question

I tried this, using SQL Server 2012: CREATE TABLE dbo.table_without_meaningful_name(id int identity(1,1) PRIMARY KEY, c1 nvarchar(max),c2 nvarchar(max)) GO CREATE TABLE dbo.log_for_table_without_meaningful_name(id int, c1 nvarchar(max), c2 nvarchar(max), dt datetime default current_timestamp) go create TRIGGER tr_without_meaningful_name ON dbo.table_without_meaningful_name FOR UPDATE,INSERT AS BEGIN set nocount on insert dbo.log_for_table_without_meaningful_name(id,c1,c2) select id,c1,c2 from inserted; END GO insert dbo.table_without_meaningful_name(c1,c2) values('123456789012345678901234567890','this will work'); GO EXECUTE sp_rename N'dbo.table_without_meaningful_name.c2', N'column2', 'COLUMN' This statement works and it does rename the column **c2** to **column2**. I don't get an error message thrown from sp_rename. Neither do I get an error message when changing the column name from the table designer in SSMS. So I can't really reproduce your error message. Can you post some sample code from a trigger, and tell us which version of SQL Server you're using? However, in my example, the actual trigger will break, because it **explicitly** accesses the column **c2**, which I have renamed to **column2**. Solving this automagically will require you to have a DDL-trigger altering the DML-trigger for you. It can be done, but I haven't looked into **how** it can be done. In a DDL-trigger, you aquire the metadata about the change from EVENTDATA(), can store that into an XML-variable and then query the variable to find out which table and which columns were changed.