Suppose I have two services, each of which launches the same Hibernate application, using the same second level cache. Now suppose that one of the services cannot connect to the level 2 cache (someone has set a firewall rule that prevents access to the cache cluster). How does this server work after that? How is data consistency between servers?
An experiment is called. But I would argue that you get an immediate exception to the Hibernate operation, which suffered a communication failure, not very different from what would happen if the database is unavailable.
Practice should be such as to handle the exception and could not get it from the cache, then it should go to the database and get the values.