Change

daemons/node/common/common_cache.go

@@ -37,43 +37,6 @@ type Score struct {
 	// Peer witnesses
 	witnessChecked    int
 	witnessConsistent int
-	// WitnessPool: up to 3 witnesses last reported by this indexer.
-	// Used for indirect probing when the indexer becomes unreachable.
-	// Oldest entry is replaced when the pool is full and a fresher witness arrives.
-	WitnessPool []WitnessCacheEntry
-}
-
-// WitnessCacheEntry holds one witness AddrInfo with its last-seen timestamp.
-const maxWitnessPool = 3
-
-type WitnessCacheEntry struct {
-	AI     pp.AddrInfo
-	SeenAt time.Time
-}
-
-// UpdateWitnessPool inserts or refreshes a witness entry.
-// If the pool is full and the witness is new, the oldest entry is replaced.
-func (s *Score) UpdateWitnessPool(w pp.AddrInfo) {
-	for i, e := range s.WitnessPool {
-		if e.AI.ID == w.ID {
-			s.WitnessPool[i].AI = w
-			s.WitnessPool[i].SeenAt = time.Now()
-			return
-		}
-	}
-	entry := WitnessCacheEntry{AI: w, SeenAt: time.Now()}
-	if len(s.WitnessPool) < maxWitnessPool {
-		s.WitnessPool = append(s.WitnessPool, entry)
-		return
-	}
-	// Replace oldest.
-	oldest := 0
-	for i, e := range s.WitnessPool {
-		if e.SeenAt.Before(s.WitnessPool[oldest].SeenAt) {
-			oldest = i
-		}
-	}
-	s.WitnessPool[oldest] = entry
 }
 
 // computeNodeSideScore computes the node's quality assessment of an indexer from raw metrics.

daemons/node/common/common_heartbeat.go

@@ -172,46 +172,6 @@ func HandleWitnessQuery(h host.Host, s network.Stream) {
 	json.NewEncoder(s).Encode(report)
 }
 
-// IndirectProbeIndexer asks each witness in the cache whether it still sees
-// the given indexer (by PeerID). Returns true if at least one witness confirms
-// it is alive — meaning our direct link is asymmetrically broken, not the indexer.
-// All probes run in parallel; the function blocks at most 5 seconds.
-func IndirectProbeIndexer(h host.Host, indexerPeerID string, pool []WitnessCacheEntry) bool {
-	if len(pool) == 0 {
-		return false
-	}
-	results := make(chan bool, len(pool))
-	for _, e := range pool {
-		go func(ai pp.AddrInfo) {
-			ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-			defer cancel()
-			s, err := h.NewStream(ctx, ai.ID, ProtocolWitnessQuery)
-			if err != nil {
-				results <- false
-				return
-			}
-			defer s.Reset()
-			s.SetDeadline(time.Now().Add(5 * time.Second))
-			if err := json.NewEncoder(s).Encode(WitnessRequest{IndexerPeerID: indexerPeerID}); err != nil {
-				results <- false
-				return
-			}
-			var rep WitnessReport
-			if err := json.NewDecoder(s).Decode(&rep); err != nil {
-				results <- false
-				return
-			}
-			results <- rep.Seen
-		}(e.AI)
-	}
-	for range pool {
-		if <-results {
-			return true
-		}
-	}
-	return false
-}
-
 // SupportsHeartbeat probes pid with a short-lived stream to verify it has
 // a ProtocolHeartbeat handler (i.e. it is an indexer, not a plain node).
 // Only protocol negotiation is performed — no data is sent.

