Gang,
I use shred to erase SCSI disks from defunct server hardware. I have a workstation set up with a SCSI card, four drives to be shrad, and boot off an IDE CDROM with an old cd of "Linux Bootable Business Card". The machine is a P4 2 ghz, 500m ram. The SCSI is an Adaptec 29160 and the drives (in this instance) are Seagate Ultra320 73 gigs.
I just do 'shred -v /dev/whatever' in a vty for each disk.
After a couple days of it working, the system comes to a standstill with each drive's lights barely/rarely showing any activity. top shows each shred using as much processor power as possible, memory not so bad.
I've heard of shred slowing down during the random stuff but IIRC this was just one of the "ones" runs. I slowly killed each shred instance, one a time, once there was only one drive going the drive light was kind of going again and showing signs of life.

Any ideas what's going on? thanks for any help.

You could try tuning each drive before shredding using hdparm and see if that speeds things up. It may just be that doing 4 at a time is too much for that machine. Or maybe you'd have better luck if you loaded the CD "toram" (some live CD's allow you to dump the entire disc to RAM for running) if it's not very big (like 50MB DSL or tiny core) -- maybe the cdrom quits accessing during the shred...