Cole Abbeduto

Added another crash report and log zip "cole21771_crash_4-25.zip". This one looks more similar to the original one uploaded by the submitter of the issue. Again, no idea what the players were doing when it happened.

I just uploaded a zip to this issue called "cole21771_crash.zip" which contains my crash report as well as the "latest.log" file leading up to the crash. We have no idea what causes the crash, just that three people were playing and all of a sudden they were kicked and the server was no longer up. Hope this helps.

For context, my server is running Ubuntu 18 with the vanilla 1.14 server.jar. Let me know if I should create a separate issue for this, or if you guys think it is the same bug.

Also, I can confirm nobody has been to the nether, nor were they trying to enter the nether when it crashed.

Spyware

@Donald Kerner

Yes, I know magma blocks are supposed to sink things (ex: players, boats, entities, etc), that is not the point of this report. The point is that if you have string on the surface above a magma block, it sinks boats immediately instead of eventually. I believe this not to be intended by the developers.

Ps: I am a developer (not at Mojang tho)

@Kumasasa, As I already said, community support is useless. I've already tried many times and nobody has anything good to suggest. I'm running out of ideas here...

@Kumasasa, I tried it and it did not fix the issue. Still nobody can join when server render distance is set anywhere above 12 (only some people can join above 12). Any other suggestions?

@Kumasasa, it's about as basic as it gets

#!/bin/bash

java -Xmx4G -Xms4G -jar server.jar nogui

java version 1.8.0_171

Kumasasa: Yes, I've spoken to the IRC chat and only one guy suggested something and it was just firewall stuff which we already tried. It seems to be related to loading terrain since that's where the logging in takes the longest and then will either let you in or remove you from the game with a "timed out" or the "An existing connection was forcibly closed by the remote host". We have gotten people able to join the game by lowering the server's render distance therefore lowering the amount of data it has to transfer. It seems as though in 1.13 the amount of time that is allocated before the connection is considered "timed out" has been decreased. It would be great if you could as least humor me and look into that when you get a chance to see if there is any truth to it. 

My assumption would be that you receive "timed out" when your client decides it has been too long and you receive the exception ("An existing connection...") when the server decides that it has been too long. 

Kumasasa: Tech support who? You guys don't offer server support and this is my own server running on my own network. 

Kumasasa, the biggest problem we have encountered so far has been inconsistency in these issues. One moment somebody can play, and then next moment they get one of the errors above. It's very strange and we've completely run out of ideas to try. 

The system is running Ubuntu Server LTS (18. whatever) with iptables that accept everything in every direction. We have tried messing with the router's firewall as well with no success (disabling it did nothing). The MC logs don't seem to agree with what they are telling the clients which is very strange. There is nothing that we've found in linux networking logs either. I have no idea what else to try, but it seems to be the new version of minecraft. 

Kumasasa, I believe this might be an issue with the server software

itself. I myself just started encountering the issue as of yesterday with our 1.13 server. We have played many servers on the exact same machine on the same network and everybody has no problems joining other servers on the machine, but many people are encountering "timed out" and "An existing connection was closed by the remote host". 

Well we are not really here to determine whether they SHOULD or SHOULD NOT be infinitely repairable. But another bug came out of this which is that they only cost 4 levels to repair every time until you get to the last repair, then it is 35 levels, then can't repair. Steps to reproduce are given by Bren Rado

But it costed 4 level to repair it every time before, then I put it in an it all of a sudden costs 35, that makes no sense on any scale at all.

Mark it as resolved, it was a rouge command block.

And it all of a sudden seems to be working fine, huh... Nevermind then, mark this as resolved and carry on

Come on Mojang, it really can't be that hard to fix a little thing like this. I just really hope they fix it before 1.9 (Monday).

Edit: Well, that obviously didn't happen...

Actually I just looked at the MC Wiki, and it says "The amount of raw chicken (or cooked chicken) dropped is not affected by the Looting enchantment.". So I guess that is what the guy is going for.

Link: http://minecraft.gamepedia.com/Chicken#Drops

So I guess I was wrong unless that that aren't suppose to drop more is a bug.

Dude, do you know what Looting does?

Looting (According to the wiki):
Mobs can drop more loot
Increases maximum loot drop by +1 per level.
Increases chance of rare drops by +1 percentage points per level (i.e., 3.5% at level I, 4.5% at level II, and 5.5% at level III).
Notably, this applies not only to kills with the sword but to any kills while holding the sword. For example, a player can fire an arrow at a target and then switch to a looting sword before the arrow hits to get the looting effect.
The exact drop chances depend on the type of drop.
Using codes to get higher levels than 3 works as expected, adding higher numbers with a weight of 1. High enough numbers however (like level 1000) may crash the game due to too many dropped item entities.

How is it incomplete?

I see how it relates the that other bug report, but why mark it as resolved if it isn't...