Subject: Re: poor nfs error recovery
To: Robert Dobbs <email@example.com>
From: Chris G Demetriou <Chris_G_Demetriou@LAGAVULIN.PDL.CS.CMU.EDU>
Date: 12/14/1994 04:06:41
> I have an i386 machine with an old 8 bit 3c503 card; this card is lame
> enough that it produces a handful of "ring buffer overruns" under heavy
> network load. This machine both serves and clients NFS. Due (I presume)
> to the ethernet card, access (read/write) of file larger than ~6K hangs.
> (5.8K works, 6.2K fails).
my guess on this is that your enet board has 8k of packet ram, which
is divided into 6k for receiving and 2k for sending. that's pretty
standard. if this is the case, and you get > 6k-worth of back-to-back
packets, then it'll have to drop some...
are you using reduced read- and write-sizes?