feat: active filter subscription management for edge nodes (#3773)

feat: active filter subscription management for edge nodes ## Subscription Manager * edgeFilterSubLoop reconciles desired vs actual filter subscriptions * edgeFilterHealthLoop pings filter peers, evicts stale ones * EdgeFilterSubState per-shard tracking of confirmed peers and health * best-effort unsubscribe on peer removal * RequestEdgeShardHealth and RequestEdgeFilterPeerCount broker providers ## WakuNode * Remove old edge health loop (loopEdgeHealth, edgeHealthEvent, calculateEdgeTopicHealth) * Register MessageSeenEvent push handler on filter client during start * startDeliveryService now returns `Result[void, string]` and propagates errors ## Health Monitor * getFilterClientHealth queries RequestEdgeFilterPeerCount via broker * Shard/content health providers fall back to RequestEdgeShardHealth when relay inactive * Listen to EventShardTopicHealthChange for health recalculation * Add missing return p.notReady() on failed edge filter peer count request * HealthyThreshold constant moved to `connection_status.nim` ## Broker types * RequestEdgeShardHealth, RequestEdgeFilterPeerCount request types * EventShardTopicHealthChange event type ## Filter Client * Add timeout parameter to ping proc ## Tests * Health monitor event tests with per-node lockNewGlobalBrokerContext * Edge (light client) health update test * Edge health driven by confirmed filter subscriptions test * API subscription tests: sub/receive, failover, peer replacement Co-authored-by: Ivan FB <128452529+Ivansete-status@users.noreply.github.com> Co-authored by Zoltan Nagy
2026-08-25 15:31:16 +00:00 · 2026-03-30 08:30:34 -03:00
parent 0623c10635
commit dc026bbff1
16 changed files with 1126 additions and 266 deletions
@@ -12,12 +12,18 @@ import
    node/health_monitor/health_status,
    node/health_monitor/connection_status,
    node/health_monitor/protocol_health,
+    node/health_monitor/topic_health,
    node/health_monitor/node_health_monitor,
+    node/delivery_service/delivery_service,
+    node/delivery_service/subscription_manager,
    node/kernel_api/relay,
    node/kernel_api/store,
    node/kernel_api/lightpush,
    node/kernel_api/filter,
+    events/health_events,
+    events/peer_events,
    waku_archive,
+    common/broker/broker_context,
  ]

 import ../testlib/[wakunode, wakucore], ../waku_archive/archive_utils
@@ -129,13 +135,12 @@ suite "Health Monitor - health state calculation":

 suite "Health Monitor - events":
  asyncTest "Core (relay) health update":
-    let
-      nodeAKey = generateSecp256k1Key()
+    var nodeA: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeAKey = generateSecp256k1Key()
      nodeA = newTestWakuNode(nodeAKey, parseIpAddress("127.0.0.1"), Port(0))
-
-    (await nodeA.mountRelay()).expect("Node A failed to mount Relay")
-
-    await nodeA.start()
+      (await nodeA.mountRelay()).expect("Node A failed to mount Relay")
+      await nodeA.start()

    let monitorA = NodeHealthMonitor.new(nodeA)

@@ -151,17 +156,15 @@ suite "Health Monitor - events":

    monitorA.startHealthMonitor().expect("Health monitor failed to start")

-    let
-      nodeBKey = generateSecp256k1Key()
+    var nodeB: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeBKey = generateSecp256k1Key()
      nodeB = newTestWakuNode(nodeBKey, parseIpAddress("127.0.0.1"), Port(0))
-
-    let driver = newSqliteArchiveDriver()
-    nodeB.mountArchive(driver).expect("Node B failed to mount archive")
-
-    (await nodeB.mountRelay()).expect("Node B failed to mount relay")
-    await nodeB.mountStore()
-
-    await nodeB.start()
+      let driver = newSqliteArchiveDriver()
+      nodeB.mountArchive(driver).expect("Node B failed to mount archive")
+      (await nodeB.mountRelay()).expect("Node B failed to mount relay")
+      await nodeB.mountStore()
+      await nodeB.start()

    await nodeA.connectToNodes(@[nodeB.switch.peerInfo.toRemotePeerInfo()])

