OldCombatMechanics

OldCombatMechanics

46.1k Downloads

Players getting kicked upon changing their world with various config settings

Aevox01 opened this issue ยท 5 comments

commented

Information

  • Server Version: 1.15.2-R0.1-SNAPSHOT
  • OldCombatMechanics version: 1.8.1
  • Server Log File: Console doesn't say anything at all.

Problem Description

This bug does not occur if I use the standard config (means without changing the generated one at all). Once I use an edited config and a player changes the world everyone gets disconnected (sometimes it gives a remotehost error, sometimes just some customized network error that says that the server went down while it didnt).

To Reproduce

Steps to reproduce the behavior:

  1. Put the config that's shown in the pastebin link
  2. Change worlds (A simple usage of multiverse is enough for that)
  3. Get kicked

Expected Behaviour

Well, I don't expect it to kick players. I just expect it to work as intended aka to not kick anyone.

Actual Behaviour

As stated above, multiple times, it kicks the whole server while the server itself doesn't crash or something (You can simply rejoin after you got kicked)

commented

looks to be a potential duplicate of #403 and the way that scoreboards are handled -- can you try disabling disable-player-collisions and see if the problem goes away? @Aevox01

also even if there don't seem to be any OCM-related errors on console, console logs often have tidbits of information that are more useful than you might expect. a console log could be quite helpful

commented

good to know, hmmm. yes we need a console log please

commented

It's pretty much what you described. I changed the config in the way that is needed for us and put disable-player-collisions on false. The bug no longer occurs now. Yet I would like to put this on true so a bug fix would still be appreciated. Is the console log still needed?

commented

I second that. I just tried your steps:
Imgur video

Do you have any plugins creating or meddling with scoreboards?

commented

@Aevox01 could you please try this test version with the new patch? We think it might fix the issue.