HP Cluster Management Utility CMU Installation Guide with Serviceguard Version 4.0 (May 2009)

6
2 CMU installation with Serviceguard
In a ”classic” CMU cluster you have a single management server.
In case of failure of that server, though the CMU cluster continues to work for the customer application
point of view, you loose management functions such as backup, cloning, booting a compute node, ssh
through CMU GUI, etc.
If the CMU cluster uses a private network for management, you also loose connection to the site
network.
Installing and configuring CMU under control of serviceguard provides redundancy to avoid this CMU
service degradation.
Figure 1 describes a “classic” CMU cluster connected to two networks: the site network and a private
cluster network where compute nodes are connected. The CMU management server is known by its IP0
address on the Site network, and by its IP1 address on the cluster network.
Figure 1
Figure 2 show the corresponding configuration where two servers can run CMU software in
active/standby mode under control of serviceguard.
Mgt server1 and Mgt server 2 are connected together to form a “CMU management cluster”. The ip
addresses IP0 and IP1 are attached to the CMU management server actually running CMU software at a
given time. The CMU management cluster is known on the Site network by its address IP0, and on the
compute network by its address IP1. IP0 and IP1 are the only addressed CMU is aware of. In case of
failure of that server, IP0 and IP1 will migrate to the other.
Each of the two servers has also one IP address per network (IP2, IP3, IP4, IP5).
The two servers are connected to a shared storage which hosts the CMU directory.
From the CMU Compute nodes point of view, as well as from the user point of view, this configuration is
perceived as a “classic” CMU configuration with a single Management Server.