Merge pull request #18398 from akka/wip-13584-WeaklyUp2-patriknw

+clu #13584 Accept joining to be WeaklyUp during network split

akka-cluster-metrics/src/main/scala/akka/cluster/metrics/ClusterMetricsCollector.scala

@@ -122,6 +122,7 @@ private[metrics] class ClusterMetricsCollector extends Actor with ActorLogging {
   // TODO collapse to ClusterEvent._ after akka-cluster metrics is gone
   import ClusterEvent.MemberEvent
   import ClusterEvent.MemberUp
+  import ClusterEvent.MemberWeaklyUp
   import ClusterEvent.MemberRemoved
   import ClusterEvent.MemberExited
   import ClusterEvent.ReachabilityEvent
@@ -174,10 +175,13 @@ private[metrics] class ClusterMetricsCollector extends Actor with ActorLogging {
     case msg: MetricsGossipEnvelope ⇒ receiveGossip(msg)
     case state: CurrentClusterState ⇒ receiveState(state)
     case MemberUp(m)                ⇒ addMember(m)
+    case MemberWeaklyUp(m)          ⇒ addMember(m)
     case MemberRemoved(m, _)        ⇒ removeMember(m)
     case MemberExited(m)            ⇒ removeMember(m)
     case UnreachableMember(m)       ⇒ removeMember(m)
-    case ReachableMember(m)         ⇒ if (m.status == MemberStatus.Up) addMember(m)
+    case ReachableMember(m) ⇒
+      if (m.status == MemberStatus.Up || m.status == MemberStatus.WeaklyUp)
+        addMember(m)
     case _: MemberEvent ⇒ // not interested in other types of MemberEvent
 
   }
@@ -207,7 +211,9 @@ private[metrics] class ClusterMetricsCollector extends Actor with ActorLogging {
    * Updates the initial node ring for those nodes that are [[akka.cluster.MemberStatus]] `Up`.
    */
   def receiveState(state: CurrentClusterState): Unit =
-    nodes = (state.members -- state.unreachable) collect { case m if m.status == MemberStatus.Up ⇒ m.address }
+    nodes = (state.members -- state.unreachable) collect {
+      case m if m.status == MemberStatus.Up || m.status == MemberStatus.WeaklyUp ⇒ m.address
+    }
 
   /**
    * Samples the latest metrics for the node, updates metrics statistics in

akka-cluster-tools/src/main/scala/akka/cluster/pubsub/DistributedPubSubMediator.scala

@@ -595,6 +595,10 @@ class DistributedPubSubMediator(settings: DistributedPubSubSettings) extends Act
       if (matchingRole(m))
         nodes += m.address
 
+    case MemberWeaklyUp(m) ⇒
+      if (matchingRole(m))
+        nodes += m.address
+
     case MemberRemoved(m, _) ⇒
       if (m.address == selfAddress)
         context stop self

akka-cluster/src/main/java/akka/cluster/protobuf/msg/ClusterMessages.java

@@ -138,6 +138,10 @@ public final class ClusterMessages {
      * <code>Removed = 5;</code>
      */
     Removed(5, 5),
+    /**
+     * <code>WeaklyUp = 6;</code>
+     */
+    WeaklyUp(6, 6),
     ;
 
     /**
@@ -164,6 +168,10 @@ public final class ClusterMessages {
      * <code>Removed = 5;</code>
      */
     public static final int Removed_VALUE = 5;
+    /**
+     * <code>WeaklyUp = 6;</code>
+     */
+    public static final int WeaklyUp_VALUE = 6;
 
 
     public final int getNumber() { return value; }
@@ -176,6 +184,7 @@ public final class ClusterMessages {
         case 3: return Exiting;
         case 4: return Down;
         case 5: return Removed;
+        case 6: return WeaklyUp;
         default: return null;
       }
     }
@@ -16783,10 +16792,11 @@ public final class ClusterMessages {
       "ort\030\003 \002(\r\022\020\n\010protocol\030\004 \001(\t\"7\n\rUniqueAdd" +
       "ress\022\031\n\007address\030\001 \002(\0132\010.Address\022\013\n\003uid\030\002" +
       " \002(\r*D\n\022ReachabilityStatus\022\r\n\tReachable\020" +
-      "\000\022\017\n\013Unreachable\020\001\022\016\n\nTerminated\020\002*T\n\014Me" +
+      "\000\022\017\n\013Unreachable\020\001\022\016\n\nTerminated\020\002*b\n\014Me" +
       "mberStatus\022\013\n\007Joining\020\000\022\006\n\002Up\020\001\022\013\n\007Leavi" +
       "ng\020\002\022\013\n\007Exiting\020\003\022\010\n\004Down\020\004\022\013\n\007Removed\020\005" +
-      "B\035\n\031akka.cluster.protobuf.msgH\001"
+      "\022\014\n\010WeaklyUp\020\006B\035\n\031akka.cluster.protobuf.",
+      "msgH\001"
     };
     akka.protobuf.Descriptors.FileDescriptor.InternalDescriptorAssigner assigner =
       new akka.protobuf.Descriptors.FileDescriptor.InternalDescriptorAssigner() {

akka-cluster/src/main/protobuf/ClusterMessages.proto

@@ -157,6 +157,7 @@ enum MemberStatus {
   Exiting = 3;
   Down = 4;
   Removed = 5;
+  WeaklyUp = 6;
 }
 
