4 ms·

Probably related to the AWS outage in Virginia

• blitzar 5y ago
  Genuine question - why not distribute over a) multiple AWS regions and b) distribute over multiple providers so that 'small' outages dont take down apps like this?
  • 0xdeadb00f 5y ago
    I always thought they did distribute among AWS, Azure, Cloudflare and GCP as the output from running their domains on runson.cloud seems to suggest. But after this outage, clearly not.
    • blitzar 5y ago
      Perhaps someone out there with skills could make a cloud of clouds to make implementing such a thing easier.
      • 0xdeadb00f 5y ago
        Hmm. Some sort of distributed or dectraslised kind of architecture might be worth exploring. What a novel idea! :D