[Pacemaker] HA DHCP3 server on Ubuntu

Simon Jansen simon.jansen1 at googlemail.com
Wed Oct 27 03:18:15 EDT 2010


Hi all,

I'm trying to configure an active/passive DHCP3 server on Ubuntu 10.04 LTS
Server test systems running pacemaker. I set up a DRBD partition to
replicate the data between the clusternodes. The DRBD partition works fine.
I moved the relevant DHCP data (dhcpd.leases and dhcpd.conf) to the DRBD
partition. So when a failover occurs the other node is able to continue the
service with the same data. I'm able to start the DHCP server manually by
running the following command: /usr/sbin/dhcpd3 -pf
/var/run/dhcp3-server/dhcpd.pid -cf
/mnt/DRBD/clusterData/drbd0/dhcp/cnf/dhcpd.conf -lf
/mnt/DRBD/clusterData/drbd0/dhcp/data/dhcpd.leases

Further I configured the following cluster resource using the antything RA:
primitive resDHCP ocf:heartbeat:anything \
params binfile="/usr/sbin/dhcpd3" cmdline_options="-pf
/var/run/dhcp3-server/dhcpd.pid -cf
/mnt/DRBD/clusterData/drbd0/dhcp/cnf/dhcpd.conf -lf
/mnt/DRBD/clusterData/drbd0/dhcp/data/dhcpd.leases"
pidfile="/var/run/dhcp3-server/dhcpd.pid" \
op monitor interval="10" timeout="50" depth="0"
But the service fails to start. The error description is 'unknown error'.
Maybe somone has a clue to fix this problem.

In addition I would like to know if it is recommended to use the anything RA
or should I change the paths in the init script and use the lsb RA?


Regards,

Simon
--
Simon Jansen
64293 Darmstadt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/pacemaker/attachments/20101027/2d67d25b/attachment.html>


More information about the Pacemaker mailing list