 /**

akka-cluster/src/main/resources/reference.conf

@@ -40,6 +40,14 @@ akka {
     # Disable with "off" or specify a duration to enable.
     down-removal-margin = off
 
+    # By default, the leader will not move 'Joining' members to 'Up' during a network
+    # split. This feature allows the leader to accept 'Joining' members to be 'WeaklyUp'
+    # so they become part of the cluster even during a network split. The leader will
+    # move 'WeaklyUp' members to 'Up' status once convergence has been reached. This
+    # feature must be off if some members are running Akka 2.3.X.
+    # WeaklyUp is an EXPERIMENTAL feature.
+    allow-weakly-up-members = off
+
     # The roles of this member. List of strings, e.g. roles = ["A", "B"].
     # The roles are part of the membership information and can be used by
     # routers or other services to distribute work to certain member types,

akka-cluster/src/main/scala/akka/cluster/ClusterDaemon.scala

@@ -829,6 +829,9 @@ private[cluster] class ClusterCoreDaemon(publisher: ActorRef) extends Actor with
         leaderActionCounter = 0
         leaderActionsOnConvergence()
       } else {
+        if (cluster.settings.AllowWeaklyUpMembers)
+          moveJoiningToWeaklyUp()
+
         leaderActionCounter += 1
         if (leaderActionCounter == firstNotice || leaderActionCounter % periodicNotice == 0)
           logInfo("Leader can currently not perform its duties, reachability status: [{}], member status: [{}]",
@@ -859,6 +862,12 @@ private[cluster] class ClusterCoreDaemon(publisher: ActorRef) extends Actor with
     }
   }
 
+  def isMinNrOfMembersFulfilled: Boolean = {
+    latestGossip.members.size >= MinNrOfMembers && MinNrOfMembersOfRole.forall {
+      case (role, threshold) ⇒ latestGossip.members.count(_.hasRole(role)) >= threshold
+    }
+  }
+
   /**
    * Leader actions are as follows:
    * 1. Move JOINING     => UP                   -- When a node joins the cluster
@@ -878,12 +887,8 @@ private[cluster] class ClusterCoreDaemon(publisher: ActorRef) extends Actor with
     val localOverview = localGossip.overview
     val localSeen = localOverview.seen
 
-    def enoughMembers: Boolean = {
+    val enoughMembers: Boolean = isMinNrOfMembersFulfilled
-      localMembers.size >= MinNrOfMembers && MinNrOfMembersOfRole.forall {
+    def isJoiningToUp(m: Member): Boolean = (m.status == Joining || m.status == WeaklyUp) && enoughMembers
-        case (role, threshold) ⇒ localMembers.count(_.hasRole(role)) >= threshold
-      }
-    }
-    def isJoiningToUp(m: Member): Boolean = m.status == Joining && enoughMembers
 
     val removedUnreachable = for {
       node ← localOverview.reachability.allUnreachableOrTerminated
@@ -967,6 +972,33 @@ private[cluster] class ClusterCoreDaemon(publisher: ActorRef) extends Actor with
     }
   }
 
+  def moveJoiningToWeaklyUp(): Unit = {
+    val localGossip = latestGossip
+    val localMembers = localGossip.members
+
+    val enoughMembers: Boolean = isMinNrOfMembersFulfilled
+    def isJoiningToWeaklyUp(m: Member): Boolean =
+      m.status == Joining && enoughMembers && latestGossip.reachabilityExcludingDownedObservers.isReachable(m.uniqueAddress)
+    val changedMembers = localMembers.collect {
+      case m if isJoiningToWeaklyUp(m) ⇒ m.copy(status = WeaklyUp)
+    }
+
+    if (changedMembers.nonEmpty) {
+      // replace changed members
+      val newMembers = changedMembers ++ localMembers
+      val newGossip = localGossip.copy(members = newMembers)
+      updateLatestGossip(newGossip)
+
+      // log status changes
+      changedMembers foreach { m ⇒
+        logInfo("Leader is moving node [{}] to [{}]", m.address, m.status)
+      }
+
+      publish(latestGossip)
+    }
+
+  }
+
   /**
    * Reaps the unreachable members according to the failure detector's verdict.
    */

