Under some very rare circumstances it's beneficial to reset frontman entirely. Up to version 1.6.1 Frontman always stores check results locally in memory until they can be delivered to the hub.
If the hub is slow on storing data your local queues can grow. The local queue is a FIFO (first in, first out). A large local queue prevents recent data to be submitted, because old data need to be sent first.
At some point is advised to delete all local unsent data to get a recent data to be processed quickly again.
A proper service shutdown would instruct frontman to sent all data to the hub before stopping the process.
With the upcoming version 1.8 of Frontman local queues will be cleaned up automatically.
Open the task manager, look for frontman.exe and kill it. After some seconds the process is re-created automatically by the service manage. If not, open the service manager, and restart "CloudRadar Frontman".
On Linux use a
kill -9 for example
sudo pkill -9 frontman
sudo systemctl restart frontman
Check, if frontman is running, by executing
ps aux|grep frontman.