Help with analysing factorio hanging up without crashing
Posted: Sun Jun 14, 2026 4:41 pm
Since I upgraded to 2.0.73, Factorio will sometimes hang without crashing.
The operating system and other apps stay fully responsive and does not feel overloaded, opening htop shows that most of the CPU cores are idling, more than enough GBs of RAM are free.
Factorio itself is irresponsive, showing a frozen screen with no feedback to any input.
The issue solves itself after a few minutes.
What could/should I investigate next times this happens?
Has anyone else experience similar issues?
Now I had this 5 times, so I want to investigate further:
- first two times, I killed factorio after waiting a minute, thinking it is stuck in a deadlocked state.
- third time, i went to drink a coffee, coming back a few minutes later to notice that factorio was working again.
- the last two times, after waiting maybe 3-4 minutes the issue solved itself again. Factorio unfreezes & continues to work without any issues. All the system info I inspected indicated that the system is not overloaded, everything else works flawlessly during this time, so I would expect the issue around factorio itself.
- Ubuntu 22.04, GNOME, x11 (non-Wayland)
- non-Steam version of Factorio
- with Space Age but no further mods
I looked into the factorio-current.log, but there was no related information logged, just the regular autosave stuff.
This is likely not caused by autosave, since it always happened between them (eg. 4:10 to next autosave, scheduled every 5 min, blocking). Also happened in an editor world while having autosave disabled.
(I inspected but forgot to secure the logfile after the game restored itself; currently only have the logfile where I killed the game after a minute wait.)
I don't think this would be caused by hardware related issues, since I don't have any crashes or unexpected behaviour in any other applications. This hang-up situations happen very rarely but repeatedly with consistently long time gaps (weeks) between occurrences. The temporal correlation with the version upgrade indicates to me that this might be a software bug.
Edit: added info about temporal correlation.
Edit2: Factorio is not completely irresponsive. After sending SIGTERM, the logfile gets updated immediately to include the shutdown reason. It's only the GUI (and probably more; I'm not sure if the game simulation progresses or not) being irresponsive.
The operating system and other apps stay fully responsive and does not feel overloaded, opening htop shows that most of the CPU cores are idling, more than enough GBs of RAM are free.
Factorio itself is irresponsive, showing a frozen screen with no feedback to any input.
The issue solves itself after a few minutes.
What could/should I investigate next times this happens?
Has anyone else experience similar issues?
More details
The issue has started about the same time when I switched from 2.0.62 (that I have played for a long time) to 2.0.73.Now I had this 5 times, so I want to investigate further:
- first two times, I killed factorio after waiting a minute, thinking it is stuck in a deadlocked state.
- third time, i went to drink a coffee, coming back a few minutes later to notice that factorio was working again.
- the last two times, after waiting maybe 3-4 minutes the issue solved itself again. Factorio unfreezes & continues to work without any issues. All the system info I inspected indicated that the system is not overloaded, everything else works flawlessly during this time, so I would expect the issue around factorio itself.
- Ubuntu 22.04, GNOME, x11 (non-Wayland)
- non-Steam version of Factorio
- with Space Age but no further mods
I looked into the factorio-current.log, but there was no related information logged, just the regular autosave stuff.
This is likely not caused by autosave, since it always happened between them (eg. 4:10 to next autosave, scheduled every 5 min, blocking). Also happened in an editor world while having autosave disabled.
(I inspected but forgot to secure the logfile after the game restored itself; currently only have the logfile where I killed the game after a minute wait.)
I don't think this would be caused by hardware related issues, since I don't have any crashes or unexpected behaviour in any other applications. This hang-up situations happen very rarely but repeatedly with consistently long time gaps (weeks) between occurrences. The temporal correlation with the version upgrade indicates to me that this might be a software bug.
Edit: added info about temporal correlation.
Edit2: Factorio is not completely irresponsive. After sending SIGTERM, the logfile gets updated immediately to include the shutdown reason. It's only the GUI (and probably more; I'm not sure if the game simulation progresses or not) being irresponsive.