If the servers never really go down, are there any issues with the network as a whole, switch ports, cables, or any messages in the logs of the servers indicating such.  If the ITA computer can not get a response from the servers during the status poll, it will see it as down.

During each Status Polling cycle, IT Assistant (ITA) sends out ICMP (ping) packets to the managed servers. It it does not get response within the timeout period, it assumes that the system is unreachable and cosequently marks it power down. Also, ITA will then generate an Alert itaAlertSystemDown. In your case, I believe, you might be having an Alert Action configured that acts on these alerts and sends out an  e-mail for each system.

 

 

One possibility of why this might happen, even though the managed system never really went down, could be network latency. To overcome it, increase the timeout in ITA

ITA-> Discovery and Monitoring-> Ranges.

Edit the Include Range. In the Discovery Wizard, go the the ICMP settings wizard pane, and increase the timeout value. See if this solves the problem.