 | |||||||||||
| |||||||||||
 |
|||||||||||
| |
| ||||||||||||||||
If your server is running an i686 architecture, you will have to use packages that is compatible with your clients so that they can share the files with the server. One most common package that is optimized is the glibc package. You will have to reinstall the glibc package that is compatible with your clients (i.e. the glibc-XXXX.i386.rpm). By default, RedHat Linux will install an optimized version of glibc on your server machine, please get the i386 version and use the command "rpm -ivh --upgrade --force glibc-XXXX.i386.rpm" to force overwrite the existing package. It is always available in your RedHat installation CD disk 1. After doing this, you will also have to remove the directory /lib/i686 to clean up incompatible libraries. You might have to do this for other packages that you want to share with clients as well.
When installing the ShaoLin Aptus, Aptus installation program will automatically change your ypbind init script /etc/rc.d/init.d/ypbind to use broadcast. Usually this happens because of an incorrect setup of /etc/hosts file . The server cannot find itself because the IP address of your Aptus server did not have a host entry in the /etc/hosts file. Correct it and you will fix your problem.
You should check your /etc/exports file to see if there an entry for /tftpboot/kernel. If you are running netgroups, please check the files /etc/netgroup and /etc/hosts with the correct host information entry. After correction, run make -C /var/yp (For RedHat Linux, other Linux may different). You may also want to run exportfs -r if you make changes to /etc/exports .
You should hit cancel here, since some distribution doesn't allow the root (superuser) to query the RPM database at Aptus clients, resulting it cannot find the RPM package. You can simply hit "Cancel". The system should work fine.
Some people will have experience with NFS problems, this is mainly due to invalid entries in the /etc/hosts file of in /etc/netgroup file. After you fix the host or netgroup entries (remember to update the NIS database "make -C /var/yp"), try to reload the NFS with the command "exportfs -r", if not success, you will have to stop the nfs completely by /etc/rc.d/init.d/nfs stop (for RedHat users),rmmod nfsd (remove the nfsd module from memory) and then start the nfs server again by "/etc/rc.d/init.d/nfs start" . If this still doesn't help, you will have to stop nfs again and unload the nfsd module, but this time clean all up the temporary files in the /var/lib/nfs directory, never do this while you have Aptus client online, otherwise they will hang and need to reboot to reconnect to the Aptus server.
Most of the daemons make use of tcpd, tcpd make use of the file /etc/hosts.allow and /etc/hosts.deny . This include NFS, sendmail, ypserv, xinetd and many other programs. You may have settings that does not allow you client connect to the server. See the manual page of hosts.allow(5) and hosts.deny(5) for more information.
Its likely you have firewall at the server. By default, some distributions have ipchains started (e.g. RedHat 7.2). Type "service ipchains stop" at the server to stop ipchains, type "chkconfig --level 2345 ipchains off" to turn it off by default. If you want to still use firewall, you have to let DHCP, and all RPC stuff to get through this includes portmap, NFS, and NIS. RPC has dynamic port numbers, it is recommend you unblock the firewall of the network adapter that is connected to the clients.
This is likely a problem of "clock skew", if your time rolls back before the date of the ntp configuration files, this will happen. On Redhat systems, the packaged ntpd may experience problem in reading the /etc/ntp/step-tickers file if there are no end of line character. Please check your ntp server's time and configuration file.
Some ntpd at the server requires running for a while (typically 15 minutes) after startup in order to be able to listen to any clients. You just have to wait until your server's ntpd will is ready for synchronization. Then your client will able to connect to it later. You may also point to another ntpd other than the Aptus server which is already functioning. You may do this during creation of the Aptus Group.
It is likely you didn't updated the NIS database after you added a user account at the server. By default, Aptus client uses NIS (see Linux NIS project for more information) for user authentication. The convinient way to solve this is to use the "User/Groups" under the "System" menu for user account management which has already configured to update the NIS database after changes. The other way is to update the NIS database manually after created new user accounts by using make -C /var/yp and then /etc/init.d/ypserv restart. If you are using other authentication systems such as LDAP or other, please check your settings.
