Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Commmand works on the host I need to monitor (its IP is denoted as $HOSTADDRESS$ here):

[root@host ~]# /usr/local/nagios/libexec/check_ssh localhost
SSH OK - OpenSSH_4.3 (protocol 2.0)

However, on the Nagios server, it does not:

root@server:~# /usr/lib/nagios/plugins/check_ssh -H $HOSTADDRESS$
Server answer: 

I can see that the port is open:

root@server:~# /usr/lib/nagios/plugins/check_tcp -H $HOSTADDRESS$ -p 22
TCP OK - 0.123 second response time on port 22|time=0.123277s;;;0.000000;10.000000

And I can verify with telnet:

root@server:/etc/nagios3/conf.d# telnet $HOSTADDRESS$ 22
Trying $HOSTADDRESS$...
Connected to $HOSTADDRESS$.
Escape character is '^]'.

After searching the web for two hours and trying different suggestions I am out of ideas. I have several other hosts with similar setup, and they don't have this problem. I have compared the configs and they look the same (I think).

I am clearly missing something here.

share|improve this question

2 Answers 2

I had the same problem to check an OpenSSH server installed on Windows (KpyM Telnet/SSH Server) and I'm checking the port using check_ssh from my Nagios linux server.

With "check_ssh", I got the answer "Server answer" With "check_tcp" on the port 22 : SSH OK

In the logs of the SSH server on Windows, the ip of the nagios server was in the banned IPs.

I removed the IP in the banned IPs and check_ssh is now ok :

check_ssh answer is now : SSH OK - cryptlib (protocol 2.0)

Maybe it helps. Probably not the same software but maybe the same behaviour.

share|improve this answer

Have you setup passwordless SSH between the Nagios server and the server your trying to check?

This needs to be setup as it cannot be prompted with a password. You can check this by getting a connection to the nagios server and attempting to SSH onto the server you are trying to check as the nagios user and see if it prompts you for a password.

Also telnet is not ssh, its a totally different protocol!

share|improve this answer
    
Thanks for your reply. The SSH itself was setup, but connection attempts were returning "ssh_exchange_identification: Connection closed by remote host". It made me search more, and I found /etc/hosts.deny, and that the Nagios server IP was there. I removed it, but it appeared there again after some time. Investigating this. –  phs Jan 22 '13 at 8:51
    
The reason its re-appearing is that the Nagios server is repeatedly attempting to connect to the server your monitoring and failing would be my guess, have you checked the /var/log/audit.log on the server you are trying to monitor? cat /var/log/audit/audit.log | grep failed –  boburob Jan 22 '13 at 13:09
    
I think it is Denyhosts script. check_ssh isn't failing after I added Nagios server IP to /etc/hosts.allow. Still kept being added to /etc/hosts.deny though, so I added same IP to /var/lib/denyhosts/allowed-hosts as well. I think it stopped now. –  phs Jan 22 '13 at 14:38
1  
Here is the process explained: denyhosts.sourceforge.net/faq.html#3_19 –  phs Jan 22 '13 at 14:44
    
I believe his comment about being able to telnet to 22 was just to show that it wasn't a firewall problem, for the record –  Keith Jan 24 '13 at 19:11

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.