When I quit openMSX while connected the the debugger, and when openMSX is breaked, openMSX gives me one of the following four results:
The issue is quite intermittent, I would say 50% of the time I get no error at all, and the actual crash only happens 1% of the time. The attached crash report is from the fourth case.
My interpretation is that there is some kind of race condition between the assertion happening, and the program terminating before the error has been fully processed. I am using a current development build on Mac OS X 10.9, the debugger is built with Qt4.
Steps to reproduce:
If I put sleep(10) at the end of the CliConnection destructor I get a reliable crash reproduction.