I just had to run through a reset of the subscriptions table in the database for the third time in six months. While it doesn't take long it does require shutting down all the services for a few minutes.
There is an existing article that references this issue: The Alert me when a node reboots alert does not reset?
The article does reference the exact issue. It is the Alert on reboot alert where I always notice the issue; it is supposed to reset in 5 minutes, but doesn't. Once I run the SQL code provided in the article the issue goes away. For a while. I haven't been able to pinpoint a trigger; we had a core firewall upgrade this past weekend, so we've had a lot of activity over the past few days.
I'd love to get some insight as to why this keeps happening, and why I need to go through this process on a regular basis. Failing that, I'd just like to know if anyone is experiencing this regularly so I know I'm not just going crazy. 
Thanks.