[CI][Python] test-conda-python-emscripten job is timing out

[raulcd]

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

The test-conda-python-emscripten job has been timing out for the last couple of months.

It fails since: db1a8a9...0ae71f8

Component(s)

Continuous Integration, Python

[raulcd]

The job probably is hanging because the last success was taking only 20 minutes instead of timing out after one hour.
See latest success here: https://github.com/ursacomputing/crossbow/actions/runs/16895765338/job/47865120178

[raulcd]

More details about the real error:

+ python scripts/run_emscripten_tests.py dist/pyarrow-22.0.0.dev178-cp312-cp312-pyodide_2024_0_wasm32.whl --dist-dir=/pyodide --runtime=node
-------------- Running emscripten tests in Node ----------------------
Script started, output log file is 'typescript'.
Welcome to Node.js v20.19.5.
Type ".help" for more information.
        const { loadPyodide } = require('/pyodide/pyodide.js');
        let pyodide = await loadPyodide();
        
await pyodide.loadPackage('/build/python/dist/pyarrow-22.0.0.dev178-cp312-cp312-pyodide_2024_0_wasm32.whl');
            python = `import sys
import micropip
if "pyarrow" not in sys.modules:
    await micropip.install("hypothesis")
    import pyodide_js as pjs
    await pjs.loadPackage("numpy")
    await pjs.loadPackage("pandas")
    import pytest
    import pandas # import pandas after pyarrow package load for pandas/pyarrow
                  # functions to work
import pyarrow
    `;
            await pyodide.loadPackagesFromImports(python);
            python_output = await pyodide.runPythonAsync(python);
        
            python = `
import pyarrow,pathlib
pyarrow_dir = pathlib.Path(pyarrow.__file__).parent
pytest.main([pyarrow_dir, '-r', 's'])
`;
            await pyodide.loadPackagesFromImports(python);
            python_output = await pyodide.runPythonAsync(python);
        
process.exit(python_output)
> 
>         const { loadPyodide } = require('/pyodide/pyodide.js');
undefined
>         let pyodide = await loadPyodide();
undefined
>         
undefined
> await pyodide.loadPackage('/build/python/dist/pyarrow-22.0.0.dev178-cp312-cp312-pyodide_2024_0_wasm32.whl');
Loading pyarrow
Loaded pyarrow
Pyodide has suffered a fatal error. Please report this to the Pyodide maintainers.
The cause of the fatal error was:
RuntimeError: null function or function signature mismatch
    at wasm://wasm/0267d6e2:wasm-function[2334]:0x211f31
    at wasm://wasm/0267d6e2:wasm-function[1836]:0x1eb349
    at wasm://wasm/0267d6e2:wasm-function[338]:0x16f5ad
    at wasm://wasm/0267d6e2:wasm-function[319]:0x16e4cc
    at wasm://wasm/0267d6e2:wasm-function[367]:0x17126e
    at Module.handle_js_error (/pyodide/pyodide.asm.js:10:51715)
    at js2python_js (/pyodide/pyodide.asm.js:10:11532)
    at wasm://wasm/0267d6e2:wasm-function[466]:0x17626c
    at python_getattr (/pyodide/pyodide.asm.js:10:75944)
    at Object.get (/pyodide/pyodide.asm.js:10:77509) {
  pyodide_fatal_error: true
}
Stack (most recent call first):
  <no Python frame>
Uncaught RuntimeError: null function or function signature mismatch
    at wasm://wasm/0267d6e2:wasm-function[2334]:0x211f31
    at wasm://wasm/0267d6e2:wasm-function[1836]:0x1eb349
    at wasm://wasm/0267d6e2:wasm-function[338]:0x16f5ad
    at wasm://wasm/0267d6e2:wasm-function[319]:0x16e4cc
    at wasm://wasm/0267d6e2:wasm-function[367]:0x17126e
    at Module.handle_js_error (/pyodide/pyodide.asm.js:10:51715)
    at js2python_js (/pyodide/pyodide.asm.js:10:11532)
    at wasm://wasm/0267d6e2:wasm-function[466]:0x17626c
    at python_getattr (/pyodide/pyodide.asm.js:10:75944)
    at Object.get (/pyodide/pyodide.asm.js:10:77509) {
  pyodide_fatal_error: true
}
> 
>             python = `import sys
... import micropip
... if "pyarrow" not in sys.modules:
...     await micropip.install("hypothesis")
...     import pyodide_js as pjs
...     await pjs.loadPackage("numpy")
...     await pjs.loadPackage("pandas")
...     import pytest
...     import pandas # import pandas after pyarrow package load for pandas/pyarrow
...                   # functions to work
... import pyarrow
...     `;
'import sys\n' +
  'import micropip\n' +
  'if "pyarrow" not in sys.modules:\n' +
  '    await micropip.install("hypothesis")\n' +
  '    import pyodide_js as pjs\n' +
  '    await pjs.loadPackage("numpy")\n' +
  '    await pjs.loadPackage("pandas")\n' +
  '    import pytest\n' +
  '    import pandas # import pandas after pyarrow package load for pandas/pyarrow\n' +
  '                  # functions to work\n' +
  'import pyarrow\n' +
  '    '
>             await pyodide.loadPackagesFromImports(python);
Uncaught Error: Pyodide already fatally failed and can no longer be used.
    at Object.get [as loadPackagesFromImports] (/pyodide/pyodide.asm.js:10:50349)
    at REPL19:1:59
>             python_output = await pyodide.runPythonAsync(python);
Uncaught Error: Pyodide already fatally failed and can no longer be used.
    at Object.get [as runPythonAsync] (/pyodide/pyodide.asm.js:10:50349)
    at REPL20:1:75
>         
undefined
> 
>             python = `
... import pyarrow,pathlib
... pyarrow_dir = pathlib.Path(pyarrow.__file__).parent
... pytest.main([pyarrow_dir, '-r', 's'])
... `;
'\n' +
  'import pyarrow,pathlib\n' +
  'pyarrow_dir = pathlib.Path(pyarrow.__file__).parent\n' +
  "pytest.main([pyarrow_dir, '-r', 's'])\n"
>             await pyodide.loadPackagesFromImports(python);
Uncaught Error: Pyodide already fatally failed and can no longer be used.
    at Object.get [as loadPackagesFromImports] (/pyodide/pyodide.asm.js:10:50349)
    at REPL28:1:59
>             python_output = await pyodide.runPythonAsync(python);
Uncaught Error: Pyodide already fatally failed and can no longer be used.
    at Object.get [as runPythonAsync] (/pyodide/pyodide.asm.js:10:50349)
    at REPL29:1:75
>         
undefined
> process.exit(python_output)
Uncaught ReferenceError: python_output is not defined

