Nagios ignoring timeperiod for one server

We have a virtual server we are trying to monitor whose host reboots every night. The host’s checks are fine, but the virtual server flat ignores the timeperiod I set for it.
This is my timeperiod configuration:

define timeperiod{
     timeperiod_name  foo
     alias                     Foo
     sunday                  03:30-24:00
     monday                 03:30-24:00
     tuesday                 03:30-24:00
     wednesday            03:30-24:00
     thursday               03:30-24:00
     friday                   03:30-24:00
     saturday               03:30-24:00
     }

And my host configuration:

define host{
     use                       windows-server
     host_name            foo
     alias                     Foo
     address                 XXX.XXX.XXX.XXX
     check_period         foo
     notification_period foo
     hostgroups             windows-servers
     parents                  bar
     }

This exact setup works for every other server in our network. However, this morning we got a critical notification that foo went down at 2:50 am. Why is it ignoring its timeperiod?

This is still not working properly, and we get several false reports that this server is going down. Does anyone know what may be going on?