Closed faissaloo closed 7 years ago
Cannot reproduce this on Oracle JRE 1.8.0.92 [oracle-jre-bin-1.8]. It looks like the makefile may have screwed up and tried to link something twice. Try running "make clean" in your enigma directory and restarting lgm afterwards.
Issue persists.
A little discouraged to see this bug has existed for 8 months and still nothing has been done about it.
More information is required. A core dump is the equivalent of "Oh no! Something is wrong!" We already know something is wrong; all software has bugs.
That said, are you also affected, and do you have more information? Or should I close this bug?
Based on your forum posts, it seems you are affected by a similar issue (or possibly the same issue Mikelle02 posted, though I doubt it). Please share a test case.
It sounds like the same bug. I can send you a gm81 file if you really want me to... But all it will contain is a single room, can't you do that yourself?
Edit:
Here's the 1 room GM81 file that crashes:
https://www.dropbox.com/s/19k9zx801mfkjhl/game.gm81?dl=0
Again, all it contains is 1 room. All I did to make this was create a new room resource in the resource tree.
If you want to see a more complex project, this one is open source on itch.io:
https://samuel-venable.itch.io/key-to-success
Be sure to download the ENIGMA - specific version of the source code.
But whether running my "Key to Success" game or a blank 1-room project, it really shouldn't make any difference, as it is the same bug and same crash that occurs, from the look of it.
If you would like me to retrieve more info by running the a blank, 1-room, project in debug mode, I'd love to help, I'll just need to know the steps to take to get more info in debug mode. (Goombert told me a while back how to do that kind of thing, but I've forgotten since then...)
Okay. My understanding is that you have been able to build other games in ENIGMA just fine (ignoring some behavioral quirks). Are you telling me that specifically creating a game with just one room causes this bug? As far as I can tell, the bug in your traceback happens when freeing old JDI contexts. I suspect this is yet another problem we would avoid by not including STL headers everywhere. The point is, you should be seeing it every time, including when updating your Definitions file—not just on weird conditions like compiling an empty room. Is that not the case?
No, I'm saying I can build on Windows just fine.
On Ubuntu is where the crash is happening. It doesn't matter what project I throw at it - it is guaranteed to crash. As I've said before, I'm using Ubuntu 16.04 LTS via wubi...
That's at least workable. I don't really have the wherewithal to deal with that, right now, but I can at least file a bug in case anyone else does, sooner.
Thank you so much for the continued help and support! :)
Closing as a duplicate of #1052 because we are dealing with the issue there. If anybody has any new problems related to this, please post there with feedback and details. If you discover any new and unrelated bugs, please post separate tickets for those. Thanks!
Happens when the user tries to compile a simple single empty room. User has tried Oracle Java and OpenJDK and updating. Neither fix it.
Here's the full thread: http://enigma-dev.org/forums/index.php?topic=2710