Connection Pooling

[Former Member]

Steps to cofigure connection pooling in J2EE Engine

i am using Web Application Server 6.40

please it is little bit URGENT

How to perform Connection Pooling in Web Application Server 6.40

With Regards

Krishna Kanth

[Former Member · Accepted solution]

Hi,

If you want to configure connection pooling for a data source that you use to connect to a relational database, you can do it in the runtime of JDBC Connector Service in Visual Administrator tool. To start the tool, run the go.bat file in <drive>\usr\sap\<SAPSID>\<instance_name>\j2ee\admin\ directory. For more information about the available connection pool parameters, see http://help.sap.com/saphelp_nw04/helpdata/en/2d/292391fa6745488f3e0e0d4b03c64e/frameset.htm.

Zornitsa

[Former Member · Accepted solution]

Thank You Zornitsa

it was informative .

i am new to this field actually.

i want to know the concepts regarding connection pooling too and how to implement in WAS 6.40.

i have an Oracle as Database.

so what .jar file i need as driver and the whole set with a step by procedure.

sorry for troubling you

With Regards

Krishna kanth

[Former Member · Accepted solution]

Hi,

First of all, is your Web AS running on the Oracle database? If it is, than you can simply use the system driver and the system data source.

If not, download the appropriate driver from Oracle's web site. Register it in JDBC Connector Service runtime in Visual Administrator. Then use it to create a data source. You can find description of these procedures in the documentation that I pointed you to in my previous post. I guess you will use a driver for JDK 1.4, so you should create a JDBC 2.0-compliant data source. Create the data source with either Native or Vendor SQL support (this is set in the Additional tab, as you will read in the documentation). Then you can set the connection pooling parameters. Unfortunately, there are no strict recommendations about this - you should decide by yourself what values you will choose for these parameters. The documentation gives you some general hints about this, and you can check how it works with your scenario.

Hope this helps at least for a start,

Zornitsa

[former_member374 · Accepted solution]

Hi Krishna,

You are thanking a poster for his answer.

Please next time also give him points.

See: /people/mark.finnern/blog/2004/08/10/spread-the-love

This time I did it for you.

Remember: When you ask a question you also assume the responsibility to assign points.

Please check your other questions and if the answers were good.

Spread the love and give some points.

Thanks in advance, Mark.

[Former Member]

Background:

The J2EE is being used to provide data to a remote cube via a UDConnect data source from Teradata.

The connection container (SDK_Teradata) parameters are set as follows

Connection Lifetime: 150 sec.

Cleanup Thread 10 sec.

Problem:

There are two problems related to the above parameters.

1. When a Bex query fails on the remote cube, it leaves a corrupt connection between J2EE and Teradata. Subsequent queries that attempt to use the connection also fail. This problem can be reduced by decreasing the Connection Lifetime parameter.

But...

2. When the Connection Lifetime parameter is too low, connections are dropped during ACTIVE queries and an incomplete result set is delivered to BW. This provides corrupt results in Bex.

Has any experienced this problem? and Corrected it???