diff --git a/lib/icinga/checkable-check.cpp b/lib/icinga/checkable-check.cpp
index efa9477a20..f0af81bdd9 100644
--- a/lib/icinga/checkable-check.cpp
+++ b/lib/icinga/checkable-check.cpp
@@ -360,21 +360,28 @@ Checkable::ProcessingResult Checkable::ProcessCheckResult(const CheckResult::Ptr
 
 	bool is_flapping = IsFlapping();
 
-	if (cr->GetActive()) {
-		UpdateNextCheck(origin);
-	} else {
-		/* Reschedule the next check for external passive check results. The side effect of
-		 * this is that for as long as we receive results for a service we
-		 * won't execute any active checks. */
-		double offset;
-		double ttl = cr->GetTtl();
-
-		if (ttl > 0)
-			offset = ttl;
-		else
-			offset = GetCheckInterval();
+	// Don't recompute the next check when the current check isn't generated by this endpoint. When the check is
+	// remotely generated we should've already received the "SetNextCheck" event before the "event::CheckResult"
+	// cluster event. Otherwise, the next check received before this check will be invalidated and cause the Checkable
+	// "next_check/next_update" in a HA setup to always be different from the other endpoint as the "m_SchedulingOffset"
+	// is randomly initialised on each node.
+	if (!origin) {
+		if (cr->GetActive()) {
+			UpdateNextCheck();
+		} else {
+			/* Reschedule the next check for external passive check results. The side effect of
+			 * this is that for as long as we receive results for a service we
+			 * won't execute any active checks. */
+			double offset;
+			double ttl = cr->GetTtl();
+
+			if (ttl > 0)
+				offset = ttl;
+			else
+				offset = GetCheckInterval();
 
-		SetNextCheck(Utility::GetTime() + offset, false, origin);
+			SetNextCheck(Utility::GetTime() + offset);
+		}
 	}
 
 	olock.Unlock();