[CI] GoogleCloudStorageBlobContainerRetriesTests testReadBlobWithReadTimeouts failing

[elasticsearchmachine]

Build Scans:

• elasticsearch-periodic #4511 / openjdk23_java-matrix
• elasticsearch-periodic #4456 / openjdk23_java-matrix

Reproduction Line:

./gradlew ':plugins:repository-gcs:test' --tests "org.elasticsearch.repositories.gcs.GoogleCloudStorageBlobContainerRetriesTests.testReadBlobWithReadTimeouts" -Dtests.seed=D91004A714FAA6C5 -Dtests.locale=lij-Latn-IT -Dtests.timezone=Pacific/Marquesas -Druntime.java=23

Applicable branches:
7.17

Reproduces locally?:
N/A

Failure History:
See dashboard

Failure Message:

java.lang.AssertionError: 
Expected: <2>
     but: was <0>

Issue Reasons:

• [7.17] 2 failures in test testReadBlobWithReadTimeouts (0.6% fail rate in 331 executions)
• [7.17] 2 failures in step openjdk23_java-matrix (40.0% fail rate in 5 executions)
• [7.17] 2 failures in pipeline elasticsearch-periodic (16.7% fail rate in 12 executions)

Note:
This issue was created using new test triage automation. Please report issues or feedback to es-delivery.

[elasticsearchmachine]

Pinging @elastic/es-distributed (Team:Distributed)

[pxsalehi]

This doesn't seem to be as clearly the same issue as #115145 (I don't see the HTTP server assertion), but seems to only fail for JDK23, which makes me thing it is a similar issue. It also fails on main.

[ywangd]

Maybe we should just remove the randomBoolean in sendIncompleteContent so it always flush. I thought about it in #115177. But was not sure whether it is too aggressive. Maybe it is the right move for jdk 23.

[pxsalehi]

ok. same time as I opened the PR. It seems this should be only temporary, so maybe not always always.