reboot in remote execution job on RHEL6 clients cause job to fail

I have an issue on RHEL6 hosts with all jobs that do a reboot in a post script.
Seems like the job task is waiting for all spawned processes to exit completely before it exits the SSH session.

If I do a reboot it means the job with fail and output:

 231:Broadcast message from someone@somewhere
 232:(/dev/pts/0) at 5:04 ...
 233:
 234:The system is going down for reboot NOW!
 235:Failed to refresh the connector
 236:Net::SSH::Disconnect connection closed by remote host
 237:Exit status: EXCEPTION

Same issue if a do a "/sbin/shutdown -r 1" or a "nohup /sbin/shutdown -r 1 &".
This problem does not occurr on RHEL7 hosts.

Any idea how to do a reboot without causing the job to fail?