I have been getting a large number of errors at boot concerning read
errors on /dev/hdb, when I have the following combination of drives:
Master: 1.2Gb IDE PALLADIUM drive (Don't buy these, they are slow)
Slave : nothing connected
Master: Hitachi ATAPI CD-ROM drive
Slave : Fujitsu 170Mb IDE drive (Model = ??)
Kernel 2.0.0 on /dev/hda3, swap on /dev/hda2
I use the Fuji drive to ship stuff to and from work, because I don't have
a zip drive, my SCSI controller blew up, and it's bigger and faster than
When the Fuji drive isn't connected, everything is fine. When it is
connected, the system attempts to read partition info from /dev/hdb
(which isn't there). All of the drives continue to work without
a problem, but I can't (as yet) find a reason why it is probing a drive
that isn't there as many times as it is.
If I connect the drive to the primary slave (which I can do) it still
works fine, although in the tower case, the primary IDE is somewhat
buried, so I use the secondary to reach that bay.
As it's not a real problem, I'm ignoring it.
Put down your spam, we have you surrounded.
This message brought to you by the number 4, and the editor VI.