on 12-18-2006 7:53 AM
Dear buddies.
My scenario is ABAPproxy-XI-JDBC to modify BD2 tables.
At first everything went fine. However once the operation of modify database occured errors, such as dupicated key words or item format mismatched, the message kept delivering status in adapter engine comparing with the successful status in integration engine.
After reaching the processing limitation, 5 maximum for instance, the following messages were set to 'To be delivered' status. The processing is extremely slow. For several hours the message would be processed and report system error or successful.
I have checked many articles on the forum but I hardly find any similar situation. Looking forward your replies and thanks in advance.
Alex
Hi,
maybe you're using EOIO quality of service
if so then the first error blocks all later messages
and you need to unblock the first one so the rest will be processed
Regards,
michal
-
<a href="/people/michal.krawczyk2/blog/2005/06/28/xipi-faq-frequently-asked-questions"><b>XI / PI FAQ - Frequently Asked Questions</b></a>
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi, Michal.
what you have told me seems to be reasonable. I use exactly once quality of service of an asynchronous message. If so would you please show me how to unblock the first error. As I have no idea which one is the first error and perhaps it is on the delivering status in which can not be changed manually.
Moreover I found that if more messages of EOIO QoS blocked in the queue the BE Qos messages could not be processed either.
Thanks for your help.
Alex
uH~~, Bhavesh.
I have read that blog however that didn't solve my problem. I only use Exactly Once QoS, not EOIO. And the messages stuck in the AE queue with the status Delivering which can not be resend or cancel. The delievring status is a temparory one described in the SAP notes and it should not appear in the RWB monitor usually.
Do you have any other ideas?
Still many thanks.
alex.
Alex,
Like I told , normally the adapter engine retires 3 times at an interval of 5 minutes . So this problem should have been sorted out within 15 minutes with an error SYSTEM ERROR.
Refer into naveen's blog I had listed earlier and it discusses the Adapter Engine Errors,
Check the IS_RETRY set in your server as shown in this blog,
Regards
Bhavesh
Bhavesh,
I have followed naveen's blog and limited the retries times to 0. According to this method, the message was reported system error after the first try failed and the error message will not stuck the adapter engine queue anymore. Seem to solve the deal.
BTW, the delivering status, describing the transfering state from adapter engine queue to the end application, is just a temprary state. Usually we can not watch these kind of messages in the monitor. Why in my situation the delivering status is keeping for such long time. Something abnormal must happen.
Thanks for your nice help.
Alex
Hi,
><i>BTW, the delivering status, describing the transfering state from adapter engine queue to the end application, is just a temprary state. Usually we can not watch these kind of messages in the monitor.</i>
very true Alex! . Maybe the load on the server is too high.
You can look into this fine tuning guide and tune your server for optimum results,
Regards
Bhavesh
Hi,
Whenever there is an error , XI tries to restart / resend the message 3 times at an interval if 5 minutes. Maybe you can turn this off.
More info on this blog,
/people/sap.user72/blog/2005/11/29/xi-how-to-re-process-failed-xi-messages-automatically
Regards
Bhavesh
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
85 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
4 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.