Subject: Re: qe0: ring overrun
To: Anders Magnusson <ragge@ludd.luth.se>
From: David Brownlee <david@mono.org>
List: port-vax
Date: 05/14/1996 18:23:50
On Mon, 13 May 1996, Anders Magnusson wrote:

> > 	I've just installed the March 13th snapshot and keep getting
> > 	qe0: ring overrun, resync'd by skipping 1
> > 	(then 2, 3,.. up to 14 then looping again)
> > 	they happen a couple of times a minute & occur even on an idle
> > 	network. The network card appears to work fine.
> > 
> > 	Has this been fixed in -current (Am I just wasting people's time
> > 	here :)
> > 
> This is a generic problem with the deqna. The receive buffer ring
> gets filled up so some packets get lost. I gets these messages
> frequently when having medium-loaded systems with NFS, but never
> when using a delqa card. It doesn't hurt to get them but you can try
> to increase the receive buffer pool in if_qe.c. I don't have any
> programming manuals for deqna so I don't know how much it can be
> det to.
> 
	I would have assumed something like this, but for the fact it
	happens when it is the only machine on the network (the network
	being a cable approx 1 metre long in this case :)

                   David/abs             (david@mono.org)

  System Manager, Southern Studios, Box 59, London N22 1AR. +44 181 888 8949 

  http://www.mono.org   ---<Monochrome>---   (** Largest UK Internet BBS **)  
  http://www.netbsd.org  __/ Free Un*x on alpha,amiga,arm32,atari,hp300,i386
  http://www.openbsd.org   \ mac,mvme68k,pc532,pica,pmax,sparc,sun3,vax,x68k