I've been setting up e-mail alerts for our various applications and I've run into a situation I can't figure out.
I have an alert that is set up to watch an application, and alert if it enters a 'critical' state. I currently have the application in that stat on a server, so I can test it's real functionality.
The application monitor is set up performance counter monitor that monitors the application, and goes critical when the application exceeds 120MB. There is also a process monitor set up to watch the executable in the process list and go critical if the process goes over 90% CPU usage, and provide a status of 'down' if the application is not running.
These alerts are working fine within Solarwinds, but the alert e-mails are what I can't figure out.
I have the alert set to trigger when the node is 'not equal to' 'down', the application name is equal to the name of the application, and the component availability is equal to Critical.
In the Trigger actions, I have two actions.
The first is as follows:
NetperfMon Event Log:
Message to send to NPM Log:
NetPerMon Event Log: Component ${N=SwisEntity;M=ComponentAlert.ComponentName} on Application ${N=SwisEntity;M=Application.ApplicationAlert.ApplicationName} on Node ${N=SwisEntity;M=Application.Node.Caption} is ${N=SwisEntity;M=ComponentAlert.ComponentAvailability} with status: ${N=SwisEntity;M=ComponentAlert.StatusOrErrorDescription}
The second action is the send E-Mail/Page alert.
In the e-mail setup, I have:
Subject:
Alert: MonitoredApp on ${N=SwisEntity;M=Application.Node.Caption} is critical.
Message:
MonitoredApp on ${N=SwisEntity;M=Application.Node.Caption} is critical:
CPU Usage: ${N=SwisEntity;M=ComponentAlert.PercentCPU}%
Memory Usage: ${N=SwisEntity;M=ComponentAlert.StatisticData}MB
Date: ${N=Generic;M=DateTime;F=DateTime}
I want the 'CPU usage line to read the APPLICATION'S CPU usage, not the server's overall CPU usage, so that line may be wrong for what I'm looking to do.
What I get is two messages. Both have identical subjects, but the content is slightly different.
The first comes through as
MonitoredApp on Servername is critical:
CPU Usage: %
Memory Usage: 202.92578125MB
Date: Thursday, May 21, 2015 4:17 PM
The second comes through as:
MonitoredApp on Servername is critical:
CPU Usage: 0%
Memory Usage: MB
Date: Thursday, May 21, 2015 4:17 PM
Does anyone have any idea why the single alert is sending the two separate, incomplete messages, and how I night fix it so it sends one e-mail with all the information?
Thanks