Package org.apache.hadoop.hbase.replication.regionserver
package org.apache.hadoop.hbase.replication.regionserver
-
ClassDescriptionReplicationSource that reads catalog WAL files -- e.g.The 'peer' used internally by Catalog Region Replicas Replication Source.This will load all the xml configuration files for the source cluster replication ID from user configured replication configuration directory.Provides information about the existing states of replication, replication peers and queues.A
ReplicationEndpointimplementation for replicating to another HBase cluster.It is used for replicating HFile entries.Provides access to gauges and counters.Hadoop2 implementation of MetricsReplicationSource.This is the metric source for table level replication metrics.This class is for maintaining the various replication statistics for a sink and publishing them through the metrics interfaces.This class is for maintaining the various replication statistics for a source and publishing them through the metrics interfaces.Noop queue storage -- does nothing.A handler for modifying replication peer in peer procedures.Class that handles the recovered source of a replication stream, which is transfered from another dead region server.Used by aRecoveredReplicationSource.The callable executed at RS side to refresh the peer config/state.AReplicationEndpointendpoint which receives the WAL edits from the WAL, and sends the edits to replicas of regions.Calls replay on the passed edits for the given set of entries belonging to the region.Gateway to Replication.Statistics task.This class is used for exporting some of the info from replication metricsThis chore is responsible to create replication marker rows with special WALEdit with family asWALEdit.METAFAMILYand column qualifier asWALEdit.REPLICATION_MARKERand empty value.An Observer to add HFile References to replication queue.This exception is thrown when a replication source is terminated and source threads got interrupted.This class is responsible for replicating the edits coming from another cluster.Maintains a collection of peers to replicate to, and randomly selects a single peer to replicate to per set of data to replicate.Wraps a replication region server sink to provide the ability to identify it.Class that handles the source of a replication stream.Constructs aReplicationSourceInterfaceNote, not used to create specialized ReplicationSourcesInterface that defines a replication sourceThis class is responsible to manage all the replication sources.This thread reads entries from a queue and ships them.Used to receive new wals.Reads and filters WAL entries, groups the filtered entries into batches, and puts the batches onto a queueIn a scenario of Replication based Disaster/Recovery, when hbase Master-Cluster crashes, this tool is used to sync-up the delta from Master to Slave using the info from ZooKeeper.Per-peer per-node throttling controller for replication: enabled if bandwidth > 0, a cycle = 100ms, by throttling we guarantee data pushed to peer within each cycle won't exceed 'bandwidth' bytesHelper class to determine whether we can push a given WAL entry without breaking the replication order.WAL reader for a serial replication peer.Interface that defines how a region server in peer cluster will get source cluster file system configurations.The callable executed at RS side to switch rpc throttle state.Holds a batch of WAL entries to replicate, along with some statisticsThis exception should be thrown from any wal filter when the filter is expected to recover from the failures and it wants the replication to backup till it fails.Implementations are installed on a Replication Sink called from inside ReplicationSink#replicateEntries to filter replicated WALEntries based off WALEntry attributes.Streaming access to WAL entries.Used by replication to prevent replicating unacked log entries.