DUMP: bread: lseek fails (why??)

[David Hinnant]

We had a strange occurance with 'dump' the other day on our 11/780 running
4.1, and before I go rumaging through the source, I thought someone might
be able to tell me what happened. Here's some of what was printed on the
console:

DUMP: bread: lseek fails
DUMP: (this should not happen)bread from /dev/rhp0h [block 175333999]:
	c=0x400, regc=0x400, &c=0x7fffe988, n=0x0
DUMP: bread: lseek fails
DUMP: (this should not happen)bread from /dev/rhp0h [block 1667328588]:
	c=0x400, regc=0x400, &c=0x7fffe988, n=0x0

This continued 13 more times with only the "[block #]" changing, and
then twice more with the "[block #]" and "&c=#" changing.

The system was rebooted, and fsck found nothing wrong with /dev/rhp0h. 
'Dd' backed up the disk just fine with no hard errors.

What happened?  There was about 3 hours worth of activity between the
time when the errors occured, and when fsck was run when the system
was rebooted, so what ever files dump was complaining about may have
vanished.  The next dump from /dev/rhp0h went just fine.  Has anybody
seen this before and have it disappear?

					David Hinnant
					ITT - Telecom
					(919) 829-3033
					...ucbvax!decvax!ittvax!ittral!hinnant