no performance graph with NRPE and CheckDriveSize

[pizzi]

I'm using nsclient++ (newest version) to check all my drives on windows servers. The command completes successfully but there is not performance graph for this check.
In the advanced service status I have the following performance data output:
Performance Data: 'C:\ %'=66%;80;90; 'C:\'=13.29G;15.99;17.99;0;19.99; 'E:\ %'=46%;80;90; 'E:\'=92.36G;159.99;179.99;0;199.99

The performance data looks a little bit strange with % sign and the space in between.

NRPE_Command:
$USER1$/check_nrpe -H $HOSTADDRESS$ -c CheckDriveSize -a ShowAll=long MaxWarnUsed=80% MaxCritUsed=90%

NagiosXI version: 2009R1.2

Any ideas?
Regards
piz

[tonyyarusso]

I think that performance data string is invalid. I believe it should have to return GB as the Unit Of Measurement rather than G, per http://nagiosplug.sourceforge.net/devel ... tml#AEN201 . If that is in fact the problem it would be a bug in NSClient++, although I'll try to confirm your behavior on a box here. The % and spaces are correct.

[tonyyarusso]

Confirming that I do not get performance graphs using that command. However, I DO get correct output for this command:

Code: Select all

[root@demo libexec]# ./check_nt -H 192.168.5.9 -s password -p 12489 -v USEDDISKSPACE -l C -w 80 -c 90
C:\ - total: 372.53 Gb - used: 190.30 Gb (51%) - free 182.23 Gb (49%) | 'C:\ Used Space'=190.30Gb;298.02;335.28;0.00;372.53

[bacilko1]

Hi all,

i have the same problem as pizzi.
After update to newest nsclient++ there are no performance graphs.

After searching on another forums, it seems that problem is somewhere between nsclient++ and pnp4nagios plugin.
There is changed performance data output format in latest nsclient++. Seems that output string should be correct but pnp4nagios version 0.4.x have a problem with trailing ";" semicolon.

In version 0.6.x this issue should be solved. Does anybody tried to upgrade it - with regards to nagios xi ?
Or do you know some workaround ?

Thanks a lot
Tomas

[mmestnik]

I don't follow nsclient++ development, so what versions are you running and were you running? Also why upgrade, was it a new feature or a security patch?

[aels]

See : http://nsclient.org/nscp/ticket/364

The problem seem to be the semicolon at the end.
We have clients using different versions of NSCient++ and this is a problem with the performance data not loaded.

Is there steps to upgrade to PNP4Nagios-0.6.6 from the version installed.

[mmestnik]

PNP4Nagios is shipped with XI, so upgrading it could cause a future upgrade to downgrade it. That said I'm sure it would be fine for you to try and upgrade it on your own.

We will consider including this in the next release.
http://go.nagios.com/tracker/82

[aels]

I started by upgrading to pnp 0.4.14 and so far all seem good.

[aels]

To apply a temporary fix I have modified the /usr/local/nagios/libexec/process_perfdata.pl script:

Changed this line in the "_parse" function

from this:
$string =~ s/^([^=]+)=([\d\.\-]+)([\w%]*);?([\d\.\-:~@]+)?;?([\d\.\-:~@]+)?;?([\d\.\-]+)?;?([\d\.\-]+)?\s*//;
to
$string =~ s/^([^=]+)=([\d\.\-]+)([\w%]*);?([\d\.\-:~@]+)?;?([\d\.\-:~@]+)?;?([\d\.\-]+)?;?([\d\.\-]+)?;?\s*//;

I just added "?;" at the end before the "?\s*//;"

This seem to fix the extra semicolon issue with the NSClient++ in the performance data.

I am busy testing this now but looking good so far.

[mmestnik]

For any one confused as I was the regex atom that was added is ";?".