org.apache.hadoop.hbase.replication

类 ReplicationPeerZKImpl

java.lang.Object

org.apache.hadoop.hbase.replication.ReplicationStateZKBase

org.apache.hadoop.hbase.replication.ReplicationPeerZKImpl

所有已实现的接口:

Closeable, AutoCloseable, Abortable, ReplicationPeer

@InterfaceAudience.Private
public class ReplicationPeerZKImpl
extends ReplicationStateZKBase
implements ReplicationPeer, Abortable, Closeable

嵌套类概要

嵌套类

限定符和类型	类和说明
class	ReplicationPeerZKImpl.PeerConfigTracker Tracker for PeerConfigNode of this peer
class	ReplicationPeerZKImpl.PeerStateTracker Tracker for state of this peer

从接口继承的嵌套类/接口 org.apache.hadoop.hbase.replication.ReplicationPeer

ReplicationPeer.PeerState

字段概要

从类继承的字段 org.apache.hadoop.hbase.replication.ReplicationStateZKBase

abortable, DISABLED_ZNODE_BYTES, ENABLED_ZNODE_BYTES, hfileRefsZNode, ourClusterKey, peerStateNodeName, peersZNode, queuesZNode, replicationZNode, tableCFsNodeName, zookeeper, ZOOKEEPER_ZNODE_REPLICATION_HFILE_REFS_DEFAULT, ZOOKEEPER_ZNODE_REPLICATION_HFILE_REFS_KEY

构造器概要

构造器

构造器和说明
ReplicationPeerZKImpl(ZooKeeperWatcher zkWatcher, org.apache.hadoop.conf.Configuration conf, String id, ReplicationPeerConfig peerConfig, Abortable abortable) Constructor that takes all the objects required to communicate with the specified peer, except for the region server addresses.

方法概要

方法

限定符和类型	方法和说明
void	abort(String why, Throwable e) Abort the server or client.
void	close()
org.apache.hadoop.conf.Configuration	getConfiguration() Get the configuration object required to communicate with this peer
String	getId() Get the identifier of this peer
long	getPeerBandwidth() Get the per node bandwidth upper limit for this peer
ReplicationPeerConfig	getPeerConfig() Get the peer config object
ReplicationPeer.PeerState	getPeerState() Returns the state of the peer
Map<TableName,List<String>>	getTableCFs() Get replicable (table, cf-list) map of this peer
boolean	isAborted() Check if the server or client was aborted.
static boolean	isStateEnabled(byte[] bytes) Parse the raw data from ZK to get a peer's state
void	removeListenerOfPeerConfig(ReplicationPeerConfigListener listener) Remove a listener when it is closed or terminated
void	startPeerConfigTracker(ZooKeeperWatcher zookeeper, String peerConfigNode) start a table-cfs tracker to listen the (table, cf-list) map change
void	startStateTracker(String peerStateNode) start a state tracker to check whether this peer is enabled or not
void	trackPeerConfigChanges(ReplicationPeerConfigListener listener) Setup a callback for chanages to the replication peer config

从类继承的方法 org.apache.hadoop.hbase.replication.ReplicationStateZKBase

getListOfReplicatorsZK, getPeerNode, getPeerStateNode, getTableCFsNode, isPeerPath, peerExists, toByteArray

从类继承的方法 java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

构造器详细资料

ReplicationPeerZKImpl

public ReplicationPeerZKImpl(ZooKeeperWatcher zkWatcher,
                     org.apache.hadoop.conf.Configuration conf,
                     String id,
                     ReplicationPeerConfig peerConfig,
                     Abortable abortable)
                      throws ReplicationException

Constructor that takes all the objects required to communicate with the specified peer, except for the region server addresses.

参数:

conf - configuration object to this peer

id - string representation of this peer's identifier

peerConfig - configuration for the replication peer

抛出:

ReplicationException

方法详细资料

startStateTracker

public void startStateTracker(String peerStateNode)
                       throws org.apache.zookeeper.KeeperException

start a state tracker to check whether this peer is enabled or not

参数:

peerStateNode - path to zk node which stores peer state

抛出:

org.apache.zookeeper.KeeperException

startPeerConfigTracker

public void startPeerConfigTracker(ZooKeeperWatcher zookeeper,
                          String peerConfigNode)
                            throws org.apache.zookeeper.KeeperException

start a table-cfs tracker to listen the (table, cf-list) map change

参数:

zookeeper -

peerConfigNode - path to zk node which stores table-cfs

抛出:

org.apache.zookeeper.KeeperException

getPeerState

public ReplicationPeer.PeerState getPeerState()

从接口复制的说明: ReplicationPeer

Returns the state of the peer

指定者:

getPeerState 在接口中 ReplicationPeer

返回:

the enabled state

getId

public String getId()

Get the identifier of this peer

指定者:

getId 在接口中 ReplicationPeer

返回:

string representation of the id (short)

getPeerConfig

public ReplicationPeerConfig getPeerConfig()

Get the peer config object

指定者:

getPeerConfig 在接口中 ReplicationPeer

返回:

the ReplicationPeerConfig for this peer

getConfiguration

public org.apache.hadoop.conf.Configuration getConfiguration()

Get the configuration object required to communicate with this peer

指定者:

getConfiguration 在接口中 ReplicationPeer

返回:

configuration object

getTableCFs

public Map<TableName,List<String>> getTableCFs()

Get replicable (table, cf-list) map of this peer

指定者:

getTableCFs 在接口中 ReplicationPeer

返回:

the replicable (table, cf-list) map

getPeerBandwidth

public long getPeerBandwidth()

从接口复制的说明: ReplicationPeer

Get the per node bandwidth upper limit for this peer

指定者:

getPeerBandwidth 在接口中 ReplicationPeer

返回:

the bandwidth up limit

trackPeerConfigChanges

public void trackPeerConfigChanges(ReplicationPeerConfigListener listener)

从接口复制的说明: ReplicationPeer

Setup a callback for chanages to the replication peer config

指定者:

trackPeerConfigChanges 在接口中 ReplicationPeer

参数:

listener - Listener for config changes, usually a replication endpoint

removeListenerOfPeerConfig

public void removeListenerOfPeerConfig(ReplicationPeerConfigListener listener)

从接口复制的说明: ReplicationPeer

Remove a listener when it is closed or terminated

指定者:

removeListenerOfPeerConfig 在接口中 ReplicationPeer

参数:

listener - Listener for config changes, usually a replication endpoint

abort

public void abort(String why,
         Throwable e)

从接口复制的说明: Abortable

Abort the server or client.

指定者:

abort 在接口中 Abortable

参数:

why - Why we're aborting.

e - Throwable that caused abort. Can be null.

isAborted

public boolean isAborted()

从接口复制的说明: Abortable

Check if the server or client was aborted.

指定者:

isAborted 在接口中 Abortable

返回:

true if the server or client was aborted, false otherwise

close

public void close()
           throws IOException

指定者:

close 在接口中 Closeable

指定者:

close 在接口中 AutoCloseable

抛出:

IOException

isStateEnabled

public static boolean isStateEnabled(byte[] bytes)
                              throws DeserializationException

Parse the raw data from ZK to get a peer's state

参数:

bytes - raw ZK data

返回:

True if the passed in bytes are those of a pb serialized ENABLED state.

抛出:

DeserializationException