🐛 webserver now replies with unknown service state (ITISFoundation#…

…5230)

Co-authored-by: Andrei Neagu <[email protected]>

api/specs/web-server/_projects_nodes.py

@@ -6,12 +6,14 @@
 
 from _common import assert_handler_signature_against_model
 from fastapi import APIRouter, status
+from models_library.api_schemas_directorv2.dynamic_services import DynamicServiceGet
 from models_library.api_schemas_long_running_tasks.tasks import TaskGet
 from models_library.api_schemas_webserver.projects_nodes import (
     NodeCreate,
     NodeCreated,
     NodeGet,
     NodeGetIdle,
+    NodeGetUnknown,
     NodeRetrieve,
     NodeRetrieved,
     ServiceResourcesDict,
@@ -44,36 +46,35 @@
     response_model=Envelope[NodeCreated],
     status_code=status.HTTP_201_CREATED,
 )
-def create_node(project_id: str, body: NodeCreate):
+def create_node(project_id: str, body: NodeCreate):  # noqa: ARG001
     ...
 
 
 @router.get(
+    # issues with this endpoint https://github.com/ITISFoundation/osparc-simcore/issues/5245
     "/projects/{project_id}/nodes/{node_id}",
-    response_model=Envelope[NodeGet | NodeGetIdle],
-    # responses={"idle": {"model": NodeGetIdle}}, TODO: check this variant
+    response_model=Envelope[NodeGetIdle | NodeGetUnknown | DynamicServiceGet | NodeGet],
 )
-def get_node(
-    project_id: str,
-    node_id: str,
-):
-    pass
+def get_node(project_id: str, node_id: str):  # noqa: ARG001
+    ...
 
 
 @router.delete(
     "/projects/{project_id}/nodes/{node_id}",
     response_model=None,
     status_code=status.HTTP_204_NO_CONTENT,
 )
-def delete_node(project_id: str, node_id: str):
-    pass
+def delete_node(project_id: str, node_id: str):  # noqa: ARG001
+    ...
 
 
 @router.post(
     "/projects/{project_id}/nodes/{node_id}:retrieve",
     response_model=Envelope[NodeRetrieved],
 )
-def retrieve_node(project_id: str, node_id: str, _retrieve: NodeRetrieve):
+def retrieve_node(
+    project_id: str, node_id: str, _retrieve: NodeRetrieve  # noqa: ARG001
+):
     ...
 
 
@@ -82,15 +83,15 @@ def retrieve_node(project_id: str, node_id: str, _retrieve: NodeRetrieve):
     status_code=status.HTTP_204_NO_CONTENT,
     response_model=None,
 )
-def start_node(project_id: str, node_id: str):
+def start_node(project_id: str, node_id: str):  # noqa: ARG001
     ...
 
 
 @router.post(
     "/projects/{project_id}/nodes/{node_id}:stop",
     response_model=Envelope[TaskGet],
 )
-def stop_node(project_id: str, node_id: str):
+def stop_node(project_id: str, node_id: str):  # noqa: ARG001
     ...
 
 
@@ -99,7 +100,7 @@ def stop_node(project_id: str, node_id: str):
     response_model=None,
     status_code=status.HTTP_204_NO_CONTENT,
 )
-def restart_node(project_id: str, node_id: str):
+def restart_node(project_id: str, node_id: str):  # noqa: ARG001
     """Note that it has only effect on nodes associated to dynamic services"""
 
 
@@ -112,16 +113,18 @@ def restart_node(project_id: str, node_id: str):
     "/projects/{project_id}/nodes/{node_id}/resources",
     response_model=Envelope[ServiceResourcesDict],
 )
-def get_node_resources(project_id: str, node_id: str):
-    pass
+def get_node_resources(project_id: str, node_id: str):  # noqa: ARG001
+    ...
 
 
 @router.put(
     "/projects/{project_id}/nodes/{node_id}/resources",
     response_model=Envelope[ServiceResourcesDict],
 )
-def replace_node_resources(project_id: str, node_id: str, _new: ServiceResourcesDict):
-    pass
+def replace_node_resources(
+    project_id: str, node_id: str, _new: ServiceResourcesDict  # noqa: ARG001
+):
+    ...
 
 
 #
@@ -134,7 +137,9 @@ def replace_node_resources(project_id: str, node_id: str, _new: ServiceResources
     response_model=Envelope[_ProjectGroupAccess],
     summary="Check whether provided group has access to the project services",
 )
-async def get_project_services_access_for_gid(project_id: ProjectID, for_gid: GroupID):
+async def get_project_services_access_for_gid(
+    project_id: ProjectID, for_gid: GroupID  # noqa: ARG001
+):
     ...
 
 
@@ -153,7 +158,7 @@ async def get_project_services_access_for_gid(project_id: ProjectID, for_gid: Gr
     response_model=Envelope[list[_ProjectNodePreview]],
     summary="Lists all previews in the node's project",
 )
-async def list_project_nodes_previews(project_id: ProjectID):
+async def list_project_nodes_previews(project_id: ProjectID):  # noqa: ARG001
     ...
 
 
@@ -166,7 +171,9 @@ async def list_project_nodes_previews(project_id: ProjectID):
     summary="Gets a give node's preview",
     responses={status.HTTP_404_NOT_FOUND: {"description": "Node has no preview"}},
 )
-async def get_project_node_preview(project_id: ProjectID, node_id: NodeID):
+async def get_project_node_preview(
+    project_id: ProjectID, node_id: NodeID  # noqa: ARG001
+):
     ...
 
