[ARROW-17501] [C++] Fatal error condition occurred in aws_thread_launch

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 9.0.0
• Fix Version/s: 10.0.0
• Component/s: C++, Python
• Labels: pull-request-available
• Environment: Python 3.10.4 in a ubuntu:22.04 docker image w/ kernel 5.4.0-90-generic
• External issue URL: https://github.com/apache/arrow/issues/32761
• Assignee: Kouhei Sutou
• Reporter: Matthew Schurr
• Created: 22/Aug/22 17:37
• Updated: 11/Jan/23 11:51
• Resolved: 17/Sep/22 04:34

Description

I am encountering this error using the latest version of pyarrow (9.0.0) w/ Python 3.10. I received pyarrow as an indirect dependency of google-cloud-bigquery.

It seems like this issue was never fully fixed:

https://issues.apache.org/jira/browse/ARROW-15141

Fatal error condition occurred in /opt/vcpkg/buildtrees/aws-c-io/src/9e6648842a-364b708815.clean/source/event_loop.c:72: aws_thread_launch(&cleanup_thread, s_event_loop_destroy_async_thread_fn, el_group, &thread_options) == AWS_OP_SUCCESS
Exiting Application
################################################################################
Stack trace:
################################################################################
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x200af06) [0x7f4719c39f06]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x20028e5) [0x7f4719c318e5]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x1f27e09) [0x7f4719b56e09]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x200ba3d) [0x7f4719c3aa3d]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x1f25948) [0x7f4719b54948]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x200ba3d) [0x7f4719c3aa3d]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x1ee0b46) [0x7f4719b0fb46]
/usr/local/lib/python3.10/dist-packages/pyarrow/libarrow.so.900(+0x194546a) [0x7f471957446a]
/lib/x86_64-linux-gnu/libc.so.6(+0x45495) [0x7f473ab81495]
/lib/x86_64-linux-gnu/libc.so.6(on_exit+0) [0x7f473ab81610]
/usr/local/bin/python(+0x2790ab) [0x55965e9a00ab]
/usr/local/bin/python(+0x265d7f) [0x55965e98cd7f]
/usr/local/bin/python(PyErr_PrintEx+0x1d) [0x55965e98ccad]
/usr/local/bin/python(_PyRun_SimpleFileObject+0x1d6) [0x55965e989576]
/usr/local/bin/python(_PyRun_AnyFileObject+0x43) [0x55965e989243]
/usr/local/bin/python(Py_RunMain+0x2be) [0x55965e979b6e]
/usr/local/bin/python(Py_BytesMain+0x2d) [0x55965e94fe6d]
/lib/x86_64-linux-gnu/libc.so.6(+0x29d90) [0x7f473ab65d90]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x80) [0x7f473ab65e40]
/usr/local/bin/python(_start+0x25) [0x55965e94fd65]
/usr/bin/bash: line 172:    20 Aborted                 (core dumped) 

I think this issue might only occur on certain kernel versions? It's strange, I cannot reproduce it running the image on my machine (ubuntu 4.15.0-180-generic), but some of our CI runners (those w/ ubuntu using kernel 5.4.0-90-generic) consistently have this issue. I will try to gather some more information.

Issue Links

• relates to: ARROW-15141 [C++] Fatal error condition occurred in aws_thread_launch
• links to: GitHub Pull Request #14157