NSClient++ Help (#1) - High CPU on nsclient process and long latency from host check tool when checking whether process is running (#861) - Message List

I cant work out why when I run this command:

check_nt -H myhost.co.uk -p 12489 -s mypassword -v PROCSTATE -d SHOWALL -l sqlwriter.exe

causes the CPU on the remote server to sit at 50 % for 8 seconds (usually timing out past 10 seconds) then returning its result.

On my nagios box I use the time command to measure:

[root@monitor libexec]# time ./check_nt -H myhost.co.uk -p 12489 -s mypassword -v PROCSTATE -d SHOWALL -l sqlwriter.exe

sqlwriter.exe: Running

real 0m8.807s user 0m0.000s sys 0m0.000s

What is even stranger is that as I am running it on Amazon Cloud, I have imaged the server with the exact same Nsclient config file and server configuration file and it is returned almost instantly.

I have restarted and run nsclient in debug mode and got the following information of which I can see nothing interesting!?

2011-09-10 23:37:58: debug:modules\CheckSystem\CheckSystem.cpp:811: PROC<<<FEnumProcesses 2011-09-10 23:37:58: debug:modules\CheckSystem\CheckSystem.cpp:805: PROC::: Failed to open process: 4: 2011-09-10 23:37:58: debug:modules\CheckSystem\CheckSystem.cpp:808: PROC>>> find_crashed_pids 2011-09-10 23:37:58: debug:modules\CheckSystem\CheckSystem.cpp:811: PROC<<<find_crashed_pids 2011-09-10 23:37:58: debug:modules\CheckSystem\CheckSystem.cpp:811: PROC<<<enumerate_processes 2011-09-10 23:37:58: debug:NSClient++.cpp:1180: Injected Result: OK 'sqlwriter.exe: Running' 2011-09-10 23:37:58: debug:NSClient++.cpp:1181: Injected Performance Result:

Any help would be greatly appreciated.