Subject: Re: Amd 6.0a16 broken? (was Re: (long) NFS misbehaving under -current?)
To: Brian C. Grayson <bgrayson@marvin.ece.utexas.edu>
From: Michael Graff <explorer@flame.org>
List: current-users
Date: 10/09/1998 13:38:44
"Brian C. Grayson" <bgrayson@marvin.ece.utexas.edu> writes:

> On Thu, Oct 08, 1998 at 01:02:09PM +0200, Urban Boquist wrote:
> > 
> > This may be related to PR bin/6037. Reverting to the previous version
> > of amd makes it work again for us (the amd upgrade occured some time
> > early august iirc).
> > 
> > If someone feels responsible for amd, _please_ have a look at PR 6037
> > and diff the -current amd against the previous version. Thanks.
> 
>   Thanks Urban!  Reverting to the previous in-tree version of
> amd (6.0a13 instead of 6.0a16) also solved the first problem
> for me (the readdir returning ERR stuff).  So it appears I was
> experiencing two distinct bugs (readdir problems with amd, and
> the still-unresolved long-timeout NFS issue).  Is there an amd
> guru in the house?  :)

No, but I found this after digging through the code:

/defaults       opts:=rw,resvport,proto=udp;

I don't know if the resvport thing works, but the proto=udp fixed my
problems.

--Michael