You can write scripts for nagios to monitor those endpoints or do end-to-end tests for you. I have done this in banking , healthcare, and gaming. On a smaller

You can write scripts for nagios to monitor those endpoints or do end-to-end tests for you. I have done this in banking , healthcare, and gaming. On a smaller scale you can probably just have cron jobs run these and email/sms you when things fall down. Forgot to add -- generally though in production you will detect very early on when your code starts crashing or gets rejection rates. In my day job we generally know when FB/Apple/Google are down/degraded before the larger internet because of scale and integration with their payments and our monitoring.