Connection refused after client restart - random

[JimmyJamez]

Hello,

We are receiving random Connection Refused errors on clients after rebooting. This happens only on a few random clients out of hundreds, but is, nonetheless rather annoying. When this happens, nscp service is running - killing and restarting the service resolves the issue. Also, the ports 12489 and 5666 are blocked when this is happening (cannot open telnet session to either). Netstat does not show either of these ports in use. No indications to problem in Event viewer. Any ideas would be much appreciated!

Nagios ver. 3.4.1 (Yes, it's an older version, but everything besides this is working wonderfully)
Windows clients: Win7 Pro SP1

[slansing]

Hmm, so if you run the following after one of the troubled servers is rebooted it shows the ports blocked?:

Code: Select all

nmap -p 12489,5666 <addr.of.windows.server> 

It sounds like this is an issue with the windows servers specifically, not nsclient or nagios.

[JimmyJamez]

I don't think it's a problem with Windows, as the service (and ports) are fine, once the service is restarted. Plus it works 99% of the time, and bugs out only on a couple of random computers (never the same computer) during restarts.

Nmap to affected computer:
PORT STATE SERVICE
5666/tcp closed nrpe
12489/tcp closed unknown

After taskkilling and restarting nscp:
PORT STATE SERVICE
5666/tcp open nrpe
12489/tcp open unknown

[slansing]

Well those ports would only be blocked or routed through by networking equipment between nagios and the server, or the server itself. NSclient does not control port forwarding or state. It would be interesting to test an nmap on a server with no client installed on it.