Using HP Serviceguard for Linux with VMware

12
Installing sg_persist on Red Hat
On Red Hat, the sg_persist command is included in sg3_utils. It is available only on RedHat 4
Update 5 (and later) and RedHat 5.
To install the rpm, use the command #
rpm –Uvh sg3_utils-version.arch.rpm
where version must be 1.223-3.1 or higher, and arch indicates the platform, for example i386
Important: You must also install the sg_persist command from sg3_utils on any physical servers
participating as nodes in the cluster, and you must ensure that the shared storage system
supports persistent reservation.
Serviceguard on VM guests
Serviceguard for Linux release A.11.18.02 and above is certified on VMware virtual machines
running 32-bit and 64-bit versions of RedHat and SUSE Linux. (VMware ESX server is supported
only on x86 (32 bit) and x86-64 platforms.)
Serviceguard for Linux release A.11.18.02 includes a new package control script and a new
package module. These add the functionality to take advantage of the sg3_utils. Packages that will
run on VMware guests MUST use the new package control script or new module. For more
information see the October 2007 revision of the HP Serviceguard for Linux Version A.11.18
Release Notes, which you can find at http://docs.hp.com
-> High Availability -> Serviceguard for
Linux.
Cluster configuration options
A Serviceguard cluster that includes VM nodes can consist of:
• Virtual machines on the same host (cluster-in-a-box; not recommended: see below)
• Virtual machines on separate hosts
• VM and physical nodes
• All of the above
If a cluster is configured with multiple VMs running on the same host, together with VMs running on
other hosts or physical servers, you need to be aware of the possibility of data corruption if an
application fails over between VMs running on the same ESX host. If one guest node that is part of
the cluster hangs and a package fails over to another guest node on the same host, there is a very
small possibility that IO pending for the first guest does not complete before the package is started on
the second guest.
Unsafe configuration:, Because of the risk of data corruption, a cluster consisting entirely of
multiple VMs running on the same ESX host is not HA safe and should be avoided in a
production environment.
Safe configuration: Serviceguard takes care of data integrity when applications fail over
between VMs running on separate physical hosts. This means that you can safely configure a cluster
in which multiple VMs are running on one host, provided packages are configured in such a way
that the failover does not happen between two VMs running on the same ESX server.