I have some problems using a swap-partition on my SCSI-drives. Everything
worked without any problem while I used a swap-partition on an IDE disk,
and the SCSI disk itself has not proved troublesome before either.
I have defined my entire second SCSI disk (80 Mb, SCSI id 6) as swap.
It has now one (primary) partition:
(bash) fdisk /dev/sdb
Device Boot Begin Start End Blocks Id System
/dev/hdb1 1 1 978 239592+ 83 Linux native
I called mkswap, and then swapon -a (after editing /etc/fstab), and it
worked like it should - until the next reboot.
Having modularized all my SCSI stuff, modules got loaded when swapon -a
was called in /etc/rc. And it ended with the following error-message:
ll_rw_block: device 08:11: only 1024-char blocks implemented (4096)
I think this was right after sd_mod was loaded - all usual messages were
printed. I have a NCR810 SCSI adaptor, and both drivers (53c7,8xx and
ncr53c8xx) cause the same error.
Worst of all, after this error the system is locked, though it does still
respond to CRTL-ALT-DEL. This is, of course, way before virtual consoles
are set up, so I couldn't switch to another console either.
But the strangest thing of all: if I boot in single-user mode (which does
not execute any rc-file at all on my system), and I use modprobe on the
SCSI modules, there is no problem at all! If I load kerneld by hand,
though, and let kerneld load the modules, the system crashes again.
I'll probably go back to using a swap partition on an IDE drive again,
but I am not really happy.
Any help is appreciated!
-- Frodo Looijaard <firstname.lastname@example.org> Find my PGP public key and more at http://huizen.dds.nl/~frodol/ Updated: a guide for installing glibc-2 as an additional target under Linux
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to email@example.com