I've found dstat and htop handy. Also, lsof has been a lifesaver many times for me.
Checking /proc/mdstat to see if a drive failed and it's rebuilding a hot spare can help--but you should've got an alert for it already... /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
