on 03-29-2012 4:46 PM
Dear experts,
We have a FTP to PI to FTP scenario. This afternoon during a period of 15 minutes we had messages that were in the queue for up to 10 minutes.
When I look at those messages in RWB the status is "Successful" but if I compare the "Start"-time with
the "End"-time I get about 10 minutes.
After 15 minutes the Start-End times look okay again. Why would messages stay in the queue for so long?
More important: How can I find out about that? Is there a log for the adapter framework (that's where they were stuck
I think)?
Thank you very much for your help and best regards,
Peter
This is difficult to monitor. But I have many questions for you. Does this happen often or you noticed it for the first time? You may check the default trace at server level. In case you had errors during that period, it must have been logged there. Were there some other scenarios running successfully at the same time without any delays? Could you find any error in communciation channel monitoring? How large were the files being processed?
Regards,
Prateek Raj Srivastava
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Prateek Raj Srivastava,
It happened the first time, other scenarios were also affected. There was no error in the communication channel monitoring. Within these 15 minutes there were about 80 messages processed according to SXMB_MONI. 10 messages have a size of around 160Kb, the rest is much smaller < 80 kB. In SXMB_MONI I cannot see big differences between "Start"-time and "End"-time (< 2seconds.)
I checked the logfiles but I could not find any errors in them. That makes me think that there was a connection problem on network level.
I have to check with network people.
Thank you for your help and best regards, Peter
I guess you message are getting stucked in the DLNG status in the messaging system queue.
Check SAP Note 1604091 & 1473299 and set Maximum Concurrency and poolWaitingTime parameter.
Additional check this blog http://scn.sap.com/community/pi-and-soa-middleware/blog/2011/01/26/tuning-the-pi-messaging-system-qu... and also increase the file adapter thread with the Messaging System parameter specified in SAP Note 791655.
Eg:
(name=File_http://sap.com/xi/XI/System, messageListener=localejbs/AFWListener, exceptionListener=localejbs/AFWListener, pollInterval=60000, pollAttempts=60, Send.maxConsumers=10, Recv.maxConsumers=10, Call.maxConsumers=10, Rqst.maxConsumers=10).
~SaNv...
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Look like the problem seems outside PI.
I would recommend you to see the below server level directory log.
/usr/sap/<SID>/D<InstanceNr>/j2ee/cluster/server<number>/log
You might want to check ICM logs too.
Couple of other possible reasons: If the file system goes space disk out of memory or any other process at OS level takes more memory which might have slowed down the process. Check with your unix adminstrators whether any network slowness or similar issues during this time.
Note: So if the problem occurs again, you might want to check the above areas.
Hope that helps.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
90 | |
10 | |
10 | |
10 | |
7 | |
7 | |
6 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.