io.aeron.cluster.ConsensusModule Maven / Gradle / Ivy
Show all versions of aeron-all Show documentation
/*
* Copyright 2014-2024 Real Logic Limited.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* https://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package io.aeron.cluster;
import io.aeron.*;
import io.aeron.archive.Archive;
import io.aeron.archive.client.AeronArchive;
import io.aeron.cluster.client.AeronCluster;
import io.aeron.cluster.client.ClusterException;
import io.aeron.cluster.codecs.AdminRequestDecoder;
import io.aeron.cluster.codecs.AdminRequestType;
import io.aeron.cluster.codecs.BackupQueryDecoder;
import io.aeron.cluster.codecs.HeartbeatRequestDecoder;
import io.aeron.cluster.codecs.MessageHeaderDecoder;
import io.aeron.cluster.codecs.StandbySnapshotDecoder;
import io.aeron.cluster.codecs.mark.ClusterComponentType;
import io.aeron.cluster.service.ClusterMarkFile;
import io.aeron.cluster.service.ClusterCounters;
import io.aeron.cluster.service.ClusteredServiceContainer;
import io.aeron.cluster.service.ClusterClock;
import io.aeron.cluster.service.SnapshotDurationTracker;
import io.aeron.config.Config;
import io.aeron.config.DefaultType;
import io.aeron.driver.DutyCycleTracker;
import io.aeron.driver.NameResolver;
import io.aeron.driver.status.DutyCycleStallTracker;
import io.aeron.exceptions.ConcurrentConcludeException;
import io.aeron.exceptions.ConfigurationException;
import io.aeron.security.Authenticator;
import io.aeron.security.AuthenticatorSupplier;
import io.aeron.security.AuthorisationService;
import io.aeron.security.AuthorisationServiceSupplier;
import io.aeron.security.DefaultAuthenticatorSupplier;
import io.aeron.version.Versioned;
import org.agrona.*;
import org.agrona.concurrent.*;
import org.agrona.concurrent.errors.DistinctErrorLog;
import org.agrona.concurrent.status.AtomicCounter;
import org.agrona.concurrent.status.CountersReader;
import java.io.File;
import java.io.IOException;
import java.io.UncheckedIOException;
import java.nio.charset.StandardCharsets;
import java.util.Random;
import java.util.concurrent.ThreadFactory;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicIntegerFieldUpdater;
import java.util.function.Function;
import java.util.function.LongConsumer;
import java.util.function.Supplier;
import static io.aeron.AeronCounters.*;
import static io.aeron.CommonContext.*;
import static io.aeron.cluster.ConsensusModule.Configuration.CLUSTER_CLIENT_TIMEOUT_COUNT_TYPE_ID;
import static io.aeron.cluster.ConsensusModule.Configuration.CLUSTER_NODE_ROLE_TYPE_ID;
import static io.aeron.cluster.ConsensusModule.Configuration.COMMIT_POSITION_TYPE_ID;
import static io.aeron.cluster.ConsensusModule.Configuration.*;
import static java.util.concurrent.atomic.AtomicIntegerFieldUpdater.newUpdater;
import static org.agrona.BitUtil.findNextPositivePowerOfTwo;
import static org.agrona.SystemUtil.*;
/**
* Component which resides on each node and is responsible for coordinating consensus within a cluster in concert
* with the lifecycle of clustered services.
*/
@Versioned
public final class ConsensusModule implements AutoCloseable
{
/**
* Default set of flags when taking a snapshot
*/
public static final int CLUSTER_ACTION_FLAGS_DEFAULT = 0;
/**
* Flag for a snapshot taken on a standby node.
*/
public static final int CLUSTER_ACTION_FLAGS_STANDBY_SNAPSHOT = 1;
/**
* Possible states for the {@link ConsensusModule}.
* These will be reflected in the {@link Context#moduleStateCounter()} counter.
*/
public enum State
{
/**
* Starting up and recovering state.
*/
INIT(0),
/**
* Active state with ingress and expired timers appended to the log.
*/
ACTIVE(1),
/**
* Suspended processing of ingress and expired timers.
*/
SUSPENDED(2),
/**
* In the process of taking a snapshot.
*/
SNAPSHOT(3),
/**
* Quitting the cluster and shutting down as soon as services ack without taking a snapshot.
*/
QUITTING(4),
/**
* In the process of terminating the node.
*/
TERMINATING(5),
/**
* Terminal state.
*/
CLOSED(6);
static final State[] STATES = values();
private final int code;
State(final int code)
{
if (code != ordinal())
{
throw new IllegalArgumentException(name() + " - code must equal ordinal value: code=" + code);
}
this.code = code;
}
/**
* Code to be stored in an {@link AtomicCounter} for the enum value.
*
* @return code to be stored in an {@link AtomicCounter} for the enum value.
*/
public final int code()
{
return code;
}
/**
* Get the {@link State} encoded in an {@link AtomicCounter}.
*
* @param counter to get the current state for.
* @return the state for the {@link ConsensusModule}.
* @throws ClusterException if the counter is not one of the valid values.
*/
public static State get(final AtomicCounter counter)
{
if (counter.isClosed())
{
return CLOSED;
}
return get(counter.get());
}
/**
* Get the {@link State} corresponding to a particular code.
*
* @param code representing a {@link State}.
* @return the {@link State} corresponding to the provided code.
* @throws ClusterException if the code does not correspond to a valid State.
*/
public static State get(final long code)
{
if (code < 0 || code > (STATES.length - 1))
{
throw new ClusterException("invalid state counter code: " + code);
}
return STATES[(int)code];
}
/**
* Get the current state of the {@link ConsensusModule}.
*
* @param counters to search within.
* @param clusterId to which the allocated counter belongs.
* @return the state of the ConsensusModule or null if not found.
*/
public static State find(final CountersReader counters, final int clusterId)
{
final int counterId = ClusterCounters.find(counters, CONSENSUS_MODULE_STATE_TYPE_ID, clusterId);
if (Aeron.NULL_VALUE != counterId)
{
return State.get(counters.getCounterValue(counterId));
}
return null;
}
}
/**
* Launch an {@link ConsensusModule} with that communicates with an out of process {@link io.aeron.archive.Archive}
* and {@link io.aeron.driver.MediaDriver} then awaits shutdown signal.
*
* @param args command line argument which is a list for properties files as URLs or filenames.
*/
public static void main(final String[] args)
{
loadPropertiesFiles(args);
try (ConsensusModule consensusModule = launch())
{
consensusModule.context().shutdownSignalBarrier().await();
System.out.println("Shutdown ConsensusModule...");
}
}
private final Context ctx;
private final ConsensusModuleAgent conductor;
private final AgentRunner conductorRunner;
private final AgentInvoker conductorInvoker;
ConsensusModule(final Context ctx)
{
try
{
ctx.conclude();
this.ctx = ctx;
conductor = new ConsensusModuleAgent(ctx);
if (ctx.useAgentInvoker())
{
conductorInvoker = new AgentInvoker(ctx.errorHandler(), ctx.errorCounter(), conductor);
conductorRunner = null;
}
else
{
conductorRunner = new AgentRunner(
ctx.idleStrategy(), ctx.errorHandler(), ctx.errorCounter(), conductor);
conductorInvoker = null;
}
}
catch (final ConcurrentConcludeException ex)
{
throw ex;
}
catch (final Exception ex)
{
if (null != ctx.clusterMarkFile())
{
ctx.clusterMarkFile().signalFailedStart();
ctx.clusterMarkFile().force();
}
CloseHelper.quietClose(ctx::close);
throw ex;
}
}
/**
* Launch an {@link ConsensusModule} using a default configuration.
*
* @return a new instance of an {@link ConsensusModule}.
*/
public static ConsensusModule launch()
{
return launch(new Context());
}
/**
* Launch an {@link ConsensusModule} by providing a configuration context.
*
* @param ctx for the configuration parameters.
* @return a new instance of an {@link ConsensusModule}.
*/
public static ConsensusModule launch(final Context ctx)
{
final ConsensusModule consensusModule = new ConsensusModule(ctx);
if (null != consensusModule.conductorRunner)
{
AgentRunner.startOnThread(consensusModule.conductorRunner, ctx.threadFactory());
}
else
{
consensusModule.conductorInvoker.start();
}
return consensusModule;
}
/**
* Get the {@link ConsensusModule.Context} that is used by this {@link ConsensusModule}.
*
* @return the {@link ConsensusModule.Context} that is used by this {@link ConsensusModule}.
*/
public Context context()
{
return ctx;
}
/**
* Get the {@link AgentInvoker} for the consensus module.
*
* @return the {@link AgentInvoker} for the consensus module.
*/
public AgentInvoker conductorAgentInvoker()
{
return conductorInvoker;
}
/**
* {@inheritDoc}
*/
public void close()
{
CloseHelper.closeAll(conductorRunner, conductorInvoker);
}
/**
* Configuration options for cluster.
*/
@Config(existsInC = false)
public static final class Configuration
{
/**
* Major version of the network protocol from consensus module to consensus module. If these don't match then
* consensus modules are not compatible.
*/
public static final int PROTOCOL_MAJOR_VERSION = 1;
/**
* Minor version of the network protocol from consensus module to consensus module. If these don't match then
* some features may not be available.
*/
public static final int PROTOCOL_MINOR_VERSION = 0;
/**
* Patch version of the network protocol from consensus module to consensus module. If these don't match then
* bug fixes may not have been applied.
*/
public static final int PROTOCOL_PATCH_VERSION = 0;
/**
* Combined semantic version for the client to consensus module protocol.
*
* @see SemanticVersion
*/
public static final int PROTOCOL_SEMANTIC_VERSION = SemanticVersion.compose(
PROTOCOL_MAJOR_VERSION, PROTOCOL_MINOR_VERSION, PROTOCOL_PATCH_VERSION);
/**
* Type of snapshot for this component.
*/
public static final long SNAPSHOT_TYPE_ID = 1;
/**
* Property name for the limit for fragments to be consumed on each poll of ingress.
*/
@Config
public static final String CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME = "aeron.cluster.ingress.fragment.limit";
/**
* Default for the limit for fragments to be consumed on each poll of ingress.
*/
@Config
public static final int CLUSTER_INGRESS_FRAGMENT_LIMIT_DEFAULT = 50;
/**
* Property name for whether IPC ingress is allowed or not.
*/
@Config
public static final String CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME = "aeron.cluster.ingress.ipc.allowed";
/**
* Default for whether IPC ingress is allowed or not.
*/
@Config
public static final String CLUSTER_INGRESS_IPC_ALLOWED_DEFAULT = "false";
/**
* Service ID to identify a snapshot in the {@link RecordingLog} for the consensus module.
*/
public static final int SERVICE_ID = Aeron.NULL_VALUE;
/**
* Property name for the identity of the cluster member.
*/
@Config
public static final String CLUSTER_MEMBER_ID_PROP_NAME = "aeron.cluster.member.id";
/**
* Default property for the cluster member identity.
*/
@Config
public static final int CLUSTER_MEMBER_ID_DEFAULT = 0;
/**
* Property name for the identity of the appointed leader. This is when automated leader elections are
* not employed.
*
* This feature is for testing and not recommended for production usage.
*/
@Config
public static final String APPOINTED_LEADER_ID_PROP_NAME = "aeron.cluster.appointed.leader.id";
/**
* Default property for the appointed cluster leader id. A value of {@link Aeron#NULL_VALUE} means no leader
* has been appointed and thus an automated leader election should occur.
*/
@Config
public static final int APPOINTED_LEADER_ID_DEFAULT = Aeron.NULL_VALUE;
/**
* Property name for the comma separated list of cluster member endpoints.
*
*
* 0,ingress:port,consensus:port,log:port,catchup:port,archive:port| \
* 1,ingress:port,consensus:port,log:port,catchup:port,archive:port| ...
*
*
* The ingress endpoints will be used as the endpoint substituted into the
* {@link io.aeron.cluster.client.AeronCluster.Configuration#INGRESS_CHANNEL_PROP_NAME} if the endpoint
* is not provided when unicast.
*/
@Config(defaultType = DefaultType.STRING, defaultString = "")
public static final String CLUSTER_MEMBERS_PROP_NAME = "aeron.cluster.members";
/**
* Property name for the comma separated list of cluster consensus endpoints used for dynamic join, cluster
* backup and cluster standby nodes.
*/
@Config
public static final String CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME = "aeron.cluster.consensus.endpoints";
/**
* Default property for the list of cluster consensus endpoints.
*/
@Config
public static final String CLUSTER_CONSENSUS_ENDPOINTS_DEFAULT = "";
/**
* Property name for whether cluster member information in snapshots should be ignored on load or not.
*/
@Config
public static final String CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME = "aeron.cluster.members.ignore.snapshot";
/**
* Default property for whether cluster member information in snapshots should be ignored or not.
*/
@Config
public static final String CLUSTER_MEMBERS_IGNORE_SNAPSHOT_DEFAULT = "false";
/**
* Channel for the clustered log.
*/
@Config
public static final String LOG_CHANNEL_PROP_NAME = "aeron.cluster.log.channel";
/**
* Channel for the clustered log. This channel can exist for a potentially long time given cluster operation
* so attention should be given to configuration such as term-length and mtu.
*/
@Config
public static final String LOG_CHANNEL_DEFAULT = "aeron:udp?term-length=64m";
/**
* Property name for the comma separated list of member endpoints.
*
*
* ingress:port,consensus:port,log:port,catchup:port,archive:port
*
*
* @see #CLUSTER_MEMBERS_PROP_NAME
*/
@Config
public static final String MEMBER_ENDPOINTS_PROP_NAME = "aeron.cluster.member.endpoints";
/**
* Default property for member endpoints.
*/
@Config
public static final String MEMBER_ENDPOINTS_DEFAULT = "";
/**
* Stream id within a channel for the clustered log.
*/
@Config
public static final String LOG_STREAM_ID_PROP_NAME = "aeron.cluster.log.stream.id";
/**
* Stream id within a channel for the clustered log.
*/
@Config
public static final int LOG_STREAM_ID_DEFAULT = 100;
/**
* Channel to be used for archiving snapshots.
*/
public static final String SNAPSHOT_CHANNEL_DEFAULT = "aeron:ipc?alias=snapshot";
/**
* Stream id for the archived snapshots within a channel.
*/
public static final int SNAPSHOT_STREAM_ID_DEFAULT = 107;
/**
* Message detail to be sent when max concurrent session limit is reached.
*/
public static final String SESSION_LIMIT_MSG = "concurrent session limit";
/**
* Message detail to be sent when a session is rejected due to authentication.
*/
public static final String SESSION_REJECTED_MSG = "session failed authentication";
/**
* Message detail to be sent when a session has an invalid client version.
