Take the time to look at your logs and search for patterns emerging over a period of time, as the source of many problems often go back quite a long time (eg. page corruptions are not always immediately noticeable). There are a few typical unimportant entries, such as
INET/inet_error: connect errno = 10061
or the Guardian restarting and of course, a routine shutdown.
There are however, a few important entries which you should take note of, should they appear in your log.
Terminated abnormally: an indication that someone has shut down your Firebird server by pulling the plug.
Modifying procedure xxx which is currently in use by active user requests: this occurs fairly often with Firebird 2. It's not critical if you modify a procedure whilst others are using it. The problem arises due to the
multi-generational architecture - when others are working with the procedure, you can only see the results of the old procedure.
Page xxx is an orphan: if this message starts to occur regularly, perform a backup and restore.
Page xxx wrong type: unfortunately this one's pretty terminal, because it's a clear indication that your database is corrupt. It is important to determine which pages are affected, because they may not be in use any more, or only store old record versions. In this case the problem will be solved by the next database sweep. On the other hand, if you're unlucky the next database sweep will turn it into a real problem!