Managing Serviceguard Nineteenth Edition, Reprinted June 2011

Table 13 Types of Changes to the Cluster Configuration (continued)
Required Cluster StateChange to the Cluster Configuration
See “What You Must Keep in Mind” (page 284). Cluster can be
running throughout.
Cluster can be running.Change NETWORK_FAILURE_DETECTION parameter
(see “Monitoring LAN Interfaces and Detecting Failure:
Link Level” (page 66))
Cluster can be running.Change NETWORK_AUTO_FAILBACK parameter
Cluster can be running.Change NETWORK_POLLING_INTERVAL
Cluster can be running. See the entries for these parameters
under “Cluster Configuration Parameters ” (page 105) for more
information.
Change IP Monitor parameters: SUBNET,
IP_MONITOR, POLLING TARGET
Cluster can be running, except in CVM environment; see the
NOTE below this table.
Change MEMBER_TIMEOUT and
AUTO_START_TIMEOUT
Cluster and package can be running.Change Access Control Policy
Cluster can be running.Change capacity and weight parameters.
A change that would cause a running package to fail will trigger
a warning, giving you a chance to cancel (unless you use
cmapplyconf -f).
NOTE: If you are using CVM or CFS, you cannot change MEMBER_TIMEOUT or
AUTO_START_TIMEOUT while the cluster is running. This is because they affect the aggregate
failover time, which is only reported to the CVM stack on cluster startup. You also cannot change
the quorum configuration while SG-CFS-pkg is running.
Previewing the Effect of Cluster Changes
Many variables affect package placement, including the availability of cluster nodes; the availability
of networks and other resources on those nodes; failover and failback policies; and package
weights, dependencies, and priorities, if you have configured them. You can preview the effect
on packages of certain actions or events before they actually occur.
For example, you might want to check to see if the packages are placed as you expect when the
cluster first comes up; or preview what happens to the packages running on a given node if the
node halts, or if the node is then restarted; or you might want to see the effect on other packages
if a currently disabled package has package switching enabled, or if a package halts and cannot
restart because none of the nodes on its node_list is available.
Serviceguard provides two ways to do this: you can use the preview mode of Serviceguard
commands, or you can use the cmeval (1m) command to simulate different cluster states.
Alternatively, you might want to model changes to the cluster as a whole; cmeval allows you to
do this; see “Using cmeval” (page 280).
What You Can Preview
You can preview any of the following, or all of them simultaneously:
• Cluster bring-up (cmruncl)
• Cluster node state changes (cmrunnode, cmhaltnode)
• Package state changes (cmrunpkg, cmhaltpkg)
• Package movement from one node to another
• Package switching changes (cmmodpkg -e)
Reconfiguring a Cluster 279