org.apache.cassandra.service.reads.repair

Class BlockingPartitionRepair

java.lang.Object

com.google.common.util.concurrent.internal.InternalFutureFailureAccess

com.google.common.util.concurrent.AbstractFuture<java.lang.Object>

org.apache.cassandra.service.reads.repair.BlockingPartitionRepair

All Implemented Interfaces:

com.google.common.util.concurrent.ListenableFuture<java.lang.Object>, java.util.concurrent.Future<java.lang.Object>, RequestCallback<java.lang.Object>

public class BlockingPartitionRepair
extends com.google.common.util.concurrent.AbstractFuture<java.lang.Object>
implements RequestCallback<java.lang.Object>

Constructor Summary

Constructors

Constructor and Description
BlockingPartitionRepair(DecoratedKey key, java.util.Map<Replica,Mutation> repairs, ReplicaPlan.ForTokenWrite writePlan)

Method Summary

Modifier and Type	Method and Description
boolean	awaitRepairsUntil(long timeoutAt, java.util.concurrent.TimeUnit timeUnit) Wait for the repair to complete util a future time If the is a past time, the method returns immediately with the repair result.
void	maybeSendAdditionalWrites(long timeout, java.util.concurrent.TimeUnit timeoutUnit) If it looks like we might not receive acks for all the repair mutations we sent out, combine all the unacked mutations and send them to the minority of nodes not involved in the read repair data read / write cycle.
void	onResponse(Message<java.lang.Object> msg)
void	sendInitialRepairs()
protected void	sendRR(Message<Mutation> message, InetAddressAndPort endpoint)

Methods inherited from class com.google.common.util.concurrent.AbstractFuture

addListener, afterDone, cancel, get, get, interruptTask, isCancelled, isDone, pendingToString, set, setException, setFuture, toString, tryInternalFastPathGetFailure, wasInterrupted

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.cassandra.net.RequestCallback

invokeOnFailure, onFailure, trackLatencyForSnitch

Constructor Detail

BlockingPartitionRepair

public BlockingPartitionRepair(DecoratedKey key,
                               java.util.Map<Replica,Mutation> repairs,
                               ReplicaPlan.ForTokenWrite writePlan)

Method Detail

onResponse

public void onResponse(Message<java.lang.Object> msg)

Specified by:

onResponse in interface RequestCallback<java.lang.Object>

Parameters:

msg - response received.

sendRR

protected void sendRR(Message<Mutation> message,
                      InetAddressAndPort endpoint)

sendInitialRepairs

public void sendInitialRepairs()

awaitRepairsUntil

public boolean awaitRepairsUntil(long timeoutAt,
                                 java.util.concurrent.TimeUnit timeUnit)

Wait for the repair to complete util a future time If the is a past time, the method returns immediately with the repair result.

Parameters:

timeoutAt, - future time

timeUnit, - the time unit of the future time

Returns:

true if repair is done; otherwise, false.

maybeSendAdditionalWrites

public void maybeSendAdditionalWrites(long timeout,
                                      java.util.concurrent.TimeUnit timeoutUnit)

If it looks like we might not receive acks for all the repair mutations we sent out, combine all the unacked mutations and send them to the minority of nodes not involved in the read repair data read / write cycle. We will accept acks from them in lieu of acks from the initial mutations sent out, so long as we receive the same number of acks as repair mutations transmitted. This prevents misbehaving nodes from killing a quorum read, while continuing to guarantee monotonic quorum reads