Uh...sorry guys, but I was on crack. It turns out, after some
tracing through the code and debugging that it was doing the right
thing and using IPv4. The problem was that nfsd wasn't registered
with rpcbind, and our mount_nfs insists that it be registered, even
though it never connects to it.

And of course the damn Linux nfsd program has no -r option....

cjs
-- 
Curt Sampson  <cjs@cynic.net>   +81 90 7737 2974   http://www.netbsd.org
    Don't you know, in this new Dark Age, we're all light.  --XTC

On Tue, 28 May 2002, Curt Sampson wrote:

> So it turns out that PR 12802 is a little more serious than just
> a CFS problem; I cannot do NFS mounts from at least one Red Hat
> 7.something box with a 2.2.9 kernel for what look like the same
> reasons. (It's asking for IPv6 mountd when it's not available.)
> It seems to me (though I might be wrong) that this could happen
> with a lot of non-IPv6 servers, not just Linux. If so, it's rather
> a shame that this bug is going out in the 1.6 release.
>
> At any rate, can I suggest that, if we can't agree on the command
> line arguments yet, we just hack in something temporary so that at
> least mount_nfs works, and then continue the discussion of the best
> way to do the command line arguments?
>
> cjs
>