Originally reported at https://github.com/adoptium/adoptium-support/issues/715

There are multiple crashes in various G1 phases (have attached 3 hs_err files). One example:

#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007fe79506d100, pid=2364926, tid=0x00007fe780a3d700
#
# JRE version: OpenJDK Runtime Environment (8.0_362-b09) (build 1.8.0_362-b09)
# Java VM: OpenJDK 64-Bit Server VM (25.362-b09 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# V [libjvm.so+0x591100] G1ParEvacuateFollowersClosure::do_void()+0x90
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# *************************
#
# If you would like to submit a bug report, please visit:
# https://github.com/adoptium/adoptium-support/issues
#
Aborted (core dumped)

I asked the end user a few Q's on the env:

1. Is there another process that could be overwriting memory (are your JVMs isolated / the only process?)

The JVM is the only process running except for a few supporting processes like monitoring, config management

2. A JNI/JNA interaction (do you have any native code elements?)

We only have a very small amount of JNI, and has been stable for quite a few years now

3. A buggy native memory allocator (recommend you're at the latest patch level for you O/S)

We have patched to the latest ubuntu20.04 and now running everything on the latest kernel: 5.4.0-139-generic. Crash still occurred

4. Bad RAM - rare, but it does happen - if you see it on a particular host then that's the issue

They are running on VMware and is happening across VM's in different clusters around the world, and random processes