Players get massive ping spikes (15k ms) and disconnect with keepalive timeout

[Sneyzi]

Description

When Orebfuscator is enabled, some players experience massive ping spikes that quickly grow until they reach ~15,000 ms, after which the player is kicked with:

[INFO]: <player> was kicked due to keepalive timeout!
[INFO]: <player> lost connection: Timed out

This can be consistently reproduced when a player uses /rtp (random teleport). The ping starts climbing smoothly after the teleport and eventually causes a disconnect.

If the player is given the Orebfuscator bypass permission and rejoins the server, the issue no longer occurs.

Disabling proximity per world does not fix the problem. I did not change other config options.

Workaround:
Giving the affected player the Orebfuscator bypass permission prevents the problem after relogging.

Actual behavior:
Teleporting with /rtp causes ping to rise rapidly until the player is disconnected with keepalive timeout.

Reproduction Steps

1. Install Orebfuscator with default configuration.
2. Teleport a player with /rtp.
3. Watch their ping steadily rise until ~15k ms.
4. The player is kicked with keepalive timeout.

Expected Behaviour

Players should not get huge ping spikes or disconnects when using /rtp with Orebfuscator enabled.

Orebfuscator Dump

https://gist.github.com/Sneyzi/753ca14705def73f4fe985ce192f6ebc

Extra Details

No response

[Ingrim4]

I tried to recreate what you are experiencing using spigot 1.21.8, with the same version of ProtocolLib and Orebfuscator, WildRTP as /rtp command but I couldn't successfully recreate what you described.

Can you please try to recreate this behavior on a fresh server with only ProtocolLib, Orebfuscator and your RTP plugin using the same versions as in the report. If you could recreate it successfully could you please provide us with a spark profiler report? Maybe this will help us extract some useful data to diagnose the issue. Here are the steps:

1. Add spark to your server if not already installed.
2. Run the command /spark profiler start.
3. Try to recreate the issue you described.
4. After the issue occurs wait a few seconds and run the command /spark profiler stop.
5. Share the link to the spark report with us.

[Sneyzi]

The issue occurs even without using /rtp. On the main server, the spawn world consists of a single island — players there experience no problems. However, in worlds with more blocks where Orebfuscator’s protection is active, players’ ping starts climbing rapidly and can spike extremely high, eventually causing a keepalive timeout.

https://spark.lucko.me/xFxKGiFhuECi

[Ingrim4]

Everything looks normal, CPU and memory also seem okay so the only thing I can think of is reducing packet sizes which can be achieved by setting layerObfuscation to true in every obfuscation config section. For more info about layer obfuscation click here.