We are interested in applying model checking techniques to the verification of communication protocols which require safe communication. Typically, in such scenarios, one desires to demonstrate that one party can reliably communicate information to another party without a third party being able to determine this information. Our approach involves using the modal logic of knowledge, which has only relatively recently been studied in the context of secure protocols. We demonstrate our approach by means of a detailed case study: The Russian cards problem. This is an example of a security protocol with nontrivial requirements on the knowledge of the agents involved. Using the Russian cards problem as an example it is shown how the satisfaction of properties involving knowledge can be verified in a standard model checker -- in our case, SPIN.