Wednesday, August 22, 2007

SSIS Troubleshoots and Bugs

ah, Now you face some troubles, before we proceed, you have to install SQL Server 2005 SP2. This Service Pack fixes alot of bugs. and also generate another. Anyway, after you install it you might face one or more of the following:

1- Could not add the DataFlow task?

check the SQL Server running service. Open the "Administrative Tools->Services-> SQL Server Integration Services " and change the logon to "Local System Account", then restart the service.

also check that most of the services uses the "Local System Account", not the network account.

this problem occurs when you are in a network, the Sql server uses this account as a general user to work with. But this is wrong

2-Error [OLE DB Destination [2923]]: The column cannot be processed because more than one code page (1252 and 1256) are specified for it.

this happens with the OleDB module that transfers the data into the destination DB.

the problem is the code page is differ, the problem has two solutions:

a. put a "derivative column", and convert all columns code page to the DB server code page.
b. go to the destination [Ole Db Destination] and check "always use default code page" to true.

3- I drawn my transformation but when I replaced it I got error?


A general bug. when you drag a component fromthe toolbox and drop it into the board. It takes a number. if this component generates an elements like the datasources, each element took a number. the problem is this number is unique. and cannot be reallocated to another unit.

the solution redraw the package....

4- Fuzzy lookup somehow took so long and other times does not take time?

the reason is simple, your data contains duplicate rows.

the Fuzzy lookup component get the data, if the data duplicated it builds another indeces. and allocate it to the rows. the do the fuzzy operation. when you try to apply the fuzzy algorithm to duplicated rows, it doubles the effort for ignoring the result of this row values.

Next time for the SSAS

1 comment:

  1. Thanks a lot. One of your solution really helped me!!

    ReplyDelete