*/
public static final String SESSION_INVALID_VERSION_MSG = "invalid client version";
/**
* Channel to be used communicating cluster consensus to each other. This can be used for default
* configuration with the endpoints replaced with those provided by {@link #CLUSTER_MEMBERS_PROP_NAME}.
*/
@Config
public static final String CONSENSUS_CHANNEL_PROP_NAME = "aeron.cluster.consensus.channel";
/**
* Channel to be used for communicating cluster consensus to each other. This can be used for default
* configuration with the endpoints replaced with those provided by {@link #CLUSTER_MEMBERS_PROP_NAME}.
*/
@Config
public static final String CONSENSUS_CHANNEL_DEFAULT = "aeron:udp?term-length=64k";
/**
* Stream id within a channel for communicating consensus messages.
*/
@Config
public static final String CONSENSUS_STREAM_ID_PROP_NAME = "aeron.cluster.consensus.stream.id";
/**
* Stream id for the communicating consensus messages.
*/
@Config
public static final int CONSENSUS_STREAM_ID_DEFAULT = 108;
/**
* Channel to be used for replicating logs and snapshots from other archives to the local one.
*/
@Config(defaultType = DefaultType.STRING, defaultString = "")
public static final String REPLICATION_CHANNEL_PROP_NAME = "aeron.cluster.replication.channel";
/**
* Channel template used for replaying logs to a follower using the {@link ClusterMember#catchupEndpoint()}.
*/
@Config
public static final String FOLLOWER_CATCHUP_CHANNEL_PROP_NAME = "aeron.cluster.follower.catchup.channel";
/**
* Default channel template used for replaying logs to a follower using the
* {@link ClusterMember#catchupEndpoint()}.
*/
@Config
public static final String FOLLOWER_CATCHUP_CHANNEL_DEFAULT = UDP_CHANNEL;
/**
* Channel used to build the control request channel for the leader Archive.
*/
@Config
public static final String LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME =
"aeron.cluster.leader.archive.control.channel";
/**
* Default channel used to build the control request channel for the leader Archive.
*/
@Config
public static final String LEADER_ARCHIVE_CONTROL_CHANNEL_DEFAULT = "aeron:udp?term-length=64k";
/**
* Counter type id for the consensus module state.
*/
public static final int CONSENSUS_MODULE_STATE_TYPE_ID = AeronCounters.CLUSTER_CONSENSUS_MODULE_STATE_TYPE_ID;
/**
* Counter type id for the cluster node role.
*/
public static final int CLUSTER_NODE_ROLE_TYPE_ID =
ClusteredServiceContainer.Configuration.CLUSTER_NODE_ROLE_TYPE_ID;
/**
* Counter type id for the control toggle.
*/
public static final int CONTROL_TOGGLE_TYPE_ID = ClusterControl.CONTROL_TOGGLE_TYPE_ID;
/**
* Counter type id of a commit position.
*/
public static final int COMMIT_POSITION_TYPE_ID =
ClusteredServiceContainer.Configuration.COMMIT_POSITION_TYPE_ID;
/**
* Type id of a recovery state counter.
*/
public static final int RECOVERY_STATE_TYPE_ID = AeronCounters.CLUSTER_RECOVERY_STATE_TYPE_ID;
/**
* Counter type id for count of snapshots taken.
*/
public static final int SNAPSHOT_COUNTER_TYPE_ID = AeronCounters.CLUSTER_SNAPSHOT_COUNTER_TYPE_ID;
/**
* Type id for election state counter.
*/
public static final int ELECTION_STATE_TYPE_ID = AeronCounters.CLUSTER_ELECTION_STATE_TYPE_ID;
/**
* Counter type id for the consensus module error count.
*/
public static final int CONSENSUS_MODULE_ERROR_COUNT_TYPE_ID =
AeronCounters.CLUSTER_CONSENSUS_MODULE_ERROR_COUNT_TYPE_ID;
/**
* Counter type id for the number of cluster clients which have been timed out.
*/
public static final int CLUSTER_CLIENT_TIMEOUT_COUNT_TYPE_ID =
AeronCounters.CLUSTER_CLIENT_TIMEOUT_COUNT_TYPE_ID;
/**
* Counter type id for the number of invalid requests which the cluster has received.
*/
public static final int CLUSTER_INVALID_REQUEST_COUNT_TYPE_ID =
AeronCounters.CLUSTER_INVALID_REQUEST_COUNT_TYPE_ID;
/**
* The number of services in this cluster instance.
*
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SERVICE_ID_PROP_NAME
*/
@Config
public static final String SERVICE_COUNT_PROP_NAME = "aeron.cluster.service.count";
/**
* The number of services in this cluster instance.
*/
@Config
public static final int SERVICE_COUNT_DEFAULT = 1;
/**
* Maximum number of cluster sessions that can be active concurrently.
*/
@Config
public static final String MAX_CONCURRENT_SESSIONS_PROP_NAME = "aeron.cluster.max.sessions";
/**
* Maximum number of cluster sessions that can be active concurrently.
*/
@Config
public static final int MAX_CONCURRENT_SESSIONS_DEFAULT = 10;
/**
* Timeout for a session if no activity is observed.
*/
@Config
public static final String SESSION_TIMEOUT_PROP_NAME = "aeron.cluster.session.timeout";
/**
* Timeout for a session if no activity is observed.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 10L * 1000 * 1000 * 1000)
public static final long SESSION_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(10);
/**
* Timeout for a leader if no heartbeat is received by another member.
*/
@Config
public static final String LEADER_HEARTBEAT_TIMEOUT_PROP_NAME = "aeron.cluster.leader.heartbeat.timeout";
/**
* Timeout for a leader if no heartbeat is received by another member.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 10L * 1000 * 1000 * 1000)
public static final long LEADER_HEARTBEAT_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(10);
/**
* Interval at which a leader will send heartbeats if the log is not progressing.
*/
@Config
public static final String LEADER_HEARTBEAT_INTERVAL_PROP_NAME = "aeron.cluster.leader.heartbeat.interval";
/**
* Interval at which a leader will send heartbeats if the log is not progressing.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 200L * 1000 * 1000)
public static final long LEADER_HEARTBEAT_INTERVAL_DEFAULT_NS = TimeUnit.MILLISECONDS.toNanos(200);
/**
* Timeout after which an election vote will be attempted after startup while waiting to canvass the status
* of members if a majority has been heard from.
*/
@Config
public static final String STARTUP_CANVASS_TIMEOUT_PROP_NAME = "aeron.cluster.startup.canvass.timeout";
/**
* Default timeout after which an election vote will be attempted on startup when waiting to canvass the
* status of all members before going for a majority if possible.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 60L * 1000 * 1000 * 1000)
public static final long STARTUP_CANVASS_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(60);
/**
* Timeout after which an election fails if the candidate does not get a majority of votes.
*/
@Config
public static final String ELECTION_TIMEOUT_PROP_NAME = "aeron.cluster.election.timeout";
/**
* Default timeout after which an election fails if the candidate does not get a majority of votes.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 1000L * 1000 * 1000)
public static final long ELECTION_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(1);
/**
* Interval at which a member will send out status updates during election phases.
*/
@Config
public static final String ELECTION_STATUS_INTERVAL_PROP_NAME = "aeron.cluster.election.status.interval";
/**
* Default interval at which a member will send out status updates during election phases.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 100L * 1000 * 1000)
public static final long ELECTION_STATUS_INTERVAL_DEFAULT_NS = TimeUnit.MILLISECONDS.toNanos(100);
/**
* Interval at which a dynamic joining member will send add cluster member and snapshot recording
* queries.
*/
@Config(hasContext = false)
public static final String DYNAMIC_JOIN_INTERVAL_PROP_NAME = "aeron.cluster.dynamic.join.interval";
/**
* Default interval at which a dynamic joining member will send add cluster member and snapshot recording
* queries.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 1000L * 1000 * 1000)
public static final long DYNAMIC_JOIN_INTERVAL_DEFAULT_NS = TimeUnit.SECONDS.toNanos(1);
/**
* Name of the system property for specifying a supplier of {@link Authenticator} for the cluster.
*/
@Config(defaultType = DefaultType.STRING, defaultString = "io.aeron.security.DefaultAuthenticatorSupplier")
public static final String AUTHENTICATOR_SUPPLIER_PROP_NAME = "aeron.cluster.authenticator.supplier";
/**
* Name of the system property for specifying a supplier of {@link AuthorisationService} for the cluster.
*/
@Config(defaultType = DefaultType.STRING, defaultString = "")
public static final String AUTHORISATION_SERVICE_SUPPLIER_PROP_NAME =
"aeron.cluster.authorisation.service.supplier";
static final AuthorisationService ALLOW_ONLY_BACKUP_QUERIES = (protocolId, actionId, type, encodedPrincipal) ->
MessageHeaderDecoder.SCHEMA_ID == protocolId && BackupQueryDecoder.TEMPLATE_ID == actionId;
/**
* Default {@link AuthorisationServiceSupplier} that returns {@link AuthorisationService} that forbids all
* command from being executed (i.e. {@link AuthorisationService#DENY_ALL}).
*/
public static final AuthorisationServiceSupplier DEFAULT_AUTHORISATION_SERVICE_SUPPLIER =
() -> ALLOW_ONLY_BACKUP_QUERIES;
/**
* Size in bytes of the error buffer for the cluster.
*/
@Config
public static final String ERROR_BUFFER_LENGTH_PROP_NAME = "aeron.cluster.error.buffer.length";
/**
* Size in bytes of the error buffer for the cluster.
*/
@Config
public static final int ERROR_BUFFER_LENGTH_DEFAULT = ClusterMarkFile.ERROR_BUFFER_MIN_LENGTH;
/**
* Timeout a leader will wait on getting termination ACKs from followers.
*/
@Config
public static final String TERMINATION_TIMEOUT_PROP_NAME = "aeron.cluster.termination.timeout";
/**
* Property name for threshold value for the consensus module agent work cycle threshold to track
* for being exceeded.
*/
@Config
public static final String CYCLE_THRESHOLD_PROP_NAME = "aeron.cluster.cycle.threshold";
/**
* Default threshold value for the consensus module agent work cycle threshold to track for being exceeded.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 1000L * 1000 * 1000)
public static final long CYCLE_THRESHOLD_DEFAULT_NS = TimeUnit.MILLISECONDS.toNanos(1000);
/**
* Property name for threshold value, which is used for tracking total snapshot duration breaches.
*/
@Config
public static final String TOTAL_SNAPSHOT_DURATION_THRESHOLD_PROP_NAME =
"aeron.cluster.total.snapshot.threshold";
/**
* Default threshold value, which is used for tracking total snapshot duration breaches.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 1000L * 1000 * 1000)
public static final long TOTAL_SNAPSHOT_DURATION_THRESHOLD_DEFAULT_NS =
TimeUnit.MILLISECONDS.toNanos(1000);
/**
* Default timeout a leader will wait on getting termination ACKs from followers.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 10L * 1000 * 1000 * 1000)
public static final long TERMINATION_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(10);
/**
* Resolution in nanoseconds for each tick of the timer wheel for scheduling deadlines.
*/
@Config
public static final String WHEEL_TICK_RESOLUTION_PROP_NAME = "aeron.cluster.wheel.tick.resolution";
/**
* Resolution in nanoseconds for each tick of the timer wheel for scheduling deadlines. Defaults to 8ms.
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 4L * 1000 * 1000)
public static final long WHEEL_TICK_RESOLUTION_DEFAULT_NS = TimeUnit.MILLISECONDS.toNanos(8);
/**
* Number of ticks, or spokes, on the timer wheel. Higher number of ticks reduces potential conflicts
* traded off against memory usage.
*/
@Config
public static final String TICKS_PER_WHEEL_PROP_NAME = "aeron.cluster.ticks.per.wheel";
/**
* Number of ticks, or spokes, on the timer wheel. Higher number of ticks reduces potential conflicts
* traded off against memory usage. Defaults to 128 per wheel.
*/
@Config
public static final int TICKS_PER_WHEEL_DEFAULT = 128;
/**
* The level at which files should be sync'ed to disk.
*
* - 0 - normal writes.
* - 1 - sync file data.
* - 2 - sync file data + metadata.
*
*/
@Config
public static final String FILE_SYNC_LEVEL_PROP_NAME = "aeron.cluster.file.sync.level";
/**
* Default file sync level of normal writes.
*/
@Config
public static final int FILE_SYNC_LEVEL_DEFAULT = 0;
/**
* {@link TimerServiceSupplier} to be used for creating the {@link TimerService} used by consensus module.
*/
@Config
public static final String TIMER_SERVICE_SUPPLIER_PROP_NAME = "aeron.cluster.timer.service.supplier";
/**
* Name of the {@link TimerServiceSupplier} that creates {@link TimerService} based on the timer wheel
* implementation.
*/
public static final String TIMER_SERVICE_SUPPLIER_WHEEL = "io.aeron.cluster.WheelTimerServiceSupplier";
/**
* Name of the {@link TimerServiceSupplier} that creates a sequence-preserving {@link TimerService} based
* on a priority heap implementation.
*/
public static final String TIMER_SERVICE_SUPPLIER_PRIORITY_HEAP =
"io.aeron.cluster.PriorityHeapTimerServiceSupplier";
/**
* Default {@link TimerServiceSupplier}.
*/
@Config
public static final String TIMER_SERVICE_SUPPLIER_DEFAULT = TIMER_SERVICE_SUPPLIER_WHEEL;
/**
* Property name for the name returned from {@link Agent#roleName()} for the consensus module.
*/
@Config(defaultType = DefaultType.STRING, defaultString = "")
public static final String CLUSTER_CONSENSUS_MODULE_AGENT_ROLE_NAME_PROP_NAME =
"aeron.cluster.consensus.module.agent.role.name";
/**
* Property name for replication progress timeout.
*
* @since 1.41.0
*/
@Config
public static final String CLUSTER_REPLICATION_PROGRESS_TIMEOUT_PROP_NAME =
"aeron.cluster.replication.progress.timeout";
/**
* Default timeout for replication progress in nanoseconds.
*
* @since 1.41.0
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 10L * 1000 * 1000 * 1000)
public static final long CLUSTER_REPLICATION_PROGRESS_TIMEOUT_DEFAULT_NS = TimeUnit.SECONDS.toNanos(10);
/**
* Property name for replication progress interval.
*
* @since 1.41.0
*/
@Config(defaultType = DefaultType.LONG, defaultLong = 0)
public static final String CLUSTER_REPLICATION_PROGRESS_INTERVAL_PROP_NAME =
"aeron.cluster.replication.progress.interval";
/**
* Property name of enabling the acceptance of standby snapshots
*/
@Config(defaultType = DefaultType.BOOLEAN, defaultBoolean = false)
public static final String CLUSTER_ACCEPT_STANDBY_SNAPSHOTS_PROP_NAME =
"aeron.cluster.accept.standby.snapshots";
/**
* Property name of setting {@link ClusterClock}. Should specify a fully qualified class name.
