I missed the thread, but found the summary interesting. One thing that I
would have added: I have an older system, and two h/ds, one of which was
scavanged. A couple of times this year I've had problems with the whole
system hanging, and the disk access LED was on.
The solution was to take the system down to single user mode (init 1), then,
and make sure that swap is off (swapoff), *and* after making sure that you
know what partition you are using for swap, do a
mkswap -c /dev/whatever
This may take a bit, since the -c switch will force a physical disk check,
and it will find and map bad blocks, before recreating swap. Once this is
done, you can do swapon -a, and go back up to multiuser (either reboot, or
init 5).
This has solved my problems. Both times, it was swap that hit a bad block,
and couldn't read.
mark
--
The truth will out: someone got it at last:
Dogs have masters; cats have staff.
--
redhat-list mailing list
unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe
https://listman.redhat.com/mailman/listinfo/redhat-list
