My company hasn't been able to deploy any updates to our ElasticBeanstalk application for about 7 hours due to this. Luckily there's nothing urgent we need to deploy, but this makes me extremely nervous about using AWS going forward. If we experienced an outage that required a rollback or forward fix we would be totally hosed.
In defense of AWS, the ddos is probably one of the most massive and targets dns servers. I fully expect the dns to be strengthened after this incident. At the same time, using another provider doesn’t resolve the problem of ddos attack, which can happen at any point in a public network, not limited to DNS servers
AWS’s track record is still good, though. I mean, there will always be problems with cloud technology. But would you rather try to mitigate a DDoS attempt yourself or would you rather Amazon do it? I think giving up on AWS because of this event would be kind of throwing the baby out with the bath water.
In my experience, you eventually piss SOMEONE off, or a competitor thinks it can drive business to them, etc. You probably won't get a huge one, but even a small DDoS can be difficult to mitigate for a company with limited resources.
