As the summary says ...
The nice_failback button (under: Cluster -> heartbeat
monitor -> configuration options) is always toggled to
'on' state during window startup; i.e. it dows not
follow the defined state in the configuration file
This is very bad, since it toggles a cluster from an
to an Active-Standby configuration; i.e. only one
operates - the other sits and waits for the partner
node to crash.
I noticed it the first time in Webmin 1.111. The bug is
still in 1.121.
Thus a WebMin update did not fix it.
(ComServe IT-Service, Essen, Germany -