problem with minecraft server | Java Community | Help. Code. Learn. | Page 1

main knot Apr 27, 2026, 3:34 PM

#

I have a problem, at random moments my server crashes and sends me this error.

📎 hs_err_pid1.log

keen skiffBOT Apr 27, 2026, 3:34 PM

#

⌛ This post has been reserved for your question.

Hey @main knot! Please use /close or the Close Post button above when your problem is solved. Please remember to follow the help guidelines. This post will be automatically marked as dormant after 720 minutes of inactivity.

TIP: Narrow down your issue to simple and precise questions to maximize the chance that others will reply in here.

shut furnace Apr 27, 2026, 4:03 PM

#

Hmm, it segfaults when trying to build a message for a ClassCastException.

raven gale Apr 27, 2026, 4:06 PM

#

and it seems to be caused by netty

#

j io.netty.buffer.AdaptivePoolingAllocator$Magazine.newBuffer()Lio/netty/buffer/AdaptivePoolingAllocator$AdaptiveByteBuf;+29

#

Can you try without -DLeaf.enableFMA -DLeaf.disable-vanilla-profiler -DLeaf.enable-io-uring?

#

Also a heap size of 60GB? wow

#

Is there any agent injected somewhere?

main knot Apr 27, 2026, 4:12 PM

#

raven gale Also a heap size of 60GB? wow

I changed it to 30 GB

raven gale Apr 27, 2026, 4:12 PM

#

No it's fine that shouldn't be the problem

#

Do you know what that is? it.ytnoos.lpx.lib.com.github.retrooper.packetevents

main knot Apr 27, 2026, 4:17 PM

#

raven gale Do you know what that is? `it.ytnoos.lpx.lib.com.github.retrooper.packetevents`

lpx is plugin AntiPacketExploit

#

raven gale Apr 27, 2026, 4:17 PM

#

yeah it might be related to that

main knot Apr 27, 2026, 4:21 PM

#

raven gale yeah it might be related to that

Do you know if this could be an issue with my new dedicated server? I upgraded it yesterday and these problems started occurring.

raven gale Apr 27, 2026, 4:22 PM

#

idk what you changed but it would be possible

main knot Apr 27, 2026, 4:22 PM

#

raven gale idk what you changed but it would be possible

entire server

raven gale Apr 27, 2026, 4:25 PM

#

raven gale Can you try without `-DLeaf.enableFMA -DLeaf.disable-vanilla-profiler -DLeaf.ena...

Can you try that?

#

Especially without IO Uring

main knot Apr 27, 2026, 4:28 PM

#

raven gale Especially without IO Uring

without IO Uring it was the same

#

i tested it before

raven gale Apr 27, 2026, 4:34 PM

#

I think the problem occurs at this line in netty: https://github.com/netty/netty/blob/7d954422936a6d1ed6bcfb2956033fe5fa80e0c9/buffer/src/main/java/io/netty/buffer/AdaptivePoolingAllocator.java#L1043

GitHub

netty/buffer/src/main/java/io/netty/buffer/AdaptivePoolingAllocator...

Netty project - an event-driven asynchronous network application framework - netty/netty

#

recycler.get() returns something that is not an AdaptiveByteBuf for some reason

#

actually not entirely sure

old sun Apr 27, 2026, 4:38 PM

#

I just want to add that the crashes occur at different uptimes, sometimes after a few hours, sometimes even immediately after turning on, and almost every time we get a different error. I'm sending here more logs that I managed to collect.

#

📎 hs_err_pid1_4.log 📎 hs_err_pid1_3.log 📎 hs_err_pid1_2.log 📎 hs_err_pid1_1.log

raven gale Apr 27, 2026, 4:43 PM

#

You are using the JDK "OpenJDK Runtime Environment Temurin-21.0.10+7 (21.0.10+7)"

#

you could try updating to 21.10.11

old sun Apr 27, 2026, 4:44 PM

#

Do you think it's definitely not because we had 60gb heap?