@@ -214,15 +217,20 @@ suite "Health Monitor - events":
    await nodeA.stop()

  asyncTest "Edge (light client) health update":
-    let
-      nodeAKey = generateSecp256k1Key()
+    var nodeA: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeAKey = generateSecp256k1Key()
      nodeA = newTestWakuNode(nodeAKey, parseIpAddress("127.0.0.1"), Port(0))
+      nodeA.mountLightpushClient()
+      await nodeA.mountFilterClient()
+      nodeA.mountStoreClient()
+      require nodeA.mountAutoSharding(1, 8).isOk
+      nodeA.mountMetadata(1, @[0'u16]).expect("Node A failed to mount metadata")
+      await nodeA.start()

-    nodeA.mountLightpushClient()
-    await nodeA.mountFilterClient()
-    nodeA.mountStoreClient()
-
-    await nodeA.start()
+    let ds =
+      DeliveryService.new(false, nodeA).expect("Failed to create DeliveryService")
+    ds.startDeliveryService().expect("Failed to start DeliveryService")

    let monitorA = NodeHealthMonitor.new(nodeA)

@@ -238,23 +246,40 @@ suite "Health Monitor - events":

    monitorA.startHealthMonitor().expect("Health monitor failed to start")

-    let
-      nodeBKey = generateSecp256k1Key()
+    var nodeB: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeBKey = generateSecp256k1Key()
      nodeB = newTestWakuNode(nodeBKey, parseIpAddress("127.0.0.1"), Port(0))
+      let driver = newSqliteArchiveDriver()
+      nodeB.mountArchive(driver).expect("Node B failed to mount archive")
+      (await nodeB.mountRelay()).expect("Node B failed to mount relay")
+      (await nodeB.mountLightpush()).expect("Node B failed to mount lightpush")
+      await nodeB.mountFilter()
+      await nodeB.mountStore()
+      require nodeB.mountAutoSharding(1, 8).isOk
+      nodeB.mountMetadata(1, toSeq(0'u16 ..< 8'u16)).expect(
+        "Node B failed to mount metadata"
+      )
+      await nodeB.start()

-    let driver = newSqliteArchiveDriver()
-    nodeB.mountArchive(driver).expect("Node B failed to mount archive")
-
-    (await nodeB.mountRelay()).expect("Node B failed to mount relay")
-
-    (await nodeB.mountLightpush()).expect("Node B failed to mount lightpush")
-    await nodeB.mountFilter()
-    await nodeB.mountStore()
-
-    await nodeB.start()
+    var metadataFut = newFuture[void]("waitForMetadata")
+    let metadataLis = WakuPeerEvent
+      .listen(
+        nodeA.brokerCtx,
+        proc(evt: WakuPeerEvent): Future[void] {.async: (raises: []), gcsafe.} =
+          if not metadataFut.finished and
+              evt.kind == WakuPeerEventKind.EventMetadataUpdated:
+            metadataFut.complete()
+        ,
+      )
+      .expect("Failed to listen for metadata")

    await nodeA.connectToNodes(@[nodeB.switch.peerInfo.toRemotePeerInfo()])

+    let metadataOk = await metadataFut.withTimeout(TestConnectivityTimeLimit)
+    WakuPeerEvent.dropListener(nodeA.brokerCtx, metadataLis)
+    require metadataOk
+
    let connectTimeLimit = Moment.now() + TestConnectivityTimeLimit
    var gotConnected = false

@@ -292,4 +317,118 @@ suite "Health Monitor - events":
      lastStatus == ConnectionStatus.Disconnected

