>> However, I have in mind a much simpler test: fill the entire "disk"
>> (it's a "hardware" (presumably really firmware) RAID array) with
>> data such that by examining a block's contents you can tell what
>> block it is.  Then read it all back and see if all the blocks'
>> contents are correct. Repeat using the block device if the raw
>> device passes this test (I expect the raw device to pass and the
>> cooked device to fail).
> =46rom your description in the same email, I expect both to pass, and
> the problem to be in the fs code layered above this, but proving it
> either way will be valuable.

Yes, after reading the other messages I'm not so sure either.  But I
definitely agree with the part after the last comma. :)

> Whatever you find, can you attempt to replicate it with a -current
> kernel?

Maybe.  Are there snapshots?  My cronjob that's been trying to update
my -current tree has been failing for some weeks now with

Write failed: Broken pipe
cvs [update aborted]: received broken pipe signal
Write failed: Broken pipe
cvs [update aborted]: received broken pipe signal

and I thus don't think I have a very current tree.  (I think I know
what's wrong, but it's not clear what there is to be done about it.
I'm trying a few things but have little confidence they'll actually
work.)

Is there any good way to switch to another cvs server?  I haven't found
a way to do that that doesn't involve refetching everything (which
seems rather wasteful).

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B