Even if this is a bit of a dirty workaround, maybe it'd be justified to add it to (for example) RFBSession >> initialize, having it check if we are running headless and, if so, evaluating that line. I'd guess many users are going to run into this issue... how else would one access a headless image if not via RFB, until Coral can handle debugging and Monticello? ;)

Thanks a lot!

2011/10/18 Henrik Sperre Johansen <henrik.s.johansen@veloxit.no>

On 18.10.2011 08:50, Dennis Schetinin wrote:
Yes, apparently I experience the same thing. What is a good way to fix it?��

2011/10/17 Carlo <snoobabk@yahoo.ie>

Hi

It looks like you're running your vm headless is that right? As far as I understand this would cause the vm to shutdown as the NonInteractiveUIManager is running...

Cheers

Carlo

http://code.google.com/p/pharo/issues/detail?id=4829 has a suggestion for a (untested) workaround, some changes to the UIManager API is needed before it can be done in a nice way though.

Cheers,
Henry

--
Bernat Romagosa.