Fix error-prone violation in trino-server-rpm

[wendigo]

fixes #21161 (comment)

[wendigo]

This is new:

#  SIGSEGV (0xb) at pc=0x00007f14fa61fcd1, pid=2495, tid=2628
#
# JRE version: OpenJDK Runtime Environment Temurin-22+36 (22.0+36) (build 22+36)
# Java VM: OpenJDK 64-Bit Server VM Temurin-22+36 (22+36, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0x81fcd1][thread 2610 also had an error]
  G1ParScanThreadState::trim_queue_to_threshold(unsigned int)+0x3861

adoptium/adoptium-support#959

[findepi]

JRE version: OpenJDK Runtime Environment Temurin-22+36 (22.0+36) (build 22+36)

is this related to #21161?

[wendigo]

@findepi rather to the reduced memory than to a jdk 22

[findepi]

might be, agreed! but did we reduce memory for maven main process or for tests (#21256)?
BTW it says SIGSEGV rather than "out of memory".

[wendigo]

@findepi the linked bug manifests itself during high-memory pressure scenarios hence the assumption. Let's monitor this

[findepi]

not sure which part of the linked bug (adoptium/adoptium-support#959) apply to our case. it says Java 21 but it also says it's resolved for some time