    await monitorA.stopHealthMonitor()
+    await ds.stopDeliveryService()
+    await nodeA.stop()
+
+  asyncTest "Edge health driven by confirmed filter subscriptions":
+    var nodeA: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeAKey = generateSecp256k1Key()
+      nodeA = newTestWakuNode(nodeAKey, parseIpAddress("127.0.0.1"), Port(0))
+      await nodeA.mountFilterClient()
+      nodeA.mountLightpushClient()
+      nodeA.mountStoreClient()
+      require nodeA.mountAutoSharding(1, 8).isOk
+      nodeA.mountMetadata(1, @[0'u16]).expect("Node A failed to mount metadata")
+      await nodeA.start()
+
+    let ds =
+      DeliveryService.new(false, nodeA).expect("Failed to create DeliveryService")
+    ds.startDeliveryService().expect("Failed to start DeliveryService")
+    let subMgr = ds.subscriptionManager
+
+    var nodeB: WakuNode
+    lockNewGlobalBrokerContext:
+      let nodeBKey = generateSecp256k1Key()
+      nodeB = newTestWakuNode(nodeBKey, parseIpAddress("127.0.0.1"), Port(0))
+      let driver = newSqliteArchiveDriver()
+      nodeB.mountArchive(driver).expect("Node B failed to mount archive")
+      (await nodeB.mountRelay()).expect("Node B failed to mount relay")
+      (await nodeB.mountLightpush()).expect("Node B failed to mount lightpush")
+      await nodeB.mountFilter()
+      await nodeB.mountStore()
+      require nodeB.mountAutoSharding(1, 8).isOk
+      nodeB.mountMetadata(1, toSeq(0'u16 ..< 8'u16)).expect(
+        "Node B failed to mount metadata"
+      )
+      await nodeB.start()
+
+    let monitorA = NodeHealthMonitor.new(nodeA)
+
+    var
+      lastStatus = ConnectionStatus.Disconnected
+      healthSignal = newAsyncEvent()
+
+    monitorA.onConnectionStatusChange = proc(status: ConnectionStatus) {.async.} =
+      lastStatus = status
+      healthSignal.fire()
+
+    monitorA.startHealthMonitor().expect("Health monitor failed to start")
+
+    var metadataFut = newFuture[void]("waitForMetadata")
+    let metadataLis = WakuPeerEvent
+      .listen(
+        nodeA.brokerCtx,
+        proc(evt: WakuPeerEvent): Future[void] {.async: (raises: []), gcsafe.} =
+          if not metadataFut.finished and
+              evt.kind == WakuPeerEventKind.EventMetadataUpdated:
+            metadataFut.complete()
+        ,
+      )
+      .expect("Failed to listen for metadata")
+
+    await nodeA.connectToNodes(@[nodeB.switch.peerInfo.toRemotePeerInfo()])
+
+    let metadataOk = await metadataFut.withTimeout(TestConnectivityTimeLimit)
+    WakuPeerEvent.dropListener(nodeA.brokerCtx, metadataLis)
+    require metadataOk
+
+    var deadline = Moment.now() + TestConnectivityTimeLimit
+    while Moment.now() < deadline:
+      if lastStatus == ConnectionStatus.PartiallyConnected:
+        break
+      if await healthSignal.wait().withTimeout(deadline - Moment.now()):
+        healthSignal.clear()
+
+    check lastStatus == ConnectionStatus.PartiallyConnected
+
+    var shardHealthFut = newFuture[EventShardTopicHealthChange]("waitForShardHealth")
+
+    let shardHealthLis = EventShardTopicHealthChange
+      .listen(
+        nodeA.brokerCtx,
+        proc(
+            evt: EventShardTopicHealthChange
+        ): Future[void] {.async: (raises: []), gcsafe.} =
+          if not shardHealthFut.finished and (
+            evt.health == TopicHealth.MINIMALLY_HEALTHY or
+            evt.health == TopicHealth.SUFFICIENTLY_HEALTHY
+          ):
+            shardHealthFut.complete(evt)
+        ,
+      )
+      .expect("Failed to listen for shard health")
+
+    let contentTopic = ContentTopic("/waku/2/default-content/proto")
+    subMgr.subscribe(contentTopic).expect("Failed to subscribe")
+
+    let shardHealthOk = await shardHealthFut.withTimeout(TestConnectivityTimeLimit)
+    EventShardTopicHealthChange.dropListener(nodeA.brokerCtx, shardHealthLis)
+
+    check shardHealthOk == true
+    check subMgr.edgeFilterSubStates.len > 0
+
+    healthSignal.clear()
+    deadline = Moment.now() + TestConnectivityTimeLimit
+    while Moment.now() < deadline:
+      if lastStatus == ConnectionStatus.PartiallyConnected:
+        break
+      if await healthSignal.wait().withTimeout(deadline - Moment.now()):
+        healthSignal.clear()
+
+    check lastStatus == ConnectionStatus.PartiallyConnected
+
+    await ds.stopDeliveryService()
+    await monitorA.stopHealthMonitor()
+    await nodeB.stop()
    await nodeA.stop()