Wow. Just Wow. "Engineering" Fail.

I want to prefix what I’m about to write with 2 statements:

  1. Please Amazon get all of these systems back up ASAP. Peoples lives are in danger.
  2. I don’t usually like to call out peoples failures, I fail often enough all by myself. But the post in question compelled me.

With that said…

I saw this link on Hacker News today.

I read it and was shocked. Horrified even.

3 instances, in the same region.

Monitoring cardiac patients at home.

What came to my mind: WTF WERE YOU THINKING!

Perhaps you should design something a little bit more fault tolerant!?!

And even if you didn’t, even if you relied on the reliability of a single region, why not update DNS and deploy to a different region?

That way you would at least be monitoring those cardiac patients?

I’d really like to find out what company this is so I can make sure that anyone I know never relies on them.

Ever.