2 DBA's, and we are on call every other month. The key to on-call is proactive automation and monitoring. Once you get that set up on all your db servers, you are emailed issues before they become major issues and you can deal with them during regular hours, thus preventing those late night calls like a T-Log that finally filled up a disk in the middle of the night. It doesn't prevent all of them mind you, but it does cut the volume down considerably. Also, having a well trained help-desk staff that can ferret out silly non-issues and simple connectivity issues before they get to you can help out considerably as well. If you have that all setup first then being on call is not such a nightmare. I have often found over the years through sheer experience that alot of after hours issues could have been caught during business hours if the checks and monitoring were already in place. There will always be issues in SQL Server that crop up quite suddenly and you must deal with them as they occur. However, I have found that 95% of most issues in SQL Server are as the result of accumulation. In other words they develop over time. This is where monitoring and automation come in. :-D
"Technology is a weird thing. It brings you great gifts with one hand, and it stabs you in the back with the other. ...:-D"