Railcraft

Railcraft

34M Downloads

Signals occasionally disconnect from each other

Spauldo opened this issue ยท 2 comments

commented

On a server running 9.5.0.0, the signal blocks occasionally "undefine" themselves. Most of the time they're fine, but I'll log into the server about twice a week and find one that's blinking red with a train waiting at it. It's ususally the same signal, placed near a villiage (with witchery hobgoblins) - it contains a 1 block rise, a liquid loader, and an item loader (both with train locking tracks). The other one that gives me trouble has a rise, but no loaders or special tracks, and is in the middle of a roofed forest with no structures around.

Both areas (and almost all my signal blocks) are defined between a block signal two blocks above the track (i.e. you can walk under it) and a signal block relay placed beside the track and one block down (for aesthetic reasons). On top of the relay is a signal receiver box tuned to the next block signal. Neither area is chunkloaded, but my signals are on chunk boundaries and the trains have anchor carts (with anchor fuel disabled - it's a private server, chunkloading isn't a problem). The signal receiver boxes stay tuned.

There's no error messages in the log that I can find that relate to Railcraft.

The server's had a lot of updates and had at least two versions of Railcraft. The signal that's been doing it lately, however, was placed after I updated to 9.5.0.0. It's not caused by server restarts - I rarely restart the server. My server's mostly tech mods - think a cut down version of the direwolf20 pack.

railcraft_block
Attached is a picture of the block in question. The arrows point to the block signal and the relay that define the block (the relay is underneath a signal receiver box).

commented

There is a debug setting in the config that you can turn on that will print information to the log about when signal pairing fails. If you can use that to further isolate the specific issue, that would be helpful.

"tweaks.blocks.signals.printDebug"

commented

This should be fixed by 6b63d55 and 1bd1246