catastrophic failure

spike decided to take an unannounced vacation today. apparently root raid under linux sucks and now i have one dead server. luckily, belle is still functioning and all the backups are intact. it took about 30 minutes to extract the needed setup files from the backups and populate belle. this experience has taught me quite a bit, and i want to record it. belle needs to have copies of all of spike’s setup and they need to be in as much of place as possible; failure should take less than 5 minutes; spike and belle need to do more rsync’ng of each other’s unique data (home,spool,etc); root raid needs to be eliminated in favor of a solid drive and another drive which will be mirrored (but not dynamically); i need serialcon access and a sweet boot/initrd image that has network tools. i’m going up to bellingham tomorrow (if bill allows it) to address some of these things.