Upon upgrading from portmap to rpcbind, autofs broke on me, which i
traced back to NFS (version 3) mounting breaking, with the somewhat
misleading

	mount.nfs: mount(2): No such file or directory
	mount.nfs: rpc.statd is not running but is required for remote locking.
	mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
	mount.nfs: an incorrect mount option was specified

In actuality, statd *was* running, but rpcbind wasn't permitting
mount.nfs to bind (or clnt_create() or whatever).  This is in spite
of it allowing me to rpcinfo -p (only as root, however)

So i figured maybe it was my tcp wrappers hosts.allow that needed tweak-
ing.  And i changed
	portmap
to
	rpcbind

But still no joy.  So then i went and changed the line from
	rpcbind: 10.10.10.0/24
to
	rpcbind: 10.10.10.0/24, 127.0.0.1
and mount worked again.  (I have no idea how rpcinfo worked, given
that, but i'll worry about filling in that knowledge deficiency later.
And i have no idea how portmap worked without localhost in there,
not to mention statd)

So i wish that, if the installer can figure out that rpcbind is re-
placing portmap, that it look for
	portmap
in /etc/hosts.{allow,deny} and either offer to modify the file(s) or
just spit out a warning, and it might also want to ensure that
127.0.0.1 is on the hosts.allow rpcbind line, subject to user assent,
unless it's only me that had that configuration brain damage

In any case, thanks, <anibal@debian.org>, for your Debian work