by Hi,
I'm having a bit of a problem getting my PPP interface up. Actually
the PPP link is up, but for some reason it interferes with NIS+.
Setup:
-------------------------------------------
Solaris 2.4 (patched with Sun recommended)
Using aspppd
NIS+ server is also the PPP client
ipdptp0 -> 204.228.154.1 Assigned by ISP
le0 -> 204.228.154.2
We have a registered Class C address (204.228.154). NIS+
works fine when ipdptp0 interface is not up. When I do a NIS+
call, I get (from snoop):
client -> server RPCBIND C GETADDR prog=100300 (NIS+) vers=3
server -> client RPCBIND R GETADDR Uaddr=204.228.154.2.128.3
204.228.154.2 is the address of the le0 interface of the NIS+ server.
Works fine.
When the the ipdptp0 interface is up. I get:
client -> server RPCBIND C GETADDR prog=100300 (NIS+) vers=3
server -> client RPCBIND R GETADDR Uaddr=204.228.154.1.128.3
^^^^^^^^^^^^^
a .1 instead
and the NIS+ client eventually times out since it can't get
a response. I can't seem to get routing up between the two
interfaces (but how to I route if both are on the same subnet?).
Before we were assigned or own subnet, the ipdptp0 interface was a
node on my ISP's subnet. (which made sense to me, but my ISP
insists that my_subnet.1 should be my PPP interface)
1. Does anyone know what might be causing rpcbind to bind to
the wrong interface on NIS+ calls?
2. How do I get both interfaces (le0 and ipdptp0) to talk to
each other? From the client machines, I'd like to be able
ping see the PPP interface. I thought I could just start up
in.routed, but that didn't do it. I even setup a static
route from the ipdptp0 interface to itself. Still nothing.
thanks for any advice,
A. Baker