* Defaults to {@link MillisecondClusterClock}.
*
* @since 1.44.0
*/
@Config(defaultType = DefaultType.STRING, defaultString = "io.aeron.cluster.MillisecondClusterClock")
public static final String CLUSTER_CLOCK_PROP_NAME = "aeron.cluster.clock";
/**
* Property name of setting {@link ConsensusModuleExtension}. Should specify a fully qualified class name.
*
* @since 1.45.0
*/
public static final String CONSENSUS_MODULE_EXTENSION_CLASS_NAME_PROP_NAME =
"aeron.cluster.consensus.module.extension";
/**
* The value {@link #CLUSTER_INGRESS_FRAGMENT_LIMIT_DEFAULT} or system property
* {@link #CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME} if set.
*
* @return {@link #CLUSTER_INGRESS_FRAGMENT_LIMIT_DEFAULT} or system property
* {@link #CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME} if set.
*/
public static int ingressFragmentLimit()
{
return Integer.getInteger(CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME, CLUSTER_INGRESS_FRAGMENT_LIMIT_DEFAULT);
}
/**
* The value {@link #CLUSTER_INGRESS_IPC_ALLOWED_DEFAULT} or system property
* {@link #CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME} if set.
*
* @return {@link #CLUSTER_INGRESS_IPC_ALLOWED_DEFAULT} or system property
* {@link #CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME} if set.
*/
public static boolean isIpcIngressAllowed()
{
return "true".equalsIgnoreCase(System.getProperty(
CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME, CLUSTER_INGRESS_IPC_ALLOWED_DEFAULT));
}
/**
* The value {@link #CLUSTER_MEMBER_ID_DEFAULT} or system property
* {@link #CLUSTER_MEMBER_ID_PROP_NAME} if set.
*
* @return {@link #CLUSTER_MEMBER_ID_DEFAULT} or system property
* {@link #CLUSTER_MEMBER_ID_PROP_NAME} if set.
*/
public static int clusterMemberId()
{
return Integer.getInteger(CLUSTER_MEMBER_ID_PROP_NAME, CLUSTER_MEMBER_ID_DEFAULT);
}
/**
* The value {@link #APPOINTED_LEADER_ID_DEFAULT} or system property
* {@link #APPOINTED_LEADER_ID_PROP_NAME} if set.
*
* This feature is for testing and not recommended for production usage.
*
* @return {@link #APPOINTED_LEADER_ID_DEFAULT} or system property
* {@link #APPOINTED_LEADER_ID_PROP_NAME} if set.
*/
public static int appointedLeaderId()
{
return Integer.getInteger(APPOINTED_LEADER_ID_PROP_NAME, APPOINTED_LEADER_ID_DEFAULT);
}
/**
* The value of system property {@link #CLUSTER_MEMBERS_PROP_NAME} if set, null otherwise.
*
* @return of system property {@link #CLUSTER_MEMBERS_PROP_NAME} if set.
*/
public static String clusterMembers()
{
return System.getProperty(CLUSTER_MEMBERS_PROP_NAME);
}
/**
* The value {@link #CLUSTER_CONSENSUS_ENDPOINTS_DEFAULT} or system property
* {@link #CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME} if set.
*
* @return {@link #CLUSTER_CONSENSUS_ENDPOINTS_DEFAULT} or system property
* {@link #CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME} it set.
*/
public static String clusterConsensusEndpoints()
{
return System.getProperty(CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME, CLUSTER_CONSENSUS_ENDPOINTS_DEFAULT);
}
/**
* The value {@link #CLUSTER_MEMBERS_IGNORE_SNAPSHOT_DEFAULT} or system property
* {@link #CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME} if set.
*
* @return {@link #CLUSTER_MEMBERS_IGNORE_SNAPSHOT_DEFAULT} or system property
* {@link #CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME} it set.
*/
public static boolean clusterMembersIgnoreSnapshot()
{
return "true".equalsIgnoreCase(System.getProperty(
CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME, CLUSTER_MEMBERS_IGNORE_SNAPSHOT_DEFAULT));
}
/**
* The value {@link #LOG_CHANNEL_DEFAULT} or system property {@link #LOG_CHANNEL_PROP_NAME} if set.
*
* @return {@link #LOG_CHANNEL_DEFAULT} or system property {@link #LOG_CHANNEL_PROP_NAME} if set.
*/
public static String logChannel()
{
return System.getProperty(LOG_CHANNEL_PROP_NAME, LOG_CHANNEL_DEFAULT);
}
/**
* The value {@link #LOG_STREAM_ID_DEFAULT} or system property {@link #LOG_STREAM_ID_PROP_NAME} if set.
*
* @return {@link #LOG_STREAM_ID_DEFAULT} or system property {@link #LOG_STREAM_ID_PROP_NAME} if set.
*/
public static int logStreamId()
{
return Integer.getInteger(LOG_STREAM_ID_PROP_NAME, LOG_STREAM_ID_DEFAULT);
}
/**
* The value {@link #MEMBER_ENDPOINTS_DEFAULT} or system property {@link #MEMBER_ENDPOINTS_PROP_NAME} if set.
*
* @return {@link #MEMBER_ENDPOINTS_DEFAULT} or system property {@link #MEMBER_ENDPOINTS_PROP_NAME} if set.
*/
public static String memberEndpoints()
{
return System.getProperty(MEMBER_ENDPOINTS_PROP_NAME, MEMBER_ENDPOINTS_DEFAULT);
}
/**
* The value {@link #SNAPSHOT_CHANNEL_DEFAULT} or system property
* {@link io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_CHANNEL_PROP_NAME} if set.
*
* @return {@link #SNAPSHOT_CHANNEL_DEFAULT} or system property
* {@link io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_CHANNEL_PROP_NAME} if set.
*/
public static String snapshotChannel()
{
return System.getProperty(
ClusteredServiceContainer.Configuration.SNAPSHOT_CHANNEL_PROP_NAME, SNAPSHOT_CHANNEL_DEFAULT);
}
/**
* The value {@link #SNAPSHOT_STREAM_ID_DEFAULT} or system property
* {@link io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_STREAM_ID_PROP_NAME} if set.
*
* @return {@link #SNAPSHOT_STREAM_ID_DEFAULT} or system property
* {@link io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_STREAM_ID_PROP_NAME} if set.
*/
public static int snapshotStreamId()
{
return Integer.getInteger(
ClusteredServiceContainer.Configuration.SNAPSHOT_STREAM_ID_PROP_NAME, SNAPSHOT_STREAM_ID_DEFAULT);
}
/**
* The value {@link #SERVICE_COUNT_DEFAULT} or system property
* {@link #SERVICE_COUNT_PROP_NAME} if set.
*
* @return {@link #SERVICE_COUNT_DEFAULT} or system property
* {@link #SERVICE_COUNT_PROP_NAME} if set.
*/
public static int serviceCount()
{
return Integer.getInteger(SERVICE_COUNT_PROP_NAME, SERVICE_COUNT_DEFAULT);
}
/**
* The value {@link #MAX_CONCURRENT_SESSIONS_DEFAULT} or system property
* {@link #MAX_CONCURRENT_SESSIONS_PROP_NAME} if set.
*
* @return {@link #MAX_CONCURRENT_SESSIONS_DEFAULT} or system property
* {@link #MAX_CONCURRENT_SESSIONS_PROP_NAME} if set.
*/
public static int maxConcurrentSessions()
{
return Integer.getInteger(MAX_CONCURRENT_SESSIONS_PROP_NAME, MAX_CONCURRENT_SESSIONS_DEFAULT);
}
/**
* Timeout for a session if no activity is observed.
*
* @return timeout in nanoseconds to wait for activity
* @see #SESSION_TIMEOUT_PROP_NAME
*/
public static long sessionTimeoutNs()
{
return getDurationInNanos(SESSION_TIMEOUT_PROP_NAME, SESSION_TIMEOUT_DEFAULT_NS);
}
/**
* Timeout for a leader if no heartbeat is received by another member.
*
* @return timeout in nanoseconds to wait for heartbeat from a leader.
* @see #LEADER_HEARTBEAT_TIMEOUT_PROP_NAME
*/
public static long leaderHeartbeatTimeoutNs()
{
return getDurationInNanos(LEADER_HEARTBEAT_TIMEOUT_PROP_NAME, LEADER_HEARTBEAT_TIMEOUT_DEFAULT_NS);
}
/**
* Interval at which a leader will send a heartbeat if the log is not progressing.
*
* @return timeout in nanoseconds to for leader heartbeats when no log being appended.
* @see #LEADER_HEARTBEAT_INTERVAL_PROP_NAME
*/
public static long leaderHeartbeatIntervalNs()
{
return getDurationInNanos(LEADER_HEARTBEAT_INTERVAL_PROP_NAME, LEADER_HEARTBEAT_INTERVAL_DEFAULT_NS);
}
/**
* Timeout waiting to canvass the status of cluster members before voting if a majority have been heard from.
*
* @return timeout in nanoseconds to wait for the status of other cluster members before voting.
* @see #STARTUP_CANVASS_TIMEOUT_PROP_NAME
*/
public static long startupCanvassTimeoutNs()
{
return getDurationInNanos(STARTUP_CANVASS_TIMEOUT_PROP_NAME, STARTUP_CANVASS_TIMEOUT_DEFAULT_NS);
}
/**
* Timeout waiting for votes to become leader in an election.
*
* @return timeout in nanoseconds to wait for votes to become leader in an election.
* @see #ELECTION_TIMEOUT_PROP_NAME
*/
public static long electionTimeoutNs()
{
return getDurationInNanos(ELECTION_TIMEOUT_PROP_NAME, ELECTION_TIMEOUT_DEFAULT_NS);
}
/**
* Interval at which a member will send out status messages during the election phases.
*
* @return interval at which a member will send out status messages during the election phases.
* @see #ELECTION_STATUS_INTERVAL_PROP_NAME
*/
public static long electionStatusIntervalNs()
{
return getDurationInNanos(ELECTION_STATUS_INTERVAL_PROP_NAME, ELECTION_STATUS_INTERVAL_DEFAULT_NS);
}
/**
* Timeout waiting for follower termination by leader.
*
* @return timeout in nanoseconds to wait followers to terminate.
* @see #TERMINATION_TIMEOUT_PROP_NAME
*/
public static long terminationTimeoutNs()
{
return getDurationInNanos(TERMINATION_TIMEOUT_PROP_NAME, TERMINATION_TIMEOUT_DEFAULT_NS);
}
/**
* Get threshold value for the consensus module agent work cycle threshold to track for being exceeded.
*
* @return threshold value in nanoseconds.
*/
public static long cycleThresholdNs()
{
return getDurationInNanos(CYCLE_THRESHOLD_PROP_NAME, CYCLE_THRESHOLD_DEFAULT_NS);
}
/**
* Get threshold value, which is used for monitoring total snapshot duration breaches of its predefined
* threshold.
*
* @return threshold value in nanoseconds.
*/
public static long totalSnapshotDurationThresholdNs()
{
return getDurationInNanos(
TOTAL_SNAPSHOT_DURATION_THRESHOLD_PROP_NAME, TOTAL_SNAPSHOT_DURATION_THRESHOLD_DEFAULT_NS);
}
/**
* Size in bytes of the error buffer in the mark file.
*
* @return length of error buffer in bytes.
* @see #ERROR_BUFFER_LENGTH_PROP_NAME
*/
public static int errorBufferLength()
{
return getSizeAsInt(ERROR_BUFFER_LENGTH_PROP_NAME, ERROR_BUFFER_LENGTH_DEFAULT);
}
/**
* The value {@link DefaultAuthenticatorSupplier#INSTANCE} or system property
* {@link #AUTHENTICATOR_SUPPLIER_PROP_NAME} if set.
*
* @return {@link DefaultAuthenticatorSupplier#INSTANCE} or system property
* {@link #AUTHENTICATOR_SUPPLIER_PROP_NAME} if set.
*/
public static AuthenticatorSupplier authenticatorSupplier()
{
final String supplierClassName = System.getProperty(AUTHENTICATOR_SUPPLIER_PROP_NAME);
if (Strings.isEmpty(supplierClassName))
{
return DefaultAuthenticatorSupplier.INSTANCE;
}
AuthenticatorSupplier supplier = null;
try
{
supplier = (AuthenticatorSupplier)Class.forName(supplierClassName).getConstructor().newInstance();
}
catch (final Exception ex)
{
LangUtil.rethrowUnchecked(ex);
}
return supplier;
}
/**
* The {@link AuthorisationServiceSupplier} specified in the
* {@link #AUTHORISATION_SERVICE_SUPPLIER_PROP_NAME} system property or the
* {@link #DEFAULT_AUTHORISATION_SERVICE_SUPPLIER}.
*
* @return system property {@link #AUTHORISATION_SERVICE_SUPPLIER_PROP_NAME} if set or
* {@link #DEFAULT_AUTHORISATION_SERVICE_SUPPLIER} otherwise.
*/
public static AuthorisationServiceSupplier authorisationServiceSupplier()
{
final String supplierClassName = System.getProperty(AUTHORISATION_SERVICE_SUPPLIER_PROP_NAME);
if (Strings.isEmpty(supplierClassName))
{
return DEFAULT_AUTHORISATION_SERVICE_SUPPLIER;
}
try
{
return (AuthorisationServiceSupplier)Class.forName(supplierClassName).getConstructor().newInstance();
}
catch (final Exception ex)
{
LangUtil.rethrowUnchecked(ex);
return null;
}
}
/**
* The value {@link #CONSENSUS_CHANNEL_DEFAULT} or system property
* {@link #CONSENSUS_CHANNEL_PROP_NAME} if set.
*
* @return {@link #CONSENSUS_CHANNEL_DEFAULT} or system property
* {@link #CONSENSUS_CHANNEL_PROP_NAME} if set.
*/
public static String consensusChannel()
{
return System.getProperty(CONSENSUS_CHANNEL_PROP_NAME, CONSENSUS_CHANNEL_DEFAULT);
}
/**
* The value {@link #CONSENSUS_STREAM_ID_DEFAULT} or system property
* {@link #CONSENSUS_STREAM_ID_PROP_NAME} if set.
*
* @return {@link #CONSENSUS_STREAM_ID_DEFAULT} or system property
* {@link #CONSENSUS_STREAM_ID_PROP_NAME} if set.
*/
public static int consensusStreamId()
{
return Integer.getInteger(CONSENSUS_STREAM_ID_PROP_NAME, CONSENSUS_STREAM_ID_DEFAULT);
}
/**
* The system property for {@link #REPLICATION_CHANNEL_PROP_NAME} if set or null.
*
* @return system property {@link #REPLICATION_CHANNEL_PROP_NAME} if set or null.
