[OpenAFS-devel] .35 sec rx delay bug?

[Tom Keiser]

On 11/2/06, Jim Rees <rees@umich.edu> wrote:
> I've got a set of file servers here that are giving extremely poor
> performance on fetchdata, less than 1% of what they should.  Tcpdump traces
> show frequent delays of 350 msec.  It looks like an rx bug.
>

I see this all the time in my test lab, especially when doing Rx
testing over gigabit links.  Unfortunately, I've never had time to
root cause the problem.  I used to think it was a race condition, but
if you've been able to repro it on lwp, then maybe it isn't...

Regards,

-Tom