I have an iflow that the data come from the external system is imported to S4H system through the java deployed on scp in which call the RFC function of the S4H system.
Here is my iflow.
1. store the request payload in the body of the content modifier.
2. call the java application deployed on scp.
So, our flow is SCPI→Java→S4H
If the large amount of the data is imported , one of the following errors receive in the log of the SCPI although the timeout setting of HTTP channel is not less than the processing run time.
javax.servlet.ServletException: Error during write of the response: , cause: java.io.IOException: Broken pipe
javax.servlet.ServletException: Error during write of the response: , cause: java.io.IOException: Connection reset by peer
Whenever the above error occurs the large amount of the data is imported, I was noticed about that
after about 10 mins, the multiple rfc calls are running in the background although only one rfc call have in our java api. After about 1 hour, the scpi is completed with fail status with the above error message and when we check the S4H
about 2 times or 3 times of the data we imported had been created or updated in the site of the S4H system.
The synchronous call of the HTTP or request-replay component may be the root cause of the 2 times or 3 times data imported when the connection time out ? Please I would like to know the root cause and how to repair our iflow.