*/
public static String replicationChannel()
{
return System.getProperty(REPLICATION_CHANNEL_PROP_NAME);
}
/**
* The value {@link #FOLLOWER_CATCHUP_CHANNEL_DEFAULT} or system property
* {@link #FOLLOWER_CATCHUP_CHANNEL_PROP_NAME} if set.
*
* @return {@link #FOLLOWER_CATCHUP_CHANNEL_DEFAULT} or system property
* {@link #FOLLOWER_CATCHUP_CHANNEL_PROP_NAME} if set.
*/
public static String followerCatchupChannel()
{
return System.getProperty(FOLLOWER_CATCHUP_CHANNEL_PROP_NAME, FOLLOWER_CATCHUP_CHANNEL_DEFAULT);
}
/**
* The value {@link #LEADER_ARCHIVE_CONTROL_CHANNEL_DEFAULT} or system property
* {@link #LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME} if set.
*
* @return {@link #LEADER_ARCHIVE_CONTROL_CHANNEL_DEFAULT} or system property
* {@link #LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME} if set.
*/
public static String leaderArchiveControlChannel()
{
return System.getProperty(LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME, LEADER_ARCHIVE_CONTROL_CHANNEL_DEFAULT);
}
/**
* The value {@link #WHEEL_TICK_RESOLUTION_DEFAULT_NS} or system property
* {@link #WHEEL_TICK_RESOLUTION_PROP_NAME} if set.
*
* @return {@link #WHEEL_TICK_RESOLUTION_DEFAULT_NS} or system property
* {@link #WHEEL_TICK_RESOLUTION_PROP_NAME} if set.
*/
public static long wheelTickResolutionNs()
{
return getDurationInNanos(WHEEL_TICK_RESOLUTION_PROP_NAME, WHEEL_TICK_RESOLUTION_DEFAULT_NS);
}
/**
* The value {@link #TICKS_PER_WHEEL_DEFAULT} or system property
* {@link #CLUSTER_MEMBER_ID_PROP_NAME} if set.
*
* @return {@link #TICKS_PER_WHEEL_DEFAULT} or system property
* {@link #TICKS_PER_WHEEL_PROP_NAME} if set.
*/
public static int ticksPerWheel()
{
return Integer.getInteger(TICKS_PER_WHEEL_PROP_NAME, TICKS_PER_WHEEL_DEFAULT);
}
/**
* The level at which files should be sync'ed to disk.
*
* - 0 - normal writes.
* - 1 - sync file data.
* - 2 - sync file data + metadata.
*
*
* @return level at which files should be sync'ed to disk.
*/
public static int fileSyncLevel()
{
return Integer.getInteger(FILE_SYNC_LEVEL_PROP_NAME, FILE_SYNC_LEVEL_DEFAULT);
}
/**
* The name of the {@link TimerServiceSupplier} to use for supplying the {@link TimerService}.
*
* @return {@link #TIMER_SERVICE_SUPPLIER_DEFAULT} or system property.
* {@link #TIMER_SERVICE_SUPPLIER_PROP_NAME} if set.
*/
public static String timerServiceSupplier()
{
return System.getProperty(TIMER_SERVICE_SUPPLIER_PROP_NAME, TIMER_SERVICE_SUPPLIER_DEFAULT);
}
/**
* The name to be used for the {@link Agent#roleName()} for the consensus module agent.
*
* @return name to be used for the {@link Agent#roleName()} for the consensus module agent.
* @see #CLUSTER_CONSENSUS_MODULE_AGENT_ROLE_NAME_PROP_NAME
*/
public static String agentRoleName()
{
return System.getProperty(CLUSTER_CONSENSUS_MODULE_AGENT_ROLE_NAME_PROP_NAME);
}
/**
* The amount of time to wait to time out an archive replication when progress has stalled.
*
* @return system property {@link #CLUSTER_REPLICATION_PROGRESS_TIMEOUT_PROP_NAME} or
* {@link #CLUSTER_REPLICATION_PROGRESS_TIMEOUT_DEFAULT_NS}.
* @since 1.41.0
*/
public static long replicationProgressTimeoutNs()
{
return SystemUtil.getDurationInNanos(
CLUSTER_REPLICATION_PROGRESS_TIMEOUT_PROP_NAME, CLUSTER_REPLICATION_PROGRESS_TIMEOUT_DEFAULT_NS);
}
/**
* Interval between checks for progress on an archive replication.
*
* @return system property {@link #CLUSTER_REPLICATION_PROGRESS_INTERVAL_PROP_NAME} or {@link Aeron#NULL_VALUE}
* if not set.
* @since 1.41.0
*/
public static long replicationProgressIntervalNs()
{
return SystemUtil.getDurationInNanos(CLUSTER_REPLICATION_PROGRESS_INTERVAL_PROP_NAME, Aeron.NULL_VALUE);
}
/**
* If this node should accept snapshots from standby nodes.
*
* @return value from property {@link #CLUSTER_ACCEPT_STANDBY_SNAPSHOTS_PROP_NAME} or false if not set.
*/
public static boolean acceptStandbySnapshots()
{
return Boolean.getBoolean(CLUSTER_ACCEPT_STANDBY_SNAPSHOTS_PROP_NAME);
}
/**
* Create a new {@link ConsensusModuleExtension} based on the configured
* {@link #CONSENSUS_MODULE_EXTENSION_CLASS_NAME_PROP_NAME}.
*
* @return a new {@link ConsensusModuleExtension} based on the configured
* {@link #CONSENSUS_MODULE_EXTENSION_CLASS_NAME_PROP_NAME}.
*/
public static ConsensusModuleExtension newConsensusModuleExtension()
{
final String className = System.getProperty(Configuration.CONSENSUS_MODULE_EXTENSION_CLASS_NAME_PROP_NAME);
if (null != className)
{
try
{
return (ConsensusModuleExtension)Class.forName(className).getConstructor().newInstance();
}
catch (final Exception ex)
{
LangUtil.rethrowUnchecked(ex);
}
}
return null;
}
}
/**
* Programmable overrides for configuring the {@link ConsensusModule} in a cluster.
*
* The context will be owned by {@link ConsensusModuleAgent} after a successful
* {@link ConsensusModule#launch(Context)} and closed via {@link ConsensusModule#close()}.
*/
public static final class Context implements Cloneable
{
/**
* Using an integer because there is no support for boolean. 1 is concluded, 0 is not concluded.
*/
private static final AtomicIntegerFieldUpdater IS_CONCLUDED_UPDATER = newUpdater(
Context.class, "isConcluded");
private volatile int isConcluded;
private boolean ownsAeronClient = false;
private String aeronDirectoryName = CommonContext.getAeronDirectoryName();
private Aeron aeron;
private boolean deleteDirOnStart = false;
private String clusterDirectoryName = ClusteredServiceContainer.Configuration.clusterDirName();
private String clusterServicesDirectoryName = ClusteredServiceContainer.Configuration.clusterServicesDirName();
private File clusterDir;
private File markFileDir;
private RecordingLog recordingLog;
private ClusterMarkFile markFile;
private NodeStateFile nodeStateFile;
private int fileSyncLevel = Archive.Configuration.fileSyncLevel();
private int appVersion = SemanticVersion.compose(0, 0, 1);
private int clusterId = ClusteredServiceContainer.Configuration.clusterId();
private int clusterMemberId = Configuration.clusterMemberId();
private int appointedLeaderId = Configuration.appointedLeaderId();
private String clusterMembers = Configuration.clusterMembers();
private String clusterConsensusEndpoints = Configuration.clusterConsensusEndpoints();
private boolean clusterMembersIgnoreSnapshot = Configuration.clusterMembersIgnoreSnapshot();
private String ingressChannel = AeronCluster.Configuration.ingressChannel();
private int ingressStreamId = AeronCluster.Configuration.ingressStreamId();
private boolean isIpcIngressAllowed = Configuration.isIpcIngressAllowed();
private int ingressFragmentLimit = Configuration.ingressFragmentLimit();
private String egressChannel = AeronCluster.Configuration.egressChannel();
private String logChannel = Configuration.logChannel();
private int logStreamId = Configuration.logStreamId();
private String memberEndpoints = Configuration.memberEndpoints();
private String replayChannel = ClusteredServiceContainer.Configuration.replayChannel();
private int replayStreamId = ClusteredServiceContainer.Configuration.replayStreamId();
private String controlChannel = ClusteredServiceContainer.Configuration.controlChannel();
private int consensusModuleStreamId = ClusteredServiceContainer.Configuration.consensusModuleStreamId();
private int serviceStreamId = ClusteredServiceContainer.Configuration.serviceStreamId();
private String snapshotChannel = Configuration.snapshotChannel();
private int snapshotStreamId = Configuration.snapshotStreamId();
private String consensusChannel = Configuration.consensusChannel();
private int consensusStreamId = Configuration.consensusStreamId();
private String replicationChannel = Configuration.replicationChannel();
private String followerCatchupChannel = Configuration.followerCatchupChannel();
private String leaderArchiveControlChannel = Configuration.leaderArchiveControlChannel();
private int logFragmentLimit = ClusteredServiceContainer.Configuration.logFragmentLimit();
private int serviceCount = Configuration.serviceCount();
private int errorBufferLength = Configuration.errorBufferLength();
private int maxConcurrentSessions = Configuration.maxConcurrentSessions();
private int ticksPerWheel = Configuration.ticksPerWheel();
private long wheelTickResolutionNs = Configuration.wheelTickResolutionNs();
private long sessionTimeoutNs = Configuration.sessionTimeoutNs();
private long leaderHeartbeatTimeoutNs = Configuration.leaderHeartbeatTimeoutNs();
private long leaderHeartbeatIntervalNs = Configuration.leaderHeartbeatIntervalNs();
private long startupCanvassTimeoutNs = Configuration.startupCanvassTimeoutNs();
private long electionTimeoutNs = Configuration.electionTimeoutNs();
private long electionStatusIntervalNs = Configuration.electionStatusIntervalNs();
private long terminationTimeoutNs = Configuration.terminationTimeoutNs();
private long cycleThresholdNs = Configuration.cycleThresholdNs();
private long totalSnapshotDurationThresholdNs = Configuration.totalSnapshotDurationThresholdNs();
private String agentRoleName = Configuration.agentRoleName();
private ThreadFactory threadFactory;
private Supplier idleStrategySupplier;
private ClusterClock clusterClock;
private EpochClock epochClock;
private Random random;
private TimerServiceSupplier timerServiceSupplier;
private Function clusterTimeConsumerSupplier;
private ConsensusModuleExtension consensusModuleExtension;
private DistinctErrorLog errorLog;
private ErrorHandler errorHandler;
private AtomicCounter errorCounter;
private CountedErrorHandler countedErrorHandler;
private Counter moduleStateCounter;
private Counter electionStateCounter;
private Counter clusterNodeRoleCounter;
private Counter commitPosition;
private Counter clusterControlToggle;
private Counter nodeControlToggle;
private Counter snapshotCounter;
private Counter timedOutClientCounter;
private Counter standbySnapshotCounter;
private Counter electionCounter;
private Counter leadershipTermId;
private ShutdownSignalBarrier shutdownSignalBarrier;
private Runnable terminationHook;
private AeronArchive.Context archiveContext;
private AuthenticatorSupplier authenticatorSupplier;
private AuthorisationServiceSupplier authorisationServiceSupplier;
private LogPublisher logPublisher;
private EgressPublisher egressPublisher;
private DutyCycleTracker dutyCycleTracker;
private SnapshotDurationTracker totalSnapshotDurationTracker;
private AppVersionValidator appVersionValidator;
private boolean isLogMdc;
private boolean useAgentInvoker = false;
private ConsensusModuleStateExport boostrapState = null;
private boolean acceptStandbySnapshots = Configuration.acceptStandbySnapshots();
/**
* Perform a shallow copy of the object.
*
* @return a shallow copy of the object.
*/
public Context clone()
{
try
{
return (Context)super.clone();
}
catch (final CloneNotSupportedException ex)
{
throw new RuntimeException(ex);
}
}
/**
* Conclude configuration by setting up defaults when specifics are not provided.
