multiconnect

multiconnect

108k Downloads

1.19 -> 1.18.2 issue with world height

wagyourtail opened this issue ยท 7 comments

commented

reproduce: go on nucleoid.xyz and try to join taterspleef

this is the main one, but it also seems some "silently" failed earlier without kicking me
Compressed packet data: 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

https://mclo.gs/ZBR0lSN

second log with -Dmulticonnect.storeBufsForHandler=true: https://mclo.gs/z2gzZ3k

commented

This seems to be caused by an issue with the world height mismatching between the client and the server

commented

yeah, ok we're back to the right error now I think...

Compressed packet data: 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
Caused by: java.lang.IndexOutOfBoundsException: varint too big
	at net.earthcomputer.multiconnect.impl.PacketIntrinsics.readVarInt(PacketIntrinsics.java:248) ~[multiconnect-1.6-beta.16.jar:?]
	at net.earthcomputer.multiconnect.generated.Protocol_1_18_2.translateSPacketChunkData_Latest(Protocol_1_18_2.java:6600)
	```
commented

@Earthcomputer is the worldheight/dimension info sent in an earlier packet, and because you're only replaying the one packet it just has an issue in your test env?

commented

It's sent in the game join packet and player respawn packet

commented

yeah, I just ran with beta.15 to confirm it's still happening
wait, hmm

[03:14:20] [Netty Epoll Client IO #0/ERROR]: Compressed packet data: H4sIAAAAAAAA/21Q0U6EMBA08QP8hdpniJwiAd79CmMuS9mDnqV7KQVsLvyUX+iiFyUefduZ6ezs3H3enqUiQ06WMqAxNMlI4od3IMvXP6pxEJjwzPAk5+iK0rZH5zVZhiZoAg3OgzbMKKPV+8uIlr+eJaiLqB+aBnu/V9R1YGsWjmAGZObBcxCxMhG8ULY0otuwaWnaM6j9kkKR9Tz0i8SH0+LWaYvKwcGXJwMB3ZK1ZjwvkiLNiiyudukhTtMsi6F6TOJklz/XyVNRFzmw1kKH324/p69Pm+c52sSv29kq7rfuCyla6MWROG4tfIui4c3CUFWFezm/rTxu/r8vwapJ7cYBAAA=
[03:14:20] [Netty Epoll Client IO #0/ERROR]: Unexpectedly disconnected from server!
io.netty.handler.codec.DecoderException: java.lang.ClassCastException: class net.earthcomputer.multiconnect.packets.v1_18_2.Text_1_18_2 cannot be cast to class net.earthcomputer.multiconnect.packets.CommonTypes$Text_Latest (net.earthcomputer.multiconnect.packets.v1_18_2.Text_1_18_2 and net.earthcomputer.multiconnect.packets.CommonTypes$Text_Latest are in unnamed module of loader net.fabricmc.loader.impl.launch.knot.KnotClassLoader @62ee68d8)
commented

It's gzip compressed. Is the world height different than vanilla on this server?

commented

yes,
image
but also that looks like an oopsie from you in beta 15 as well, my reconfirming was a class cast exception on Text_1_18_2
yeah, so if the worldheight matching code is already there and working properly, I can't actually test if it's fixed until #389