Browse Prior Art Database

Framework for achieving effective connection retries for applications using SAP JCo

IP.com Disclosure Number: IPCOM000197256D
Publication Date: 2010-Jun-30
Document File: 4 page(s) / 2M

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a framework for connection retries to the SAP system for applications using JCo middleware to connect to the SAP system. The framework proposes a set of components, an algorithm which constitute a retry mechanism to the SAP system. When the connection to the SAP system is interrupted, the retry mechanism comes into action and applications can leverage all the benefits of the retry mechanism.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 39% of the total text.

Page 1 of 4

Framework for achieving effective connection retries for applications using SAP JCo

Framework for achieving effective connection retries for applications using SAP JCoFramework for achieving effective connection retries for applications using SAP JCo

Disclosed is a framework for connection retries to the SAP system for applications using JCo middleware to connect to the SAP system. The framework proposes a set of components, an algorithm which constitute a retry mechanism to the SAP system.

SAP's Java middleware, the SAP Java Connector (JCo) allows users to easily build SAP-enabled components and applications in Java. SAP JCo supports communication with the AS ABAP in both directions: inbound (Java calls ABAP) and outbound calls (ABAP calls Java).

For outbound, the JCo Server program used implements functions that are called by the RFC Client. The JCo server program is registered via the SAP Gateway and waits for inbound RFC calls.

An RFC server program registers itself under a program ID to an SAP gateway
If an RFC call is passed on from any SAP system to this SAP gateway with the option "Connection with a registered program" (with the same program ID), the connection takes place with the corresponding JCo server program.

Once an RFC function has been executed, the JCo Server waits for further RFC calls from the same or other SAP systems.

If a RFC connection is interrupted or terminated either due to the connection to SAP being lost or the SAP system being shutdown, the JCo server tries to register itself again on the same SAP gateway under the same program ID. JCo would try to connect back infinitely. The time after which it would try to connect would change from 1, 2, 4, 8, 16…..200 s in exponentially varying intervals. This internal re-connecting to SAP is the only method applications have to connect back to the SAP system.

Using this default behavior has several limitations.

Firstly, it is not under the control of applications as to how the application should retry connecting to SAP and they are forced to conform to JCo's default behavior. Applications cannot override the infinite connection retries to the SAP system. Applications may want to retry only a finite number of times, and if they are still not able to connect to the SAP system then take actions appropriately for example, throwing an exception.

The time interval after which the application would try connecting back to the SAP system is also not fixed and varies exponentially. When the application is started the time interval increases exponentially from 1, 2, 4….seconds. Suppose the application connects back to the SAP system in the fourth attempt. The next time the connection gets interrupted, the application would try connecting back after a time interval of 8, 16… seconds rather than starting back from 1, 2, 4….seconds. Once this reaches 200s, the interval would remain fixed at 200 seconds. Applications may not desire this kind of retry time intervals.

The...