Portmapper problems

[Jack N. Fenner]

In article <1991Feb27.145926.14813 at ccu1.aukuni.ac.nz> gmoff at ccu1.aukuni.ac.nz (Moffat) writes:
>Could someone give me a clue as to what's wrong here - when NFS is started
>on our 730 I get the following at regular intervals:
>  Cannot register service: RPC: Timed out
>  rpc.statd: unable to register (SM_PROG, SM_VERS, udp)
>  Cannot register service: RPC: Timed out
>  unable to register (NLM_PROG, NLM_VERS, tcp)
>The command 'rpcinfo -p' produces:
>  Cant contact portmapper: RPC: Remote system error - No route to host
>which is the same as when portmap is not even running!  

[rest of note deleted]

I had these exact symptoms on my machine.  The problem was that the 
"local" RPC host lo0 was out of sync with my machine.  That is, an attempt
to send an RPC message to the portmap on the local machine fails.  To
solve the problem:

1) Make sure that your /etc/hosts file contains an entry for the local 
   host.  A typical entry looks like "127.0.0.1		localhost loopback".

2) Make sure that the ifconfig line in /etc/rc.net for lo0 is
   uncommented and uses the correct name.  A typical entry looks like
   "/etc/ifconfig lo0 inet localhost up >> $LOGFILE 2>&1".

3) Reboot the machine.

Note that in the above, "localhost" must NOT be your regular host name, such
as is returned by the hostname command.  You should be able to ping localhost
if lo0 is working.  By the way, if you've been getting the RPC messages for
a long time, you probably want to clean them out of your error log.  Enter
"smit error" to clean it.

Regards,
-- 

Jack Fenner           IBM Federal Sector Division Advanced Programs
		      Temporarily assigned to Supercomputing Research Center
		      All opinions expressed are my own, not my company's