SNMP Monitoring just stopped on two systems

[stevecalderoni]

So I am confused by this one.

I have numerous systems using SNMP to monitor. Two of them at the same time just stopped working. In Nagios the error is "ERROR: Description/Type table : No response from remote host" and "ERROR: Alarm signal (Nagios time-out)". Ping, URL checking both are fine and a TCP port we use for the app is fine. Just the service specific stuff that uses SNMP. I have removed and re-added SNMP and even (just to test) set the SNMP service to allow any host to interact with it. Still nothing, I deleted the services and the server and tried to add again. The message received is that Nagios doesn't have permission to SNMP walk. At that time the SNMP service was set to allow ALL so that message is confusing as well.

Any guidance would be awesome

[tgriep]

Can you provide more details on what the 2 systems are and how SNMP is setup on those systems?
From what you describe, it sounds like either SNMP was disabled on those devices or that the community string was changed and the SNMP check commands cannot poll them anymore.

[stevecalderoni]

For the benefit of others that may run across this with the same issue.

The root cause was that when the SNMP role is installed to a Windows server the TCP firewall rule created has it active for all networks where as the UDP rule is only for Private networks. For what ever reason these two servers decided that they were on Guest networks which cause the packets to drop at the firewall. Setting the UDP rule to all networks (these are private network isolated servers in AWS) got it working again. Pretty easy to miss but there it is.

[cdienger]

Sounds like the problem has been resolved. Can the thread be closed or did you have any more questions or issues with SNMP?

[stevecalderoni]

Oh yeah, for sure. I did figure it out. Just commenting the solutions for others to benefit