[BUG] Frame-tool loader omits a selected process-owner helper #35
Labels
No labels
alpha-1
bug
ci
docs
duplicate
enhancement
help wanted
invalid
performance
phase-6
question
refactor
security
wontfix
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
erix/vfsd#35
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Summary
Vfsd CI stops before frame analysis because the selected process owner imports
command_live_output, which the frame-tool loader does not load.Reproduction
Run
tests/frame_tool_authority_test.pyat Vfsdd1bb30e0with the workflow-selected Integration helpers at74d8e383.Expected and actual behavior
The eight authority controls should load their selected helpers. Six instead raise
ModuleNotFoundError: No module named 'command_live_output'; the two early refusal controls pass.Environment and identity
Vfsd CI 387, test job 48596, attempt 1. The import fails before disassembler execution.
Evidence and prior checks
The complete log identifies the missing import in
cargo_source_metadata.py:25. The loader preloads only two modules and omits this dependency. This is separate from the stack exhaustion tracked in #33.Tracking
0830021cloads the complete selected helper set through the existing bootstrap and restores the caller’s module cache and importer after success or failure. All 11 authority controls pass, including the original eight. The existing frame step in the new CI run remains pending.CI388 passes at
0830021c. The complete test log shows all 11 authority controls passing and both runtime and smoke frame checks completing at 135,440 bytes against the existing 245,760-byte budget. Both jobs pass without warnings. This closes the helper import regression; the separate guest stack issue in #33 remains open.