Hi,
The trap for connectivityFailure has a standard to use http://downloads.ironport.com which is not obvious.
Start with SNMP disabled, then use snmpconfig to enable the agent, and select the option to enable all traps. Because trap connectivityFailure is enabled, every 5 to 7 seconds (not configurable) the management interface (by default) will try to connect to downloads.ironport.com on port 80 (configurable). It will then try an HTTP GET request, and it will expect a response code between 200 and 300. If fail, then trap. That's the only test for this trap. The interval is not configurable, but the path is. Let's say alerts should be sent when the appliance cannot connect somewhere, but the firewall is blocking the management port from opening port 80 at downloads.ironport.com. Or the appliance is supposed to use the configured proxy and connectivityFailure trap should be sent if proxy is not reachable. The URL should be changed.
1. Assuming all traps are enabled, use snmpconfig to ENABLE trap #3 connectivityFailure in the list
AGAIN, or disable it and re-enable it, it makes no difference.
2. An extra question will pop up every time it is set to ENABLE i(except when ALL traps are enabled):
What URL would you like to check for connectivity failure?
[http://downloads.ironport.com]>
3. Enter a local URL, for example http://example.com:8080
ConnectivityFailure traps can be easily simulated by using dnsconfig and entering a non-working DNS server (lookups for downloads.ironport.com will fail, traps will be sent every 5-7 seconds.)
