check_udp on port 9000 (clistener)

[nagios_aws]

Hi,

I've surched in all existing topics without result, even if some topics are similar but don't solve my problem.
I want to check a service with an UDP connection on port 9000. The problem is that whatever I put in the option -s and -e, I have the error " OUTPUT: CRITICAL-Socket timeout after-party 10 second ". The command is launch from the NAGIOS Server : COMMAND: /usr/local/nagios/libexec/check_udp -H **.***.**.** -p 9000 -s '***' -e '***'.
Even if I add time with the option -t **, it is the same result.

I test the UDP connection with nmap from NAGIOS server, I have the following result :
nmap -sU **.***.**.** -p 9000
Starting Nmap 6.47 ( http://nmap.org ) at 2017-11-17 08:15 CET
Nmap scan report for *************** (**.***.**.**)
Host is up (0.0010s latency).
PORT STATE SERVICE
9000/udp open|filtered cslistener
MAC Address: **:**:**:**:**:** (VMware)
Nmap done: 1 IP address (1 host up) scanned in 6.78 seconds

How can I solve this problem ? The command seems to be good

Thanks,
Stéphane.

[kyang]

When you added the option for -t did you increase the value?

Something like -t 60

Code: Select all

/usr/local/nagios/libexec/check_udp -H **.***.**.** -p 9000 -s '***' -e '*** -t 60

[nagios_aws]

Hi,

Yes -t 60, -t 90, the result is the same ...

[kyang]

PORT STATE SERVICE
9000/udp open|filtered cslistener

According to your nmap, is it actually opened or is it still behind a firewall?

Code: Select all

open|filtered

Nmap places ports in this state when it is unable to determine whether  
a port is open or filtered. This occurs for scan types in which open
ports give no response. The lack of response could also mean that a 
packet filter dropped the probe or any response it elicited. So Nmap 
does not know for sure whether the port is open or being filtered. The
UDP, IP protocol, FIN, NULL, and Xmas scans classify ports this way.

You could do a tcpdump to check whether packets are reaching your machine.

Code: Select all

sudo tcpdump -A 'udp and port 9000' 

[nagios_aws]

Hi,

First of all, thank you for having answered.
For "open|filtered", I knew that.

I cannot make of tcpdump on the machine, the server is a Windows server. Furthermore, I have no authorizations to add applications as windump on servers...

On the other hand I test a process on the machine and it works perfectly ($USER1$ / check_nt - a hour $HOSTADDRESS-p $USER7-v $ARG1 ARG2$)
Thus, the server Nagios can reach it.
Maybe it is simply due to the service listening on port 9000 who don't send back anything ? I'm trying to find out on that way too.

[kyang]

Ah, I see, thanks for letting us know.

Maybe it is simply due to the service listening on port 9000 who don't send back anything ? I'm trying to find out on that way too.

Possibly, you could also check to see if there's another plugin on the Nagios Exchange that does not require a string to be passed.

https://exchange.nagios.org/index.php?o ... chword=udp

Otherwise, let us know.

[nagios_aws]

Hi,

This script is not usefull for that particular case because the nmap command return :
PORT STATE SERVICE
9000/udp open|filtered cslistener
But "cslistener" is not the service listening on UDP port as expected ...

I'm still waiting for vendor information about the service used.

I will let you know about this.

[kyang]

Thanks!

We will be waiting to hear back then.

[nagios_aws]

Hi,
We have finally just received information on behalf of the vendor concerning the service to be monitored. Unfortunately, technical knowledge of the commercial are not sufficient enough to understand our needs and send us back towards the utility delivered with the software … The only solution would thus be that I develop a specific plugin for this software.
Thanks to all for your help.
We can close this topic then.

[kyang]

Sounds good! I'll be closing this thread!

We do offer custom development if you are interested, you can email [email protected]

If you have any more questions, feel free to create another thread.

Thanks for using the Nagios Support Forum!