OK, then let's build out a test case to reproduce the problem. Can you post some sample data of rows that were successfully updated and some that were not...

Then your code should go through 4686 iterations, one for each row. If you are left with rows with NULL...

When you say, 'I have xxxx records', do you mean that this is the count of rows that match the IS NULL criteria in your cursor?

Any ideas? - it's hard...

Yea, Garadin is correct. You need a second FETCH statement in your while loop. Also, something doesn't look right to me here. Your code does an update...

That's the problem then. If you've written your stored procedure to only insert one row at a time, you are stuck with RBAR unless you re-create the SP to...

So you are looking for a stored procedure that can do a set based insert into one table from another. Quickly to you means good performance?

Here's an example of...

Can you explain what you mean by 'quickly call' the SP? The code for inserting from one table to another is simple so I'm a bit stuck here as...

Well, you need some way to make sure the data does not exist in the target table prior to inserting it. You could accomplish this by using an IF...

Can you post your code?

I am a bit confused as to what the problem is here. Can you explain a bit more on what your stored procedure is doing?

Well, this is a bit out of the box and I've not done this before, but it may be an option for you to try. You may be able...

Is this a production DB? If so, you'll want to wait for down time to run the BACKUP LOG command with TRUNCATEONLY. Unless, of course, you don't care...

Once you've backed up the log, you should be able to use DBCC SHRIINKFILE() on the log file. This will reduce the physical size of the log. How...

Yes.

Good explanation of the self-blocking processes. Also, for your deadlocks, here's a good link to help :http://www.sqlservercentral.com/articles/Performance+Tuning+and+Scaling/tracingdeadlocks/1324/