merge(r2): r2-plan03

src/ZB.MOM.WW.ScadaBridge.SiteRuntime/Actors/AlarmActor.cs

@@ -186,6 +186,10 @@ public class AlarmActor : ReceiveActor
 
         // Handle the off-dispatcher trigger-expression evaluation result (P1).
         Receive<ExpressionEvalResult>(HandleExpressionEvalResult);
+
+        // Handle a faulted off-dispatcher evaluation task (N2) — clears in-flight
+        // instead of stranding the alarm trigger with an unhandled Status.Failure.
+        Receive<ExpressionEvalFailed>(HandleExpressionEvalFailed);
     }
 
     /// <inheritdoc />
@@ -551,7 +555,8 @@ public class AlarmActor : ReceiveActor
             }
         }, CancellationToken.None, TaskCreationOptions.DenyChildAttach,
             ScriptExecutionScheduler.Shared(_options)).Unwrap().PipeTo(self,
-            success: r => new ExpressionEvalResult(r));
+            success: r => new ExpressionEvalResult(r),
+            failure: ex => new ExpressionEvalFailed(ex));
     }
 
     /// <summary>
@@ -576,6 +581,22 @@ public class AlarmActor : ReceiveActor
         if (_evalPending) { _evalPending = false; StartExpressionEvaluation(); }
     }
 
+    /// <summary>
+    /// Applies a faulted off-dispatcher evaluation task (N2): logs the fault, then
+    /// delegates to the false result (per the class contract — a throwing expression
+    /// is treated as false so the state machine still runs, an Active alarm clears)
+    /// which clears in-flight and drains any pending evaluation. Without this the
+    /// alarm trigger would be permanently parked with _evalInFlight stuck true.
+    /// </summary>
+    private void HandleExpressionEvalFailed(ExpressionEvalFailed msg)
+    {
+        _healthCollector?.IncrementAlarmError();
+        _logger.LogError(msg.Cause,
+            "Alarm {Alarm} trigger-expression evaluation task faulted on {Instance}; treated as false",
+            _alarmName, _instanceName);
+        HandleExpressionEvalResult(new ExpressionEvalResult(false));
+    }
+
     /// <summary>
     /// HiLo level evaluator: returns the most-severe matching band for the
     /// given value. Severity order checked from highest to lowest so that a
@@ -795,6 +816,15 @@ public class AlarmActor : ReceiveActor
     /// actor thread.
     /// </summary>
     private sealed record ExpressionEvalResult(bool Result);
+
+    /// <summary>
+    /// Piped back to self when the off-dispatcher evaluation TASK itself faults
+    /// (e.g. ObjectDisposedException from a disposed ScriptExecutionScheduler
+    /// during shutdown) — the inner body's catch never sees that. Without this
+    /// mapping the actor receives an unhandled Status.Failure and _evalInFlight
+    /// stays true forever, permanently parking the expression trigger (N2).
+    /// </summary>
+    internal sealed record ExpressionEvalFailed(Exception Cause);
 }
 
 internal enum RateOfChangeDirection { Either, Rising, Falling }

src/ZB.MOM.WW.ScadaBridge.SiteRuntime/Actors/DeploymentManagerActor.cs

@@ -518,7 +518,11 @@ public class DeploymentManagerActor : ReceiveActor, IWithTimers
         // delete/disable commands, breaking that ordering. A deploy is an
         // infrequent admin command, so briefly holding the singleton for a pure
         // Roslyn compile is acceptable; the central deployer already Asks and
-        // waits for the DeploymentStatusResponse.
+        // waits for the DeploymentStatusResponse. Redeploys and multi-instance
+        // deploys of unchanged scripts hit the process-wide compile cache
+        // (SiteScriptCompileCache), so the synchronous gate recompiles only
+        // genuinely new code and the Instance Actor's PreStart reuses the gate's
+        // compile (N4).
         var errors = _deployCompileValidator.Validate(command.FlattenedConfigurationJson);
         if (errors.Count > 0)
         {

src/ZB.MOM.WW.ScadaBridge.SiteRuntime/Actors/ScriptActor.cs

@@ -159,6 +159,10 @@ public class ScriptActor : ReceiveActor, IWithTimers
 
         // Handle the off-dispatcher trigger-expression evaluation result (P1).
         Receive<ExpressionEvalResult>(HandleExpressionEvalResult);
+
+        // Handle a faulted off-dispatcher evaluation task (N2) — clears in-flight
+        // instead of stranding the trigger with an unhandled Status.Failure.
+        Receive<ExpressionEvalFailed>(HandleExpressionEvalFailed);
     }
 
     /// <inheritdoc />
@@ -310,7 +314,8 @@ public class ScriptActor : ReceiveActor, IWithTimers
             }
         }, CancellationToken.None, TaskCreationOptions.DenyChildAttach,
             ScriptExecutionScheduler.Shared(_options)).Unwrap().PipeTo(self,
-            success: r => new ExpressionEvalResult(r));
+            success: r => new ExpressionEvalResult(r),
+            failure: ex => new ExpressionEvalFailed(ex));
     }
 
     /// <summary>
@@ -333,6 +338,18 @@ public class ScriptActor : ReceiveActor, IWithTimers
         if (_evalPending) { _evalPending = false; StartExpressionEvaluation(); }
     }
 
+    /// <summary>
+    /// Applies a faulted off-dispatcher evaluation task (N2): logs the fault via the
+    /// existing error path (which also increments the health counter), then reuses the
+    /// false-result path — clearing in-flight, applying the false edge, and draining any
+    /// pending evaluation — so a transient scheduler fault cannot permanently park the trigger.
+    /// </summary>
+    private void HandleExpressionEvalFailed(ExpressionEvalFailed msg)
+    {
+        LogExpressionError(msg.Cause);
+        HandleExpressionEvalResult(new ExpressionEvalResult(false));
+    }
+
     /// <summary>
     /// Applies a WhileTrue trigger's condition-state transition: on the
     /// false→true edge, fire once and start the re-fire timer; on the
@@ -621,6 +638,15 @@ public class ScriptActor : ReceiveActor, IWithTimers
     /// the actor thread.
     /// </summary>
     private sealed record ExpressionEvalResult(bool Result);
+
+    /// <summary>
+    /// Piped back to self when the off-dispatcher evaluation TASK itself faults
+    /// (e.g. ObjectDisposedException from a disposed ScriptExecutionScheduler
+    /// during shutdown) — the inner body's catch never sees that. Without this
+    /// mapping the actor receives an unhandled Status.Failure and _evalInFlight
+    /// stays true forever, permanently parking the expression trigger (N2).
+    /// </summary>
+    internal sealed record ExpressionEvalFailed(Exception Cause);
 }
 
 // ── Trigger config types ──