anyone running with large server parms on linux afs servers?

[Bill Zumach]

Russ Allbery wrote:

> Nathan Neulinger <nneul@umr.edu> writes:
>
> > Why the -nojumbo? Do you actually have routers that couldn't handle it,
> > or some other reason?
>
> We had major problems with it originally, but "originally" in this context
> may be somewhere around 1992 to 1994 (it was before my time).  Now it's
> there for "if it's not broken, don't fix it" reasons.

Actually it's still there if one needs to stop AFS 3.4 clients from
sending jumobgrams and one can't do it on a per client basis. It will
however stop all jumobgrams; even to AFS 3.5 and 3.6 clients.

The current version does slow start and congestion control as well as allowing
resending data in smaller packets that was initally set as jumograms.

The jumobgram problem affected many sites including Transarc. If a fragment
was dropped during periods of congestion, the entire jumobgram was resent,
causing further congestion. Fileserver could get to the point where they
_never_ got a complete jumbogram.

Bill Zumach