Segfaults-a-gogo
The lightbulb turned out to be a dud and the server process fell over fairly rapidly. After doing some old-school debugging by prints and commenting sections of code out, it seems that the failure occurs when I save the updated map file. I was saving to the DB, so after seeing that it ran for about 5000 updates with the save turned off, I decided to do a simple save to file rather than spend forever working out why the DB save as BLOB wasn’t working. Good plan, except that the save to file caused a segfault too.
I’ve tried a few variations on the theme, but it does seem as if there’s something messing up with the pickling process (common to the BLOB and file saves) in combination with my code. I’m very curious to discover the problem, because if I can’t save the updated map information then I’m going to have a bit of a hard time with the rest of the project.
Outside of this, when the server is up, you will see the larger map with newer, more in-your-face rivers. I’ll put the other rendering modes in tomorrow so the underlying model is viewable by way of demonstrating the possibility that it might look a bit more lifelike at some point. This crash bug has to be a priority though.
