Posted in Amazon, AWS, Cloud, Google, Facebook, Networks

What constitutes the Internet

The recent hysteria about the massive and unfortunate AWS outage in US-EAST-1 and their S2 storage issues. Has raised the discussion about the vulnerability of the Internet. First lets be clear here, the Internet is NOT services like Amazon, or Google, or Dropbox, or any one of thousands of ‘Sites’ ON the internet. The Internet did not fail during the AWS outage, Sites on the Internet were offline, as in, “not on the Internet”, or at best unavailable as a facility there on.

The internet is a web, which can be fragile, but is mostly fairly resilient to most things, including facilities being disabled, or unavailable. So when you listen to talk about ‘LOSING the INTERNET’ take it with a grain of salt. It’s probably more about loosing connectivity with someones favorite destination on the Internet, facebook, Netflix whatever and less likely about the Internet actually being down.

Posted in Cloud, history, Internet, Networks

The Expenditure of TCPv4

While working on an TCP/IP problem today, I was finally struck by the fact that we have for all intents and purposes expended the entire TCPv4 addressing space. I knew it was coming, years ago, but now while testing IP addresses, it dawned on me.

You can now pick any arbitrary set of numbers nnn.nnn.nnn.nnn and expect a response. Ping them, probe them, something will be there, or it’s being held. All gone, this is the equivalent of spitting in the middle of an ocean while swimming,  you are going to hit ocean.

4,294,967,296 (232) addresses gone, 4 Billion addresses in use…..