Merge pull request #2452 from langbot-app/ci/immutable-core-images

ci: publish immutable Core image tags

.github/workflows/build-dev-image.yaml

@@ -7,23 +7,42 @@ on:
 jobs:
   build-dev-image:
     runs-on: ubuntu-latest
-    # 如果是tag则跳过
     if: ${{ !startsWith(github.ref, 'refs/tags/') }}
+    permissions:
+      contents: read
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
         with:
           persist-credentials: false
 
-      - name: Generate Tag
-        id: generate_tag
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Generate image metadata
+        id: image
+        shell: bash
         run: |
-          # 获取分支名称，把/替换为-
-          echo ${{ github.ref }} | sed 's/refs\/heads\///g' | sed 's/\//-/g'
-          echo ::set-output name=tag::$(echo ${{ github.ref }} | sed 's/refs\/heads\///g' | sed 's/\//-/g')
-      - name: Login to Registry
-        run: docker login --username=${{ secrets.DOCKER_USERNAME }} --password ${{ secrets.DOCKER_PASSWORD }}
-      - name: Build Docker Image
-        run: |
-          docker buildx create --name mybuilder --use
-          docker build -t rockchin/langbot:${{ steps.generate_tag.outputs.tag }} . --push
+          set -euo pipefail
+          branch_tag="${GITHUB_REF#refs/heads/}"
+          branch_tag="${branch_tag//\//-}"
+          echo "branch_tag=${branch_tag}" >> "$GITHUB_OUTPUT"
+          echo "sha_tag=sha-${GITHUB_SHA}" >> "$GITHUB_OUTPUT"
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_PASSWORD }}
+
+      - name: Build and push immutable Core image
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          push: true
+          tags: |
+            rockchin/langbot:${{ steps.image.outputs.branch_tag }}
+            rockchin/langbot:${{ steps.image.outputs.sha_tag }}
+          labels: |
+            org.opencontainers.image.revision=${{ github.sha }}
+            org.opencontainers.image.source=${{ github.server_url }}/${{ github.repository }}

src/langbot/pkg/core/app.py

@@ -301,11 +301,36 @@ class Application:
     async def initialize(self):
         pass
 
+    async def _initialize_plugin_runtime(self) -> None:
+        try:
+            await self.plugin_connector.initialize()
+        except asyncio.CancelledError:
+            raise
+        except Exception as exc:
+            self.logger.warning(f'Plugin runtime unavailable during startup; reconnecting in background: {exc}')
+            self.plugin_connector.schedule_reconnect()
+
+    def _start_plugin_runtime_initialization(self) -> asyncio.Task | None:
+        task = getattr(self, '_plugin_runtime_initialization_task', None)
+        if task is not None and not task.done():
+            return task
+        # This is application lifecycle work, not a request side effect. It must
+        # not wait on PersistenceManager's after-commit gate at boot.
+        task = asyncio.create_task(
+            self._initialize_plugin_runtime(),
+            name='plugin-runtime-initialization',
+        )
+        self._plugin_runtime_initialization_task = task
+        return task
+
     async def run(self):
         self.event_loop_monitor.start()
         try:
-            if self.directory_projection_service is not None:
-                self.task_mgr.create_task(
+            if (
+                self.directory_projection_service is not None
+                and getattr(self, 'directory_projection_task', None) is None
+            ):
+                self.directory_projection_task = self.task_mgr.create_task(
                     self.directory_projection_service.run(),
                     name='cloud-directory-projection',
                     scopes=[core_entities.LifecycleControlScope.APPLICATION],
@@ -322,7 +347,6 @@ class Application:
                     name='cloud-manifest-refresh',
                     scopes=[core_entities.LifecycleControlScope.APPLICATION],
                 )
-            await self.plugin_connector.initialize_plugins()
 
             # 后续可能会允许动态重启其他任务
             # 故为了防止程序在非 Ctrl-C 情况下退出，这里创建一个不会结束的协程
@@ -348,6 +372,7 @@ class Application:
                 name='http-api-controller',
                 scopes=[core_entities.LifecycleControlScope.APPLICATION],
             )
+            self._start_plugin_runtime_initialization()
 
