The only truly safe formats for DATETIME/SMALLDATETIME in SQL Server are:
yyyyMMdd
yyyyMMdd hh:nn:ss[.mmmmmmm]
yyyy-MM-ddThh:nn:ss[.mmmmmmm]
----------^ yes, that T is important!
Anything else is subject to incorrect interpretation by SQL Server, Windows, the provider, the application code, end users, etc. For example, the following always breaks:*
SET LANGUAGE FRENCH;
SELECT CONVERT(DATETIME, '2013-11-13');
Result:
Le paramètre de langue est passé à Français.
Msg 242, Level 16, State 3, Line 2
La conversion d'un type de données varchar en type de données datetime a créé une valeur hors limites.
Just changing the language (which any of your user sessions can do) forced SQL Server to interpret that as YYYY-DD-MM
instead of YYYY-MM-DD
. Similar things can happen with setting like DATEFORMAT
. But these settings are literally ignored when using the above two formats.
Always, always, always use one of the above two formats. If you are passing a variable as a string, stop doing that. If you can't, check to make sure it passes ISDATE()
first. If you are letting people type any date string into a form field, stop doing that, too. Use a date-picker or calendar control and dictate the format of the string before you pass it to SQL Server. Well, depending on the language, just keep it as a datetime value and don't convert it to a string at all.
Please read this post:
There is an exception: SELECT CONVERT(DATE, 'yyyy-mm-dd');
will not break. But I err on the side of consistency rather than using a format only in the one place where I know it doesn't break, and having to use a safer format everywhere else.
Here is what you need to do
- Open the table in definition view
- Select a cell in the column which is empty.
- Add an entry of that data type (I picked the datetime in an example where this happened to me)
When done, you should be able to click the data button and see all the data.
Best Answer
The problem isn't the number of rows, the problem is that you have a row which does not fit into the decimal(18, 0) data type. If you're on SQL 2012 you can identify the rows fairly easily (and can do your own number checking function on other versions).
Once they're identified you can fix the data on those rows and then convert the table over as you originally attempted.