Factorio Forums

Hello there ! I got a desync for the first time ever so I am gonna report it.
When I got desynchronized I was AFK and alone, so I have no idea what could have triggered the desync.
I am running 1.0.0 Windows factorio (no mods) (Windows 10 pro 2004, 19041.50, x64) and the server is running 1.0.0 linux headless factorio server (Ubuntu 18.04.5 LTS, x64).

Here is the desync report :
https://www.dropbox.com/s/2unzj5tkd5qe8 ... 5.zip?dl=1

Here is a backup I got like an hour before the desync. I was disusing about some trains failing to repath in a stacker on the forum.
https://www.dropbox.com/s/jx8vwg65szp9v ... e.zip?dl=1
The forum discussion in question, but i really dont think it is related anyway.
viewtopic.php?f=18&t=89399&p=513033#p513033

No idea if this report will be of any use but yeah

There is a lot of differences around electricity (like steam turbine frame positions, fluid differences) that seems to be a secondary differences. There is 1 interesting difference: inserter flag HAD_VALID_TARGET for the long handed inserter at {-1052.5, 407.5} near the centrifuges. Looking at the electric network statistics i see there is a power usage difference for long handed inserters around 0.5s before the desync. Do you have any clues if there was something happening here? Right now for me it looks pretty much as bitflip but i will look if i can reproduce the desync.

No idea, this kovarex setup has been running for hours. The only thing that changed is the electric consumption.
The setup is running very slowly since it is only used for the nuclear reactor, I am not even sure this inserter activate even once in an hour.
I'll try to replicate it by cloning the server with the backup, if I find some time.

I got another desync, here is the desync report :
https://www.dropbox.com/s/bi0slrxyvlm8w ... 2.zip?dl=1

This time I was not afk, but I was not doing anything when the desync occurred, but some robots were building things.

That last desync is really clean, it points onto the same desync as the previous one: long handed inserter taking from blue belt to the centrifuge has HAD_VALID_TARGET flag differing. I have some idea what may be happening here and it may be related to the belt loop.

I fixed the desync we were able to find which is likely to fix this issue.

Nice ! Thank you ! I did not expect it to be fixed that fast