Hi experts,

I have a flow running which converts an incoming INVOIC-IDoc into a Flatfile (simple line by line-structure).

When the invoice is small, it works perfectly, if it’s big, the IS4.6 says “bye-bye” without further notice (program disappears completely from memory and server needs to be restarted).

I need to be able to process IDocs with 1000-2000 positions. Right now it crashes at position 450.

My configuration: SAP BC 4.6, Servicepack 3 with EDI-Adapter, 512MB RAM, 1GHz PIII, W2k-Server, JDK 1.2.2. Javamem: 384 - 768MB (tried 256MB first).

Any ideas how to free memory or best practice infos on how to handle large Idocs?

Thanks,
Gunter

For Large Doc(XML) or IDOC handling
try with these 2 built in IS services to process your flows with less memory usage.

pub.web:getNodeIterator

by specifying your criteria of IDOC(segments)in the i/p parameter which gives the o/p Iterator and map it to I/p of

pub.web.getNextNode.

So by utilization of both these services may help your problem.

Cheers…