[frs-181] The Enterprise Status Monitor sometimes reports the monitoring instance as up or down.

Symptoms

The FusionReactor installer adds the current installation to the Enterprise Dashboard. All servers within the Enterprise Dashboard are automatically monitored for their status, and action can be taken if a server goes offline, and when it becomes available again.

  • If notification is enabled, an email is sent when a monitored server's status changes
  • A script may be optionally run.
  • A line similar to the following is written to the CrashProtection log.
2007-12-06 16:41:35.246 1196955695246 ServerUP NOTIFY "" "" "" "" "" "" coldfusion.cfmx7.int000b "" ""

This message is written not only for all remotely monitored servers, but also for the local (Dashboard) instance.

Analysis

The issue arises from the fact that the Enterprise Status Monitor was designed to be independent, and has no notion of what the local server is: it is simply treated as another instance to be monitored.

Problems can arise if you assign a script to be run when this system's state changes. The message may appear in the ServerDOWN form when the monitoring instance is shut down. If a script were assigned to this server in Manage Machines, it would be run.

Solution

This behaviour is by design. Administrators should not assign scripts in Manage Machines to the monitoring instance, unless the consequences of this are fully understood, i.e. the script takes into account the fact that the instance being monitored may be the master monitoring instance.

Issue Details

Type: Technote
Issue Number: FRS-181
Components: Crash Protection
Environment:
Resolution: Fixed
Last Updated: 21/Feb/08 11:29 AM
Affects Version: 2.0, 2.0.3, 2.0.4, 3.0
Fixed Version:
Server:
Platform:
Related Issues:

Comments are closed.