Based on what I know of chrony the combination of stratum (how many hops are we away from a reliable 'local' clock source) and the dispersion / epsilon will determine whether or not a server, reachable or not, is declared unreliable/unreachable for synchronization purposes.
If you're that curious as to why, trace the NTP exchanges when you turn on NTP between client and server.
FWIW, while there is a KB article on this XIQ-SE is technically not configured to be a NTP server. If 'dateconfig' is ran at any time customized settings of this nature will be lost. NTP should point to a more reliable clock source than the SE VM hosted on a hypervisor which is by its nature subject to clock drift.
