on 04-05-2010 4:01 PM
Hi,
We have developed a BPM collect scenario for Collecting the IDOCs. The IDOCs will be collected until an Collect end IDOC is received by the BPM. If the End Collect IDOC is not received, then I have a time out of say 3 mins after wich the N:1 transformation is done in the transform set and finally a SEND step in the BPM.
Im using correlation in the above mentioned BPM.
The Issue Im facing here is that, say for some reason the End Collect IDOC got stuck in R3 system and the actuall Idocs are collected in the BPM, the BPM waits for the END Idoc for 3 mins and then the time out is triggered and the file is written in the FTP folder. What happens after this is that if the END collect Idoc reaches the BPM after this (after 3 mins), it fails in the QUEUE (SMQ2/outbound).
My understand is that, it fails coz of the correlation that have been used in the BPM, is that correct and is there any solution to this.
Ofcourse we have schedule the RSOUT pgm (background job)in the R3 system, which will ensure that no IDOC is stuck in R3, but the problem is the time out. Any suggestions pls.
Thanks
Prasanna
Hi Michal,
I didn't know that the BPM collect is the most inefficient way to handle IDOC packaging.
Thanks for the blog it is very informative and I will try to figure out someother way of doing it.
Thanks
Prasanna
Edited by: Prasanna Shanmugasundaram on Apr 30, 2010 8:43 PM
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
hi,
why not use an alternative ? which will not be causing errors (also like this) ?
/people/michal.krawczyk2/blog/2009/05/21/xipi-collecting-idocs--possible-ways-with-pros-and-cons--5-ways
it really pays off to change it before you start using BPM collect pattern on PRD
you will save yourself issues later on
Regards,
Michal Krawczyk
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
87 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.