Hey Cryptic, here is a suggestion. How about you cut the shard in half. Run 1/2 the servers and when they crash (not IF, but WHEN) then you quickly turn on the other half. Then you fix/reboot the broken ones and put them back on stand by. Keep repeating until you get a real IT team to manage things.

I take it you have no idea how servers or server farms work with there hotswap redundant parts?

EDIT: Also if you half the server capacity then you loose player slots and performance, making the game even more unstable than it is now.

Suggestion: You are not EVE and apparently cannot do the one shard thing

So, how's that hot swap working out for you this weekend eh? Hot swap parts only work if there is a hardware failure, not a software crash.

So halving the servers so one is at a out of date state when the other crashes is a much better solution?

Yay for rollbacks...

You can't possibly believe that the database is on the same server that handles the requests and is running the software do you?