akka-cluster/src/main/scala/akka/cluster/ClusterEvent.scala

@@ -121,6 +121,16 @@ object ClusterEvent {
     def member: Member
   }
 
+  /**
+   * Member status changed to WeaklyUp.
+   * A joining member can be moved to `WeaklyUp` if convergence
+   * cannot be reached, i.e. there are unreachable nodes.
+   * It will be moved to `Up` when convergence is reached.
+   */
+  final case class MemberWeaklyUp(member: Member) extends MemberEvent {
+    if (member.status != WeaklyUp) throw new IllegalArgumentException("Expected WeaklyUp status, got: " + member)
+  }
+
   /**
    * Member status changed to Up.
    */
@@ -268,6 +278,7 @@ object ClusterEvent {
         case (_, newMember :: oldMember :: Nil) if newMember.status != oldMember.status ⇒ newMember
       }
       val memberEvents = (newMembers ++ changedMembers) collect {
+        case m if m.status == WeaklyUp ⇒ MemberWeaklyUp(m)
         case m if m.status == Up       ⇒ MemberUp(m)
         case m if m.status == Exiting  ⇒ MemberExited(m)
         // no events for other transitions

akka-cluster/src/main/scala/akka/cluster/ClusterHeartbeat.scala

@@ -111,6 +111,7 @@ private[cluster] final class ClusterHeartbeatSender extends Actor with ActorLogg
     case HeartbeatTick                ⇒ heartbeat()
     case HeartbeatRsp(from)           ⇒ heartbeatRsp(from)
     case MemberUp(m)                  ⇒ addMember(m)
+    case MemberWeaklyUp(m)            ⇒ addMember(m)
     case MemberRemoved(m, _)          ⇒ removeMember(m)
     case UnreachableMember(m)         ⇒ unreachableMember(m)
     case ReachableMember(m)           ⇒ reachableMember(m)
@@ -120,7 +121,7 @@ private[cluster] final class ClusterHeartbeatSender extends Actor with ActorLogg
 
   def init(snapshot: CurrentClusterState): Unit = {
     val nodes: Set[UniqueAddress] = snapshot.members.collect {
-      case m if m.status == MemberStatus.Up ⇒ m.uniqueAddress
+      case m if m.status == MemberStatus.Up || m.status == MemberStatus.WeaklyUp ⇒ m.uniqueAddress
     }(collection.breakOut)
     val unreachable: Set[UniqueAddress] = snapshot.unreachable.map(_.uniqueAddress)
     state = state.init(nodes, unreachable)

akka-cluster/src/main/scala/akka/cluster/ClusterMetricsCollector.scala

@@ -24,6 +24,7 @@ import akka.actor.Address
 import akka.actor.DynamicAccess
 import akka.actor.ExtendedActorSystem
 import akka.cluster.MemberStatus.Up
+import akka.cluster.MemberStatus.WeaklyUp
 import akka.event.Logging
 import java.lang.management.MemoryUsage
 
@@ -89,10 +90,13 @@ private[cluster] class ClusterMetricsCollector(publisher: ActorRef) extends Acto
     case msg: MetricsGossipEnvelope ⇒ receiveGossip(msg)
     case state: CurrentClusterState ⇒ receiveState(state)
     case MemberUp(m)                ⇒ addMember(m)
+    case MemberWeaklyUp(m)          ⇒ addMember(m)
     case MemberRemoved(m, _)        ⇒ removeMember(m)
     case MemberExited(m)            ⇒ removeMember(m)
     case UnreachableMember(m)       ⇒ removeMember(m)
-    case ReachableMember(m)         ⇒ if (m.status == Up) addMember(m)
+    case ReachableMember(m) ⇒
+      if (m.status == MemberStatus.Up || m.status == MemberStatus.WeaklyUp)
+        addMember(m)
     case _: MemberEvent ⇒ // not interested in other types of MemberEvent
 
   }
@@ -122,7 +126,7 @@ private[cluster] class ClusterMetricsCollector(publisher: ActorRef) extends Acto
    * Updates the initial node ring for those nodes that are [[akka.cluster.MemberStatus]] `Up`.
    */
   def receiveState(state: CurrentClusterState): Unit =
-    nodes = state.members collect { case m if m.status == Up ⇒ m.address }
+    nodes = state.members collect { case m if m.status == Up || m.status == WeaklyUp ⇒ m.address }
 
   /**
    * Samples the latest metrics for the node, updates metrics statistics in

akka-cluster/src/main/scala/akka/cluster/ClusterRemoteWatcher.scala

@@ -9,6 +9,7 @@ import akka.cluster.ClusterEvent.CurrentClusterState
 import akka.cluster.ClusterEvent.MemberEvent
 import akka.cluster.ClusterEvent.MemberUp
 import akka.cluster.ClusterEvent.MemberRemoved
+import akka.cluster.ClusterEvent.MemberWeaklyUp
 import akka.remote.FailureDetectorRegistry
 import akka.remote.RemoteWatcher
 
