test(cli): skip install.sh tests on Windows runner (#1099) by mayankbharati-ops · Pull Request #1101 · Tracer-Cloud/opensre

mayankbharati-ops · 2026-04-30T06:56:09Z

Closes #1099.

Root cause

The post-merge test (windows-latest) job on 67dc64d1 (the merge of #1090) is failing with two distinct symptoms in the install.sh test suite added by #1064:

1. `tests/cli/test_install_sh_path.py` — 12 failures

The helper _run calls subprocess.run(["bash", "-c", script]). On the GitHub Actions windows-latest runner, bash resolves to wsl.exe, and that runner has no installed WSL distribution. Every _run exits with code 1 and the literal message:

Windows Subsystem for Linux has no installed distributions.
You can resolve this by installing a distribution with the instructions below

So none of the asserted rc files (.zshrc, .bashrc, config.fish) ever get written, and assertions like assert zshrc.exists() and assert result.returncode == 0 fail.

2. `tests/cli/test_install_sh_resolution.py` — module collection error

@pytest.mark.skipif(os.geteuid() == 0, reason="sudo fallback is only meaningful for non-root users")

os.geteuid does not exist on Windows. The decorator condition is evaluated at decorator-application time (i.e. module import), so AttributeError: module 'os' has no attribute 'geteuid' fires before pytest gets a chance to read any markers. The whole module fails to collect.

Why skip rather than fix the test scaffolding

Both files exclusively test install.sh, which is a POSIX shell script driving zsh/bash/fish rc-file behaviour. There's no Windows analogue (the project ships a separate Windows installer / pip install path), so installing WSL or porting the test scaffolding to PowerShell would be busywork. Skipping the module on Windows matches what the codebase actually claims to support.

Changes

# test_install_sh_path.py
pytestmark = pytest.mark.skipif(
    sys.platform == "win32",
    reason="install.sh is POSIX-only; the Windows runner has no usable bash "
           "(resolves to unconfigured WSL), so this module's subprocess-driven "
           "tests cannot run there. See issue #1099.",
)

test_install_sh_resolution.py gets the same pytestmark, plus a defensive replacement of the bare os.geteuid() access:

# Evaluated once at module load. hasattr short-circuits the `and` on
# Windows, so os.geteuid() is never called there — the module imports
# cleanly even before pytest reads ``pytestmark``.
_RUNNING_AS_ROOT = hasattr(os, "geteuid") and os.geteuid() == 0

@pytest.mark.skipif(_RUNNING_AS_ROOT, reason="sudo fallback is only meaningful for non-root users")

Without that defensive guard the module still wouldn't import on Windows, even with pytestmark defined — and pytestmark only kicks in once pytest has collected the module.

Verification

	macOS / Linux	Windows
Module imports	✅ unchanged	✅ no more AttributeError
Tests run	✅ all 15 pass on local macOS	Skipped cleanly with reason text
Existing root-skip	✅ still active on Unix when running as root	n/a (whole module skipped)
`ruff check` / `ruff format --check`	✅ clean	✅ clean
Full unit suite	4164 / 2 skipped / 1 xfailed	(will skip 13 from these modules + still fail count = 0)

Scope

File	Change
`tests/cli/test_install_sh_path.py`	+13 / -0 (module-level pytestmark + comment block)
`tests/cli/test_install_sh_resolution.py`	+28 / -1 (pytestmark + defensive `_RUNNING_AS_ROOT` constant; existing decorator references the new constant)

No production code touched. No test logic changed for the platforms where these tests do run.

cc @muddlebee — this is the post-#1090 Windows CI fix. Both files were added by #1064 and have been failing on Windows since that landed; this PR makes the Windows runner green again without changing any Unix-side behaviour.

