According to Jukka Marin:
>
>Uh, are you talking about failing disks or other _hardware_ problems?
>

Yes I was/am.

>I can well understand that with malfunctioning hardware, the system is
>likely to crash, not even panic (it doesn't have 16-bit CRC's for every
>disk block used for swap, does it?),

Not at the kernel level - the disk drive/controller actually do have
fairly sophisticated error recovery built in, capable of recovering
from long bursts of bit errors.  Don't ask me how long - I cannot
remember now.

> but how about the situations of
>running out of swap etc.?
>

It should not panic - I would expect things to slow to a crawl or
appear to stop but the machine should not panic.  If it does then,
IMHO, it is a bug in the kernel.

>
>Is bad swap == running out of swap?
>

No, I meant bad swap to mean hardware errors _reading_ back swap data.
Running out of swap should be handled robustly - this is not a
situation where I would expect a machine to panic.

>I agree, but maybe some panics could be avoided if the system was trying
>to protect itself a bit harder?
>

I don't know - I have not looked at all the places that panic is
called.  My expectation is that panic is only called when there is no
other way out, without looking at all the code that calls panic I
cannot say if this is true or not.

-- 
Brett Lymn, Computer Systems Administrator, AWA Defence Industries
===============================================================================
  "Upgrading your memory gives you MORE RAM!" - ad in MacWAREHOUSE catalogue.