• My experience is that a monitoring tool is preferable to 'roll your own' solutions. Setting up operators means having to keep track of every person who might get an alert ever. and maintaining them as people come and go or switch job functons. It's much easier when everything is in one central location. We user Foglight, and it monitors everything else besides job failures. Although tools cost money, keep in mind that staff is getting paid real money for the time spent on developing and maintaining home grown solutions. The OS manager at one of my past positions commented once 'that we finally realized that we are a bank, not a software company'.