@@ -72,13 +73,20 @@ private[cluster] class ClusterRemoteWatcher(
       clusterNodes = state.members.collect { case m if m.address != selfAddress ⇒ m.address }
       clusterNodes foreach takeOverResponsibility
       unreachable --= clusterNodes
-    case MemberUp(m) ⇒
+    case MemberUp(m)                      ⇒ memberUp(m)
+    case MemberWeaklyUp(m)                ⇒ memberUp(m)
+    case MemberRemoved(m, previousStatus) ⇒ memberRemoved(m, previousStatus)
+    case _: MemberEvent                   ⇒ // not interesting
+  }
+
+  def memberUp(m: Member): Unit =
     if (m.address != selfAddress) {
       clusterNodes += m.address
       takeOverResponsibility(m.address)
       unreachable -= m.address
     }
-    case MemberRemoved(m, previousStatus) ⇒
+
+  def memberRemoved(m: Member, previousStatus: MemberStatus): Unit =
     if (m.address != selfAddress) {
       clusterNodes -= m.address
       if (previousStatus == MemberStatus.Down) {
@@ -86,8 +94,6 @@ private[cluster] class ClusterRemoteWatcher(
       }
       publishAddressTerminated(m.address)
     }
-    case _: MemberEvent ⇒ // not interesting
-  }
 
   override def watchNode(watchee: InternalActorRef) =
     if (!clusterNodes(watchee.path.address)) super.watchNode(watchee)

akka-cluster/src/main/scala/akka/cluster/ClusterSettings.scala

@@ -75,6 +75,8 @@ final class ClusterSettings(val config: Config, val systemName: String) {
     }
   }
 
+  val AllowWeaklyUpMembers = cc.getBoolean("allow-weakly-up-members")
+
   val Roles: Set[String] = immutableSeq(cc.getStringList("roles")).toSet
   val MinNrOfMembers: Int = {
     cc.getInt("min-nr-of-members")

akka-cluster/src/main/scala/akka/cluster/Member.scala

@@ -108,6 +108,8 @@ object Member {
       case (_, Exiting)         ⇒ true
       case (Joining, _)         ⇒ false
       case (_, Joining)         ⇒ true
+      case (WeaklyUp, _)        ⇒ false
+      case (_, WeaklyUp)        ⇒ true
       case _                    ⇒ ordering.compare(a, b) <= 0
     }
   }
@@ -150,6 +152,8 @@ object Member {
     case (_, Leaving)  ⇒ m2
     case (Joining, _)  ⇒ m2
     case (_, Joining)  ⇒ m1
+    case (WeaklyUp, _) ⇒ m2
+    case (_, WeaklyUp) ⇒ m1
     case (Up, Up)      ⇒ m1
   }
 
@@ -164,6 +168,11 @@ abstract class MemberStatus
 
 object MemberStatus {
   @SerialVersionUID(1L) case object Joining extends MemberStatus
+  /**
+   * WeaklyUp is an EXPERIMENTAL feature and is subject to change until
+   * it has received more real world testing.
+   */
+  @SerialVersionUID(1L) case object WeaklyUp extends MemberStatus
   @SerialVersionUID(1L) case object Up extends MemberStatus
   @SerialVersionUID(1L) case object Leaving extends MemberStatus
   @SerialVersionUID(1L) case object Exiting extends MemberStatus
@@ -175,6 +184,13 @@ object MemberStatus {
    */
   def joining: MemberStatus = Joining
 
+  /**
+   * Java API: retrieve the “weaklyUp” status singleton.
+   * WeaklyUp is an EXPERIMENTAL feature and is subject to change until
+   * it has received more real world testing.
+   */
+  def weaklyUp: MemberStatus = WeaklyUp
+
   /**
    * Java API: retrieve the “up” status singleton
    */
@@ -205,7 +221,8 @@ object MemberStatus {
    */
   private[cluster] val allowedTransitions: Map[MemberStatus, Set[MemberStatus]] =
     Map(
-      Joining -> Set(Up, Down, Removed),
+      Joining -> Set(WeaklyUp, Up, Down, Removed),
+      WeaklyUp -> Set(Up, Down, Removed),
       Up -> Set(Leaving, Down, Removed),
       Leaving -> Set(Exiting, Down, Removed),
       Down -> Set(Removed),

akka-cluster/src/main/scala/akka/cluster/protobuf/ClusterMessageSerializer.scala

@@ -168,7 +168,8 @@ class ClusterMessageSerializer(val system: ExtendedActorSystem) extends BaseSeri
     MemberStatus.Leaving -> cm.MemberStatus.Leaving_VALUE,
     MemberStatus.Exiting -> cm.MemberStatus.Exiting_VALUE,
     MemberStatus.Down -> cm.MemberStatus.Down_VALUE,
-    MemberStatus.Removed -> cm.MemberStatus.Removed_VALUE)
+    MemberStatus.Removed -> cm.MemberStatus.Removed_VALUE,
+    MemberStatus.WeaklyUp -> cm.MemberStatus.WeaklyUp_VALUE)
 
   private val memberStatusFromInt = memberStatusToInt.map { case (a, b) ⇒ (b, a) }
 

akka-cluster/src/main/scala/akka/cluster/routing/ClusterRouterConfig.scala

@@ -382,7 +382,7 @@ private[akka] trait ClusterRouterActor { this: RouterActor ⇒
   }
 