 

packages/models-library/src/models_library/api_schemas_webserver/projects_nodes.py

@@ -91,6 +91,10 @@ class NodeGetIdle(OutputSchema):
     service_state: Literal["idle"]
     service_uuid: NodeID
 
+    @classmethod
+    def from_node_id(cls, node_id: NodeID) -> "NodeGetIdle":
+        return cls(service_state="idle", service_uuid=node_id)
+
     class Config:
         schema_extra: ClassVar[dict[str, Any]] = {
             "example": {
@@ -100,6 +104,23 @@ class Config:
         }
 
 
+class NodeGetUnknown(OutputSchema):
+    service_state: Literal["unknown"]
+    service_uuid: NodeID
+
+    @classmethod
+    def from_node_id(cls, node_id: NodeID) -> "NodeGetUnknown":
+        return cls(service_state="unknown", service_uuid=node_id)
+
+    class Config:
+        schema_extra: ClassVar[dict[str, Any]] = {
+            "example": {
+                "service_uuid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
+                "service_state": "unknown",
+            }
+        }
+
+
 class NodeRetrieve(InputSchemaWithoutCameCase):
     port_keys: list[ServicePortKey] = []
 

packages/service-library/src/servicelib/rabbitmq/_errors.py

@@ -23,4 +23,7 @@ class RemoteMethodNotRegisteredError(BaseRPCError):
 
 
 class RPCServerError(BaseRPCError):
-    msg_template = "Raised '{exc_type}' while running '{method_name}' method: {msg}"
+    msg_template = (
+        "While running method '{method_name}' raised "
+        "'{exc_type}': '{exc_message}'\n{traceback}"
+    )

packages/service-library/src/servicelib/rabbitmq/_rpc_router.py

@@ -72,7 +72,8 @@ async def _wrapper(*args, **kwargs):
                         raise RPCServerError(
                             method_name=func.__name__,
                             exc_type=f"{exc.__class__.__module__}.{exc.__class__.__name__}",
-                            msg=f"{formatted_traceback}",
+                            exc_message=f"{exc}",
+                            traceback=f"{formatted_traceback}",
                         ) from None
 
             self.routes[RPCMethodName(func.__name__)] = _wrapper

packages/service-library/tests/rabbitmq/test_rabbitmq_rpc_router.py

@@ -96,6 +96,8 @@ async def test_exposed_methods(
             router_namespace,
             RPCMethodName(raising_unexpected_error.__name__),
         )
+    assert "This is not good!" in f"{exc_info.value}"
+    assert "builtins.ValueError" in f"{exc_info.value}"
 
     # This error was classified int he interface
     with pytest.raises(MyBaseError) as exc_info:

services/dynamic-scheduler/src/simcore_service_dynamic_scheduler/services/director_v2/_public_client.py

@@ -51,7 +51,7 @@ async def get_status(
                 e.response.status_code  # pylint:disable=no-member # type: ignore
                 == status.HTTP_404_NOT_FOUND
             ):
-                return NodeGetIdle(service_state="idle", service_uuid=node_id)
+                return NodeGetIdle.from_node_id(node_id)
             raise
 
     async def run_dynamic_service(

services/dynamic-scheduler/tests/unit/api_rpc/test_api_rpc__services.py

@@ -160,7 +160,7 @@ async def test_get_state(
 
     # node not tracked any of the two directors
     result = await services.get_service_status(rpc_client, node_id=node_not_found)
-    assert result == NodeGetIdle(service_state="idle", service_uuid=node_not_found)
+    assert result == NodeGetIdle.from_node_id(node_not_found)
 
 
 @pytest.fixture
@@ -401,7 +401,9 @@ async def test_stop_dynamic_service_serializes_generic_errors(
     simcore_user_agent: str,
     save_state: bool,
 ):
-    with pytest.raises(RPCServerError, match="while running 'stop_dynamic_service'"):
+    with pytest.raises(
+        RPCServerError, match="While running method 'stop_dynamic_service'"
+    ):
         await services.stop_dynamic_service(
             rpc_client,
             node_id=node_id,

services/payments/tests/unit/test_rpc_payments.py

@@ -87,7 +87,8 @@ async def test_rpc_init_payment_fail(
     assert isinstance(error, RPCServerError)
     assert error.exc_type == "httpx.ConnectError"
     assert error.method_name == "init_payment"
-    assert error.msg
+    assert error.exc_message
+    assert error.traceback
 
 
 async def test_webserver_one_time_payment_workflow(

services/static-webserver/client/source/class/osparc/data/model/Node.js

@@ -1290,6 +1290,7 @@ qx.Class.define("osparc.data.model.Node", {
           break;
         }
         case "stopping":
+        case "unknown":
         case "starting":
         case "pulling": {
           status.setInteractive(serviceState);

services/static-webserver/client/source/class/osparc/viewer/NodeViewer.js

@@ -127,6 +127,7 @@ qx.Class.define("osparc.viewer.NodeViewer", {
           qx.event.Timer.once(() => this.__nodeState(), this, interval);
           break;
         }
+        case "unknown":
         case "starting":
         case "connecting":
         case "pulling": {