[Nagiosplug-help] How to interpret CRC32 errors?

[Ralph.Grothe at itdz-berlin.de]

Hi,

I'm currently writing a custom shell script plug-in that simply
checks the state of an HP Virtual Array.

I sourced utils.sh in it and return the proposed
STATE_{OK|CRITICAL|UNKNOWN} return codes
depending on the outcom.
I didn't care for WARNING as I only distinguish between OK and
CRITICAL.

I would submit the script hadn't I received contemptuous
reactions by some expert here on last occasion.

The script isn't doing much more than parsing the States of all
FRUs of the array and returning CRITICAL
if something other than state Good is found, OK elsewise.

The script doesn't require higher privileges than the nrpe daemon
has,
and it doesn't produce any output (apart from the final oneliner
indicating the state),
nor does it attempt to write anywhere.

If I run it at the shell under the uid of the nrpe daemon it
works fine.


$ /usr/local/nagios/libexec/check_hpva.sh

OK:     HP Virtual Array "va" on inwo2 all FRUs State Good

Whereas when I run it via check_nrpe I receive the to me
meaningless CRC32 error.

$ /usr/local/nagios/libexec/check_nrpe -H 127.0.0.1 -c
check_hpva_INWO2
CHECK_NRPE: Response packet had invalid CRC32.

In nrpe.cfg it's defined as

$ grep check_hpva /usr/local/nagios/etc/nrpe.cfg 
command[check_hpva_INWO2]=/usr/local/nagios/libexec/check_hpva.sh


What's going on here?

Regards
--
Ralph Grothe