Maybe it's something unique to our setup, but we find that our Amazon EBS volumes fail approximately once every 1 - 2 months. We are running a pretty normal Mysql server through RightScale with fairly low traffic. And yet, without fail every month or two our main database will die because the EBS store "goes away"
We had a similar experience with failure every 3 months or so. We had a small portion of our services on Amazon and it was the main reason we moved it back to our own Servers.
I really don't know how they can continue to offer a service with a failure rate as high as EBS. Amazon should be ashamed of themselves for having such a glaring gap in their otherwise strong collection of products.
I've been told about this sort of thing, but I wasn't sure how widespread it was. Reading the latest RFO from Amazon, I guess I'm not terribly suprised. Its unfortunate that its so difficult to get 3rd party storage into amazon's infrastructure without paying an arm and a leg, maybe someone will do this some day.
We had two disappear in quick succession shortly after migrating to ec2 a couple of months ago (one running postgres and one varnish), but nothing since.