   def isAvailable(m: Member): Boolean =
-    m.status == MemberStatus.Up &&
+    (m.status == MemberStatus.Up || m.status == MemberStatus.WeaklyUp) &&
       satisfiesRole(m.roles) &&
       (settings.allowLocalRoutees || m.address != cluster.selfAddress)
 

akka-cluster/src/multi-jvm/scala/akka/cluster/MemberWeaklyUpSpec.scala

@@ -0,0 +1,119 @@
+/**
+ * Copyright (C) 2009-2014 Typesafe Inc. <http://www.typesafe.com>
+ */
+package akka.cluster
+
+import scala.language.postfixOps
+import scala.concurrent.duration._
+import akka.remote.testkit.MultiNodeConfig
+import akka.remote.testkit.MultiNodeSpec
+import akka.remote.transport.ThrottlerTransportAdapter.Direction
+import akka.testkit._
+import com.typesafe.config.ConfigFactory
+import akka.cluster.MemberStatus.WeaklyUp
+
+object MemberWeaklyUpSpec extends MultiNodeConfig {
+  val first = role("first")
+  val second = role("second")
+  val third = role("third")
+  val fourth = role("fourth")
+  val fifth = role("fifth")
+
+  commonConfig(debugConfig(on = false).
+    withFallback(ConfigFactory.parseString("""
+        akka.remote.retry-gate-closed-for = 3 s
+        akka.cluster.allow-weakly-up-members = on
+        """)).
+    withFallback(MultiNodeClusterSpec.clusterConfig))
+
+  testTransport(on = true)
+}
+
+class MemberWeaklyUpMultiJvmNode1 extends MemberWeaklyUpSpec
+class MemberWeaklyUpMultiJvmNode2 extends MemberWeaklyUpSpec
+class MemberWeaklyUpMultiJvmNode3 extends MemberWeaklyUpSpec
+class MemberWeaklyUpMultiJvmNode4 extends MemberWeaklyUpSpec
+class MemberWeaklyUpMultiJvmNode5 extends MemberWeaklyUpSpec
+
+abstract class MemberWeaklyUpSpec
+  extends MultiNodeSpec(MemberWeaklyUpSpec)
+  with MultiNodeClusterSpec {
+
+  import MemberWeaklyUpSpec._
+
+  muteMarkingAsUnreachable()
+
+  val side1 = Vector(first, second)
+  val side2 = Vector(third, fourth, fifth)
+
+  "A cluster of 3 members" must {
+
+    "reach initial convergence" taggedAs LongRunningTest in {
+      awaitClusterUp(first, third, fourth)
+
+      enterBarrier("after-1")
+    }
+
+    "detect network partition and mark nodes on other side as unreachable" taggedAs LongRunningTest in within(20 seconds) {
+      runOn(first) {
+        // split the cluster in two parts (first, second) / (third, fourth, fifth)
+        for (role1 ← side1; role2 ← side2) {
+          testConductor.blackhole(role1, role2, Direction.Both).await
+        }
+      }
+      enterBarrier("after-split")
+
+      runOn(first) {
+        awaitAssert(clusterView.unreachableMembers.map(_.address) should be(Set(address(third), address(fourth))))
+      }
+
+      runOn(third, fourth) {
+        awaitAssert(clusterView.unreachableMembers.map(_.address) should be(Set(address(first))))
+      }
+
+      enterBarrier("after-2")
+    }
+
+    "accept joining on each side and set status to WeaklyUp" taggedAs LongRunningTest in within(20 seconds) {
+      runOn(second) {
+        Cluster(system).join(first)
+      }
+      runOn(fifth) {
+        Cluster(system).join(fourth)
+      }
+      enterBarrier("joined")
+
+      runOn(side1: _*) {
+        awaitAssert {
+          clusterView.members.size should be(4)
+          clusterView.members.exists { m ⇒ m.address == address(second) && m.status == WeaklyUp } should be(true)
+        }
+      }
+
+      runOn(side2: _*) {
+        awaitAssert {
+          clusterView.members.size should be(4)
+          clusterView.members.exists { m ⇒ m.address == address(fifth) && m.status == WeaklyUp } should be(true)
+        }
+      }
+
+      enterBarrier("after-3")
+    }
+
+    "change status to Up after healed network partition" taggedAs LongRunningTest in within(20 seconds) {
+      runOn(first) {
+        for (role1 ← side1; role2 ← side2) {
+          testConductor.passThrough(role1, role2, Direction.Both).await
+        }
+      }
+      enterBarrier("after-passThrough")
+
+      awaitAllReachable()
+      awaitMembersUp(5)
+
+      enterBarrier("after-4")
+    }
+
+  }
+
+}

akka-cluster/src/multi-jvm/scala/akka/cluster/MinMembersBeforeUpSpec.scala

@@ -26,6 +26,17 @@ object MinMembersBeforeUpMultiJvmSpec extends MultiNodeConfig {
     withFallback(MultiNodeClusterSpec.clusterConfigWithFailureDetectorPuppet))
 }
 
