[Nagiosplug-help] check_snmp_process & check_snmp_proc

[Stephen Berg (Contractor)]

I'm seeing some peculiar behavior from both of these scripts when 
checking processes on Fedora 15.

When checking a F15 system that has multiple nfsd threads, the scripts 
return a critical condition even though I have checked and the proper 
number of nfsd are running.

[root at nagiossvr plugins]# ./check_snmp_process -H targethost -C public 
-2 -n nfsd
No process matching nfsd found : CRITICAL

If I add in the "[]" that show up in a "ps auxf" on the targethost it 
finds them with no problems:

 From the target server:
root     29654  0.0  0.0      0     0 ?        S    Jun13   0:00  \_ [nfsd]
root     29655  0.0  0.0      0     0 ?        S    Jun13   0:00  \_ [nfsd]
root     29656  0.0  0.0      0     0 ?        S    Jun13   0:00  \_ [nfsd]
root     29657  0.0  0.0      0     0 ?        S    Jun13   0:00  \_ [nfsd]

[root at nagiossvr plugins]# ./check_snmp_process -H targethost -C public 
-2 -n \[nfsd\]
118 process matching [nfsd] (> 0)

The same behavior happens when I try to use check_snmp_proc. 
/etc/snmp/snmpd.conf has:
proc    nfsd    256 64

I've tried adding the escaped brackets to snmpd.conf but it still fails:

[root at nagiossvr plugins]# ./check_snmp_proc -H targethost -C public
No nfsd process running

Nagios Core Version 3.2.3 running on Scientific Linux 5.5
Both target hosts are Fedora 15, fully up to date.

Anyone have an idea about what could cause this and how to fix?  The 
same checks work as expected to Fedora 13, 14, SL5.x and SL6.x.

-- 
Stephen Berg
Systems Administrator
NRL Code: 7320
Office: 228-688-5738
stephen.berg.ctr at nrlssc.navy.mil