Crash on long large simulations

[alonfnt]

Hi,
I keep encountering an issue when I run large simulations (target being 200k cells), where suddenly the residual will explode (see log below) and after a long pause the program will fail.
I understand that the issue may arise from overlapping cells (maybe?), but I was wondering if there'd be an obvious solution to it. I've tried lowering the time step size but that doesn't seem to avoid the crashes.

Output:
1270 96368 cells 2232381 cts 674576 iterations residual = 40957.245665

Other than these cases, the program is great at running simulations! and I typically obverse good residuals, e.g.
1720 140704 cells 2624289 cts 120 iterations residual = 0.004888

System info:
I'm running the simulations on a Nvidia RTX A5000 and only using the Bacterium model (without integrator nor signalling model).