Shit happens. My last employer is now mostly on AWS, and mostly in that region, so things broke all over the place yesterday (oddly enough the only service that

Shit happens. My last employer is now mostly on AWS, and mostly in that region, so things broke all over the place yesterday (oddly enough the only service that is multi region was auth). In the past when we were mostly in our own datacenters we suffered as well, so that's not better either. One time a bad DNS update killed every single service worldwide and it took most of a day to properly restart thousands of services. Build complex systems with thousands or millions of things and it's bound to blow up sometime.