[OpenAFS] Re: Performance problems seem to be coming back

Ken Dreyer ktdreyer@ktdreyer.com
Mon, 12 Sep 2011 08:41:12 -0600


On Mon, Sep 12, 2011 at 5:46 AM, Dale Pontius <pontius@btv.ibm.com> wrote:
> In this case, I'm wanting something that acts more like normal AFS packet=
s,
> and I believe that's what rxdebug is doing in this case. =C2=A0Some peopl=
e around
> here have been using ping as an indicator of network latency, and while i=
t
> can be an indicator, I don't believe it's terribly representative of what=
 is
> happening to AFS packets. =C2=A0I believe the current rxdebug is more ind=
icative,
> though I can accept evidence otherwise.

A while back I looked into writing an iperf probe for Smokeping, which
would at least provide the metrics for UDP rather than ICMP, for the
reason you describe. I got pulled onto other things, but it would be
nice to have.

- Ken