*/
@SuppressWarnings("MethodLength")
public void conclude()
{
if (0 != IS_CONCLUDED_UPDATER.getAndSet(this, 1))
{
throw new ConcurrentConcludeException();
}
validateLogChannel();
if (null == clusterDir)
{
clusterDir = new File(clusterDirectoryName);
}
if (null == markFileDir)
{
final String dir = ClusteredServiceContainer.Configuration.markFileDir();
markFileDir = Strings.isEmpty(dir) ? clusterDir : new File(dir);
}
try
{
clusterDir = clusterDir.getCanonicalFile();
clusterDirectoryName = clusterDir.getAbsolutePath();
markFileDir = markFileDir.getCanonicalFile();
if (Strings.isEmpty(clusterServicesDirectoryName))
{
clusterServicesDirectoryName = clusterDirectoryName;
}
else
{
clusterServicesDirectoryName = new File(clusterServicesDirectoryName).getCanonicalPath();
}
}
catch (final IOException ex)
{
throw new UncheckedIOException(ex);
}
if (null == clusterMembers)
{
throw new ClusterException("ConsensusModule.Context.clusterMembers must be set");
}
if (deleteDirOnStart)
{
IoUtil.delete(clusterDir, false);
}
IoUtil.ensureDirectoryExists(clusterDir, "cluster");
IoUtil.ensureDirectoryExists(markFileDir, "mark file");
if (startupCanvassTimeoutNs / leaderHeartbeatTimeoutNs < 2)
{
throw new ClusterException("startupCanvassTimeoutNs=" + startupCanvassTimeoutNs +
" must be a multiple of leaderHeartbeatTimeoutNs=" + leaderHeartbeatTimeoutNs);
}
if (null == clusterClock)
{
final String clockClassName = System.getProperty(
CLUSTER_CLOCK_PROP_NAME, MillisecondClusterClock.class.getName());
try
{
clusterClock = (ClusterClock)Class.forName(clockClassName).getConstructor().newInstance();
}
catch (final Exception e)
{
throw new ClusterException("failed to instantiate ClusterClock " + clockClassName, e);
}
}
if (null == epochClock)
{
epochClock = SystemEpochClock.INSTANCE;
}
if (null == appVersionValidator)
{
appVersionValidator = AppVersionValidator.SEMANTIC_VERSIONING_VALIDATOR;
}
if (null == clusterTimeConsumerSupplier)
{
clusterTimeConsumerSupplier = (ctx) -> (timestamp) -> {};
}
if (null == markFile)
{
markFile = new ClusterMarkFile(
new File(markFileDir, ClusterMarkFile.FILENAME),
ClusterComponentType.CONSENSUS_MODULE,
errorBufferLength,
epochClock,
ClusteredServiceContainer.Configuration.LIVENESS_TIMEOUT_MS);
}
MarkFile.ensureMarkFileLink(
clusterDir,
new File(markFile.parentDirectory(), ClusterMarkFile.FILENAME),
ClusterMarkFile.LINK_FILENAME);
if (null == nodeStateFile)
{
try
{
nodeStateFile = new NodeStateFile(clusterDir, true, fileSyncLevel());
}
catch (final IOException ex)
{
throw new ClusterException("unable to create node-state file", ex);
}
}
if (Aeron.NULL_VALUE == nodeStateFile.candidateTerm().candidateTermId() &&
Aeron.NULL_VALUE != markFile.candidateTermId())
{
nodeStateFile.updateCandidateTermId(markFile.candidateTermId(), Aeron.NULL_VALUE, epochClock.time());
}
if (null == errorLog)
{
errorLog = new DistinctErrorLog(markFile.errorBuffer(), epochClock, StandardCharsets.US_ASCII);
}
errorHandler = CommonContext.setupErrorHandler(errorHandler, errorLog);
if (null == recordingLog)
{
recordingLog = new RecordingLog(clusterDir, true);
}
if (Strings.isEmpty(agentRoleName))
{
agentRoleName = "consensus-module-" + clusterId + "-" + clusterMemberId;
}
final ExpandableArrayBuffer buffer = new ExpandableArrayBuffer();
if (null == aeron)
{
ownsAeronClient = true;
aeron = Aeron.connect(
new Aeron.Context()
.aeronDirectoryName(aeronDirectoryName)
.errorHandler(errorHandler)
.epochClock(epochClock)
.useConductorAgentInvoker(true)
.subscriberErrorHandler(RethrowingErrorHandler.INSTANCE)
.awaitingIdleStrategy(YieldingIdleStrategy.INSTANCE)
.clientLock(NoOpLock.INSTANCE)
.clientName(agentRoleName));
if (null == errorCounter)
{
errorCounter = ClusterCounters.allocateVersioned(
aeron,
buffer,
"Cluster Errors",
CONSENSUS_MODULE_ERROR_COUNT_TYPE_ID,
clusterId,
ConsensusModuleVersion.VERSION,
ConsensusModuleVersion.GIT_SHA);
}
}
else
{
if (!aeron.context().useConductorAgentInvoker())
{
throw new ClusterException(
"Supplied Aeron client instance must set Aeron.Context.useConductorInvoker(true)");
}
}
if (null == ingressChannel)
{
throw new ClusterException("ingressChannel must be specified");
}
if (!(aeron.context().subscriberErrorHandler() instanceof RethrowingErrorHandler))
{
throw new ClusterException("Aeron client must use a RethrowingErrorHandler");
}
if (null == aeron.conductorAgentInvoker())
{
throw new ClusterException("Aeron client must use conductor agent invoker");
}
if (null == errorCounter)
{
throw new ClusterException("error counter must be supplied if aeron client is");
}
if (null == countedErrorHandler)
{
countedErrorHandler = new CountedErrorHandler(errorHandler, errorCounter);
if (ownsAeronClient)
{
aeron.context().errorHandler(countedErrorHandler);
}
}
if (null == moduleStateCounter)
{
final CountersReader counters = aeron.countersReader();
if (Aeron.NULL_VALUE != ClusterCounters.find(counters, CONSENSUS_MODULE_STATE_TYPE_ID, clusterId))
{
throw new ClusterException("existing consensus module detected for clusterId=" + clusterId);
}
moduleStateCounter = ClusterCounters.allocate(
aeron, buffer, "Consensus Module state", CONSENSUS_MODULE_STATE_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, moduleStateCounter, CONSENSUS_MODULE_STATE_TYPE_ID);
if (null == electionStateCounter)
{
electionStateCounter = ClusterCounters.allocate(
aeron, buffer, "Cluster election state", ELECTION_STATE_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, electionStateCounter, ELECTION_STATE_TYPE_ID);
if (null == electionCounter)
{
electionCounter = ClusterCounters.allocate(
aeron, buffer, "Cluster election count", CLUSTER_ELECTION_COUNT_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, electionCounter, CLUSTER_ELECTION_COUNT_TYPE_ID);
if (null == leadershipTermId)
{
leadershipTermId = ClusterCounters.allocate(
aeron, buffer, "Cluster leadership term id", CLUSTER_LEADERSHIP_TERM_ID_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, leadershipTermId, CLUSTER_LEADERSHIP_TERM_ID_TYPE_ID);
if (null == clusterNodeRoleCounter)
{
clusterNodeRoleCounter = ClusterCounters.allocate(
aeron, buffer, "Cluster node role", CLUSTER_NODE_ROLE_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, clusterNodeRoleCounter, CLUSTER_NODE_ROLE_TYPE_ID);
if (null == commitPosition)
{
commitPosition = ClusterCounters.allocate(
aeron, buffer, "Cluster commit-pos:", COMMIT_POSITION_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, commitPosition, COMMIT_POSITION_TYPE_ID);
if (null == clusterControlToggle)
{
clusterControlToggle = ClusterCounters.allocate(
aeron, buffer, "Cluster control toggle", CONTROL_TOGGLE_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, clusterControlToggle, CONTROL_TOGGLE_TYPE_ID);
if (null == nodeControlToggle)
{
nodeControlToggle = ClusterCounters.allocate(
aeron, buffer, "Node control toggle", NODE_CONTROL_TOGGLE_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, nodeControlToggle, NODE_CONTROL_TOGGLE_TYPE_ID);
if (null == snapshotCounter)
{
snapshotCounter = ClusterCounters.allocate(
aeron, buffer, "Cluster snapshot count", SNAPSHOT_COUNTER_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, snapshotCounter, SNAPSHOT_COUNTER_TYPE_ID);
if (null == timedOutClientCounter)
{
timedOutClientCounter = ClusterCounters.allocate(
aeron, buffer, "Cluster timed out client count", CLUSTER_CLIENT_TIMEOUT_COUNT_TYPE_ID, clusterId);
}
validateCounterTypeId(aeron, timedOutClientCounter, CLUSTER_CLIENT_TIMEOUT_COUNT_TYPE_ID);
// TODO: Disable with configuration... (Mike)
if (acceptStandbySnapshots)
{
if (null == standbySnapshotCounter)
{
standbySnapshotCounter = ClusterCounters.allocate(
aeron,
buffer,
"Cluster standby snapshots received",
CLUSTER_STANDBY_SNAPSHOT_COUNTER_TYPE_ID,
clusterId);
}
validateCounterTypeId(aeron, standbySnapshotCounter, CLUSTER_STANDBY_SNAPSHOT_COUNTER_TYPE_ID);
}
if (null == dutyCycleTracker)
{
dutyCycleTracker = new DutyCycleStallTracker(
ClusterCounters.allocate(
aeron, buffer, "Cluster max cycle time in ns",
AeronCounters.CLUSTER_MAX_CYCLE_TIME_TYPE_ID, clusterId),
ClusterCounters.allocate(
aeron, buffer, "Cluster work cycle time exceeded count: threshold=" + cycleThresholdNs + "ns",
AeronCounters.CLUSTER_CYCLE_TIME_THRESHOLD_EXCEEDED_TYPE_ID, clusterId),
cycleThresholdNs);
}
if (null == totalSnapshotDurationTracker)
{
totalSnapshotDurationTracker = new SnapshotDurationTracker(
ClusterCounters.allocate(
aeron,
buffer,
"Total max snapshot duration in ns",
AeronCounters.CLUSTER_TOTAL_MAX_SNAPSHOT_DURATION_TYPE_ID,
clusterId),
ClusterCounters.allocate(
aeron,
buffer,
"Total max snapshot duration exceeded count: threshold=" +
totalSnapshotDurationThresholdNs + "ns",
AeronCounters.CLUSTER_TOTAL_SNAPSHOT_DURATION_THRESHOLD_EXCEEDED_TYPE_ID,
clusterId),
totalSnapshotDurationThresholdNs);
}
if (null == threadFactory)
{
threadFactory = Thread::new;
}
if (null == idleStrategySupplier)
{
idleStrategySupplier = ClusteredServiceContainer.Configuration.idleStrategySupplier(null);
}
if (null == timerServiceSupplier)
{
timerServiceSupplier = getTimerServiceSupplierFromSystemProperty();
}
if (null == archiveContext)
{
archiveContext = new AeronArchive.Context()
.controlRequestChannel(AeronArchive.Configuration.localControlChannel())
.controlResponseChannel(AeronArchive.Configuration.localControlChannel())
.controlRequestStreamId(AeronArchive.Configuration.localControlStreamId());
}
if (!archiveContext.controlRequestChannel().startsWith(CommonContext.IPC_CHANNEL))
{
throw new ClusterException("local archive control must be IPC");
}
if (!archiveContext.controlResponseChannel().startsWith(CommonContext.IPC_CHANNEL))
{
throw new ClusterException("local archive control must be IPC");
}
if (null == replicationChannel)
{
throw new ClusterException("replicationChannel must be set");
}
archiveContext
.aeron(aeron)
.errorHandler(countedErrorHandler)
.ownsAeronClient(false)
.lock(NoOpLock.INSTANCE);
if (null == shutdownSignalBarrier)
{
shutdownSignalBarrier = new ShutdownSignalBarrier();
}
if (null == terminationHook)
{
terminationHook = () -> shutdownSignalBarrier.signalAll();
}
if (null == authenticatorSupplier)
{
authenticatorSupplier = Configuration.authenticatorSupplier();
}
if (null == authorisationServiceSupplier)
{
authorisationServiceSupplier = Configuration.authorisationServiceSupplier();
}
if (null == random)
{
random = new Random();
}
if (null == logPublisher)
{
logPublisher = new LogPublisher(logChannel());
}
if (null == egressPublisher)
{
egressPublisher = new EgressPublisher();
}
final ChannelUri channelUri = ChannelUri.parse(logChannel());
isLogMdc = channelUri.isUdp() && null == channelUri.get(ENDPOINT_PARAM_NAME);
if (null == consensusModuleExtension)
{
consensusModuleExtension = Configuration.newConsensusModuleExtension();
}
if (null != consensusModuleExtension && 0 != serviceCount)
{
throw new ClusterException("Service count must be zero when ConsensusModuleExtension installed");
}
concludeMarkFile();
if (io.aeron.driver.Configuration.printConfigurationOnStart())
{
System.out.println(this);
}
}
/**
* Has the context had the {@link #conclude()} method called.
*
* @return true of the {@link #conclude()} method has been called.
*/
public boolean isConcluded()
{
return 1 == isConcluded;
}
/**
* Should the consensus module attempt to immediately delete {@link #clusterDir()} on startup.
*
* @param deleteDirOnStart Attempt deletion.
* @return this for a fluent API.
*/
public Context deleteDirOnStart(final boolean deleteDirOnStart)
{
this.deleteDirOnStart = deleteDirOnStart;
return this;
}
/**
* Will the consensus module attempt to immediately delete {@link #clusterDir()} on startup.
*
* @return true when directory will be deleted, otherwise false.
*/
public boolean deleteDirOnStart()
{
return deleteDirOnStart;
}
/**
* Set the directory name to use for the consensus module directory.
*
* @param clusterDirectoryName to use.
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_DIR_PROP_NAME
*/
public Context clusterDirectoryName(final String clusterDirectoryName)
{
this.clusterDirectoryName = clusterDirectoryName;
return this;
}
/**
* The directory name to use for the consensus module directory.
*
* @return directory name for the consensus module directory.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_DIR_PROP_NAME
*/
public String clusterDirectoryName()
{
return clusterDirectoryName;
}
/**
* Set the directory used by clustered service containers. By default, the cluster services will share a
* directory with the consensus module. However, sometimes it will be useful for these values to be different.
* Setting this value allows that location to be tracked in the ClusterMarkFile to allow for the ClusterTool to
* resolve their location when using action like describe or errors. There is an
* expectation that all clustered service containers will use the same directory.
*
* @param clusterServicesDirectoryName to use.
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_SERVICES_DIR_PROP_NAME
*/
public Context clusterServicesDirectoryName(final String clusterServicesDirectoryName)
{
this.clusterServicesDirectoryName = clusterServicesDirectoryName;
return this;
}
/**
* The directory used by the clustered service containers.
*
* @return directory used by the clustered service containers.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_SERVICES_DIR_PROP_NAME
* @see #clusterServicesDirectoryName(String)
*/
@Config(id = "CLUSTER_SERVICES_DIR")
public String clusterServicesDirectoryName()
{
return clusterServicesDirectoryName;
}
/**
* Set the directory to use for the consensus module directory.
*
* @param clusterDir to use.
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_DIR_PROP_NAME
*/
public Context clusterDir(final File clusterDir)
{
this.clusterDir = clusterDir;
return this;
}
/**
* The directory used for the consensus module directory.
*
* @return directory for the consensus module directory.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_DIR_PROP_NAME
*/
public File clusterDir()
{
return clusterDir;
}
/**
* Get the directory in which the ConsensusModule will store mark file (i.e. {@code cluster-mark.dat}). It
* defaults to {@link #clusterDir()} if it is not set explicitly via the
* {@link io.aeron.cluster.service.ClusteredServiceContainer.Configuration#MARK_FILE_DIR_PROP_NAME}.
*
* @return the directory in which the ConsensusModule will store mark file (i.e. {@code cluster-mark.dat}).
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#MARK_FILE_DIR_PROP_NAME
* @see #clusterDir()
*/
public File markFileDir()
{
return markFileDir;
}
/**
* Set the directory in which the ConsensusModule will store mark file (i.e. {@code cluster-mark.dat}).
*
* @param markFileDir the directory in which the Archive will store mark file (i.e. {@code cluster-mark.dat}).
* @return this for a fluent API.
*/
public ConsensusModule.Context markFileDir(final File markFileDir)
{
this.markFileDir = markFileDir;
return this;
}
/**
* Set the {@link RecordingLog} for the log terms and snapshots.
*
* @param recordingLog to use.
* @return this for a fluent API.
*/
public Context recordingLog(final RecordingLog recordingLog)
{
this.recordingLog = recordingLog;
return this;
}
/**
* The {@link RecordingLog} for the log terms and snapshots.
*
* @return {@link RecordingLog} for the log terms and snapshots.
*/
public RecordingLog recordingLog()
{
return recordingLog;
}
/**
* User assigned application version which appended to the log as the appVersion in new leadership events.
*
* This can be validated using {@link org.agrona.SemanticVersion} to ensure only application nodes of the same
* major version communicate with each other.
*
* @param appVersion for user application.
* @return this for a fluent API.
*/
public Context appVersion(final int appVersion)
{
this.appVersion = appVersion;
return this;
}
/**
* User assigned application version which appended to the log as the appVersion in new leadership events.
*
* This can be validated using {@link org.agrona.SemanticVersion} to ensure only application nodes of the same
* major version communicate with each other.
*
* @return appVersion for user application.
