Concluded we better just start a new VM... and it worked.
Next time there is such scheduled event, I guess being proactive and creating a new VM beforehand is the better solution.