[1.0.0] Crashed twice, no stack trace

[Berdes]

Hi,

My game crashed twice today. I think both time, I had a temporary blueprint (from Ctrl+V) in hand, although I'm not 100% sure about the first time. I didn't think much about the first time, assuming the devs could take a look at the automated crash report, so I didn't save the log file.

I did overclock my memory a bit yesterday (going from 3600MHz to 3733MHz with the same CL timings), but it had no issues passing a memory test which I left running for about 2.5 hours. That was long enough to do 1 pass and a half. I plan on doing another memory test for the full night, just in case and I will report the result here tomorrow. I should note that I didn't experience any crash or unexpected behavior with other software running on my computer (mainly Chrome, Steam and Spotify).

Here is the save file, in case it can help: https://drive.google.com/file/d/1AZet6Y ... sp=sharing. The only thing that might not be super common is the relatively heavy use of signals and circuitry to force the train pathfinding to do better stuff.

I'm only using 2 mods (Power Grid Comb and Dynamic Construction Queue), but I doubt they would be the cause of the crash.

I managed to find the two coredump files, but they seem to be truncated to 2GB (from ~8GB and ~5.7GB) and I cannot get a backtrace from any thread ("Cannot access memory at address ...), so I'm not sure if these files could be useful in any way.

Please let me know if there is anything I can to do help debug this.

[Loewchen]

Can you reproduce the crash?

[Berdes]

Since I wasn't doing anything that felt special when the crashes happened, I didn't try to reproduce them actively. That being said, I still continued what I was doing from the last autosave each time and the game didn't crash from that.

In each case, the game was running for at least 30 minutes before a crash happened.

[Rseding91]

Thanks for the report however without any reliable way to reproduce the issue, without any stack trace, and with it happening randomly there's nothing we can do.

All of that in conjunction with every past bug report that follows this same pattern says: you have bad/unstable hardware/software on your end and there's nothing we can do about it.

[Berdes]

Small update: the memory test had no issues going through 8 passes for a total of ~10 hours of runtime. So I think I'm safe on that side.

While trying to understand what's going on, I discovered a CPU overclock that I forgot about from a previous test. After disabling it, I had no further crash from Factorio. Based on this, I think the bug can be closed (if that's something you do).

It's amazing how the game is so stable that crashes are more likely caused by faulty hardware/software than by the game itself!