I thought something was amiss when my workstations reported only 2% load this weekend.

It appears that SETI@Home’s science database machine, Thumper, has forgotten how to identify it’s own hard drives.  Since the disk controllers for the server are built into the motherboard (let this be a lesson to anyone running mission-critical servers on integrated components!) and major repairs are expected to be necessary.  So until this server is up and distributing work again, it seems that we’ll all be running with lots of extra clock cycles.

Luckily, the great people at Sun Microsystems are going to be replacing the database server.  As of this writing, they should have the machine running through its final preparations or enroute to UC Berkeley for immediate deployment.  I am curious to know what kind of system Sun will provide, though.  These machines often remind me of the Mars rovers, in that they exceed their expected lifespan by ridiculous margins.

When the new server is up and running (likely Monday evening) there will likely be a huge rush to go out and collect large sums of work.  Hopefully people will adjust their cache to grab no more than two days of work (my PCs typically only cache 3 days of work) so that the few available workunits can be evenly distributed among the community.

There’s right now a discussion in the forums to determine a new name for the server (I’m going with Helios), as well as updates on the progress of the replacement.  Feel free to stop by and add your voice.