…racer-Cloud#1099) The post-merge ``test (windows-latest)`` job on commit 67dc64d (PR Tracer-Cloud#1090 landing) is failing with two distinct symptoms in the install.sh test suites added by PR Tracer-Cloud#1064: 1. ``tests/cli/test_install_sh_path.py`` — 12 tests fail because the subprocess call ``subprocess.run(["bash", "-c", script])`` resolves ``bash`` to ``wsl.exe`` on the GitHub Actions ``windows-latest`` runner, and that runner has no installed WSL distribution. Every ``_run`` call exits 1 with a literal "Windows Subsystem for Linux has no installed distributions" message, so none of the asserted rc files (``.zshrc``/``.bashrc``/``config.fish``) ever get written. 2. ``tests/cli/test_install_sh_resolution.py`` — module fails to import on Windows: the existing ``@pytest.mark.skipif(os.geteuid() == 0, reason=...)`` decorator evaluates ``os.geteuid()`` at decorator-application time, but ``os.geteuid`` does not exist on Windows. ``AttributeError`` fires *before* pytest can read any markers, blocking collection of every test in the module. Both files exclusively test ``install.sh``, which is a POSIX shell script (zsh/bash/fish RC files) with no Windows analogue, so the correct fix for the Windows runner is to skip them. Changes: - ``test_install_sh_path.py``: module-level ``pytestmark = pytest.mark.skipif(sys.platform == "win32", ...)`` with a comment that names the failure mode. - ``test_install_sh_resolution.py``: same module-level skipif, *and* defensively replace ``os.geteuid() == 0`` with ``_RUNNING_AS_ROOT = hasattr(os, "geteuid") and os.geteuid() == 0`` evaluated once at module load. ``hasattr`` short-circuits the ``and`` on Windows so ``os.geteuid()`` is never called there — the module imports cleanly even before pytest reads ``pytestmark``. No source-checkout behaviour change: skipif is a no-op on macOS/Linux, and the existing root-only test still skips correctly on those platforms (verified by running ``pytest tests/cli/test_install_sh_path.py tests/cli/test_install_sh_resolution.py`` locally — 15 / 15 pass). Full unit suite: 4164 pass / 2 skipped / 1 xfailed. Closes Tracer-Cloud#1099.

greptile-apps · 2026-04-30T06:57:27Z

Greptile Summary

Adds module-level pytestmark skips to both install.sh test files so the windows-latest CI runner no longer fails when bash resolves to an unconfigured WSL. test_install_sh_resolution.py also replaces the bare os.geteuid() == 0 decorator argument with a hasattr-guarded _RUNNING_AS_ROOT constant, which correctly prevents AttributeError at import time on Windows (before pytestmark can take effect). No production code or Unix-side test behaviour is changed.

Confidence Score: 5/5

Safe to merge — minimal, targeted fix with no production code changes and no Unix-side test regressions.

Both changes are correct: pytestmark is the idiomatic pytest mechanism for module-level skips, and the hasattr short-circuit is necessary and sufficient to prevent the import-time AttributeError. No logic is altered for macOS/Linux. No P0 or P1 findings.

No files require special attention.

Important Files Changed

Filename	Overview
tests/cli/test_install_sh_path.py	Adds `import sys`, `import pytest`, and a module-level `pytestmark` that skips all 12 tests on `win32`; no test logic changed.
tests/cli/test_install_sh_resolution.py	Adds module-level `pytestmark` Windows skip and replaces bare `os.geteuid() == 0` with a `hasattr`-guarded `_RUNNING_AS_ROOT` constant to prevent `AttributeError` at import time on Windows.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[pytest collects module] --> B{sys.platform == win32?}
    B -- Yes --> C[pytestmark: skip all tests\nin module with reason text]
    B -- No --> D[Evaluate _RUNNING_AS_ROOT\nhasattr guard]
    D --> E{os.geteuid == 0?}
    E -- Yes --> F[skip test_uses_sudo_for_non_writable_system_path_dir]
    E -- No --> G[Run all tests via\nsubprocess bash -c]
    G --> H[Assert rc-file writes\nand return codes]

_{Reviews (2): Last reviewed commit: "test(cli): skip install.sh tests on Wind..." | Re-trigger Greptile}

muddlebee · 2026-04-30T07:00:06Z

@mayankbharati-ops I hope only these two were the affected regions?

muddlebee · 2026-04-30T07:18:52Z

@mayankbharati-ops I just added a custom workflow check(already in main), after your PR. so maintainers can trigger window CI checks during PR only.. have triggered it here, pls see if any more windows issues..

looks like it passed #1102

github-actions · 2026-04-30T07:20:30Z

🛸 Aliens watching our repo just upgraded @mayankbharati-ops's threat level to: do not engage — too competent. 👽

👋 Join us on Discord - OpenSRE : hang out, contribute, or hunt for features and issues. Everyone's welcome.

muddlebee · 2026-04-30T07:20:31Z

@mayankbharati-ops thanks for the quick fix. window checks green at #1102

mayankbharati-ops · 2026-04-30T07:27:15Z

@muddlebee yes — those two were the only affected modules. I just re-audited the test tree to be sure, and PR #1102's now-green Windows checks corroborate it:

Audit results across tests/:

Pattern	Hits	Files
`subprocess.run(["bash"...])` / `["sh"...]` / `["zsh"...]` / `["fish"...]`	2	both fixed in #1101
`os.geteuid` / `os.getuid` / `os.fork` / `os.symlink` / `os.getlogin`	1	only `test_install_sh_resolution.py` (fixed in #1101)
`import pwd` / `import grp` / `import fcntl` / `import termios` / `import resource`	0	—
Hard-coded Unix paths (`/tmp/...`, `/usr/...`) actually opened from disk	0	the only `/tmp/...` references in `test_investigate_service_flag.py:86` and `test_health_view.py:75` are test-parameter strings the CLI rejects/serializes without touching the filesystem — safe on Windows

So the surface area for Windows-specific test failures was exactly those two files. The "windows test SUCCESS" on #1102 is the empirical confirmation.

