For the last couple of years, I’ve been doing virtually all of my development on an Amazon EC2 server I’ve called “the forge”.
In order to make sure my backup system was 100%, I just “tested” having a catastrophic failure. I brought up a new instance, cloned the EBS volume (disk), and switched the elastic IP.
New server, identical to the first.
Total time? Less than 5 minutes, and absolutely zero detectable downtime (I’m sure there was a few seconds during the elastic IP switch).
Now, for the second time, I feel reluctant to “terminate” the instance of the server that has served so faithfully for nearly a year:
16:43:23 up 298 days, 18:36, 2 users, load average: 0.00, 0.00, 0.00
And, again, I re-iterate – I’ve never had a hardware solution (shared or dedicated) from any hosting company that has been as reliable as the “virtual” servers from amazon – nor any that I could replicate and get back up so quickly in the event that there was a failure.
Good work Amazon!
{ 1 comment… read it below or add one }
Lucky escape. Good work Amazon!