As abishek pointed out, this is the readpage bug.
-eric
On Jan 13, 2010, at 2:33 PM, ron minnich wrote:
> On Wed, Jan 13, 2010 at 12:12 PM, Eric W. Biederman
> <[email protected]> wrote:
>
>> Are this bugs in the 9p implementation or are you tickling another
>> kernel bug?
>
> I am pretty sure it's a kernel bug. Actually, it looks like it went to
> do the exec and exploded on the 9p I/O to the server back on the node
> we were running xrx from.
>
> one other issue with xcpufs is that its mount setup requires writeable
> root. We need to set up a tmpfs mount I guess on / to fix that.
>
> Overall though the system as a whole is not functional and we're
> trying to evaluate our next steps.
>
> ron
