Kinda new here for forgive if I'm doing this incorrectly.
Scenario (current Process and Issue):
1. Add new Monitor for Windows Server

      a. New Monitor >> Add new monitor >> Servers_Windows

      b. Provide relevant info and finish with "Add Monitor(s)"

2. Select Monitor from Home and select "Configure Alarms" under Host Details.

      a. Click "Configure Health"

      b. Select "Configure actions at Attribute level"

      c. Select CRITICAL and WARNING "Associated Actions"

      d. Select "Show Advanced Options" >> Use default "Selected Dependent Attributes"

3. Click SAVE

Issue, When Monitored Server is RESTARTED and/or a Network Interface goes DOWN, a CRITICAL ALERT is generated automatically. Because we have our "CRITICAL Associated Action i Configure Alarms set to email and SMS, we are notified every time this occurs as an CRITICAL Issue. In our case a SERVER Re-start once a week is OK and is part of the regular site maintenance plan; however if this was done more 3 times in a single 7-day span, it would suggest that the site is possibly re-starting the server to remedy another issue.

      How can we set the AVAILABILITY (specifically "Server Re-started" to ERROR as a WARNING and not a CRITICAL (This appears to be a "default")?

            Secondly, is it possible that we can do this without impacting the AVAILABILITY attribute as a whole (obviously if the server was to go down for any other reason aside from Re-start - which AM seems to be able to differentiate based on the information provided in the ALERT - this would be a CRITICAL Issue)?

Second issue is that some sites are in locations where network connectivity is occasionally sporadic. In such situations we are finding that we occasionally get CRITICAL Alerts stating that an interface is DOWN. Often within about 30 minutes or so, the interface begins responding again. Unfortunately, it is not possible for us to demand that the site upgrade their network/host environment; however, if we could change these alerts to WARNING it would be helpful as we can simply utilize the information for reporting back to the site potential "outage times" and not ALERT all members of the CRITICAL Monitoring Group unless the interface is DOWN for extended amount of time (example 8+ hours).

Thanks for your assistance and detailed instruction.