daemons/node/common/common_indexer_hb.go

@@ -368,7 +368,6 @@ func SendHeartbeat(ctx context.Context, proto protocol.ID, name string, h host.H
 
 				resp, rtt, err := sendHeartbeat(ctx, h, proto, ai.Info, hb, directory.Streams, interval*time.Second)
 				if err != nil { // Heartbeat fails
-					fmt.Println("EERR", err)
 					HeartbeatFailure(h, proto, directory, ai.Addr, ai.Info, isIndexerHB, maxPool, err)
 					continue
 				}
@@ -377,6 +376,7 @@ func SendHeartbeat(ctx context.Context, proto protocol.ID, name string, h host.H
 				// even if the indexer does not support bidirectional heartbeat (Fix 1).
 				if isIndexerHB && score != nil {
 					score.UptimeTracker.RecordHeartbeat()
+					score.UptimeTracker.ConsecutiveFails = 0 // reset on success
 
 					maxRTT := BaseRoundTrip * 10
 					latencyScore := 1.0 - float64(rtt)/float64(maxRTT)
@@ -442,11 +442,6 @@ func SendHeartbeat(ctx context.Context, proto protocol.ID, name string, h host.H
 							}
 						}
 
-						// Refresh local witness cache for indirect probing on future failure.
-						for _, w := range resp.Witnesses {
-							score.UpdateWitnessPool(w)
-						}
-
 						// Launch witness cross-check asynchronously (must not hold lock).
 						if len(resp.Witnesses) > 0 {
 							go queryWitnesses(h, ai.Info.ID.String(), resp.BornAt, resp.FillRate, resp.Witnesses, score)
@@ -550,16 +545,22 @@ func HeartbeatFailure(h host.Host, proto protocol.ID, directory *Directory,
 					Msg("[pool] seed heartbeat failed — keeping in pool, ticker will retry " + err.Error())
 				return
 			}
-			// Indirect probe: query cached witnesses before declaring the indexer dead.
-			// If a witness confirms it is alive, the failure is a local asymmetric
-			// link — not the indexer. Skip eviction; next tick will retry directly.
-			if len(score.WitnessPool) > 0 {
-				pool := append([]WitnessCacheEntry(nil), score.WitnessPool...)
-				if IndirectProbeIndexer(h, info.ID.String(), pool) {
+			// Indirect probing via other alive indexers:
+			// If other indexers in the pool are still responding, they act as implicit
+			// third-party witnesses confirming our connectivity is fine — the failed
+			// indexer is genuinely dead, evict immediately.
+			// If this is the last indexer, there is no third party. Retry up to 3 times
+			// (consecutive failures tracked in UptimeTracker) before declaring it dead.
+			if len(directory.GetAddrs()) <= 1 {
+				score.UptimeTracker.ConsecutiveFails++
+				if score.UptimeTracker.ConsecutiveFails < 3 {
 					logger.Warn().Str("peer", info.ID.String()).
-						Msg("[indirect] witness confirms indexer alive — asymmetric link, skipping eviction " + err.Error())
+						Int("attempt", score.UptimeTracker.ConsecutiveFails).
+						Msg("[indirect] last indexer failed, retrying before eviction")
 					return
 				}
+				logger.Warn().Str("peer", info.ID.String()).
+					Msg("[indirect] last indexer failed 3 times consecutively, evicting")
 			}
 		}
 	}

daemons/node/common/common_scoring.go

@@ -18,9 +18,10 @@ const MaxPayloadChallenge = 2048
 const BaseRoundTrip = 400 * time.Millisecond
 
 type UptimeTracker struct {
-	FirstSeen   time.Time
-	LastSeen    time.Time
-	TotalOnline time.Duration
+	FirstSeen        time.Time
+	LastSeen         time.Time
+	TotalOnline      time.Duration
+	ConsecutiveFails int // incremented on each heartbeat failure; reset to 0 on success
 }
 
 // RecordHeartbeat accumulates online time gap-aware: only counts the interval if

daemons/node/common/interface.go

@@ -12,6 +12,6 @@ type HeartBeatStreamed interface {
 }
 
 type DiscoveryPeer interface {
-	GetPeerRecord(ctx context.Context, key string, search bool) ([]*peer.Peer, error)
+	GetPeerRecord(ctx context.Context, key string) ([]*peer.Peer, error)
 	GetPubSub(topicName string) *pubsub.Topic
 }