It is currently possible for /etc/init.d/netmount to hang due to a stale/dead NFS mount, preventing the system from rebooting. The solution to this problem is to use the "-l" lazy option when unmounting filesystems.

The entire script might need to be reworked, but a simple workaround to fix this issue is to change "umount -at" to "umount -lat" in the script. This allows a client system to reboot even if the remote NFS server is down. According to "man umount", the "-l" option appears to be designed for this purpose.