Since updating to Accpac 5.6A we have had numerous occasions where the Day end Proces "falls over"

It appear to always do it on process in a sales transaction (Record already exists is nornmal) and this then requires our Business Partner to dial in and sort the underlying table then re run the Day End. After this we need to manually post through the AR invoice batch and then clear down any zero batches i GL transactions and post any batches with transactions therein.

Feedback from Accpac themselves is said to be problem with ODBC links but these have all been reset and........ still we have this recurring problem

So are we a one off or is there anyone else having a similar problems?
Back in the earlier versions over the last 3.5 years we never had this problem so it appears to be a 5.6 only occurrance

Thanks
Ian

Comments

  • I believe we found the problem - we had a version of 5.6 on the same server and believe the 2 were not good bedfellows - result is that the day end would "fail" on a random basis

    Ian

  • Nope still falling over

    Damn and the fix for the day end falling over appears to be deleting a line out of the database to allow the posting to continue - now we are having problems with IC inventpry movement reports as transactions do not "show"

  • Strangely enough this is now reported by Sage as being a known issue and they issued a small fix in a sub upgrade. But the initial "Fix" where Sage / accpac tech support advised our business partner to delete a line from the posting source has resulted in inventory reports being screwed up.
    They then said here are 3 pieces of script to run to sort the problem - but this screwed it up even more

    Net effect is months later inventory reports are totally fictitious and we cannot obtain a definative stock report from the system.

    Now suggesting to run more scripts, and of that fails move to V6 as that will cure allills - my money is on not!

  • Do you ever get a "iloc mismatch error" when the day end fails? If so then we have a similar problem. If so it is not a problem in 5.6. Our error started several versions before that.