Resolution
The default configuration results in a 14.5 minute interval between rule updates. This can be changed easily from the MMC.
Rule Update Times | |||||
Poling Interval | Deliberate Pause | Cache Load Time | Heartbeat Interval | Agent Download or Activation Time | Total Time |
5 min | 2.5 min | 1 min | 5 min | 1 min | 14.5 min |
NOTE: In Operations Manager 3.3, the Agent Heartbeat Interval is 5 minutes by default. Thus, default configuration results in a 14.5 minute interval between rule updates.
The polling interval is how often the consolidator will attempt to pick up a change and load it into its cache.
The deliberate pause is a hard-coded two and a half minute delay. The consolidator waits for at least 2.5 minutes after a change is made before loading up changes in preparation for sending them out to agents. This helps prevent bandwidth consumption due to excessive rule updates.
The cache load time is a rough number representing how long it takes to load rules from the database into the consolidator's cache in memory for distribution to agents.
The heartbeat interval is how often the agent will check for new rules.
The download/activation is a rough number representing how long the agent may take to download the rules and begin running with them.
It is possible to lengthen or shorten the total time it takes for rule updates to occur by modifying one of the sub-intervals listed above. For example, you may want to speed up rule updates by decreasing the heartbeat interval and the consolidator rule change polling interval. These dialog pages are available in the Configuration node at the Global Settings folder. Select Agents | Properties for the heartbeat and Consolidators | Properties for the Rule Change Polling interval.
The 2.5 minute delay can be tweaked in the registry at:
HKLM \ Software \ Mission Critical Software \ OnePoint \ Configurations \ <your configuration> \ Operations \ Consolidator \ AgentConfig \ InactivitySeconds
NOTE: This formula may be useful when allowing Operations Manager to update the configuration change, rather than forcing an immediate change by right-clicking the Rules node. The best practice is to schedule your configuration changes. You should not use the Force Configuration Changes option for every change or high volume frequent changes, as this option consumes network bandwidth.
Settings to consider revising are as follows:
- Review your need for heartbeating and service availability reporting. These settings are configurable globally from the Configuration | Global Settings | Agent properties page or individually, per Agent, from Configuration | Agent Manager | Managed Computers tab. To configure an individual Agent, you highlight its machine name and click Settings.
- Consider increasing the buffering time slightly. Significantly increase the communication failure retry times, and simultaneously reduce the number of retry attempts. These settings are also configured from the Agent property pages, a.
s described above. - Make configuration and rule changes in batches. All such changes are pushed out dynamically, at the next heartbeat interval. If you only make changes once a day, for example, new configurations will go out only once a day to all Agents.