Troubleshooting NIS

ypbind crashes

If ypbind crashes almost immediately each time it is started, look for a problem in some other part of the system. Check for the presence of the rpcbind(1Mtcp) daemon by entering:

ps -eaf | grep rpcbind

If the message returned states that rpcbind is not running, restart the rpcbind daemon.

If rpcbind itself will not stay up or behaves strangely, look for more fundamental problems. Check the network software as suggested in ``Troubleshooting TCP/IP''.

You may be able to communicate with rpcbind on the problem client from a machine operating normally. From the functioning machine, enter:

rpcinfo client | grep ypbind

If rpcbind on the problem machine is running normally, rpcinfo(1tcp) produces an output similar to the following:

   100007    3     tcp  0.0.0.0.12.169  ypbind  superuser
   100007    3  ticlts  Q 00 00 00  ypbind  superuser
   100007    3  ticots   07 00 00 00  ypbind  superuser
   100007    3  ticotsord   07 00 00 00  ypbind  superuser

There should be one entry per transport; in the preceding example, the entry for udp is missing. If a transport is missing, ypbind has been unable to register its services with it. As long as there are other transports to run on, ypbind should run, but the omission may indicate some kind of a problem. Stop and start NIS using sh /etc/inid.d/nis stop and sh /etc/inid.d/nis start, and then run rpcinfo again. If ypbind is still not registered over all transports, reboot the machine.