+object MinMembersBeforeUpWithWeaklyUpMultiJvmSpec extends MultiNodeConfig {
+  val first = role("first")
+  val second = role("second")
+  val third = role("third")
+
+  commonConfig(debugConfig(on = false).withFallback(ConfigFactory.parseString("""
+      akka.cluster.min-nr-of-members = 3
+      akka.cluster.allow-weakly-up-members = on""")).
+    withFallback(MultiNodeClusterSpec.clusterConfigWithFailureDetectorPuppet))
+}
+
 object MinMembersOfRoleBeforeUpMultiJvmSpec extends MultiNodeConfig {
   val first = role("first")
   val second = role("second")
@@ -46,6 +57,10 @@ class MinMembersBeforeUpMultiJvmNode1 extends MinMembersBeforeUpSpec
 class MinMembersBeforeUpMultiJvmNode2 extends MinMembersBeforeUpSpec
 class MinMembersBeforeUpMultiJvmNode3 extends MinMembersBeforeUpSpec
 
+class MinMembersBeforeUpWithWeaklyUpMultiJvmNode1 extends MinMembersBeforeUpSpec
+class MinMembersBeforeUpWithWeaklyUpMultiJvmNode2 extends MinMembersBeforeUpSpec
+class MinMembersBeforeUpWithWeaklyUpMultiJvmNode3 extends MinMembersBeforeUpSpec
+
 class MinMembersOfRoleBeforeUpMultiJvmNode1 extends MinMembersOfRoleBeforeUpSpec
 class MinMembersOfRoleBeforeUpMultiJvmNode2 extends MinMembersOfRoleBeforeUpSpec
 class MinMembersOfRoleBeforeUpMultiJvmNode3 extends MinMembersOfRoleBeforeUpSpec
@@ -63,6 +78,19 @@ abstract class MinMembersBeforeUpSpec extends MinMembersBeforeUpBase(MinMembersB
   }
 }
 
+abstract class MinMembersBeforeUpWithWeaklyUpSpec extends MinMembersBeforeUpBase(MinMembersBeforeUpMultiJvmSpec) {
+
+  override def first: RoleName = MinMembersBeforeUpWithWeaklyUpMultiJvmSpec.first
+  override def second: RoleName = MinMembersBeforeUpWithWeaklyUpMultiJvmSpec.second
+  override def third: RoleName = MinMembersBeforeUpWithWeaklyUpMultiJvmSpec.third
+
+  "Cluster leader" must {
+    "wait with moving members to UP until minimum number of members have joined with weakly up enabled" taggedAs LongRunningTest in {
+      testWaitMovingMembersToUp()
+    }
+  }
+}
+
 abstract class MinMembersOfRoleBeforeUpSpec extends MinMembersBeforeUpBase(MinMembersOfRoleBeforeUpMultiJvmSpec) {
 
   override def first: RoleName = MinMembersOfRoleBeforeUpMultiJvmSpec.first

akka-docs/rst/common/cluster.rst

@@ -273,12 +273,26 @@ leader, also *auto-down* a node after a configured time of unreachability..
    follows from the fact that the ``unreachable`` node will likely see the rest of
    the cluster as ``unreachable``, become its own leader and form its own cluster.
 
+As mentioned before, if a node is ``unreachable`` then gossip convergence is not
+possible and therefore any ``leader`` actions are also not possible. By enabling
+``akka.cluster.allow-weakly-up-members`` it is possible to let new joining nodes be
+promoted while convergence is not yet reached. These ``Joining`` nodes will be
+promoted as ``WeaklyUp``. Once gossip convergence is reached, the leader will move
+``WeaklyUp`` members to ``Up``.
 
-State Diagram for the Member States
+Note that members on the other side of a network partition have no knowledge about 
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+the existence of the new members. You should for example not count ``WeaklyUp`` 
+members in quorum decisions.
+
+State Diagram for the Member States (``akka.cluster.allow-weakly-up-members=off``)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 .. image:: ../images/member-states.png
 
+State Diagram for the Member States (``akka.cluster.allow-weakly-up-members=on``)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. image:: ../images/member-states-weakly-up.png
 
 Member States
 ^^^^^^^^^^^^^
@@ -286,6 +300,9 @@ Member States
 - **joining**
     transient state when joining a cluster
 
+- **weakly up**
+    transient state while network split (only if ``akka.cluster.allow-weakly-up-members=on``)
+
 - **up**
     normal operating state
 

akka-docs/rst/java/cluster-metrics.rst

@@ -33,6 +33,9 @@ and add the following configuration stanza to your ``application.conf``
 Make sure to disable legacy metrics in akka-cluster: ``akka.cluster.metrics.enabled=off``,
 since it is still enabled in akka-cluster by default (for compatibility with past releases).
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_java>`, if that feature is enabled,
+will participate in Cluster Metrics collection and dissemination.
+ 
 Metrics Collector
 -----------------
 

akka-docs/rst/java/cluster-sharding.rst

@@ -22,6 +22,9 @@ the sender to know the location of the destination actor. This is achieved by se
 the messages via a ``ShardRegion`` actor provided by this extension, which knows how
 to route the message with the entity id to the final destination.
 
