Data anomaloy after upgrade
Posted: Wed Jul 01, 2009 8:56 pm
In the process of exporting table data from PC 10 with the latest updates, I have run into what appears to be a problem with a couple of tables. in both faprior and faprevyr, the data has been static since the upgrade from PC 8.5 in April, 2007 - not a single addition or deletion. This condition is consistent in every monthly backup since the changeover date. Additionally, the famstrd table contains all transactions since 1 Jan 2007 while the PC 8.5 tables contained full transaction data back to 1997.
The purpose of going back to previous backups was to re-assemble some long term historical data. Based on PC 8.5 as well as earlier PC 6 and PC 7x tables led me to expect to find the famastr and or famstr/famstrd tables contained all the historical data I was seeking since - to the best of my knowledge there has never been a data purge on the system. I readily concede that such a purge could have been run, but the static table contents of the two historical tables has me questioning the validity of a couple of years worth of reporting - at the very least.
Fortunately, reconstruction of the tables in an independent exported system was relatively simple and the methods could be verified against seemingly consistent V8.5 tables but that still leaves me pretty uncomfortable about our current production system. I'm not that familiar with PC but it bears all the hallmarks of failed end of period/year closing. What should the tables faprior and faprevyr have in terms of date ranges? Are these tables really significant given the ease with which they can be constructed given the appropriate historical ledger tables? Should it have been detected by the system and reported?
The purpose of going back to previous backups was to re-assemble some long term historical data. Based on PC 8.5 as well as earlier PC 6 and PC 7x tables led me to expect to find the famastr and or famstr/famstrd tables contained all the historical data I was seeking since - to the best of my knowledge there has never been a data purge on the system. I readily concede that such a purge could have been run, but the static table contents of the two historical tables has me questioning the validity of a couple of years worth of reporting - at the very least.
Fortunately, reconstruction of the tables in an independent exported system was relatively simple and the methods could be verified against seemingly consistent V8.5 tables but that still leaves me pretty uncomfortable about our current production system. I'm not that familiar with PC but it bears all the hallmarks of failed end of period/year closing. What should the tables faprior and faprevyr have in terms of date ranges? Are these tables really significant given the ease with which they can be constructed given the appropriate historical ledger tables? Should it have been detected by the system and reported?