invoke-ai/InvokeAI · error

Old and new paths are identical for {move.image_name}

Error message

Old and new paths are identical for {move.image_name}

What it means

A move where old_path == new_path is a no-op and likely a naming/planning bug, so preflight_moves raises ValueError identifying the offending image_name. It only fires when the source exists (checked earlier) and the path isn't already taken by another file check ordering.

Source

Thrown at invokeai/app/services/image_moves/image_moves_default.py:437

                    str(move.old_thumbnail_path),
                    str(move.new_thumbnail_path),
                    message,
                ),
            )
            return job_id

    def preflight_moves(self, moves: Sequence[PlannedImageMove]) -> None:
        destinations: set[Path] = set()
        thumbnail_destinations: set[Path] = set()
        for move in moves:
            if not move.old_path.exists():
                if not move.is_intermediate:
                    raise FileNotFoundError(f"Source image does not exist: {move.old_path}")
                continue
            if move.new_path.exists():
                raise FileExistsError(f"Destination image already exists: {move.new_path}")
            if move.old_path == move.new_path:
                raise ValueError(f"Old and new paths are identical for {move.image_name}")
            if move.new_path in destinations:
                raise ValueError(f"Duplicate destination path: {move.new_path}")
            destinations.add(move.new_path)
            if move.new_thumbnail_path in thumbnail_destinations:
                raise ValueError(f"Duplicate destination thumbnail path: {move.new_thumbnail_path}")
            thumbnail_destinations.add(move.new_thumbnail_path)
            if self._has_active_job_for_image(move.image_name):
                raise ValueError(f"Image {move.image_name} already has an active image move job")
            self._assert_same_filesystem(move.old_path, move.new_path)
            if move.old_thumbnail_path.exists():
                if move.new_thumbnail_path.exists():
                    raise FileExistsError(f"Destination thumbnail already exists: {move.new_thumbnail_path}")
                self._assert_same_filesystem(move.old_thumbnail_path, move.new_thumbnail_path)

    def _record_missing_source_errors(self, moves: Sequence[PlannedImageMove]) -> tuple[list[PlannedImageMove], int]:
        remaining_moves: list[PlannedImageMove] = []
        errors = 0
        for move in moves:

View on GitHub (pinned to 0b6a024f2f)

Solutions

  1. In the caller, skip or filter moves whose resolved old_path equals new_path (treat as success/no-op)
  2. Fix destination path computation so the target differs from the source
  3. Normalize both paths before planning and drop identity moves

Example fix

// before
service.preflight_moves(moves)
// after
moves = [m for m in moves if m.old_path != m.new_path]
service.preflight_moves(moves)
Defensive patterns

Strategy: validation

Validate before calling

identical = [m for m in moves if m.old_path == m.new_path]
if identical:
    raise ValueError(f"identity moves: {[m.image_name for m in identical]}")
service.preflight_moves(moves)

Type guard

def is_real_move(move: PlannedImageMove) -> bool:
    return move.old_path != move.new_path

Try / catch

try:
    service.preflight_moves(moves)
except ValueError as e:
    if "identical" in str(e):
        moves = [m for m in moves if m.old_path != m.new_path]
        service.preflight_moves(moves)
    else:
        raise

Prevention

When it happens

Trigger: Computing a destination subfolder that resolves to the image's current location (e.g. moving to the same folder with the same name); passing identical old/new paths by mistake when constructing PlannedImageMove; path normalization differences resolved to the same path.

Common situations: UI allowing the user to select the image's current folder as the target; scripts that build target paths with empty suffix components; symlinks or relative paths that resolve identically.

Related errors

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/4ad47ddaa3ab78f2. Report an issue: GitHub.