fix: v0.3.3 — delete_container params (error 114) + delete_project orphan guard

Bug 1 — delete_container (DSM error 114):
SYNO.Docker.Container/delete requires three parameters: name
(JSON-encoded), force=false, and preserve_profile=false. Previously
only a bare `name` string was sent, causing DSM to reject the call
with error 114. Added the two missing fields and JSON-encode name to
match the DSM convention. The connector-side running-container guard
is unchanged; force stays hard-coded to false.

Bug 2 — delete_project orphan containers:
Production test revealed that DSM does NOT reject Project/delete on a
running project — it silently removes the registration and leaves the
containers running without any project context. The previous
implementation tried to handle this via a caught SynologyError that
never actually fires. Fix: check the project status from _find_project
connector-side before issuing any DSM call; if RUNNING, return an
error pointing at stop_project. The delete request is never sent for
a running project.

The corresponding unit test (test_delete_project_running_returns_stop_hint)
was a false positive — it mocked a DSM rejection that real DSM never
produces. Replaced with test_delete_project_running_blocked_connector_side
which asserts that client.request("delete") is never called when the
project is RUNNING.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

CHANGELOG.md

@@ -2,7 +2,26 @@
 
 All notable changes to this project will be documented in this file.
 
-## [0.3.2] - 2026-05-18
+## [0.3.3] - 2026-05-18
+
+### Fixed
+
+- `delete_container`: `SYNO.Docker.Container/delete` requires three
+  parameters — `name` (JSON-encoded), `force=false`, and
+  `preserve_profile=false`. Previously only `name` was sent (without
+  JSON-encoding), causing DSM to reject the call with error 114.
+- `delete_project`: DSM does **not** reject `Project/delete` on a
+  running project — it silently removes the registration and leaves
+  the containers running as orphans. The connector now blocks the call
+  itself when the project status is `RUNNING`, before issuing any DSM
+  request, and tells the user to `stop_project` first. The previous
+  implementation relied on a DSM-level rejection that never occurs in
+  practice; the corresponding unit test was a false positive (mocked
+  an error that real DSM never returns) and has been replaced with a
+  test that asserts `Project/delete` is never called for a `RUNNING`
+  project.
+
+
 
 ### Added
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "mcp-synology-container"
-version = "0.3.2"
+version = "0.3.3"
 description = "MCP server for Synology Container Manager"
 requires-python = ">=3.12"
 dependencies = [

src/mcp_synology_container/modules/containers.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import json
 import logging
 import re
 from typing import TYPE_CHECKING, Any
@@ -319,7 +320,11 @@ def register_containers(mcp: FastMCP, config: AppConfig, client: DsmClient) -> N
             await client.request(
                 "SYNO.Docker.Container",
                 "delete",
-                params={"name": resolved_name},
+                params={
+                    "name": json.dumps(resolved_name),
+                    "force": "false",
+                    "preserve_profile": "false",
+                },
             )
             return f"Deleted container '{display_name}'."
         except Exception as e:

src/mcp_synology_container/modules/projects.py

@@ -403,6 +403,15 @@ def register_projects(mcp: FastMCP, config: AppConfig, client: DsmClient) -> Non
                 f"Call this tool again with confirmed=True to proceed."
             )
 
+        # Guard: DSM does NOT reject a delete call on a running project — it
+        # removes the registration silently and leaves the containers running
+        # as orphans. Reject here so we never put the NAS into that state.
+        if status == "RUNNING":
+            return (
+                f"Error: project '{project_name}' is RUNNING. "
+                f"Stop it first with stop_project, then delete_project."
+            )
+
         try:
             await client.request(
                 "SYNO.Docker.Project",
@@ -411,14 +420,6 @@ def register_projects(mcp: FastMCP, config: AppConfig, client: DsmClient) -> Non
                 params={"id": json.dumps(project_id)},
             )
         except SynologyError as e:
-            # DSM refuses to delete a running project. We deliberately do NOT
-            # auto-stop — that would be too destructive for a delete tool —
-            # but we tell the user how to proceed.
-            if status == "RUNNING":
-                return (
-                    f"Cannot delete project '{project_name}' while it is running ({e}).\n"
-                    f"Stop the project first with stop_project."
-                )
             return f"Error deleting project '{project_name}': {e}"
 
         return (

tests/test_modules/test_containers.py

@@ -317,6 +317,32 @@ async def test_delete_container_stopped_confirmed():
     assert "myapp_web" in result
 
 
+@pytest.mark.asyncio
+async def test_delete_container_sends_required_params():
+    """SYNO.Docker.Container/delete must include name (json-encoded), force=false,
+    and preserve_profile=false — without them DSM returns error 114."""
+    from mcp_synology_container.modules.containers import register_containers
+
+    client = AsyncMock()
+    client.request.return_value = {
+        "details": {"State": {"Running": False, "Status": "exited"}},
+        "profile": {"image": "nginx:latest"},
+    }
+
+    mcp, tools = make_mock_mcp()
+    register_containers(mcp, make_config(), client)
+
+    await tools["delete_container"]("myapp_web", confirmed=True)
+
+    # Find the delete call (second call — first is Container/get)
+    delete_calls = [c for c in client.request.call_args_list if c.args[1] == "delete"]
+    assert len(delete_calls) == 1
+    params = delete_calls[0].kwargs.get("params") or {}
+    assert params["name"] == '"myapp_web"'
+    assert params["force"] == "false"
+    assert params["preserve_profile"] == "false"
+
+
 @pytest.mark.asyncio
 async def test_container_stats_cpu_calculation():
     """CPU% is computed via the standard Docker formula."""

tests/test_modules/test_projects.py

@@ -898,20 +898,17 @@ async def test_delete_project_happy_path():
 
 
 @pytest.mark.asyncio
-async def test_delete_project_running_returns_stop_hint():
+async def test_delete_project_running_blocked_connector_side():
-    """DSM refusing to delete a running project produces a clean 'stop_project' hint
+    """Live test showed that DSM does NOT reject Project/delete on a running project —
-    rather than a raw error."""
+    it silently orphans the containers. The connector must therefore block the call
-    from mcp_synology_container.dsm_client import SynologyError
+    itself when the project is RUNNING, without ever calling client.request(delete)."""
-
+    client, calls = make_delete_project_client(project=SAMPLE_PROJECT_RUNNING)
-    client, calls = make_delete_project_client(
-        project=SAMPLE_PROJECT_RUNNING,
-        delete_raises=SynologyError("Project is running", code=2103),
-    )
     tools = make_projects_tools(client)
 
     result = await tools["delete_project"]("myapp", confirmed=True)
 
+    assert "RUNNING" in result
     assert "stop_project" in result
-    assert "running" in result.lower()
+    # The delete endpoint must NOT have been called — no orphaned containers.
-    # No "deleted" success line
+    delete_calls = [m for _, m, _ in calls if m == "delete"]
-    assert "deleted (registration removed)" not in result
+    assert delete_calls == []

uv.lock

@@ -362,7 +362,7 @@ wheels = [
 
 [[package]]
 name = "mcp-synology-container"
-version = "0.3.2"
+version = "0.3.3"
 source = { editable = "." }
 dependencies = [
     { name = "click" },