Re: the pre-merge Windows trigger — that's a nice addition. With it, this whole regression class would have been caught at PR-review time on #1064 (where these tests landed) instead of post-merge. Glad it's in.

Thanks for the quick merge!

muddlebee · 2026-04-30T07:29:07Z

@mayankbharati-ops in case you are interested how we added the new ci windows workflows

base PR: #1100
follow up PR: #1103

…racer-Cloud#1099) (Tracer-Cloud#1101) The post-merge ``test (windows-latest)`` job on commit 67dc64d (PR Tracer-Cloud#1090 landing) is failing with two distinct symptoms in the install.sh test suites added by PR Tracer-Cloud#1064: 1. ``tests/cli/test_install_sh_path.py`` — 12 tests fail because the subprocess call ``subprocess.run(["bash", "-c", script])`` resolves ``bash`` to ``wsl.exe`` on the GitHub Actions ``windows-latest`` runner, and that runner has no installed WSL distribution. Every ``_run`` call exits 1 with a literal "Windows Subsystem for Linux has no installed distributions" message, so none of the asserted rc files (``.zshrc``/``.bashrc``/``config.fish``) ever get written. 2. ``tests/cli/test_install_sh_resolution.py`` — module fails to import on Windows: the existing ``@pytest.mark.skipif(os.geteuid() == 0, reason=...)`` decorator evaluates ``os.geteuid()`` at decorator-application time, but ``os.geteuid`` does not exist on Windows. ``AttributeError`` fires *before* pytest can read any markers, blocking collection of every test in the module. Both files exclusively test ``install.sh``, which is a POSIX shell script (zsh/bash/fish RC files) with no Windows analogue, so the correct fix for the Windows runner is to skip them. Changes: - ``test_install_sh_path.py``: module-level ``pytestmark = pytest.mark.skipif(sys.platform == "win32", ...)`` with a comment that names the failure mode. - ``test_install_sh_resolution.py``: same module-level skipif, *and* defensively replace ``os.geteuid() == 0`` with ``_RUNNING_AS_ROOT = hasattr(os, "geteuid") and os.geteuid() == 0`` evaluated once at module load. ``hasattr`` short-circuits the ``and`` on Windows so ``os.geteuid()`` is never called there — the module imports cleanly even before pytest reads ``pytestmark``. No source-checkout behaviour change: skipif is a no-op on macOS/Linux, and the existing root-only test still skips correctly on those platforms (verified by running ``pytest tests/cli/test_install_sh_path.py tests/cli/test_install_sh_resolution.py`` locally — 15 / 15 pass). Full unit suite: 4164 pass / 2 skipped / 1 xfailed. Closes Tracer-Cloud#1099.

mayankbharati-ops mentioned this pull request Apr 30, 2026

[BUG] : CI failing for opensre tests #1099

Closed

muddlebee added the ci:windows label Apr 30, 2026

muddlebee marked this pull request as draft April 30, 2026 07:04

muddlebee marked this pull request as ready for review April 30, 2026 07:04

muddlebee mentioned this pull request Apr 30, 2026

test(cli): cherry-pick #1101 to verify ci:windows label CI #1102

Closed

muddlebee merged commit d09b7c2 into Tracer-Cloud:main Apr 30, 2026
7 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

test(cli): skip install.sh tests on Windows runner (#1099)#1101

test(cli): skip install.sh tests on Windows runner (#1099)#1101
muddlebee merged 1 commit intoTracer-Cloud:mainfrom
mayankbharati-ops:fix/1099-skip-install-sh-tests-on-windows

mayankbharati-ops commented Apr 30, 2026

Uh oh!

greptile-apps Bot commented Apr 30, 2026 •

edited

Loading

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

Uh oh!

github-actions Bot commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

mayankbharati-ops commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

mayankbharati-ops commented Apr 30, 2026

Root cause

1. tests/cli/test_install_sh_path.py — 12 failures

2. tests/cli/test_install_sh_resolution.py — module collection error

Why skip rather than fix the test scaffolding

Changes

Verification

Scope

Uh oh!

greptile-apps Bot commented Apr 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 5/5

Important Files Changed

Flowchart

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

Uh oh!

github-actions Bot commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

mayankbharati-ops commented Apr 30, 2026

Uh oh!

muddlebee commented Apr 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

1. `tests/cli/test_install_sh_path.py` — 12 failures

2. `tests/cli/test_install_sh_resolution.py` — module collection error

greptile-apps Bot commented Apr 30, 2026 •

edited

Loading