Chad Callihan talks about a potential problem:
Of all the problems that can cause a database transaction log to fill up, perhaps one lesser-known cause is a large clustered index rebuild. Let’s look at a scenario where this issue could cause a headache.
Click through for the scenario. I will say that I disagree with reorganizing the index under pretty much any circumstances. Jeff Moden, in a long and deep but extremely interesting presentation he gave to us at TriPASS three years ago, laid out why that is. I do agree that rebuilding a clustered index should be uncommon and may be a sign of a data modeling problem.
The other thing to note is that this problem is bigger when you rebuild online and your data drive is the same as your log drive. When you rebuild an index online, SQL Server needs enough space to make a copy of the data, so the rule of thumb is that you’ll have 2x the normal disk space requirements during online rebuild. It’s not necessarily exactly 2x because of fragmentation, compression, and other factors, but that’s why it’s a rule of thumb and not a “guarantee or your money back” type of thing.