[frs-86] I keep getting alerts from the main FusionReactor dashboard about remote servers, informing that the remote servers are unavailable, followed by another alert (within a minute) stating that it is available again.


These messages are triggered when FusionReactor fails to get a response from ColdFusion within a timely fashion, likely caused by the ColdFusion server being temporarily under heavy load, or the machine not being available due to network saturation or DNS problems.

You can reduce the number of false positives that you receive by changing the amount of time FusionReactor waits before deciding that the ColdFusion server is unavailable.
To change the FusionReactor poll timeout, change the configuration key within your reactor.conf file. This file can be found in the following location:

[FusionReactor Root dir]/instance/[instance name]/conf

Within that file you will need to add the following line:

remoting.server.poll_timeout=15

NOTE: It is not required to restart CF to pickup the setting from the .conf file. You can get FusionReactor to pick up the setting by running the following example http://127.0.0.1:8088/fusionreactor/freset.cfm.

By doing this, FusionReactor will reload the config on the fly.

Issue Details

Type: Technote
Issue Number: FRS-86
Components: Enterprise Dashboard
Environment:
Resolution: Fixed
Last Updated: 13/Jul/07 2:32 PM
Affects Version:
Fixed Version:
Server:
Platform:
Related Issues: