BI Aggregation based on services ignores downtimes of their hosts in availability reports?

Hi all,
I am using CRE 2.0.0p24 on Debian/GNU Linux 11

Currently I don’t know if this is bug or feature: I tried to test the problem and setup testing aggregation from test service (TestSVC1) on two hosts (ser1, ser2). The service is simple MRPE

zito@ser1:~$ cat /etc/check_mk/mrpe.cfg 
TestSVC1        /usr/local/bin/testsvc1
zito@ser1:~$ cat /usr/local/bin/testsvc1
echo OK; exit 0
echo CRIT; exit 2

dtto on ser2…

So I can simply generate CRIT state, by commenting out echo OK; exit 0

The aggregation is simply the worst state of the state of service TestSVC1 on ser1 and TestSVC1 on ser2.

When I schedule downtime on services themselves. The availability report works as I expected. Downtime is computed, and OK state otherwise.

When I schedule downtime on hosts ser1, ser2 and not on services. The availability report counts OK and CRIT, and no DT. Shouldn’t be DT of host applied on its services too?

Is this bug or feature?