

prateek
Active Contributor
0 Kudos

Would the messsage be resent again, till successful delivery

The default behaviour of adapter engine is to try sending data to receiver 3 times at an interval of 5 minutes after which the message goes into system error. You may change this behavior through visual admin. After this, you may also manually restart the message.

Do we need to set the Advance Mode, parameter "Number of Re-tries of database Transaction on SQL error"

This is not mandatory. As I mentioned above, it is handled to a certain exten automatically.

Is this only after a post-commit system crash?

The persistance factors when defined could help in error handling cases. This could also be used when a system crash occurs at the time of message processing. Then the message processing has to be done for the second time and one of the two conflict resolution methods are used.

Regards,

Prateek