I think I have it figured out. There was something wrong with the networking on the box I was using for the master. I switched to a different one and everything started working perfectly. It appears my problem had nothing to do with ECJ.
Thanks for your timely response. And I'd like to thank you and everyone who helped develop this package. I have found it to be a valuable tool in my research.
Michael, I'm not sure what the root of the problem is that you're
having (perhaps Sean Paus may be able to give you more definitive an
answer), but I can say that the reference to islands is cruft that
shouldn't be there. Could you give us a single replicable setup that
causes the problem? The OS being used would also be helpful. I'll see
if I can nail it down (though not today or tomorrow probably).
On Aug 5, 2005, at 2:24 PM, Michael Solano wrote:
> I have been trying to use the master-slave evaluation functionality
> which was recently added to ECJ. I seem to have problems using the
> example problem (ant) in the star directory. Each time I set it up,
> the master stalls shortly after the first slave connects to it. The
> master outputs a message "Waiting for the name of the island...." and
> then halts. The slave outputs the message "The name of the island is
> slave-1" and then halts. I've tried this on both a network and a
> single computer (3 slave threads connecting to 127.0.0.1) and the
> master hangs each time. I've tried letting the system create a name
> and naming the slave myself but neither way seems to work. Does
> anyone know why this might be happening? Thank you for your help.