On Tue, Dec 2, 2008 at 11:50 PM, Andreas Ericsson wrote:
> James King wrote:
>> Hello all,
>>
>> I'm attempting to write a NEB module that will enable forwarding of
>> service state transitions to another NMS platform. I'm registering
>> for NEBCALLBACK_SERVICE_STATUS_DATA, but my callback is being invoked
>> twice for each status update, with identical timestamp, type code and
>> identical current_state and last_state in the (service *)
>> (data->object_ptr). There's only a single event type associated with
>> this hook (NEBTYPE_SERVICESTATUS_UPDATE). Is this normal behavior for
>> this hook (and if so, what is the reasoning behind this, or is there
>> at least a difference between the 2 invocations that I can use to
>> filter on)?
>
> Here's how Merlin does it. Note the first check there, making sure that
> if (ds->type != NEBTYPE_SERVICECHECK_PROCESSED), it returns immediately.
Thanks Andreas, I've changed to SERVICE_CHECK_DATA instead, and I can
properly filter my events now. I was just using SERVICE_STATUS_DATA
so that I would get the direct pointer to the service object, but
find_service() works just as well.
Nevertheless, the double invocation for SERVICE_STATUS_DATA still
seems a bit fishy to me.
Thanks,
James
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]