[raulcd]

I tried to identify whether there are symbols missing or why is it failing. I tried to build locally the latest working build and had the same issue so it seems is not code related but infrastructure/dependencies related. If someone with emscripten/pyodide experience wants to keep investigating.

[raulcd]

This somehow is fixed (node tests are successful) but what seems to be failing to load the wheel is on chrome tests.
See one of the latest builds:
https://github.com/ursacomputing/crossbow/actions/runs/18826277374/job/53709491953
And the current error

+ echo '-------------- Running emscripten tests in Chrome --------------------'
+ python scripts/run_emscripten_tests.py dist/pyarrow-23.0.0.dev3-cp312-cp312-pyodide_2024_0_wasm32.whl --dist-dir=/pyodide --runtime=chrome
-------------- Running emscripten tests in Chrome --------------------
127.0.0.1 - - [27/Oct/2025 01:07:14] code 404, message File not found
Traceback (most recent call last):
dist dir=/pyodide
  File "/build/python/scripts/run_emscripten_tests.py", line 331, in <module>
Starting server for /pyodide at: ('0.0.0.0', 40639)
    _load_pyarrow_in_runner(driver, Path(args.wheel).name)
Load pyodide in browser
Load pyarrow in browser
  File "/build/python/scripts/run_emscripten_tests.py", line 276, in _load_pyarrow_in_runner
    driver.load_arrow()
  File "/build/python/scripts/run_emscripten_tests.py", line 209, in load_arrow
    self.execute_python(
  File "/build/python/scripts/run_emscripten_tests.py", line 216, in execute_python
    self.driver.execute_async_script(
  File "/opt/conda/envs/arrow/lib/python3.12/site-packages/selenium/webdriver/remote/webdriver.py", line 426, in execute_async_script
    return self.execute(command, {"script": script, "args": converted_args})["value"]
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/envs/arrow/lib/python3.12/site-packages/selenium/webdriver/remote/webdriver.py", line 347, in execute
    self.error_handler.check_response(response)
  File "/opt/conda/envs/arrow/lib/python3.12/site-packages/selenium/webdriver/remote/errorhandler.py", line 229, in check_response
    raise exception_class(message, screen, stacktrace)
selenium.common.exceptions.TimeoutException: Message: script timeout
  (Session info: chrome=134.0.6998.165)
Stacktrace: