GH-36331: [C++][CI] Sporadic errors in AsofJoinTest

[rtpsw]

Rationale for this change

Skipping a flaky test until a robust way for it is found.

What changes are included in this PR?

The AsofJoinTest.BackpressureWithBatchesGen test is skipped.

Are these changes tested?

N/A

Are there any user-facing changes?

No.

• Closes: [C++][CI] Sporadic errors in arrow-acero-asof-join-node-test #36331

[github-actions]

⚠️ GitHub issue #36311 has been automatically assigned in GitHub to PR creator.

[rtpsw]

cc @icexelloss, @pitrou, @bkietz, @kou

[pitrou]

@rtpsw Looks like you mentioned the wrong issue number in the PR description?

[github-actions]

⚠️ GitHub issue #36311 has been automatically assigned in GitHub to PR creator.

[github-actions]

⚠️ GitHub issue #36331 has been automatically assigned in GitHub to PR creator.

[kou]

Are the followings correct?

• We disable the test and close [C++][CI] Sporadic errors in arrow-acero-asof-join-node-test #36331 as a workaround for now.
• We will add a real fix and close [C++] Fix and re-enable Asof Join Backpresure test flakiness #36248 later.

[rtpsw]

Yes, this is my understanding of this post.

[kou]

I see. I'll merge this.

[conbench-apache-arrow]

Conbench analyzed the 5 benchmark runs on commit 5cc6aa35.

There were 8 benchmark results indicating a performance regression:

• Commit Run on arm64-m6g-linux-compute at 2023-06-29 08:58:30Z
  • params=1048576/1, source=cpp-micro, suite=arrow-acero-aggregate-benchmark
  • params=1048576/10000, source=cpp-micro, suite=arrow-acero-aggregate-benchmark
• and 6 more (see the report linked below)

The full Conbench report has more details.