🚀 go-pugleaf

In article <7541@swan.ulowell.edu>, page@swan.ulowell.edu (Bob Page) writes:
> I wrote:
> > Oh, in case you think this is just an AmigaDOSism ... here's a line
> > from the UNIX (4BSD) man page for mount(1):
> >      Mounting file systems full of garbage will crash the system.

> peter@sugar.UUCP (Peter da Silva) replied:
> >Bob, that's misleading. ... Once you mount the disk successfully,
> >UNIX will stay up even in the face of *rampant* disk corruption

> No way.  Look at the source code.

Love to. Will you email me a copy? :->

> The 4.3 file system has at least 49
> system panics (read: crashes) alone.  Duplicate inodes, freeing free
> blocks/inodes/frags, corrupted bitmaps, block reads that return zero
> bytes, corrupted cache, and run-of-the-mill things like linked
> directories and lost directory entries.

We get some of those things fairly often in our environment without
crashing. Of course we don't have the good fortune to have the latest and
greatest Berkeley product... just 30 intel boxes running that bastard Xenix
and a couple of SYSV machines.

Also, at least some of those problems are not attributable to defective
media: corrupted caches, for example.

Finally, UNIX is much more dependent on disks than AmigaDOS. Are you sure
that some of those panics aren't restricted to the root partition?

> My point is ... UNIX is not the answer to everything.

No, it's abominably poor at real-time work. It does do a better job than
AmigaDOS of dealing with defective media, though. It recognises that it
exists.
-- 
-- Peter da Silva      `-_-'      ...!hoptoad!academ!uhnix1!sugar!peter
-- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter
-- Disclaimer: These may be the official opinions of Hackercorp.