HDDS-10208. Simplify tests using assertThrows in hadoop-hdds

[wzhallright]

What changes were proposed in this pull request?

Simplify tests using assertThrows in hadoop-hdds

What is the link to the Apache JIRA

HDDS-10208

CI

https://github.com/wzhallright/ozone/actions/runs/7737861963

[fapifta]

Thank you for the path @wzhallright the changes look good except that one change I left an inline comment on, can you please elaborate on that?

[fapifta]

May I ask what is the rationale behind changing this long to AtomicLong? The rest of the patch is really just changing exception testing to assertThrows, is it by chance a leftover from some other work, or intentional?

[adoroszlai]

I assume it is changed to be able to use in this lambda expression:

ozone/hadoop-hdds/server-scm/src/test/java/org/apache/hadoop/hdds/scm/pipeline/TestPipelineStateManagerImpl.java

Lines 322 to 325 in 482ec1a

	IOException e =
	assertThrows(IOException.class,
	() -> stateManager.addContainerToPipeline(finalPipeline.getId(),
	ContainerID.valueOf(containerID.incrementAndGet())));

I think it's fine. If we want to minimize the diff, we could keep long containerID and change the code above to something like this:

    ContainerID cid = ContainerID.valueOf(++containerID);
    IOException e = assertThrows(IOException.class,
        () -> stateManager.addContainerToPipeline(finalPipeline.getId(), cid));

[wzhallright]

Thanks for your review! the IntelliJ prompts me variable used in lambda expression should be final or effectively final, so i change long to AtomicLong. Perhaps there is a more elegant solution to this problem?

[wzhallright]

@adoroszlai That's a good idea. I will modify it. Thanks

[fapifta]

thank you for the explanation, I did not catch that point. my bad.

[adoroszlai]

Thanks @wzhallright for the patch, @fapifta for the review.