Test failure with latest code

[Ton Voon]

Hi!

I've noticed that there is a test failure in the Nagios Tinderbox  
builds.

This started on 28th August: http://tinderbox.nagios.org/cgi-bin/tinder.cgi?tree=nagios&start-time=1283080801&display-hours=100

The test case is test_events, which does various tests on the events  
code.

This is the test:

        /* This test should have both services moved forward due to  
not executing any service checks */
        /* Will make svc2 move forward by a retry_interval amount */
        execute_service_checks=0;
        sigshutdown=FALSE;
        currently_running_service_checks=0;
        max_parallel_service_checks=1;
        setup_events(now);
        svc2->current_state=STATE_CRITICAL;
        event_execution_loop();
        ok(svc1->next_check == now+300, "svc1 rescheduled ahead -  
normal interval" );
        ok(svc2->next_check == now+60,  "svc2 rescheduled ahead -  
retry interval" );

svc1's next_check is moved forward because it executes. svc2 should  
also get rescheduled, but doesn't - the next_check stays the same.

The change where run_event=TRUE was set in events.c is the culprit  
(from http://tracker.nagios.org/view.php?id=152), but I don't  
understand what it is trying to do and if the test is still valid.

In the tracker item, the other suggestion was to change the subsequent  
if() to an else if(). This makes the test pass, which makes me believe  
that resetting run_event to TRUE is causing different side effects.

I changed the max_parallel_service_checks to 2 and svc2 is still not  
getting rescheduled.

So can anyone confirm if either the current code:
  * causes breakage or
  * means that the test is now invalid and svc2->next_check should ==  
now

Ton


------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd