10.3. Configuring PC/NFSThe steps for installing and using a PC/NFS client will vary from vendor to vendor. You can expect that they will offer simple GUI-based installation that is compatible with Windows and NT norms, such as Installshield installation technology. The installer will walk you through most, if not all, of the necessary configuration. At install time or connect time, you should be asked to state how you will be authenticated, via NIS or PCNFSD, and you might be asked if you want to cache your username and password.10.3.1. Server-side PC/NFS configurationThere should not be any additional configuration for a PC/NFS client other than that needed for a Unix-based NFS client, unless the client requires the use of the PCNFSD protocol (either because you do not run NIS, or because you want to give your PCs access to Unix-connected printers). You may find that the PC/NFS client does not use reserved source ports (IP address port values less than 1024), and if so, you may have to disable "port monitoring" on the server as we'll discuss in Section 12.4.6, "Port monitoring". If you need to run a PCNFSD daemon on the server, you will want to add it to the rc scripts that get started when the Unix server boots up. For Solaris, you would add a script to /etc/init.d:and then link this script to a hard or soft link in /etc/rc3.d to start it before the NFS server.#!/bin/sh PCNFSD_NAME=hclnfsd # in /opt/pcnfs/bin PATH=/opt/pcnfsd/bin:$PATH export PATH case "$1" in start ) # The named directory is used as a temporary area for print spool files. $PCNFSD -A /var/run ;; stop ) pkill $PCNFSD ;; esac exit 0
Copyright © 2002 O'Reilly & Associates. All rights reserved. |
|