langflow-ai/langflow · error · HTTPException

Name conflict — another role already uses this name

Error message

Name conflict — another role already uses this name

What it means

Raised by PATCH /api/v1/authz/roles/{role_id} when session.commit() raises an IntegrityError. The route rolls back and maps any integrity failure to a 409 with this message; in practice it means the new name collides with another role's unique name constraint (a parent_role_id cycle or other constraint violation would surface here too, since the handler is not constraint-specific).

Source

Thrown at src/backend/base/langflow/api/v1/authz_roles.py:235

    if "permissions" in fields_set:
        # permissions column is nullable=False (default_factory=list). An empty
        # list is the natural "clear" — None would violate the constraint at
        # commit, so reject it up front.
        if payload.permissions is None:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail="permissions cannot be null; pass an empty list to clear",
            )
        role.permissions = list(payload.permissions)

    role.updated_at = datetime.now(timezone.utc)

    try:
        await session.commit()
    except IntegrityError as exc:
        await session.rollback()
        raise HTTPException(
            status_code=status.HTTP_409_CONFLICT,
            detail="Name conflict — another role already uses this name",
        ) from exc
    await session.refresh(role)
    await safe_invalidate_role(get_authorization_service(), role.id, op="role:update")
    await audit_decision(
        user_id=current_user.id,
        action="role:update",
        obj=f"role:{role.id}",
        result="allow",
        details={
            "role_name": role.name,
            "fields_changed": sorted(fields_set),
        },
    )
    logger.info("Updated role %s (id=%s)", role.name, role.id)
    return RoleRead.model_validate(role)

View on GitHub (pinned to 976ec789d2)

Solutions

1. Pick a different, unique name for the role
2. GET /api/v1/authz/roles first and check the desired name is not taken (note: still racy under concurrency)
3. Treat 409 as retryable with a new name — catch it in the client and prompt the user for a different name
4. If you were not renaming, inspect the server log for the underlying IntegrityError (parent_role_id cycle, etc.)

Example fix

// before
await updateRole(id, { name: 'admin' }); // another role already called 'admin'

// after
const roles = await listRoles();
const name = roles.some(r => r.name === 'admin') ? 'admin-2' : 'admin';
await updateRole(id, { name });

async function isRoleNameFree(name: string) {
  const roles = await (await fetch('/api/v1/authz/roles')).json();
  return !roles.some(r => r.name === name);
}

try {
  await patchRole(id, body);
} catch (e) {
  if (e instanceof HttpException && e.status === 409) {
    // prompt user for a new name; do not blind-retry the same name
  } else throw e;
}

When it happens

Trigger: PATCH /authz/roles/{id} renaming a role to a name already used by another role; creating a parent_role_id cycle that the DB rejects at commit.

Common situations: Case-sensitive vs case-insensitive unique indexes surprising callers ("Admin" vs "admin"); renaming during concurrent role administration where another request claimed the name between your GET and PATCH; seeding scripts that rename built-in roles to colliding names.

Related errors

• Role still has active assignments — revoke them before delet
• Share could not be created: it may already exist or conflict
• Role with name {payload.name!r} already exists
• name cannot be null
• permissions cannot be null; pass an empty list to clear

AI-assisted analysis of langflow-ai/langflow@976ec789d2 (2026-08-14). Data as JSON: /api/errors/2e7dc0e0e4979454. Report an issue: GitHub.