reboot in update.sh kills virtualbox.sh provisionning script for fedora28

[olberger]

Maybe this is due to some change with the fedora28 provisionning, but I've noticed that the virtualbox.sh doesn't run completely, and gets killed by the VM rebooting.

I suspect this is caused by the reboot that is made at the end of update.sh (and is required whenever the Kernel changes).

Anyone experienced this too ?

[olberger]

I think the sleep after the reboot in update.sh is useless. What is needed is separating the scripts before the reboot at the end of update.sh, and the rest which will pause due to pause_before
I guess this solves the is olberger@b1aaacd

There maybe another syntax for the packer JSON config... I've not RTFM ;-)

[olberger]

See my Pull Request that should be fixing this: #39