First I would disagree with "this is a known issue"

In the vast majority of the cases you do not need to do this.

I had used transactional replication _a lot_ and time and time again time outs occurring one way or another and fixing those takes care of the problem.

In other cases very large transactions could cause this too and once again there are many things that can be done, like adding another log file temporarily, etc.

My suggestion is that you should clarify in the article that such procedure is to be performed as a last resort!!