Getting a Table Column value in Catch section while using Output clause for Update Statement

It would have helped to see some example code, preferably with sample data. That could have helped to clarify what you are looking for.

But if the question is "I'm updating a thousand rows in a single UPDATE statement and I get an error. How do I determine which row(s) that caused the error?", the answer is that this information is not available in the general case. Sometimes it can be deduced from the error message, but far from always.

In SQL Server, when using the OUTPUT clause with an UPDATE statement inside a TRY...CATCH block, capturing the ID of a specific row that causes an error is not straightforward, because if the error occurs during the statement execution, the entire update operation fails — and SQL Server does not commit any changes, nor does it populate the OUTPUT clause.

To track the ID of a row that causes an error during the update, especially when updating multiple rows, you can approach this in a two-step manner:

Approach: Update in a Cursor or Loop with TRY-CATCH Around Each Row

This method processes each row individually so that any error can be caught along with the ID.

Step-by-step solution:

CREATE PROCEDURE dbo.UpdateWithErrorTracking
AS
BEGIN
    SET NOCOUNT ON;
    DECLARE @ID INT;
    DECLARE @ErrorMessage NVARCHAR(4000);
    DECLARE @ErrorNumber INT;
    DECLARE @ErrorSeverity INT;
    DECLARE @ErrorState INT;
    DECLARE @ErrorLine INT;
    DECLARE @ErrorProcedure NVARCHAR(128);
    -- Temp table to hold IDs of rows to update
    DECLARE @IDsToUpdate TABLE (ID INT);
    -- Fill with IDs to process
    INSERT INTO @IDsToUpdate (ID)
    SELECT ID FROM YourTable WHERE <YourConditions>;
    DECLARE cur CURSOR LOCAL FAST_FORWARD FOR
    SELECT ID FROM @IDsToUpdate;
    OPEN cur;
    FETCH NEXT FROM cur INTO @ID;
    WHILE @@FETCH_STATUS = 0
    BEGIN
        BEGIN TRY
            UPDATE YourTable
            SET Col1 = 'NewValue'
            OUTPUT inserted.ID, inserted.Col1
            WHERE ID = @ID;
        END TRY
        BEGIN CATCH
            -- Capture error info
            SELECT 
                @ErrorMessage = ERROR_MESSAGE(),
                @ErrorNumber = ERROR_NUMBER(),
                @ErrorSeverity = ERROR_SEVERITY(),
                @ErrorState = ERROR_STATE(),
                @ErrorLine = ERROR_LINE(),
                @ErrorProcedure = ERROR_PROCEDURE();
            -- Log or raise the error with the ID that failed
            RAISERROR('Error updating ID = %d: %s', 16, 1, @ID, @ErrorMessage);
            -- Optionally insert into error log table
            -- INSERT INTO ErrorLog (ID, ErrorMessage, ...)
            -- VALUES (@ID, @ErrorMessage, ...);
        END CATCH;
        FETCH NEXT FROM cur INTO @ID;
    END
    CLOSE cur;
    DEALLOCATE cur;
END;

Key Notes:

This method ensures fine-grained error handling per row.

You sacrifice some performance (because of row-by-row operation), but gain error traceability.

If performance is critical and the error frequency is low, you could try the bulk update first, and fall back to this method only in the CATCH.