+Cluster sharding will not be active on members with status :ref:`WeaklyUp <weakly_up_java>` 
+if that feature is enabled.
+
 An Example
 ----------
 

akka-docs/rst/java/cluster-singleton.rst

@@ -53,6 +53,9 @@ It's worth noting that messages can always be lost because of the distributed na
 As always, additional logic should be implemented in the singleton (acknowledgement) and in the
 client (retry) actors to ensure at-least-once message delivery.
 
+The singleton instance will not run on members with status :ref:`WeaklyUp <weakly_up_java>` if that feature
+is enabled.
+
 Potential problems to be aware of
 ---------------------------------
 

akka-docs/rst/java/cluster-usage.rst

@@ -169,6 +169,41 @@ leaving member will be shutdown after the leader has changed status of the membe
 automatically, but in case of network failures during this process it might still be necessary
 to set the node’s status to ``Down`` in order to complete the removal.
 
+.. _weakly_up_java:
+
+WeaklyUp Members
+^^^^^^^^^^^^^^^^
+
+If a node is ``unreachable`` then gossip convergence is not possible and therefore any
+``leader`` actions are also not possible. However, we still might want new nodes to join
+the cluster in this scenario.
+
+.. warning::
+
+  The WeaklyUp feature is marked as **“experimental”** as of its introduction in Akka 2.4.0. We will continue to
+  improve this feature based on our users’ feedback, which implies that while we try to keep incompatible
+  changes to a minimum the binary compatibility guarantee for maintenance releases does not apply this feature.
+
+This feature is disabled by default. With a configuration option you can allow this behavior::
+
+    akka.cluster.allow-weakly-up-members = on
+
+When ``allow-weakly-up-members`` is enabled and there is no gossip convergence,
+``Joining`` members will be promoted to ``WeaklyUp`` and they will become part of the
+cluster. Once gossip convergence is reached, the leader will move ``WeaklyUp``
+members to ``Up``.
+
+You can subscribe to the ``WeaklyUp`` membership event to make use of the members that are
+in this state, but you should be aware of that members on the other side of a network partition
+have no knowledge about the existence of the new members. You should for example not count 
+``WeaklyUp`` members in quorum decisions.
+
+.. warning::
+
+  This feature is only available from Akka 2.4.0 and cannot be used if some of your
+  cluster members are running an older version of Akka.
+
+
 .. _cluster_subscriber_java:
 
 Subscribe to Cluster Events
@@ -437,6 +472,9 @@ automatically unregistered from the router. When new nodes join the cluster addi
 routees are added to the router, according to the configuration. Routees are also added 
 when a node becomes reachable again, after having been unreachable.
 
+Cluster aware routers make use of members with status :ref:`WeaklyUp <weakly_up_java>` if that feature
+is enabled.
+
 There are two distinct types of routers. 
 
 * **Group - router that sends messages to the specified path using actor selection** 

akka-docs/rst/java/distributed-data.rst

@@ -40,6 +40,10 @@ with a specific role. It communicates with other ``Replicator`` instances with t
 (without address) that are running on other nodes . For convenience it can be used with the
 ``akka.cluster.ddata.DistributedData`` extension.
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_java>`, if that feature is enabled,
+will currently not participate in Distributed Data, but that is something that should be possible to
+add in a future release.
+
 Below is an example of an actor that schedules tick messages to itself and for each tick 
 adds or removes elements from a ``ORSet`` (observed-remove set). It also subscribes to
 changes of this. 

akka-docs/rst/java/distributed-pub-sub.rst

@@ -79,6 +79,11 @@ Successful ``Subscribe`` and ``Unsubscribe`` is acknowledged with
 ``DistributedPubSubMediator.SubscribeAck`` and ``DistributedPubSubMediator.UnsubscribeAck``
 replies.
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_java>`, if that feature is enabled,
+will participate in Distributed Publish Subscribe, i.e. subscribers on nodes with 
+``WeaklyUp`` status will receive published messages if the publisher and subscriber are on
+same side of a network partition.
+
 A Small Example
 ---------------
 

akka-docs/rst/scala/cluster-metrics.rst

@@ -29,6 +29,9 @@ and add the following configuration stanza to your ``application.conf``
 Make sure to disable legacy metrics in akka-cluster: ``akka.cluster.metrics.enabled=off``,
 since it is still enabled in akka-cluster by default (for compatibility with past releases).
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_scala>`, if that feature is enabled,
+will participate in Cluster Metrics collection and dissemination.
+
 Metrics Collector
 -----------------
 

akka-docs/rst/scala/cluster-sharding.rst

@@ -22,6 +22,9 @@ the sender to know the location of the destination actor. This is achieved by se
 the messages via a ``ShardRegion`` actor provided by this extension, which knows how
 to route the message with the entity id to the final destination.
 
