Wants to check after installing service pack 4. so that we can confirm it should not happen in future.
Any clues to reprodue it.
If the following conditions are all true, Microsoft SQL Server may store floating point data with an exponent lower than -308, which may cause floating point underflow exceptions that terminate a clients connection to SQL Server:
• The client application is using stored procedures or server side cursors to perform data modification
and passes the request to the SQL Server server as a remote procedure call (RPC) event.
• The client application is passing parameters for the RPC event to the SQL Server server.
• The column of the table affected by the parameter that is passed is defined as a float datatype.
• If a stored procedure is called, the parameter is defined as a float datatype.
And plz check the previous link that I gave you...You could get you answer there.
Last edited by rudra; 10-26-06 at 23:12.
Success is the ability to go from one failure to another with no loss of enthusiasm.
- Sir Winston Churchill
Thanks for your information. I have tried following ways to reproduce it
1. By using Index Tuning Wizard Execution
2. By passing the expression 0/0 (zero divided by zero) to SQL Server as a floating point value for a stored procedure parameter
3. By trying query with aggregate function
4. By running a Complex Query
5. By Query optimization
But i couldnt able to reproduce it. Do you have any stored procedure or SQL Query to stimulate this problem.