A bolshevik plot....taken straight out of the book.
That could be a big part of it. DOS kiddies like to screw with us.
By the way, for those who are wondering about FR's hardware, my understanding is that we have about a dozen servers at the data center. About half of them are dedicated to handling Free Republic back-end processing. We also have a couple front-end servers directing the traffic and balancing the load, a couple database servers, an administrative server, a secure server, and a real-time database backup server. We have two main IP's, one as a standby if the first goes down. We also have some older servers that John uses for backup and development.
Theoretically, we have a fail-safe system. The "fail" seems to be working great, now we're working on perfecting the "safe" part.
It appears that our current series of crashes have to do with our traffic load. It appears that our servers begin overloading and first one crashes (by becoming completely jammed and failing to respond) then the next, and the next, etc, until the entire system and all backups fail to respond. That's when we have to start rebooting. And we've had some problems recently with our rebooting process.
We do not know yet why we are overloading. We have been able to handle several times our normal load with no problem. We do not rule out DOS attack, but cannot find anything in the logs to identify it. John will continue trying to isolate the problem, but we do not feel at this time that it is for a lack of server power.
All I can say is, please bear with us. We've experienced these kinds of systems problems before whether hardware, software, balancing or Bolshevik, and have always been able to work through or around them eventually.
God willing, we will persevere!