[1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Things that has been reported already before.
Post Reply
Kalixt
Inserter
Inserter
Posts: 30
Joined: Tue May 12, 2020 1:59 pm
Contact:

[1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Kalixt »

Hello,

I was playing on 1.1.5 from the day it was released and only now I am getting desync every time after I send train to one of the defense outposts. I can upload save also if you need it.
Attachments
factorio-current.log
(22.46 KiB) Downloaded 94 times

Loewchen
Global Moderator
Global Moderator
Posts: 8308
Joined: Wed Jan 07, 2015 5:53 pm
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Loewchen »

Post the desync report, see my signature.


Kalixt
Inserter
Inserter
Posts: 30
Joined: Tue May 12, 2020 1:59 pm
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Kalixt »

I am able to reproduce it in multiplayer, however if I use /toggle-heavy-mode command in single player, game became unplayable. I cant move, delay is huge and it use up all hardware resources.

Twinsen
Factorio Staff
Factorio Staff
Posts: 1330
Joined: Tue Sep 23, 2014 7:10 am
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Twinsen »

What are the steps to reproduce it in multiplayer?

Kalixt
Inserter
Inserter
Posts: 30
Joined: Tue May 12, 2020 1:59 pm
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Kalixt »

It is somehow related to train stop limit, I built new defence outpost which was destroyed and rebuilt again but then I noticed that trains which supply defence outposts with ammo ignores this new one. I checked if there is clear path to it, if there isn't missing rail somewhere. Everything was right so I tried to force train to go there and then desync happened.

Steps to reproduce:

1. This train showing "Destination full" even if there is one empty spot. Send him back to train stacker.
Image Image


2. New train who take his place can be sent to "Defense Outpost". After this step desync occurs.
Image

You must act fast as this happens only if there is train with status "Destination full" when there is one empty slot. This situation is present at game load and it will fix later somehow. Last built defense outpost is the one with empty spot. Enable condition on train stop is true, but train ignores it anyway.

Outpost with the empty slot for train.
Image

Save:
https://drive.google.com/file/d/1dCPdMy ... sp=sharing

I was able to reproduce it but I can't now as I fixed it in singleplayer and load it to multiplayer server. I still have save above where there is exact condictions for desync but the time I connect to the server it is already fixed and trains found path to outpost. I was trying to decrease game speed in order to join in right moment but it didn't help. Best would be if I could pause the game after load.
Last edited by Kalixt on Thu Dec 10, 2020 2:38 pm, edited 1 time in total.

Kalixt
Inserter
Inserter
Posts: 30
Joined: Tue May 12, 2020 1:59 pm
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Kalixt »

I forget about "auto_pause", unfortunately I am still unable to reproduce it. Its not the last defense outpost I built, I will be watching it and report if the problem occurs again.

Twinsen
Factorio Staff
Factorio Staff
Posts: 1330
Joined: Tue Sep 23, 2014 7:10 am
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by Twinsen »

Regarding the crash, I can't seem to reproduce it. You seem to have managed to press quit twice, but I can't make the game crash no matter how I try to quit the game during/after a desync.

User avatar
boskid
Factorio Staff
Factorio Staff
Posts: 2242
Joined: Thu Dec 14, 2017 6:56 pm
Contact:

Re: [1.1.5] Crash after desync (ClientMultiplayerManager::~ClientMultiplayerManager)

Post by boskid »

From the description of the desync itself it is duplicate of 92761 which was fixed in 1.1.6

Post Reply

Return to “Duplicates”