             # Telemetry instance heartbeat (startup + daily); respects
             # space.disable_telemetry via TelemetryManager.send().
@@ -529,6 +554,11 @@ class Application:
 
             if self.task_mgr is not None:
                 self.task_mgr.cancel_by_scope(core_entities.LifecycleControlScope.APPLICATION)
+            plugin_runtime_task = getattr(self, '_plugin_runtime_initialization_task', None)
+            if plugin_runtime_task is not None and not plugin_runtime_task.done():
+                plugin_runtime_task.cancel()
+                with contextlib.suppress(asyncio.CancelledError):
+                    await plugin_runtime_task
             with contextlib.suppress(Exception):
                 await self.event_loop_monitor.stop()
             mcp_mount = getattr(self.http_ctrl, 'mcp_mount', None)

src/langbot/pkg/core/stages/build_app.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from .. import stage, app
+from .. import stage, app, entities as core_entities
 from ...utils import version, proxy, constants
 from ...pipeline import pool, controller, pipelinemgr
 from ...pipeline import aggregator as message_aggregator
@@ -292,14 +292,17 @@ class BuildAppStage(stage.BootingStage):
         async def runtime_disconnect_callback(connector: plugin_connector.PluginRuntimeConnector) -> None:
             connector.schedule_reconnect()
 
+        if ap.directory_projection_service is not None:
+            # Keep the projection fresh while shared Runtime cold restore runs.
+            # BuildApp initializes the connector before Application.run() starts
+            # its long-lived tasks, so start the single refresh task here.
+            ap.directory_projection_task = ap.task_mgr.create_task(
+                ap.directory_projection_service.run(),
+                name='cloud-directory-projection',
+                scopes=[core_entities.LifecycleControlScope.APPLICATION],
+            )
+
         plugin_connector_inst = plugin_connector.PluginRuntimeConnector(ap, runtime_disconnect_callback)
-        try:
-            await plugin_connector_inst.initialize()
-        except Exception as exc:
-            # Keep the API/UI available while an external or managed runtime is
-            # starting, then recover in the background with bounded backoff.
-            ap.logger.warning(f'Plugin runtime unavailable during startup; reconnecting in background: {exc}')
-            plugin_connector_inst.schedule_reconnect()
         ap.plugin_connector = plugin_connector_inst
         workspace_service_inst.release_startup_execution_bindings()
 

src/langbot/pkg/plugin/connector.py

@@ -701,7 +701,13 @@ class PluginRuntimeConnector(ManagedRuntimeConnector):
             }
             self._known_desired_states.update({state.binding.installation_uuid: state for state in desired_states})
 
-        result = await runtime_handler.reconcile_plugin_installations(tuple(self._known_desired_states.values()))
+        reconcile_timeout_seconds = max(
+            300.0, self._runtime_connect_timeout(self.ap.instance_config.data.get('plugin', {}))
+        )
+        result = await runtime_handler.reconcile_plugin_installations(
+            tuple(self._known_desired_states.values()),
+            timeout=reconcile_timeout_seconds,
+        )
         await self._repair_reconcile_missing_artifacts(self._known_desired_states, result)
         self._record_reconcile_failures(self._known_desired_states, result)
 
@@ -736,7 +742,13 @@ class PluginRuntimeConnector(ManagedRuntimeConnector):
                     if state.binding.installation_uuid in all_states:
                         raise ValueError('Duplicate plugin installation UUID across projected Workspaces')
                     all_states[state.binding.installation_uuid] = state
-            result = await runtime_handler.reconcile_plugin_installations(tuple(all_states.values()))
+            reconcile_timeout_seconds = max(
+                300.0, self._runtime_connect_timeout(self.ap.instance_config.data.get('plugin', {}))
+            )
+            result = await runtime_handler.reconcile_plugin_installations(
+                tuple(all_states.values()),
+                timeout=reconcile_timeout_seconds,
+            )
             await self._repair_reconcile_missing_artifacts(all_states, result)
             self._record_reconcile_failures(all_states, result)
             for installation_uuid, previous in tuple(self._known_desired_states.items()):

src/langbot/pkg/plugin/handler.py

@@ -1677,13 +1677,15 @@ class RuntimeConnectionHandler(handler.Handler):
     async def reconcile_plugin_installations(
         self,
         installations: tuple[PluginInstallationDesiredState, ...],
+        *,
+        timeout: float = 300,
     ) -> dict[str, Any]:
         request = ReconcilePluginInstallationsRequest(installations=installations)
         with self.installation_scope(None):
             return await self.call_action(
                 LangBotToRuntimeAction.RECONCILE_PLUGIN_INSTALLATIONS,
                 request.model_dump(),
-                timeout=300,
+                timeout=timeout,
             )
 
