This is with a distributed Icinga 1 environment.
I have about 100 hosts on an Icinga 1 client/satellite that are stuck with UNREACHABLE status. All four checks for each host are returning OK state but the overall state of the device is UNREACHABLE.
The problem may have been caused by me leaving Icinga 1 running with the wrong permissions for /usr/lib64/nagios/plugins/check_icmp. (check_icmp did not have suid bit set.)
So I stopped Icinga and emptied the state retention file (state_retention_file=/var/spool/icinga/retention.dat) on the satellite and that didn't help. If I empty that same file on the master might it help?
ps shows my submit_check_result.sh submit_host_check.sh scripts running as zombies but they don't live very long.