In <199409130308.XAA20172@duality.gnu.ai.mit.edu> 
	"Charles M. Hannum" <mycroft@gnu.ai.mit.edu> writes:
>In <199409130201.VAA00203@packrat.vorpal.com>
>	Michael Graff <explorer@vorpal.com> writes:
>>My roomates and a friend of mine have been getting a lot of DUP errors
>>when writing a large number of files and then rebooting.

>What kind of machine (CPU, etc.)?  How much memory?  How are the disks
>partitioned?  Old or new format file systems?

There is a variety of machines here, of varying hardware and setup.

  1) 486DX/40, 16meg RAM. Adaptec 1542C SCSI: sd0=1033Mb, sd1=412Mb.
  2) 386DX/40,  8meg RAM. IDE: wd0=340Mb.
  3) 386SX/25,  6meg RAM. IDE: wd0=130Mb, wd1=85Mb.
  4) 386SX/16,  6meg RAM. IDE: wd0=425Mb.

The first two machines use OS-BS Operating System Boot Switcher to boot
to either an MS-DOS partition or NetBSD. The last two are completely
NetBSD.

No's 2 and 4 experienced a great deal of problems with the automatic
fsck failing, and a manual fsck resulting in literally PAGES of DUP
errors. In one instance approximately 15 files from /usr/bin (!) were
removed by fsck on #3, files that had been recently installed from a
build, but had survived a reboot or two before.

All machines are built from recent (less than a week or two old)
sources.

#1 had been spared the problem until a recent rebuild (it's about a week
old). At first we thought it was only IDE that fell prey .. not so.

#4 does nothing but go idle for days at a time and serve SLIP and NFS. ;)

>What *is* the phase of the moon, anyway?

If you must know, last night was the beginning of the last quarter.
(see, we can *all* miss a joke here and there .. ;)

Any help or information would be greatly appreciated.

Cheers,
Phil
----------
Phil Knaack, WGA		flipk@iastate.edu
Sociology Department		flipk@vorpal.com
311 East Hall			flipk@cs.iastate.edu
Iowa State University, Ames IA