Package org.apache.cassandra.service.reads

Class ReadCallback<E extends Endpoints<E>,P extends ReplicaPlan.ForRead<E,P>>

java.lang.Object

org.apache.cassandra.service.reads.ReadCallback<E,P>

All Implemented Interfaces:

RequestCallback<ReadResponse>

public class ReadCallback<E extends Endpoints<E>,P extends ReplicaPlan.ForRead<E,P>>
extends Object
implements RequestCallback<ReadResponse>

Field Summary

Fields

Modifier and Type	Field	Description
protected static final org.slf4j.Logger	logger
final ResponseResolver<E,P>	resolver

Constructor Summary

Constructors

Constructor	Description
ReadCallback(ResponseResolver<E,P> resolver, ReadCommand command, ReplicaPlan.Shared<E,P> replicaPlan, Dispatcher.RequestTime requestTime)

Method Summary

Modifier and Type	Method	Description
boolean	await(long commandTimeout, TimeUnit unit)
void	awaitResults()
boolean	awaitUntil(long deadline)	In case of speculation, we want to time out the request immediately if we have _also_ hit a deadline.
int	blockFor()
boolean	invokeOnFailure()	Returns true if the callback handles failure reporting - in which case the remove host will be asked to report failures to us in the event of a problem processing the request.
void	onFailure(InetAddressAndPort from, RequestFailureReason failureReason)	Called when there is an exception on the remote node or timeout happens
void	onResponse(Message<ReadResponse> message)
protected P	replicaPlan()
void	response(ReadResponse result)
boolean	trackLatencyForSnitch()

Methods inherited from class java.lang.Object

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

Field Details

logger

protected static final org.slf4j.Logger logger

resolver

public final ResponseResolver<E extends Endpoints<E>,P extends ReplicaPlan.ForRead<E,P>> resolver

Constructor Details

ReadCallback

public ReadCallback(ResponseResolver<E,P> resolver,
 ReadCommand command,
 ReplicaPlan.Shared<E,P> replicaPlan,
 Dispatcher.RequestTime requestTime)

Method Details

replicaPlan

protected P replicaPlan()

await

public boolean await(long commandTimeout,
 TimeUnit unit)

awaitUntil

public boolean awaitUntil(long deadline)

In case of speculation, we want to time out the request immediately if we have _also_ hit a deadline. For example, we have a read timeout of 10s, 99% latency of 5 seconds, native_transport_timeout of 12s, and time base is QUEUE: * Request has spent 3 seconds in the queue. Here, we will wait for 2 seconds and try to speculate * Request has spent 10 seconds in the queue. Here, we will wait for 0 seconds and try to speculate If the time base is REQUEST: * Request has spent 10 seconds in the queue. Here, we will only wait 2 seconds and then try to speculate We should _not_ speculate in all these cases, since by that time we are already past request deadline.

awaitResults

public void awaitResults()
                  throws ReadFailureException,
ReadTimeoutException

Throws:

ReadFailureException

ReadTimeoutException

blockFor

public int blockFor()

onResponse

public void onResponse(Message<ReadResponse> message)

Specified by:

onResponse in interface RequestCallback<E extends Endpoints<E>>

Parameters:

message - response received.

response

public void response(ReadResponse result)

trackLatencyForSnitch

public boolean trackLatencyForSnitch()

Specified by:

trackLatencyForSnitch in interface RequestCallback<E extends Endpoints<E>>

Returns:

true if this callback is on the read path and its latency should be given as input to the dynamic snitch.

onFailure

public void onFailure(InetAddressAndPort from,
 RequestFailureReason failureReason)

Description copied from interface: RequestCallback

Called when there is an exception on the remote node or timeout happens

Specified by:

onFailure in interface RequestCallback<E extends Endpoints<E>>

invokeOnFailure

public boolean invokeOnFailure()

Description copied from interface: RequestCallback

Returns true if the callback handles failure reporting - in which case the remove host will be asked to report failures to us in the event of a problem processing the request. TODO: this is an error prone method, and we should be handling failures everywhere so we should probably just start doing that, and remove this method

Specified by:

invokeOnFailure in interface RequestCallback<E extends Endpoints<E>>

Returns:

true if the callback should be invoked on failure