Introduction to HA and HP Unix Service guard cluster

What is HP UX cluster?

A Serviceguard cluster is a networked grouping of HP 9000 or HP Integrity servers (or both), known as nodes, having sufficient redundancy of software and hardware that a single point of failure will not significantly disrupt service. A package groups application services (individual HP-UX processes) together.  Serviceguard has existed since 1990, which HP claims to have been the first high availability solution for UNIX. HP Serviceguard, formerly known as MC/ServiceGuard, is a high-availability cluster software produced by HP that runs on HP-UX and Linux.

High availability (HA) is a component of a technology system that eliminates single points of failure to ensure continuous operations or uptime for an extended period. High availability clusters are groups of servers that support business-critical applications that require minimal downtime and continuous availability.

Here we are going to tell how to check the status of two node cluster and its package-

Suppose we have two node cluster and 3 packages on each node. And our server rebooted accidently, so here few commands are listed for how can check and run cluster services -

To check cluster status -

#cmviewcl

To run cluster service -

#cmruncl

If above command is showing errors, so you need to check cluster status again -

#cmviewcl

If any nodes or packages still in down state then we need to start them manually -

#cmrunnode nodename

If all packages are in halt state, so we need to login individual nodes to run their packages

#cmrunpkg pkgname

To make enable package up automatically after rebooting -

#cmmodepkg pkgname

In case, we need to set our one node on maintenance and we need to run all service with single node.

Then we need to run all packages in one node and need to halt another node. After that we can put another node in maintenance -

To halt  and shift packages to one node -

#cmhaltpkg pkgname

#cmrunpkg pkgname nodename

To halt node

cmhaltnode nodename

After that we can check all packages will be moved and in up state in one node, while another node will be in haltstate-

#cmviewcl

Congratulations, Now you have run the cluster in single node. 

you can also check our page for 

Import/Export volume group map in HP UNIX cluster

https://lalganjrbl.blogspot.com/search/label/Import%2FExport%20volume%20group%20map%20in%20HP%20UNIX%20cluster