*/
public int appVersion()
{
return appVersion;
}
/**
* User assigned application version validator implementation used to check version compatibility.
*
* The default validator uses {@link org.agrona.SemanticVersion} semantics.
*
* @param appVersionValidator for user application.
* @return this for fluent API.
*/
public Context appVersionValidator(final AppVersionValidator appVersionValidator)
{
this.appVersionValidator = appVersionValidator;
return this;
}
/**
* User assigned application version validator implementation used to check version compatibility.
*
* The default is to use {@link org.agrona.SemanticVersion} major version for checking compatibility.
*
* @return AppVersionValidator in use.
*/
public AppVersionValidator appVersionValidator()
{
return appVersionValidator;
}
/**
* Get level at which files should be sync'ed to disk.
*
* - 0 - normal writes.
* - 1 - sync file data.
* - 2 - sync file data + metadata.
*
*
* @return the level to be applied for file write.
* @see Configuration#FILE_SYNC_LEVEL_PROP_NAME
*/
@Config
int fileSyncLevel()
{
return fileSyncLevel;
}
/**
* Set level at which files should be sync'ed to disk.
*
* - 0 - normal writes.
* - 1 - sync file data.
* - 2 - sync file data + metadata.
*
*
* @param syncLevel to be applied for file writes.
* @return this for a fluent API.
* @see Configuration#FILE_SYNC_LEVEL_PROP_NAME
*/
public Context fileSyncLevel(final int syncLevel)
{
this.fileSyncLevel = syncLevel;
return this;
}
/**
* Set the id for this cluster instance. This must match with the service containers.
*
* @param clusterId for this clustered instance.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_ID_PROP_NAME
*/
public Context clusterId(final int clusterId)
{
this.clusterId = clusterId;
return this;
}
/**
* Get the id for this cluster instance. This must match with the service containers.
*
* @return the id for this cluster instance.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CLUSTER_ID_PROP_NAME
*/
public int clusterId()
{
return clusterId;
}
/**
* This cluster member identity.
*
* @param clusterMemberId for this member.
* @return this for a fluent API.
* @see Configuration#CLUSTER_MEMBER_ID_PROP_NAME
*/
public Context clusterMemberId(final int clusterMemberId)
{
this.clusterMemberId = clusterMemberId;
return this;
}
/**
* This cluster member identity.
*
* @return this cluster member identity.
* @see Configuration#CLUSTER_MEMBER_ID_PROP_NAME
*/
@Config
public int clusterMemberId()
{
return clusterMemberId;
}
/**
* The cluster member id of the appointed cluster leader.
*
* -1 means no leader has been appointed and an automated leader election should occur.
*
* This feature is for testing and not recommended for production usage.
*
* @param appointedLeaderId for the cluster.
* @return this for a fluent API.
* @see Configuration#APPOINTED_LEADER_ID_PROP_NAME
*/
public Context appointedLeaderId(final int appointedLeaderId)
{
this.appointedLeaderId = appointedLeaderId;
return this;
}
/**
* The cluster member id of the appointed cluster leader.
*
* -1 means no leader has been appointed and an automated leader election should occur.
*
* @return cluster member id of the appointed cluster leader.
* @see Configuration#APPOINTED_LEADER_ID_PROP_NAME
*/
@Config
public int appointedLeaderId()
{
return appointedLeaderId;
}
/**
* String representing the cluster members.
*
*
* 0,ingress:port,consensus:port,log:port,catchup:port,archive:port| \
* 1,ingress:port,consensus:port,log:port,catchup:port,archive:port| ...
*
*
* The ingress endpoints will be used as the endpoint substituted into the {@link #ingressChannel()}
* if the endpoint is not provided unless it is multicast.
*
* @param clusterMembers which are all candidates to be leader.
* @return this for a fluent API.
* @see Configuration#CLUSTER_MEMBERS_PROP_NAME
*/
public Context clusterMembers(final String clusterMembers)
{
this.clusterMembers = clusterMembers;
return this;
}
/**
* The endpoints representing members of the cluster which are all candidates to be leader.
*
* The ingress endpoints will be used as the endpoint in {@link #ingressChannel()} if the endpoint is
* not provided in that when it is not multicast.
*
* @return members of the cluster which are all candidates to be leader.
* @see Configuration#CLUSTER_MEMBERS_PROP_NAME
*/
@Config
public String clusterMembers()
{
return clusterMembers;
}
/**
* String representing the cluster members consensus endpoints used to request to join the cluster.
*
* {@code "endpoint,endpoint,endpoint"}
*
* @param endpoints which are to be contacted for joining the cluster.
* @return this for a fluent API.
* @see Configuration#CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME
* @deprecated the dynamic join feature that uses this configuration in order to join the cluster is now
* deprecated and will be removed in a later release.
*/
@Deprecated
public Context clusterConsensusEndpoints(final String endpoints)
{
this.clusterConsensusEndpoints = endpoints;
return this;
}
/**
* The endpoints representing cluster members of the cluster to attempt to contact to join the cluster.
*
* @return members of the cluster to attempt to request to join from.
* @see Configuration#CLUSTER_CONSENSUS_ENDPOINTS_PROP_NAME
* @deprecated the dynamic join feature that uses this configuration in order to join the cluster is now
* deprecated and will be removed in a later release.
*/
@Deprecated
@Config
public String clusterConsensusEndpoints()
{
return clusterConsensusEndpoints;
}
/**
* Whether the cluster members in the snapshot should be ignored or not.
*
* @param ignore or not the cluster members in the snapshot.
* @return this for a fluent API.
* @see Configuration#CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME
*/
public Context clusterMembersIgnoreSnapshot(final boolean ignore)
{
this.clusterMembersIgnoreSnapshot = ignore;
return this;
}
/**
* Whether the cluster members in the snapshot should be ignored or not.
*
* @return ignore or not the cluster members in the snapshot.
* @see Configuration#CLUSTER_MEMBERS_IGNORE_SNAPSHOT_PROP_NAME
*/
@Config
public boolean clusterMembersIgnoreSnapshot()
{
return clusterMembersIgnoreSnapshot;
}
/**
* Set the channel parameter for the ingress channel.
*
* @param channel parameter for the ingress channel.
* @return this for a fluent API.
* @see io.aeron.cluster.client.AeronCluster.Configuration#INGRESS_CHANNEL_PROP_NAME
*/
public Context ingressChannel(final String channel)
{
ingressChannel = channel;
return this;
}
/**
* Get the channel parameter for the ingress channel.
*
* @return the channel parameter for the ingress channel.
* @see io.aeron.cluster.client.AeronCluster.Configuration#INGRESS_CHANNEL_PROP_NAME
*/
public String ingressChannel()
{
return ingressChannel;
}
/**
* Set the stream id for the ingress channel.
*
* @param streamId for the ingress channel.
* @return this for a fluent API
* @see io.aeron.cluster.client.AeronCluster.Configuration#INGRESS_STREAM_ID_PROP_NAME
*/
public Context ingressStreamId(final int streamId)
{
ingressStreamId = streamId;
return this;
}
/**
* Get the stream id for the ingress channel.
*
* @return the stream id for the ingress channel.
* @see io.aeron.cluster.client.AeronCluster.Configuration#INGRESS_STREAM_ID_PROP_NAME
*/
public int ingressStreamId()
{
return ingressStreamId;
}
/**
* Set limit for fragments to be consumed on each poll of ingress.
*
* @param ingressFragmentLimit for the ingress channel.
* @return this for a fluent API
* @see Configuration#CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME
*/
public Context ingressFragmentLimit(final int ingressFragmentLimit)
{
this.ingressFragmentLimit = ingressFragmentLimit;
return this;
}
/**
* The limit for fragments to be consumed on each poll of ingress.
*
* @return the limit for fragments to be consumed on each poll of ingress.
* @see Configuration#CLUSTER_INGRESS_FRAGMENT_LIMIT_PROP_NAME
*/
@Config(id = "CLUSTER_INGRESS_FRAGMENT_LIMIT")
public int ingressFragmentLimit()
{
return ingressFragmentLimit;
}
/**
* Set the template channel that is used to refine the response channels for a Cluster, i.e. egress channel,
* backup query response channel, heartbeat response channel etc. Defaults to {@code null} in which case the
* provided response channel will be used. The main use-case is the ability to define the `interface` parameter
* to steer the response traffic via a specific network interface.
*
*
Note: each URI parameter that is defined on this channel template will
* override the corresponding URI parameter in the provided response channel!
*
* @param channel the channel template for response channels.
* @return this for a fluent API.
* @see io.aeron.cluster.client.AeronCluster.Configuration#EGRESS_CHANNEL_PROP_NAME
*/
public Context egressChannel(final String channel)
{
egressChannel = channel;
return this;
}
/**
* Get the template channel that is used to refine the response channels for a Cluster, i.e. egress channel,
* backup query response channel, heartbeat response channel etc. Defaults to {@code null} in which case the
* provided response channel will be used. The main use-case is the ability to define the `interface` parameter
* to steer the response traffic via a specific network interface.
*
*
Note: each URI parameter that is defined on this channel template will
* override the corresponding URI parameter in the provided response channel!
*
* @return the channel template for response channels.
* @see io.aeron.cluster.client.AeronCluster.Configuration#EGRESS_CHANNEL_PROP_NAME
*/
public String egressChannel()
{
return egressChannel;
}
/**
* Set whether IPC ingress is allowed or not.
*
* @param isIpcIngressAllowed or not.
* @return this for a fluent API
* @see Configuration#CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME
*/
public Context isIpcIngressAllowed(final boolean isIpcIngressAllowed)
{
this.isIpcIngressAllowed = isIpcIngressAllowed;
return this;
}
/**
* Get whether IPC ingress is allowed or not.
*
* @return whether IPC ingress is allowed or not.
* @see Configuration#CLUSTER_INGRESS_IPC_ALLOWED_PROP_NAME
*/
@Config(id = "CLUSTER_INGRESS_IPC_ALLOWED")
public boolean isIpcIngressAllowed()
{
return isIpcIngressAllowed;
}
/**
* Set the channel parameter for the cluster log channel.
*
* @param channel parameter for the cluster log channel.
* @return this for a fluent API.
* @see Configuration#LOG_CHANNEL_PROP_NAME
*/
public Context logChannel(final String channel)
{
logChannel = channel;
return this;
}
/**
* Get the channel parameter for the cluster log channel.
*
* @return the channel parameter for the cluster channel.
* @see Configuration#LOG_CHANNEL_PROP_NAME
*/
@Config
public String logChannel()
{
return logChannel;
}
/**
* Set the stream id for the cluster log channel.
*
* @param streamId for the cluster log channel.
* @return this for a fluent API
* @see Configuration#LOG_STREAM_ID_PROP_NAME
*/
public Context logStreamId(final int streamId)
{
logStreamId = streamId;
return this;
}
/**
* Get the stream id for the cluster log channel.
*
* @return the stream id for the cluster log channel.
* @see Configuration#LOG_STREAM_ID_PROP_NAME
*/
public int logStreamId()
{
return logStreamId;
}
/**
* Set the endpoints for this cluster node.
*
* @param endpoints for the cluster node.
* @return this for a fluent API.
* @see Configuration#MEMBER_ENDPOINTS_PROP_NAME
*/
public Context memberEndpoints(final String endpoints)
{
memberEndpoints = endpoints;
return this;
}
/**
* Get the endpoints for this cluster node.
*
* @return the endpoints for the cluster node.
* @see Configuration#MEMBER_ENDPOINTS_PROP_NAME
*/
@Config
public String memberEndpoints()
{
return memberEndpoints;
}
/**
* Set the channel parameter for the cluster log and snapshot replay channel.
*
* @param channel parameter for the cluster log replay channel.
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#REPLAY_CHANNEL_PROP_NAME
*/
public Context replayChannel(final String channel)
{
replayChannel = channel;
return this;
}
/**
* Get the channel parameter for the cluster log and snapshot replay channel.
*
* @return the channel parameter for the cluster replay channel.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#REPLAY_CHANNEL_PROP_NAME
*/
public String replayChannel()
{
return replayChannel;
}
/**
* Set the stream id for the cluster log and snapshot replay channel.
*
* @param streamId for the cluster log replay channel.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#REPLAY_STREAM_ID_PROP_NAME
*/
public Context replayStreamId(final int streamId)
{
replayStreamId = streamId;
return this;
}
/**
* Get the stream id for the cluster log and snapshot replay channel.
*
* @return the stream id for the cluster log replay channel.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#REPLAY_STREAM_ID_PROP_NAME
*/
public int replayStreamId()
{
return replayStreamId;
}
/**
* Set the channel parameter for bidirectional communications between the consensus module and services.
*
* @param channel parameter for bidirectional communications between the consensus module and services.
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CONTROL_CHANNEL_PROP_NAME
*/
public Context controlChannel(final String channel)
{
controlChannel = channel;
return this;
}
/**
* Get the channel parameter for bidirectional communications between the consensus module and services.
*
* @return the channel parameter for bidirectional communications between the consensus module and services.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CONTROL_CHANNEL_PROP_NAME
*/
public String controlChannel()
{
return controlChannel;
}
/**
* Set the stream id for communications from the consensus module and to the services.
*
* @param streamId for communications from the consensus module and to the services.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SERVICE_STREAM_ID_PROP_NAME
*/
public Context serviceStreamId(final int streamId)
{
serviceStreamId = streamId;
return this;
}
/**
* Get the stream id for communications from the consensus module and to the services.
*
* @return the stream id for communications from the consensus module and to the services.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SERVICE_STREAM_ID_PROP_NAME
*/
public int serviceStreamId()
{
return serviceStreamId;
}
/**
* Set the stream id for communications from the services to the consensus module.
*
* @param streamId for communications from the services to the consensus module.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CONSENSUS_MODULE_STREAM_ID_PROP_NAME
*/
public Context consensusModuleStreamId(final int streamId)
{
consensusModuleStreamId = streamId;
return this;
}
/**
* Get the stream id for communications from the services to the consensus module.
*
* @return the stream id for communications from the services to the consensus module.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#CONSENSUS_MODULE_STREAM_ID_PROP_NAME
*/
public int consensusModuleStreamId()
{
return consensusModuleStreamId;
}
/**
* Set the channel parameter for snapshot recordings.
*
* @param channel parameter for snapshot recordings
* @return this for a fluent API.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_CHANNEL_PROP_NAME
*/
public Context snapshotChannel(final String channel)
{
snapshotChannel = channel;
return this;
}
/**
* Get the channel parameter for snapshot recordings.
*
* @return the channel parameter for snapshot recordings.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_CHANNEL_PROP_NAME
*/
public String snapshotChannel()
{
return snapshotChannel;
}
/**
* Set the stream id for snapshot recordings.
*
* @param streamId for snapshot recordings.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_STREAM_ID_PROP_NAME
*/
public Context snapshotStreamId(final int streamId)
{
snapshotStreamId = streamId;
return this;
}
/**
* Get the stream id for snapshot recordings.
