Class RoutingNodes

    • Constructor Detail

      • RoutingNodes

        public RoutingNodes​(ClusterState clusterState)
      • RoutingNodes

        public RoutingNodes​(ClusterState clusterState,
                            boolean readOnly)
    • Method Detail

      • getIncomingRecoveries

        public int getIncomingRecoveries​(java.lang.String nodeId)
      • getOutgoingRecoveries

        public int getOutgoingRecoveries​(java.lang.String nodeId)
      • iterator

        public java.util.Iterator<RoutingNode> iterator()
        Specified by:
        iterator in interface java.lang.Iterable<RoutingNode>
      • mutableIterator

        public java.util.Iterator<RoutingNode> mutableIterator()
      • node

        public RoutingNode node​(java.lang.String nodeId)
      • nodesPerAttributesCounts

        public com.carrotsearch.hppc.ObjectIntHashMap<java.lang.String> nodesPerAttributesCounts​(java.lang.String attributeName)
      • hasUnassignedPrimaries

        public boolean hasUnassignedPrimaries()
        Returns true iff this RoutingNodes instance has any unassigned primaries even if the primaries are marked as temporarily ignored.
      • hasInactivePrimaries

        public boolean hasInactivePrimaries()
      • hasInactiveShards

        public boolean hasInactiveShards()
      • getRelocatingShardCount

        public int getRelocatingShardCount()
      • assignedShards

        public java.util.List<ShardRouting> assignedShards​(ShardId shardId)
        Returns all shards that are not in the state UNASSIGNED with the same shard ID as the given shard.
      • activePrimary

        public ShardRouting activePrimary​(ShardId shardId)
        Returns the active primary shard for the given shard id or null if no primary is found or the primary is not active.
      • activeReplicaWithHighestVersion

        public ShardRouting activeReplicaWithHighestVersion​(ShardId shardId)
        Returns one active replica shard for the given shard id or null if no active replica is found. Since replicas could possibly be on nodes with a older version of ES than the primary is, this will return replicas on the highest version of ES.
      • allReplicasActive

        public boolean allReplicasActive​(ShardId shardId,
                                         MetaData metaData)
        Returns true iff all replicas are active for the given shard routing. Otherwise false
      • toString

        public java.lang.String toString()
        toString in class java.lang.Object
      • initializeShard

        public ShardRouting initializeShard​(ShardRouting unassignedShard,
                                            java.lang.String nodeId,
                                            java.lang.String existingAllocationId,
                                            long expectedSize,
                                            RoutingChangesObserver routingChangesObserver)
        Moves a shard from unassigned to initialize state
        existingAllocationId - allocation id to use. If null, a fresh allocation id is generated.
        the initialized shard
      • relocateShard

        public Tuple<ShardRouting,​ShardRouting> relocateShard​(ShardRouting startedShard,
                                                                    java.lang.String nodeId,
                                                                    long expectedShardSize,
                                                                    RoutingChangesObserver changes)
        Relocate a shard to another node, adding the target initializing shard as well as assigning it.
        pair of source relocating and target initializing shards.
      • startShard

        public ShardRouting startShard​(org.apache.logging.log4j.Logger logger,
                                       ShardRouting initializingShard,
                                       RoutingChangesObserver routingChangesObserver)
        Applies the relevant logic to start an initializing shard. Moves the initializing shard to started. If the shard is a relocation target, also removes the relocation source. If the started shard is a primary relocation target, this also reinitializes currently initializing replicas as their recovery source changes
        the started shard
      • failShard

        public void failShard​(org.apache.logging.log4j.Logger logger,
                              ShardRouting failedShard,
                              UnassignedInfo unassignedInfo,
                              IndexMetaData indexMetaData,
                              RoutingChangesObserver routingChangesObserver)
        Applies the relevant logic to handle a cancelled or failed shard. Moves the shard to unassigned or completely removes the shard (if relocation target). - If shard is a primary, this also fails initializing replicas. - If shard is an active primary, this also promotes an active replica to primary (if such a replica exists). - If shard is a relocating primary, this also removes the primary relocation target shard. - If shard is a relocating replica, this promotes the replica relocation target to a full initializing replica, removing the relocation source information. This is possible as peer recovery is always done from the primary. - If shard is a (primary or replica) relocation target, this also clears the relocation information on the source shard.
      • size

        public int size()
        Returns the number of routing nodes
      • assertShardStats

        public static boolean assertShardStats​(RoutingNodes routingNodes)
        Calculates RoutingNodes statistics by iterating over all ShardRoutings in the cluster to ensure the book-keeping is correct. For performance reasons, this should only be called from asserts
        this method always returns true or throws an assertion error. If assertion are not enabled this method does nothing.
      • nodeInterleavedShardIterator

        public java.util.Iterator<ShardRouting> nodeInterleavedShardIterator()
        Creates an iterator over shards interleaving between nodes: The iterator returns the first shard from the first node, then the first shard of the second node, etc. until one shard from each node has been returned. The iterator then resumes on the first node by returning the second shard and continues until all shards from all the nodes have been returned.