#

On the previous server we had 48GB and there were no such problems. Our hosting provider also says that if it was a physical problem with RAM, the entire system would crash, not just Java.

raven gale Apr 27, 2026, 4:48 PM

#

old sun Do you think it's definitely not because we had 60gb heap?

I don't think the heap size itself causes the issue

#

But https://github.com/adoptium/adoptium-support/issues/344 is a similar error and was caused by a hardware/memory failure

GitHub

Intermittent OpenJDK 11.0.8 (and 11.0.12) Crash in GC · Issue #344...

Summary During the JVM runtime excessive of 24 hours JVM intermittently crashes on G1GC Thread. The below logs are from A1 node # A fatal error has been detected by the Java Runtime Environment: # ...

old sun Apr 27, 2026, 4:50 PM

#

Okay, so far we have 40 minutes of uptime and there is no crash after changing the heap to 30GB. If it crashes again, I will do some memory tests, maybe it will show something.

raven gale Apr 27, 2026, 4:51 PM

#

And it seems like the crash logs you provided are all segmentation fauls (except one SIGILL) from different location

#

so a memory failure sounds like a plausible option

#

and SIGILL means there's an illegal machine instruction which doesn't easily happen without a memory failure

#

Theoretically you could also try to download one of the coredumps and try to analyze them

old sun Apr 27, 2026, 4:54 PM

#

yea it crashed again

old sun Apr 27, 2026, 4:54 PM

#

raven gale Theoretically you could also try to download one of the coredumps and try to ana...

where are they?

raven gale Apr 27, 2026, 4:55 PM

#

The logs mention

Core dump will be written. Default location: /home/container/core.1

#

But these files will probably be big

#

I think they contain the entire memory

old sun Apr 27, 2026, 4:55 PM

#

yes, but it's nowhere in the pterodactyl panel

#

only thing i have is that

📎 hs_err_pid1_5.log

raven gale Apr 27, 2026, 4:56 PM

#

Since it seems to be using some containerization, the file is probably removed once the program stops

#

If you can do that, try to check for memory failures or similar

old sun Apr 27, 2026, 4:58 PM

#

Do you think coredump would actually help with this? If so, I can try to change something in the pterodactyl so that it doesn't delete itself

#

for now i will try running memory tests

raven gale Apr 27, 2026, 4:58 PM

#

Maybe maybe not. I think it would be possible to read some things but core dumps are annoying to analyze

#

They basically allow "reviving" the JVM into a state where it cannot run any code but you can inspect some things

raven gale Apr 27, 2026, 5:00 PM

#

old sun

btw I think the last two log files there are exactly the same (but that doesn't make a difference)

#

The error occured in regular (JITed) Java code, native code and even in GC code

old sun Apr 27, 2026, 5:15 PM

#

I ran memory test and the whole dedicate turned off, so it looks to me like it's a physical RAM problem.

keen skiffBOT Apr 28, 2026, 5:20 AM

#

💤 Post marked as dormant

This post has been inactive for over 720 minutes, thus, it has been archived.
If your question was not answered yet, feel free to re-open this post or create a new one.
In case your post is not getting any attention, you can try to use /help ping.
Warning: abusing this will result in moderative actions taken against you.

old sun Apr 28, 2026, 7:02 PM

#

Hello, it turned out that the RAM was actually physically damaged, so I have another question, apart from these crashes, what would be the appropriate heap size for a Minecraft server with +- 200 players or is it better to have closer to 30 or 40GB?

raven gale Apr 28, 2026, 7:04 PM

#

I don't know much about Minecraft servers

old sun Apr 28, 2026, 7:05 PM

#

I'm rather concerned about the Java itself whether the heap size has any greater significance

raven gale Apr 28, 2026, 7:05 PM

#

but there are some general tricks about heap sizing

#

A bigger heap size will typically make it easier for the JVM

#

but ofc you don't have infinite memory available

old sun Apr 28, 2026, 7:10 PM

#

I have 110 GB of RAM available, but I want to assign a reasonable amount of RAM to the heap, max 45 GB, so that the GC doesn't have problems.

#problem with minecraft server