*
* @return the stream id for snapshot recordings.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SNAPSHOT_STREAM_ID_PROP_NAME
*/
public int snapshotStreamId()
{
return snapshotStreamId;
}
/**
* Set the channel parameter for the consensus communication channel.
*
* @param channel parameter for the consensus communication channel.
* @return this for a fluent API.
* @see Configuration#CONSENSUS_CHANNEL_PROP_NAME
*/
public Context consensusChannel(final String channel)
{
consensusChannel = channel;
return this;
}
/**
* Get the channel parameter for the consensus communication channel.
*
* @return the channel parameter for the consensus communication channel.
* @see Configuration#CONSENSUS_CHANNEL_PROP_NAME
*/
@Config
public String consensusChannel()
{
return consensusChannel;
}
/**
* Set the stream id for the consensus channel.
*
* @param streamId for the consensus channel.
* @return this for a fluent API
* @see Configuration#CONSENSUS_STREAM_ID_PROP_NAME
*/
public Context consensusStreamId(final int streamId)
{
consensusStreamId = streamId;
return this;
}
/**
* Get the stream id for the consensus channel.
*
* @return the stream id for the consensus channel.
* @see Configuration#CONSENSUS_STREAM_ID_PROP_NAME
*/
@Config
public int consensusStreamId()
{
return consensusStreamId;
}
/**
* Set the channel parameter for the replication communication channel. This is channel that the local
* will send to src archives to replay their recordings back to. It should contain an endpoint that other
* members in the cluster will be able to reach. Using port 0 for the endpoint is valid for this channel to
* simplify port allocation.
*
* @param channel replication communication channel to be used by the consensus module.
* @return this for a fluent API
* @see Configuration#REPLICATION_CHANNEL_PROP_NAME
*/
public Context replicationChannel(final String channel)
{
replicationChannel = channel;
return this;
}
/**
* Get the replication channel for logs and snapshots.
*
* @return channel to receive replication responses from other node's archives when using log and snapshot
* replication to catch up.
* @see Configuration#REPLICATION_CHANNEL_PROP_NAME
*/
@Config
public String replicationChannel()
{
return replicationChannel;
}
/**
* Set a channel template used for replaying logs to a follower using the
* {@link ClusterMember#catchupEndpoint()}.
*
* @param channel to do a catch replay to a follower.
* @return this for a fluent API
* @see Configuration#FOLLOWER_CATCHUP_CHANNEL_PROP_NAME
*/
public Context followerCatchupChannel(final String channel)
{
this.followerCatchupChannel = channel;
return this;
}
/**
* Gets the channel template used for replaying logs to a follower using the
* {@link ClusterMember#catchupEndpoint()}.
*
* @return channel used for replaying older data during a catchup phase.
* @see Configuration#FOLLOWER_CATCHUP_CHANNEL_PROP_NAME
*/
@Config
public String followerCatchupChannel()
{
return followerCatchupChannel;
}
/**
* Set a channel template used to build the control request channel for the leader Archive using the
* {@link ClusterMember#archiveEndpoint()}.
*
* @param channel for the Archive control requests.
* @return this for a fluent API
* @see Configuration#LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME
*/
public Context leaderArchiveControlChannel(final String channel)
{
this.leaderArchiveControlChannel = channel;
return this;
}
/**
* Gets the channel template used to build the control request channel for the leader Archive using the
* {@link ClusterMember#archiveEndpoint()}.
*
* @return channel used for replaying older data during a catchup phase.
* @see Configuration#LEADER_ARCHIVE_CONTROL_CHANNEL_PROP_NAME
*/
@Config
public String leaderArchiveControlChannel()
{
return leaderArchiveControlChannel;
}
/**
* Set the fragment limit to be used when polling the log {@link Subscription}.
*
* @param logFragmentLimit for this clustered service.
* @return this for a fluent API
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#LOG_FRAGMENT_LIMIT_DEFAULT
*/
public Context logFragmentLimit(final int logFragmentLimit)
{
this.logFragmentLimit = logFragmentLimit;
return this;
}
/**
* Get the fragment limit to be used when polling the log {@link Subscription}.
*
* @return the fragment limit to be used when polling the log {@link Subscription}.
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#LOG_FRAGMENT_LIMIT_PROP_NAME
*/
public int logFragmentLimit()
{
return logFragmentLimit;
}
/**
* Resolution in nanoseconds for each tick of the timer wheel for scheduling deadlines.
*
* @param wheelTickResolutionNs the resolution in nanoseconds of each tick on the timer wheel.
* @return this for a fluent API
* @see Configuration#WHEEL_TICK_RESOLUTION_PROP_NAME
*/
public Context wheelTickResolutionNs(final long wheelTickResolutionNs)
{
this.wheelTickResolutionNs = wheelTickResolutionNs;
return this;
}
/**
* Resolution in nanoseconds for each tick of the timer wheel for scheduling deadlines.
*
* @return the resolution in nanoseconds for each tick on the timer wheel.
* @see Configuration#WHEEL_TICK_RESOLUTION_PROP_NAME
*/
@Config
public long wheelTickResolutionNs()
{
return wheelTickResolutionNs;
}
/**
* Number of ticks, or spokes, on the timer wheel. Higher number of ticks reduces potential conflicts
* traded off against memory usage.
*
* @param ticksPerWheel the number of ticks on the timer wheel.
* @return this for a fluent API
* @see Configuration#TICKS_PER_WHEEL_PROP_NAME
*/
public Context ticksPerWheel(final int ticksPerWheel)
{
this.ticksPerWheel = ticksPerWheel;
return this;
}
/**
* Number of ticks, or spokes, on the timer wheel. Higher number of ticks reduces potential conflicts
* traded off against memory usage.
*
* @return the number of ticks on the timer wheel.
* @see Configuration#TICKS_PER_WHEEL_PROP_NAME
*/
@Config
public int ticksPerWheel()
{
return ticksPerWheel;
}
/**
* Set the number of clustered services in this cluster instance.
*
* @param serviceCount the number of clustered services in this cluster instance.
* @return this for a fluent API
* @see Configuration#SERVICE_COUNT_PROP_NAME
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SERVICE_ID_PROP_NAME
*/
public Context serviceCount(final int serviceCount)
{
this.serviceCount = serviceCount;
return this;
}
/**
* Get the number of clustered services in this cluster instance.
*
* @return the number of clustered services in this cluster instance.
* @see Configuration#SERVICE_COUNT_PROP_NAME
* @see io.aeron.cluster.service.ClusteredServiceContainer.Configuration#SERVICE_ID_PROP_NAME
*/
@Config
public int serviceCount()
{
return serviceCount;
}
/**
* Set the limit for the maximum number of concurrent cluster sessions.
*
* @param maxSessions after which new sessions will be rejected.
* @return this for a fluent API
* @see Configuration#MAX_CONCURRENT_SESSIONS_PROP_NAME
*/
public Context maxConcurrentSessions(final int maxSessions)
{
this.maxConcurrentSessions = maxSessions;
return this;
}
/**
* Get the limit for the maximum number of concurrent cluster sessions.
*
* @return the limit for the maximum number of concurrent cluster sessions.
* @see Configuration#MAX_CONCURRENT_SESSIONS_PROP_NAME
*/
@Config
public int maxConcurrentSessions()
{
return maxConcurrentSessions;
}
/**
* Timeout for a session if no activity is observed.
*
* @param sessionTimeoutNs to wait for activity on a session.
* @return this for a fluent API.
* @see Configuration#SESSION_TIMEOUT_PROP_NAME
*/
public Context sessionTimeoutNs(final long sessionTimeoutNs)
{
this.sessionTimeoutNs = sessionTimeoutNs;
return this;
}
/**
* Timeout for a session if no activity is observed.
*
* @return the timeout for a session if no activity is observed.
* @see Configuration#SESSION_TIMEOUT_PROP_NAME
*/
@Config
public long sessionTimeoutNs()
{
return CommonContext.checkDebugTimeout(sessionTimeoutNs, TimeUnit.NANOSECONDS);
}
/**
* Timeout for a leader if no heartbeat is received by another member.
*
* @param heartbeatTimeoutNs to wait for heartbeat from a leader.
* @return this for a fluent API.
* @see Configuration#LEADER_HEARTBEAT_TIMEOUT_PROP_NAME
*/
public Context leaderHeartbeatTimeoutNs(final long heartbeatTimeoutNs)
{
this.leaderHeartbeatTimeoutNs = heartbeatTimeoutNs;
return this;
}
/**
* Timeout for a leader if no heartbeat is received by another member.
*
* @return the timeout for a leader if no heartbeat is received by another member.
* @see Configuration#LEADER_HEARTBEAT_TIMEOUT_PROP_NAME
*/
@Config
public long leaderHeartbeatTimeoutNs()
{
return leaderHeartbeatTimeoutNs;
}
/**
* Interval at which a leader will send heartbeats if the log is not progressing.
*
* @param heartbeatIntervalNs between leader heartbeats.
* @return this for a fluent API.
* @see Configuration#LEADER_HEARTBEAT_INTERVAL_PROP_NAME
*/
@Config
public Context leaderHeartbeatIntervalNs(final long heartbeatIntervalNs)
{
this.leaderHeartbeatIntervalNs = heartbeatIntervalNs;
return this;
}
/**
* Interval at which a leader will send heartbeats if the log is not progressing.
*
* @return the interval at which a leader will send heartbeats if the log is not progressing.
* @see Configuration#LEADER_HEARTBEAT_INTERVAL_PROP_NAME
*/
public long leaderHeartbeatIntervalNs()
{
return leaderHeartbeatIntervalNs;
}
/**
* Timeout to wait for hearing the status of all cluster members on startup after recovery before commencing
* an election if a majority of members has been heard from.
*
* @param timeoutNs to wait on startup after recovery before commencing an election.
* @return this for a fluent API.
* @see Configuration#STARTUP_CANVASS_TIMEOUT_PROP_NAME
*/
public Context startupCanvassTimeoutNs(final long timeoutNs)
{
this.startupCanvassTimeoutNs = timeoutNs;
return this;
}
/**
* Timeout to wait for hearing the status of all cluster members on startup after recovery before commencing
* an election if a majority of members has been heard from.
*
* @return the timeout to wait on startup after recovery before commencing an election.
* @see Configuration#STARTUP_CANVASS_TIMEOUT_PROP_NAME
*/
@Config
public long startupCanvassTimeoutNs()
{
return startupCanvassTimeoutNs;
}
/**
* Timeout to wait for votes in an election before declaring the election void and starting over.
*
* @param timeoutNs to wait for votes in an elections.
* @return this for a fluent API.
* @see Configuration#ELECTION_TIMEOUT_PROP_NAME
*/
public Context electionTimeoutNs(final long timeoutNs)
{
this.electionTimeoutNs = timeoutNs;
return this;
}
/**
* Timeout to wait for votes in an election before declaring the election void and starting over.
*
* @return the timeout to wait for votes in an elections.
* @see Configuration#ELECTION_TIMEOUT_PROP_NAME
*/
@Config
public long electionTimeoutNs()
{
return electionTimeoutNs;
}
/**
* Interval at which a member will send out status messages during the election phases.
*
* @param electionStatusIntervalNs between status message updates.
* @return this for a fluent API.
* @see Configuration#ELECTION_STATUS_INTERVAL_PROP_NAME
* @see Configuration#ELECTION_STATUS_INTERVAL_DEFAULT_NS
*/
public Context electionStatusIntervalNs(final long electionStatusIntervalNs)
{
this.electionStatusIntervalNs = electionStatusIntervalNs;
return this;
}
/**
* Interval at which a member will send out status messages during the election phases.
*
* @return the interval at which a member will send out status messages during the election phases.
* @see Configuration#ELECTION_STATUS_INTERVAL_PROP_NAME
* @see Configuration#ELECTION_STATUS_INTERVAL_DEFAULT_NS
*/
@Config
public long electionStatusIntervalNs()
{
return electionStatusIntervalNs;
}
/**
* Timeout to wait for follower termination by leader.
*
* @param terminationTimeoutNs to wait for follower termination.
* @return this for a fluent API.
* @see Configuration#TERMINATION_TIMEOUT_PROP_NAME
* @see Configuration#TERMINATION_TIMEOUT_DEFAULT_NS
*/
public Context terminationTimeoutNs(final long terminationTimeoutNs)
{
this.terminationTimeoutNs = terminationTimeoutNs;
return this;
}
/**
* Timeout to wait for follower termination by leader.
*
* @return timeout to wait for follower termination by leader.
* @see Configuration#TERMINATION_TIMEOUT_PROP_NAME
* @see Configuration#TERMINATION_TIMEOUT_DEFAULT_NS
*/
@Config
public long terminationTimeoutNs()
{
return terminationTimeoutNs;
}
/**
* Set a threshold for the consensus module agent work cycle time which when exceed it will increment the
* counter.
*
* @param thresholdNs value in nanoseconds
* @return this for fluent API.
* @see Configuration#CYCLE_THRESHOLD_PROP_NAME
* @see Configuration#CYCLE_THRESHOLD_DEFAULT_NS
*/
public Context cycleThresholdNs(final long thresholdNs)
{
this.cycleThresholdNs = thresholdNs;
return this;
}
/**
* Threshold for the consensus module agent work cycle time which when exceed it will increment the
* counter.
*
* @return threshold to track for the consensus module agent work cycle time.
*/
@Config
public long cycleThresholdNs()
{
return cycleThresholdNs;
}
/**
* Set a duty cycle tracker to be used for tracking the duty cycle time of the consensus module agent.
*
* @param dutyCycleTracker to use for tracking.
* @return this for fluent API.
*/
public Context dutyCycleTracker(final DutyCycleTracker dutyCycleTracker)
{
this.dutyCycleTracker = dutyCycleTracker;
return this;
}
/**
* The duty cycle tracker used to track the consensus module agent duty cycle.
*
* @return the duty cycle tracker.
*/
public DutyCycleTracker dutyCycleTracker()
{
return dutyCycleTracker;
}
/**
* Set a threshold for total snapshot duration which when exceeded will result in a counter increment.
*
* @param thresholdNs value in nanoseconds
* @return this for fluent API.
* @see ConsensusModule.Configuration#TOTAL_SNAPSHOT_DURATION_THRESHOLD_PROP_NAME
* @see ConsensusModule.Configuration#TOTAL_SNAPSHOT_DURATION_THRESHOLD_DEFAULT_NS
*/
public Context totalSnapshotDurationThresholdNs(final long thresholdNs)
{
this.totalSnapshotDurationThresholdNs = thresholdNs;
return this;
}
/**
* Threshold for total snapshot duration which when exceeded it will increment the counter.
*
* @return threshold value in nanoseconds.
