[OpenAFS] ADS communications issue?

[Derrick Brashear]

On Fri, Sep 11, 2009 at 1:46 PM, Eric Chris Garrison <ecgarris@iupui.edu> w=
rote:
> Okay, we "solved" the ADS/MTU problem for one supercomputer... what it
> turned up was that our AFS servers were only able to do 1500 MTU, but the
> router they were on accepted jumbo (9000 MTU) frames and passed them on,
> so they fragmented. =A0We had them set up our VLAN to not accept jumbo
> frames, and that supercomputer cluster was up and running fine.
>
> However, there's another supercomputer that had similar problems that wer=
e
> NOT solved by this. =A0In fact, the problem there turns out to be
> fragmentation during aklog when talking to the AD servers, not when
> talking to our AFS servers.
>
> The traceroute shows that the DF (do not fragment) flag is set, and a
> packet of 2441 was being sent, which is bigger than 1500. =A0It's
> fragmenting somewhere closer to the ADS servers, which are themselves set
> to 1500 MTU, according to their admins.
>
> So is the DF flag necessary? =A0If not, how can we change that?

AFS isn't setting DF in Kerberos traffic, so necessary or not, it's
not directly our fault, at least, assuming this is as I read it, aklog
failing.