Il 09/05/2011 19:53, Jochen Bern ha scritto:
> On 05/09/2011 02:55 PM, Roberto Oppedisano wrote:
>> It appears that the problem happens on services member of both
>> Contact-Center, INS_Backbone servicegroups; i did not verify
>> all of them (only a visual inspect) but did not find a contrary case
>> 'till now.
> And those are *exactly* the two servicegroups with the longest "member"
> line in your objects.cache diff, which lends weight to the environment
> overload theory.
>
> Do you use environment variables in the first place? (I.e., a plugin,
> event handler, notification command etc. which reads a $NAGIOSsomething
> variable from its environment, rather than having it passed on the
> command line through a $something$ macro in the Nagios config.) If not,
AFAIK we don't use environment variables (and grepping for NAGIOS in the
config file led me only to commented out strings).
> try setting "enable_environment_macros=0" in your nagios.cfg (it's
> usually a performance boost, anyway).
Yep, this did the trick. After setting "enable_environment_macros=0" the
missing plugin notification did disappear.
Also if this workarounds the problem I'll be glad to do further
testing/analysis to catch it down (this is a pre-prod system so testing
is easy).
Kind regards
R
PS: I'll be @Nagios World Conference in Bolzano next Tuesday so if
someone wants to give a closer look...
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]