2.3.2 Examining the availability of monitoring managers
Two methods are available to increase the availability of monitoring managers: cluster configuration and multiple-monitoring configuration. Compare the features of the two configurations and employ the type that is appropriate for your operating environment.
For details about cluster configurations, see 2.3.3 Examining cluster configurations and the chapter that describes the cluster system configuration and operation in the JP1/Performance Management User's Guide.
In a multiple-monitoring configuration, two managers, the primary and the secondary, monitor the same agents. The secondary manager can take over and continue monitoring if the primary manager fails. This eliminates the downtime for monitoring, thus increasing the availability of the system. In addition, by placing the paired managers in different locations, you can restore information assets if a disaster hits either location.
For details about multiple-monitoring configurations, see the chapter that describes the multiple-monitoring system configuration and operation in the JP1/Performance Management User's Guide.
The following table describes the differences between a cluster configuration and a multiple-monitoring configuration.
|
Item |
Cluster configuration |
Multiple-monitoring configuration |
|
|---|---|---|---|
|
Configuration |
Number of servers that are functionally identical |
2 |
2 |
|
Cluster software |
Required |
Not required |
|
|
Shared disk |
Required |
Not required |
|
|
Version of Performance Management programs |
Not limited |
10-10 or later (Monitoring agents must be 10-00 or later.) |
|
|
Failover |
Enabled (active and standby) |
Not enabled (Use a command to switch nodes.) |
|
|
Failures |
Downtime |
Several minutes (Monitoring is disabled during node switching because services stop.) |
0 seconds |
|
Execution of actions if a failure occurs |
Possible (You can select the host on which actions are performed when you configure alarms. However, actions can be performed only on one host.) |
Possible (You can select the node on which actions are performed if you want to perform them on a manager host.) |
|
|
Application of Performance Management program settings |
For failover |
Not required (The settings are automatically passed when cluster software performs node switching.) |
Required (Use a command to manually pass the settings.) |
|
For failback |
Not required (The settings are automatically passed when cluster software performs node switching.) |
Required (Use a command to manually pass the settings.) |
|
|
Linkage with JP1 products |
JP1/AJS3 |
Possible |
Possible with limitations For details, see the section that explains how to set up linking with other systems in a multiple-monitoring environment in the JP1/Performance Management User's Guide. |
|
JP1/IM |
Possible |
||
|
JP1/SLM |
Possible |
||