Stabilize QUIC pressure snapshots

agents/rap-node-agent/internal/mesh/fabric_quic_transport.go

@@ -7,6 +7,7 @@ import (
 	"crypto/x509"
 	"encoding/hex"
 	"fmt"
+	"sort"
 	"strings"
 	"sync"
 	"time"
@@ -412,6 +413,15 @@ func (t *QUICFabricTransport) Snapshot() QUICFabricTransportSnapshot {
 			snapshot.CapacityPressurePercent = (snapshot.ActiveStreams * 100) / capacity
 		}
 	}
+	sort.Slice(snapshot.Connections, func(i, j int) bool {
+		if snapshot.Connections[i].PeerID != snapshot.Connections[j].PeerID {
+			return snapshot.Connections[i].PeerID < snapshot.Connections[j].PeerID
+		}
+		if snapshot.Connections[i].Endpoint != snapshot.Connections[j].Endpoint {
+			return snapshot.Connections[i].Endpoint < snapshot.Connections[j].Endpoint
+		}
+		return snapshot.Connections[i].CertSHA256 < snapshot.Connections[j].CertSHA256
+	})
 	return snapshot
 }
 

docs/architecture/DISTRIBUTED_FABRIC_NODE_PROTOCOL_PLAN.md

@@ -427,6 +427,8 @@ QUIC fabric snapshots now include per cached connection pressure, endpoint, and
 saturation state; VPN fabric endpoint ranking consumes that live local pressure
 before stream-limit rejection, spreading new sessions away from already busy
 QUIC carriers.
+Per-connection QUIC snapshot entries are sorted by peer and endpoint so
+heartbeats and diagnostics stay stable across reports.
 Endpoint ranking treats `capacity_limited` observations as a soft pressure
 penalty instead of a hard recent failure, enabling load spreading without
 marking the carrier unhealthy.