emv@math.lsa.umich.edu (Edward Vielmetti) (08/08/90)
Ultrix 4.0, Dec 5000s & Dec 3100s, with far too many NFS mount
points. I was hoping to squash out Yellow Pages entirely --
not run it at all -- but it looks like that's going to be hard.
Hesiod/Kerberos will take care of all of the services (aliases, auth,
group, hosts, networks, passwd) except for "netgroup". Quoth the
man page,
Restrictions
You must have yp as the entry for the netgroup database.
So the question: is there something fundamental about the structure of
the "getnetgrent()" call or data that makes it impossible to wedge that
information into Hesiod, or is it simply a matter of something else?
[a little looking later]
Oh, I see that Ultrix 4.0 /etc/mountd has an option to verify that the
mount requests come from a host in the server's domain or subdomain;
that might work just as well for the case when the only netgroup that
I want to define ("export") consists exactly of machines in the server's
domain. So '/etc/mountd -d' and a wide-open /etc/exports might just
do the trick....
More when I actually run some tests.
--Ed
Edward Vielmetti, U of Michigan math dept <emv@math.lsa.umich.edu>