     async def apply_plugin_installation(

tests/unit_tests/core/test_app_shutdown.py

@@ -144,3 +144,39 @@ async def test_runtime_resource_stats_are_aggregate_and_constant_time() -> None:
     assert stats['models']['providers'] == 1
     assert stats['runtimes']['plugin_installations'] == 1
     assert stats['runtimes']['plugin_runtime_connected'] is True
+
+
+@pytest.mark.asyncio
+async def test_start_plugin_runtime_initialization_bypasses_after_commit_gate() -> None:
+    app = Application()
+    app.plugin_connector = SimpleNamespace(initialize=AsyncMock())
+    app.task_mgr = SimpleNamespace(create_task=AsyncMock())
+
+    task = app._start_plugin_runtime_initialization()
+    await task
+
+    app.plugin_connector.initialize.assert_awaited_once_with()
+    app.task_mgr.create_task.assert_not_called()
+
+
+@pytest.mark.asyncio
+async def test_shutdown_cancels_plugin_runtime_initialization_task() -> None:
+    app = Application()
+    app._plugin_runtime_initialization_task = asyncio.create_task(asyncio.sleep(60))
+    app.task_mgr = SimpleNamespace(cancel_by_scope=lambda *_: None, tasks=[])
+    app.event_loop_monitor = SimpleNamespace(stop=AsyncMock())
+    app.http_ctrl = SimpleNamespace(mcp_mount=None)
+    app.platform_mgr = None
+    app.tool_mgr = None
+    app.model_mgr = None
+    app.box_service = None
+    app.plugin_connector = None
+    app.telemetry = None
+    app.vector_db_mgr = None
+    app.storage_mgr = None
+    app.persistence_mgr = SimpleNamespace(db=SimpleNamespace(engine=SimpleNamespace(dispose=AsyncMock())))
+    app.deployment = None
+
+    await app.shutdown()
+
+    assert app._plugin_runtime_initialization_task.cancelled()

tests/unit_tests/plugin/test_connector_reconcile.py

@@ -107,6 +107,19 @@ def shared_connector(
     return connector
 
 
+@pytest.mark.asyncio
+async def test_shared_reconcile_uses_configured_cold_start_timeout():
+    binding = execution_binding("workspace-a")
+    setting = plugin_setting("01", "a" * 64)
+    connector = shared_connector([[binding]], {"workspace-a": [setting]})
+    connector.ap.instance_config.data["plugin"]["connect_timeout_seconds"] = 900
+    connector.handler = runtime_handler()
+
+    await connector._prepare_connected_runtime()
+
+    assert connector.handler.reconcile_plugin_installations.await_args.kwargs["timeout"] == 900
+
+
 @pytest.mark.asyncio
 async def test_shared_reconnect_replays_two_workspaces_and_removes_missing_projection():
     binding_a = execution_binding('workspace-a')
@@ -150,7 +163,7 @@ async def test_empty_projected_workspaces_do_not_retain_installation_sets():
 
     assert connector._workspace_installations == {}
     assert connector._known_desired_states == {}
-    connector.handler.reconcile_plugin_installations.assert_awaited_once_with(())
+    connector.handler.reconcile_plugin_installations.assert_awaited_once_with((), timeout=300.0)
 
 
 @pytest.mark.asyncio

tests/unit_tests/plugin/test_handler.py

@@ -81,6 +81,18 @@ async def test_reconcile_plugin_installations_allows_cloud_cold_start_to_finish(
     assert runtime_handler.call_action.await_args.kwargs['timeout'] == 300
 
 
+@pytest.mark.asyncio
+async def test_reconcile_plugin_installations_accepts_configured_cold_start_timeout():
+    runtime_handler = make_handler(SimpleNamespace())
+    runtime_handler.call_action = AsyncMock(return_value={})
+    binding = next(iter(runtime_handler._installation_bindings.values()))[0]
+    desired = PluginInstallationDesiredState(binding=binding, enabled=True)
+
+    await runtime_handler.reconcile_plugin_installations((desired,), timeout=900)
+
+    assert runtime_handler.call_action.await_args.kwargs["timeout"] == 900
+
+
 class TestHandlerQueryVariables:
     """Tests for handler query variable logic."""