I have succesfully configured a clustered DHCP running on NSS volumes under Linux OES2 like described in Novell Linux OES2 documentation and it seems to work fine except that sometimes the DHCP service stop to distribute adresses, but still appear up in iManager : clustered resource is up, binded ip adress with the clustered resources up too. Well... I called this strange thing a "Stop emitting" :-)

No error messages appears in /var/log/dhcpd.log nor in /var/log/dhcp-ldap-startup.log nor in /var/log/messages.

Log files only shows that at one time the dhcp service runs correctly by responding to a dhcp dialog (DHCPINFORM, DHCPACK, DHCPREQUEST etc...) and the moment no more messages.

To bring the service works properly again I have to unload/reload the corresponding resource or migrate him to another server.

Note that the DHCP information was previously imported from a Netware Nss clustered DHCP server (Netware 6.5 sp7 migration to OES2 sp1) : I try to use miggui software but it did not work properly so I have exported all records from the previous dhcp, re-format them to Linux Dhcp format and imported them through the java management console. Well it seemed to be fine.

Do anyone have an Idea ? Do I miss something ?

I notice that the Dhcp stop to answer every 5h...

I attached the corresponding file to this thread.

When the server first "stop emitting" : 03/31-07:16
When I first restart the resources : 03/31-09:11

When the server secondly "stop emitting" : 03/31-12:40
When I secondly restart : 03/31-13:51

When the server Thirdly "stop emitting" : 03/31-17:07
When I thirdly restart : 03/31-17:21