Occasional SYSFAIL tRFC entries for WORKFLOW_LOCAL_100 since ECC 6 Upgrade with Event Queue lock message

[Mike Gambier]

Fellow WUGgers,
 
Since we upgraded to EEC 6 the new Event delivery framework has thrown up an odd situation to do with the table lock placed on SWEQUEUE.
 
Whenever Work Items being executed by users try to trigger events asynchronously whilst at the same time the background SWEQSRV job just happens to be running as well, the tRFC for the event raise is set to status SYSFAIL in ARFCSSTATE with the message 'Object requested is locked by <whatever userid scheduled the job>'. 
 
Later on, when the next RSARFCEX job picks up these tRFCs, the error messages are then changed to 'Data has been changed in the meantime - the action is canceled'. 
 
We're currently unsure as to whether anything happens with these tRFCS afterwards, but their date and time values continue to change with each new RSARFCEX job run (although their keys do not) so presumably they are re-processed in some way, again and again, unless we trash them via SM58.
 
We've opened a ticket to ask for some advice on this one, but I wondered if anyone else had encountered this?
 
As a precaution we're trying to increase the delay between each SWEQSRV job run to see if we can reduce the number of errored tRFCs but that's merely a patch on a bigger problem I think.
 
Regards,
 
Mike GT
 
_________________________________________________________________
Twice the fun—Share photos while you chat with Windows Live Messenger. Learn more.
http://www.microsoft.com/uk/windows/windowslive/products/messenger.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/pipermail/sap-wug/attachments/20090212/def1f11a/attachment.htm