arrow: [C++] Test failures for libarrow in conda-forge

Describe the bug, including details regarding any error messages, version, and platform.

As I explain in https://github.com/conda-forge/arrow-cpp-feedstock/pull/1058, I wanted to enable the gtest suite for libarrow. The result is as follows:

linux-64:

The following tests FAILED:
	 33 - arrow-utility-test (Failed)
	 51 - arrow-cuda-test (Failed)
	 65 - arrow-gcsfs-test (Failed)
	 66 - arrow-s3fs-test (Failed)
	 69 - arrow-flight-test (Failed)
	 70 - arrow-flight-transport-ucx-test (Failed)

osx-64:

The following tests FAILED:
	 64 - arrow-gcsfs-test (Failed)
	 65 - arrow-s3fs-test (Failed)
	 68 - arrow-flight-test (Failed)
	 69 - arrow-flight-sql-test (Failed)

win-64:

The following tests FAILED:
	 19 - arrow-compute-scalar-cast-test (Failed)
	 22 - arrow-compute-scalar-temporal-test (Failed)
	 39 - arrow-substrait-substrait-test (Failed)
	 51 - arrow-cuda-test (Exit code 0xc0000135)
	 62 - arrow-dataset-file-orc-test (Failed)
	 65 - arrow-gcsfs-test (Failed)
	 66 - arrow-s3fs-test (Failed)
	 68 - arrow-flight-internals-test (Exit code 0xc0000135)
	 69 - arrow-flight-test (Exit code 0xc0000135)
	 70 - arrow-flight-sql-test (Exit code 0xc0000135)
	 77 - arrow-orc-adapter-test (Failed)
	 87 - gandiva-internals-test (Failed)
	 96 - gandiva-date-time-test (Failed)

The arrow-cuda-test failure is probably unavoidable (as we have no CUDA drivers in CI), but the others shouldn’t have to fail AFAICT.

More detailed logs can be found here

Component(s)

C++, Packaging

About this issue

  • Original URL
  • State: closed
  • Created 4 years ago
  • Comments: 26 (26 by maintainers)

Commits related to this issue

Most upvoted comments

Well, we’ve tried several times to run the C++ S3 tests on macOS, but have for now bumped into issues we weren’t able to diagnose: see https://github.com/apache/arrow/issues/40410 and https://github.com/apache/arrow/pull/44211

Interestingly, the PyArrow tests with S3 run fine.

You could perhaps try disabling the nonexistent bucket tests on macOS and see if that gets macOS CI to pass reliably on conda-forge.