Hey,
thanks for reporting.
The dump itself hasn't really helped, except that I'm now quite sure that somewhere, some actions write to invalid memory.
I know that you're deploying a lot of server snapshots. My guess is that it's somehow related to that, especially with users on the server, but I've not managed to reproduce that yet.
HAHA Well nice, I actually changed a few things in relation to the file transfer.
As I remember right I also fixed a small crash I've detected on the fly but not sure anymore.
I think if you're not experiencing any more crashes we could close this in a few days.