Exceeding external_command_buffer_slots?

[grimm26]

Is there a way to tell if I am exceeding external_command_buffer_slots? I have a setup where I am feeding a lot of passive check results into the command_file (one check puts in over 1000 when it runs). Some of the passive checks seem to not be processed, though, so I'm trying to figure out why . I do have command_check_interval set to -1.

[abrist]

You can use mrtg and nagiostats to monitor the external command buffer. Otherwise, just try increasing the buffer's value in the nagios.cfg and wait to see if you have any lost checks.

[grimm26]

OK, so it turned out I was well under my buffer slot ceiling. So I started poking around some more and doing some more Google searching. My service check is a perl script, and what finally helped me out was unbuffering the filehandle to the nagios command file (named pipe). Once I did that, everything started coming through fine. For some reason the buffer for that filehandle was not always getting cleared, I guess. I was closing it properly, so it is a bit of a mystery.

[slansing]

Hmm, well thanks for the fix you used for the problem, can we help you with anything else?