I am using RHEL-5 with nagios latest version. I want to monitor sendmail queue which is on host2 ,through nagios(host1).
Below is what I did and at last getting some error.Please check where I am missing.


[mf1@host2 ~]$ file `which mailq`
/usr/bin/mailq: symbolic link to /etc/alternatives/mta-mailq

[mf1@host2 ~]$ /usr/local/nagios/libexec/check_mailq -w2 -c4 -t7
ok: mailq (1)is below threshold (2/4)unsent=1;2;3;0

[mf1@host2 ~]$ grep mailq /etc/snmp/snmpd.conf
exec .1.3.6.1.4.1.2021.53 mailq-check /usr/local/nagios/libexec/check_mailq -w2 -c4 -t7

[mf1@host1 ~]$ snmpget -v 2c host2 -c community .1.3.6.1.4.1.2021.53
timeout: No response from host2

What does the SNMP daemon log say on host2?

You might also want to investigate on the ACL/VACM config. From the few information given it could possibly be that host1 is not allowed to query the oid on host2.

Normally ACLs/VACM is explained on top of the provided /etc/snmp/snmpd.conf. And by default RHEL disables most oid queries.

As you are not getting a reply from the other host, and not an unknown OID error, check your firewall on both sides to make sure that is not blocking (tracepath is a good tool for this as well). Often when I am not getting an error having to do with the OID or authentication, the problem is with either the firewall or services. Make sure your snmp daemon is running, and you might want to run the check local on host2 to see if it works locally first.

The other suggestions are also something to always keep in the back of your mind. It is true that a lot of distros ship a "crippled" version of SNMP as a security measure. Anything potentially harmful is disabled, and it is up to the administrator to understand how to enable the functionality and authentication. Due to the potential for disaster if somebody got RW access via SNMP I think that the added work is a great idea. Keeps you from accidentally throwing the switch unless you know what you are doing.

$.02