My SQL Server at Discount ASP is mission critical and it was a little disappointing to have it go down during the recent attack on one of the web servers. I had some confidence in the fact that the database servers were on different hardware than the web servers, but in this particular case it seems like it was of no benefit. There was no explanation as to why the database server(s) went down, so would it be possible to explain in here once you guys get back on your feet? I'm not complaining (you guys are rock solid) - I just want a better understanding of your network setup so that I can be better prepared next time. My web server also went down but I have a backup in the UK data center so everything worked according to plan there. I was happy when the 2007 attack happened as you guys learned so much and that's why I have stuck with you. I'm even happier this time as I've been curious for some time to see how well you'd do. My network relies heavily (but not 100%) on your network and I've got plenty of work to do myself over the next few weeks to overcome the problems that became very apparent during this attack.
The attack was so heavy at some points that it actually disabled (system administrators say "killed") one of our Juniper ISG firewalls. Not overwhelmed it - broke it. That's why the entire network suffered. Also the DDoS wasn't just on two customer sites, but also on one of our name servers. The name servers are all in different locations, as they should be, but the truth of the matter is if someone takes out one of your name servers, the problems it causes are going to cascade down through the network (and for a lot of users). The SQL servers are separate from the web, mail and other servers, but all of those servers are on the same network. All behind the same firewalls and switches. This Internet that we all rely on is very fragile, and there isn't anything special about us that allows us to avoid problems (if there was I would steal it and sell it to healthcare.gov). We take every precaution that we can, and we spend millions of dollars to ensure that things are as good as they can be. But as anyone who has ever suffered from a wide outage will tell you, that ain't enough. Nothing you ever do is enough, and sometimes we just have to live through these things. We're sorry you have to live through them with us, but we also appreciate your business.
Because you guys have kept everything going for so long without problems, I think I got a little complacent and had it in my mind that there's no way both my web and database servers would go down at the same time. It seems like all attacks these days are so heavy, that I need to change my attitude to one of "if a data center or one of it's customers gets attacked - be prepared for the whole data center to go down". I totally agree that whatever you do it can never be enough, so we just have to keep getting smarter and be better prepared next time. The thing that sets you guys apart from the others is your honesty of what is going on and your real world attitude. Keep up the great work.