+Cluster sharding will not be active on members with status :ref:`WeaklyUp <weakly_up_scala>` 
+if that feature is enabled.
+
 An Example
 ----------
 

akka-docs/rst/scala/cluster-singleton.rst

@@ -53,6 +53,9 @@ It's worth noting that messages can always be lost because of the distributed na
 As always, additional logic should be implemented in the singleton (acknowledgement) and in the
 client (retry) actors to ensure at-least-once message delivery.
 
+The singleton instance will not run on members with status :ref:`WeaklyUp <weakly_up_scala>` if that feature
+is enabled.
+
 Potential problems to be aware of
 ---------------------------------
 

akka-docs/rst/scala/cluster-usage.rst

@@ -163,6 +163,41 @@ leaving member will be shutdown after the leader has changed status of the membe
 automatically, but in case of network failures during this process it might still be necessary
 to set the node’s status to ``Down`` in order to complete the removal.
 
+.. _weakly_up_scala:
+
+WeaklyUp Members
+^^^^^^^^^^^^^^^^
+
+If a node is ``unreachable`` then gossip convergence is not possible and therefore any
+``leader`` actions are also not possible. However, we still might want new nodes to join
+the cluster in this scenario.
+
+.. warning::
+
+  The WeaklyUp feature is marked as **“experimental”** as of its introduction in Akka 2.4.0. We will continue to
+  improve this feature based on our users’ feedback, which implies that while we try to keep incompatible
+  changes to a minimum the binary compatibility guarantee for maintenance releases does not apply this feature.
+
+This feature is disabled by default. With a configuration option you can allow this behavior::
+
+    akka.cluster.allow-weakly-up-members = on
+
+When ``allow-weakly-up-members`` is enabled and there is no gossip convergence,
+``Joining`` members will be promoted to ``WeaklyUp`` and they will become part of the
+cluster. Once gossip convergence is reached, the leader will move ``WeaklyUp``
+members to ``Up``.
+
+You can subscribe to the ``WeaklyUp`` membership event to make use of the members that are
+in this state, but you should be aware of that members on the other side of a network partition
+have no knowledge about the existence of the new members. You should for example not count 
+``WeaklyUp`` members in quorum decisions.  
+
+.. warning::
+
+  This feature is only available from Akka 2.4.0 and cannot be used if some of your
+  cluster members are running an older version of Akka.
+
+
 .. _cluster_subscriber_scala:
 
 Subscribe to Cluster Events
@@ -434,6 +469,9 @@ automatically unregistered from the router. When new nodes join the cluster, add
 routees are added to the router, according to the configuration. Routees are also added 
 when a node becomes reachable again, after having been unreachable.
 
+Cluster aware routers make use of members with status :ref:`WeaklyUp <weakly_up_scala>` if that feature
+is enabled.
+
 There are two distinct types of routers. 
 
 * **Group - router that sends messages to the specified path using actor selection** 

akka-docs/rst/scala/distributed-data.rst

@@ -40,6 +40,10 @@ with a specific role. It communicates with other ``Replicator`` instances with t
 (without address) that are running on other nodes . For convenience it can be used with the
 ``akka.cluster.ddata.DistributedData`` extension.
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_scala>`, if that feature is enabled,
+will currently not participate in Distributed Data, but that is something that should be possible to
+add in a future release.
+
 Below is an example of an actor that schedules tick messages to itself and for each tick 
 adds or removes elements from a ``ORSet`` (observed-remove set). It also subscribes to
 changes of this. 

akka-docs/rst/scala/distributed-pub-sub.rst

@@ -79,6 +79,11 @@ Successful ``Subscribe`` and ``Unsubscribe`` is acknowledged with
 ``DistributedPubSubMediator.SubscribeAck`` and ``DistributedPubSubMediator.UnsubscribeAck``
 replies.
 
+Cluster members with status :ref:`WeaklyUp <weakly_up_scala>`, if that feature is enabled,
+will participate in Distributed Publish Subscribe, i.e. subscribers on nodes with 
+``WeaklyUp`` status will receive published messages if the publisher and subscriber are on
+same side of a network partition.
+
 A Small Example
 ---------------
 

project/MiMa.scala

@@ -566,7 +566,10 @@ object MiMa extends AutoPlugin {
       FilterAnyProblemStartingWith("akka.remote.serialization.DaemonMsgCreateSerializer"),
       FilterAnyProblemStartingWith("akka.remote.testconductor.TestConductorProtocol"),
       FilterAnyProblemStartingWith("akka.cluster.protobuf.msg.ClusterMessages"),
-      FilterAnyProblemStartingWith("akka.cluster.protobuf.ClusterMessageSerializer")
+      FilterAnyProblemStartingWith("akka.cluster.protobuf.ClusterMessageSerializer"),
+      
+      // #13584 change in internal actor
+      ProblemFilters.exclude[IncompatibleMethTypeProblem]("akka.cluster.ClusterCoreDaemon.akka$cluster$ClusterCoreDaemon$$isJoiningToUp$1")
 
      )
   }