[Rseding91] [0.16.37] Inserters stop working (angels + bobs modded)

[themroc]

We play an Angels/Bobs game and after 200+ hours we experience an odd bug:

• An inserter is sitting above a full chest and is in idle, empty handed. The barreling pump is empty too. So the inserter should do a swing, but it actually does not.
• I assume it started after the last stack inserter capacity bonus research to level 20 (48 per swing). However, this could be unrelated.
• We have a huge base and only spotted this between chests and barreling pumps.
• The bug does persist after saving and reloading.
• The bug was found with version 0.16.37 but still occurs in 0.16.38
• Another bug was introduced with 0.16.38: When opening a chest, a custom barrel (like Ammonia gas barrel) looks like an empty barrel. However, that bug does not change the behavior of this bug (in case anyone checks the save). Edit: See viewtopic.php?f=7&t=59811
• In the save the inserter between the ammonia gas request chest and the barreling pump is in idle.
• Maybe it helps: This happened a lot with ammonia gas barrels.

Here are some relevant images: https://imgur.com/a/O1SH8qm
Savefile: http://factorio.gap.io/bugreport/inserter_bug.zip
Mods: http://factorio.gap.io/bugreport/mods.zip

[TruePikachu]

Did you pick up and put the inserter back down since switching to 0.16.38?

Both bugs you mention would be considered duplicates; the unrelated barrel fluid thing is viewtopic.php?f=7&t=59811, while the inserter bug is probably going to be the one mentioned in the changelog.

[themroc]

The barrel display bug is not the bug I report here. I only mentioned it for people to be aware of if they check the save file.

The "fixed" inserter bug issue was slightly different: viewtopic.php?f=7&t=59316 In that case the inserter always had an item in hand and woudn't drop it. In my case the inserter is sitting above a full chest not picking up anything. I found it in 0.16.36 and confirmed that it is still present in 0.16.37 with other inserter instances.

We were "fixing" the inserters all the time with replacing it, or simply turning it 4 times. But the bug would reappear...

Before posting this bug I talked to the Discord staff to verify that this is not a duplicate. Of course I can't know 100%.

[bobingabout]

Since 0.16.37, I've actually received quite a few complaints about inserters locking up.

I thought this version was supposed to fix this issue (viewtopic.php?t=59316) , not cause more issues.

I guess RSeding was having a bad day when he tried to fix that bug.

[Lubricus]

it's not viewtopic.php?t=59316
In 59316 the inserters wake up call stuff... is coupled with the wrong entity.
In this bug the waky uppy stuff is coupled with the unbarreler, but the line is yellow so the unbarreler doesn't make the wake up call to the inserter even if it should.

[Rseding91]

Since 0.16.37, I've actually received quite a few complaints about inserters locking up.

I thought this version was supposed to fix this issue (viewtopic.php?t=59316) , not cause more issues.

I guess RSeding was having a bad day when he tried to fix that bug.

The fix I did wouldn't ever cause inserters to break in more ways.

Does anyone have any way to reproduce what's happening?

[themroc]

Does anyone have any way to reproduce what's happening?

You can check my savefile (1st post...)
Edit: If the character remains where he was when I saved the game (dunno about that since it was a multiplay game), you're standing right next to it...

[Jap2.0]

Does anyone have any way to reproduce what's happening?

You can check my savefile (1st post...)
Edit: If the character remains where he was when I saved the game (dunno about that since it was a multiplay game), you're standing right next to it...

What he means is a way to reliably get it into that state.

[themroc]

Does anyone have any way to reproduce what's happening?

You can check my savefile (1st post...)
Edit: If the character remains where he was when I saved the game (dunno about that since it was a multiplay game), you're standing right next to it...

What he means is a way to reliably get it into that state.

Oops! No this happened in our gas area mostly with Ammonia Gas Barrel requester chests. And we were unable to find any pattern...

[TruePikachu]

Just throwing out ideas right now, are the inserters sitting across any chunk boundaries?

[themroc]

Does anyone have any way to reproduce what's happening?

I have found a way!

• Savegame: http://factorio.gap.io/bugreport/bugged ... ircuit.zip
• It happens 40 seconds in the future.
• It happens with filtered inserters and with normal ones. In this sample it's a filter inserter.
• It happens with circuit conditions attached to the inserter and without. In this sample it is connected to a circuit condition, which is enabled all the time.

Bug location:

I hope this helps

[Rseding91]

Does anyone have any way to reproduce what's happening?

I have found a way!

• Savegame: http://factorio.gap.io/bugreport/bugged ... ircuit.zip
• It happens 40 seconds in the future.
• It happens with filtered inserters and with normal ones. In this sample it's a filter inserter.
• It happens with circuit conditions attached to the inserter and without. In this sample it is connected to a circuit condition, which is enabled all the time.

Bug location:

I hope this helps

That's not a way to reproduce it That's just a broken save that it happens in. I need a blank map (50x50 size) and steps to get it into the broken state.

[themroc]

That's not a way to reproduce it That's just a broken save that it happens in. I need a blank map (50x50 size) and steps to get it into the broken state.

I tried reproducing it in sandbox mode with same science/circumstances but coudn't get it bugged...

[daydev]

Hello, I seem to have stumbled into the same or a related issue with insterters stalling, this time with direct insertion between assemblers and a different mod pack (Pyanodon's).

As you can see on the second and third image, the machine on the bottom has all the products evacuated, except coal which is to be taken to the machine on the top which is completely empty and ought to request coal to be inserted, but the outlined long handed inserters won't move. This doesn't happen every time, but about one in five pair of machines (in a row of 30 pairs) stalls in this manner from what I've seen so far. If I rebuild or rotate one of the red inserters it starts working and after a few swings the other starts working as well. Also if I wire them together to work in sync (read hand content, hold, enable iff everything = 0) they also start swinging. But after some time they stop again.

Unfortunately, I don't know a reliable way to get it to bug either. I notice that in both cases assemblers outputting fluids are involved.

I'm running the latest version, 16.41.

UPD: Interestingly, if change the state of the assemblers by either taking stuff out of the bottom one or putting stuff into the top one (either into the ingredient or product slot), the inserters start swinging too.

Images
First save
Second save (about 2 hours later, after I wired the insterters in an attempt to fix them, but they are broken again)
Mods

[themroc]

So will this bug not be investigated in its present state of information?

[Rseding91]

So will this bug not be investigated in its present state of information?

I still need a reliable way to reproduce the problem. Broken saves just show the result of the problem but not how it got into that state.

[theMightyMan]

hey Rsesing,

I am currently working in a reproducible map, but what I found out by testing on my playthrough-map is that the inserters are subscribed to the wrong wakeup-list.
it seems to be that after a longer period of time the inserters get subscribed to the outer side of the belt, while they are set to put items on the near side of the belt(this usually happens when I'm replying with information to this thread )
I've not found any particular reason for this happening, but I got a save where it happens 30 seconds before the autosave.