*/
@Config
public long totalSnapshotDurationThresholdNs()
{
return totalSnapshotDurationThresholdNs;
}
/**
* Set snapshot duration tracker used for monitoring total snapshot duration.
*
* @param snapshotDurationTracker snapshot duration tracker.
* @return this for fluent API.
*/
public Context totalSnapshotDurationTracker(final SnapshotDurationTracker snapshotDurationTracker)
{
this.totalSnapshotDurationTracker = snapshotDurationTracker;
return this;
}
/**
* Get snapshot duration tracker used for monitoring total snapshot duration.
*
* @return snapshot duration tracker
*/
public SnapshotDurationTracker totalSnapshotDurationTracker()
{
return totalSnapshotDurationTracker;
}
/**
* Get the {@link Agent#roleName()} to be used for the consensus module agent. If {@code null} then one will
* be generated.
*
* @return the {@link Agent#roleName()} to be used for the consensus module agent.
*/
@Config(id = "CLUSTER_CONSENSUS_MODULE_AGENT_ROLE_NAME")
public String agentRoleName()
{
return agentRoleName;
}
/**
* Set the {@link Agent#roleName()} to be used for the consensus module agent.
*
* @param agentRoleName to be used for the consensus module agent.
* @return this for a fluent API.
*/
public Context agentRoleName(final String agentRoleName)
{
this.agentRoleName = agentRoleName;
return this;
}
/**
* Get the thread factory used for creating threads.
*
* @return thread factory used for creating threads.
*/
public ThreadFactory threadFactory()
{
return threadFactory;
}
/**
* Set the thread factory used for creating threads.
*
* @param threadFactory used for creating threads
* @return this for a fluent API.
*/
public Context threadFactory(final ThreadFactory threadFactory)
{
this.threadFactory = threadFactory;
return this;
}
/**
* Provides an {@link IdleStrategy} supplier for the idle strategy for the agent duty cycle.
*
* @param idleStrategySupplier supplier for the idle strategy for the agent duty cycle.
* @return this for a fluent API.
*/
public Context idleStrategySupplier(final Supplier idleStrategySupplier)
{
this.idleStrategySupplier = idleStrategySupplier;
return this;
}
/**
* Get a new {@link IdleStrategy} based on configured supplier.
*
* @return a new {@link IdleStrategy} based on configured supplier.
*/
public IdleStrategy idleStrategy()
{
return idleStrategySupplier.get();
}
/**
* Set the {@link ClusterClock} to be used for timestamping messages and timers.
*
* @param clock {@link ClusterClock} to be used for timestamping messages and timers.
* @return this for a fluent API.
*/
public Context clusterClock(final ClusterClock clock)
{
this.clusterClock = clock;
return this;
}
/**
* Get the {@link ClusterClock} to used for timestamping messages and timers.
*
* @return the {@link ClusterClock} to used for timestamping messages and timers.
*/
@Config
public ClusterClock clusterClock()
{
return clusterClock;
}
/**
* Set the {@link EpochClock} to be used for tracking wall clock time.
*
* @param clock {@link EpochClock} to be used for tracking wall clock time.
* @return this for a fluent API.
*/
public Context epochClock(final EpochClock clock)
{
this.epochClock = clock;
return this;
}
/**
* Get the {@link EpochClock} to used for tracking wall clock time.
*
* @return the {@link EpochClock} to used for tracking wall clock time.
*/
public EpochClock epochClock()
{
return epochClock;
}
/**
* Set the supplier of a consumer of timestamps which can be used for testing time progress in a cluster. The
* timestamp passed to the consumer is the timestamp of the last completed {@link Agent#doWork()} cycle by the
* consensus module {@link Agent}. The supplier will be called after the context is concluded and can be
* referenced during the construction of the consumer.
*
* @param clusterTimeConsumerSupplier to which the latest timestamp will be passed.
* @return this for a fluent API
*/
public Context clusterTimeConsumerSupplier(final Function clusterTimeConsumerSupplier)
{
this.clusterTimeConsumerSupplier = clusterTimeConsumerSupplier;
return this;
}
/**
* Get the supplier of a consumer of timestamps which can be used for testing time progress in a cluster. The
* timestamp passed to the consumer is the timestamp of the last completed {@link Agent#doWork()} cycle by the
* consensus module {@link Agent}.
*
* @return the consumer of timestamps for completed work cycles.
*/
public Function clusterTimeConsumerSupplier()
{
return clusterTimeConsumerSupplier;
}
/**
* Get the {@link ErrorHandler} to be used by the Consensus Module.
*
* @return the {@link ErrorHandler} to be used by the Consensus Module.
*/
public ErrorHandler errorHandler()
{
return errorHandler;
}
/**
* Set the {@link ErrorHandler} to be used by the Consensus Module.
*
* @param errorHandler the error handler to be used by the Consensus Module.
* @return this for a fluent API
*/
public Context errorHandler(final ErrorHandler errorHandler)
{
this.errorHandler = errorHandler;
return this;
}
/**
* The {@link #errorHandler()} that will increment {@link #errorCounter()} by default.
*
* @param countedErrorHandler to override the default.
* @return this for a fluent API.
*/
public Context countedErrorHandler(final CountedErrorHandler countedErrorHandler)
{
this.countedErrorHandler = countedErrorHandler;
return this;
}
/**
* The {@link #errorHandler()} that will increment {@link #errorCounter()} by default.
*
* @return {@link #errorHandler()} that will increment {@link #errorCounter()} by default.
*/
public CountedErrorHandler countedErrorHandler()
{
return countedErrorHandler;
}
/**
* Get the error counter that will record the number of errors observed.
*
* @return the error counter that will record the number of errors observed.
*/
public AtomicCounter errorCounter()
{
return errorCounter;
}
/**
* Set the error counter that will record the number of errors observed.
*
* @param errorCounter the error counter that will record the number of errors observed.
* @return this for a fluent API.
*/
public Context errorCounter(final AtomicCounter errorCounter)
{
this.errorCounter = errorCounter;
return this;
}
/**
* Get the counter for the current state of the consensus module.
*
* @return the counter for the current state of the consensus module.
* @see ConsensusModule.State
*/
public Counter moduleStateCounter()
{
return moduleStateCounter;
}
/**
* Set the counter for the current state of the consensus module.
*
* @param moduleState the counter for the current state of the consensus module.
* @return this for a fluent API.
* @see ConsensusModule.State
*/
public Context moduleStateCounter(final Counter moduleState)
{
this.moduleStateCounter = moduleState;
return this;
}
/**
* Get the counter for the current state of an election.
*
* @return the counter for the current state of an election.
* @see ElectionState
*/
public Counter electionStateCounter()
{
return electionStateCounter;
}
/**
* Set the counter for the current state of an election.
*
* @param electionStateCounter for the current state of an election.
* @return this for a fluent API.
* @see ElectionState
*/
public Context electionStateCounter(final Counter electionStateCounter)
{
this.electionStateCounter = electionStateCounter;
return this;
}
/**
* Get the counter for the commit position the cluster has reached for consensus.
*
* @return the counter for the commit position the cluster has reached for consensus.
*/
public Counter commitPositionCounter()
{
return commitPosition;
}
/**
* Set the counter for the commit position the cluster has reached for consensus.
*
* @param commitPosition counter for the commit position the cluster has reached for consensus.
* @return this for a fluent API.
*/
public Context commitPositionCounter(final Counter commitPosition)
{
this.commitPosition = commitPosition;
return this;
}
/**
* Get the counter for representing the current {@link io.aeron.cluster.service.Cluster.Role} of the
* consensus module node.
*
* @return the counter for representing the current {@link io.aeron.cluster.service.Cluster.Role} of the
* cluster node.
* @see io.aeron.cluster.service.Cluster.Role
*/
public Counter clusterNodeRoleCounter()
{
return clusterNodeRoleCounter;
}
/**
* Set the counter for representing the current {@link io.aeron.cluster.service.Cluster.Role} of the
* cluster node.
*
* @param nodeRole the counter for representing the current {@link io.aeron.cluster.service.Cluster.Role}
* of the cluster node.
* @return this for a fluent API.
* @see io.aeron.cluster.service.Cluster.Role
*/
public Context clusterNodeRoleCounter(final Counter nodeRole)
{
this.clusterNodeRoleCounter = nodeRole;
return this;
}
/**
* Get the counter for the control toggle for triggering actions for the cluster.
*
* @return the counter for triggering cluster node actions.
* @see ClusterControl
*/
public Counter controlToggleCounter()
{
return clusterControlToggle;
}
/**
* Set the counter for the control toggle for triggering actions for the cluster.
*
* @param controlToggle the counter for triggering cluster node actions.
* @return this for a fluent API.
* @see ClusterControl
*/
public Context controlToggleCounter(final Counter controlToggle)
{
this.clusterControlToggle = controlToggle;
return this;
}
/**
* Get the counter for the control toggle for triggering actions on the cluster node.
*
* @return the counter for triggering cluster node actions.
* @see ClusterControl
*/
public Counter nodeControlToggleCounter()
{
return nodeControlToggle;
}
/**
* Set the counter for the control toggle for triggering actions on the cluster node.
*
* @param nodeControlToggle the counter for triggering cluster node actions.
* @return this for a fluent API.
* @see ClusterControl
*/
public Context nodeControlToggleCounter(final Counter nodeControlToggle)
{
this.nodeControlToggle = nodeControlToggle;
return this;
}
/**
* Get the counter for the count of snapshots taken.
*
* @return the counter for the count of snapshots taken.
*/
public Counter snapshotCounter()
{
return snapshotCounter;
}
/**
* Set the counter for the count of snapshots taken.
*
* @param snapshotCounter the count of snapshots taken.
* @return this for a fluent API.
*/
public Context snapshotCounter(final Counter snapshotCounter)
{
this.snapshotCounter = snapshotCounter;
return this;
}
/**
* Get the counter for the count of clients that have been timed out and disconnected.
*
* @return the counter for the count of clients that have been timed out and disconnected.
*/
public Counter timedOutClientCounter()
{
return timedOutClientCounter;
}
/**
* Set the counter for the count of clients that have been timed out and disconnected.
*
* @param timedOutClientCounter the count of clients that have been timed out and disconnected.
* @return this for a fluent API.
*/
public Context timedOutClientCounter(final Counter timedOutClientCounter)
{
this.timedOutClientCounter = timedOutClientCounter;
return this;
}
/**
* {@link Aeron} client for communicating with the local Media Driver.
*
* This client will be closed when the {@link ConsensusModule#close()} or {@link #close()} methods are called
* if {@link #ownsAeronClient()} is true.
*
* The method is mostly here in order to test with a custom Aeron instance. The recommended approach is to set
* the aeronDirectoryName and allow the ConsensusModule to construct the Aeron client. The supplied Aeron
* client must be set to {@code Aeron.Context.useConductorInvoker(true)} to work correctly with the
* ConsensusModule.
*
* @param aeron client for communicating with the local Media Driver.
* @return this for a fluent API.
* @see io.aeron.Aeron#connect()
* @see io.aeron.Aeron.Context#useConductorAgentInvoker(boolean)
* @see #aeronDirectoryName(String)
*/
public Context aeron(final Aeron aeron)
{
this.aeron = aeron;
return this;
}
/**
* {@link Aeron} client for communicating with the local Media Driver.
*
* If not provided then a default will be established during {@link #conclude()} by calling
* {@link Aeron#connect()}.
*
* @return client for communicating with the local Media Driver.
*/
public Aeron aeron()
{
return aeron;
}
/**
* Set the top level Aeron directory used for communication between the Aeron client and Media Driver.
*
* @param aeronDirectoryName the top level Aeron directory.
* @return this for a fluent API.
*/
public Context aeronDirectoryName(final String aeronDirectoryName)
{
this.aeronDirectoryName = aeronDirectoryName;
return this;
}
/**
* Get the top level Aeron directory used for communication between the Aeron client and Media Driver.
*
* @return The top level Aeron directory.
*/
public String aeronDirectoryName()
{
return aeronDirectoryName;
}
/**
* Does this context own the {@link #aeron()} client and this takes responsibility for closing it?
*
* @param ownsAeronClient does this context own the {@link #aeron()} client.
* @return this for a fluent API.
*/
public Context ownsAeronClient(final boolean ownsAeronClient)
{
this.ownsAeronClient = ownsAeronClient;
return this;
}
/**
* Does this context own the {@link #aeron()} client and this takes responsibility for closing it?
*
* @return does this context own the {@link #aeron()} client and this takes responsibility for closing it?
*/
public boolean ownsAeronClient()
{
return ownsAeronClient;
}
/**
* Set the {@link io.aeron.archive.client.AeronArchive.Context} that should be used for communicating with the
* local Archive.
*
* @param archiveContext that should be used for communicating with the local Archive.
* @return this for a fluent API.
*/
public Context archiveContext(final AeronArchive.Context archiveContext)
{
this.archiveContext = archiveContext;
return this;
}
/**
* Get the {@link io.aeron.archive.client.AeronArchive.Context} that should be used for communicating with
* the local Archive.
*
* @return the {@link io.aeron.archive.client.AeronArchive.Context} that should be used for communicating
* with the local Archive.
*/
public AeronArchive.Context archiveContext()
{
return archiveContext;
}
/**
* Get the {@link AuthenticatorSupplier} that should be used for the consensus module.
*
* @return the {@link AuthenticatorSupplier} to be used for the consensus module.
*/
@Config
public AuthenticatorSupplier authenticatorSupplier()
{
return authenticatorSupplier;
}
/**
* Set the {@link AuthenticatorSupplier} that will be used for the consensus module.
*
* @param authenticatorSupplier {@link AuthenticatorSupplier} to use for the consensus module.
* @return this for a fluent API.
*/
public Context authenticatorSupplier(final AuthenticatorSupplier authenticatorSupplier)
{
this.authenticatorSupplier = authenticatorSupplier;
return this;
}
/**
* Get the {@link AuthorisationServiceSupplier} that should be used for the consensus module.
*
* @return the {@link AuthorisationServiceSupplier} to be used for the consensus module.
*/
@Config
public AuthorisationServiceSupplier authorisationServiceSupplier()
{
return authorisationServiceSupplier;
}
/**
*
Set the {@link AuthorisationServiceSupplier} that will be used for the consensus module.
*
* When using an authorisation service for the ConsensusModule, then the following values for protocolId,
* actionId, and type should be considered.
*
* Parameters for authorisation service queries from the Consensus Module
*
* Description protocolId actionId type(s)
* Admin requests made through the client API
* {@link MessageHeaderDecoder#SCHEMA_ID}
* {@link AdminRequestDecoder#TEMPLATE_ID}
* {@link AdminRequestType#SNAPSHOT}
*
*
* Backup queries from Cluster Backup & Standby
* {@link BackupQueryDecoder#TEMPLATE_ID}
* (null)
*
* Heartbeat requests from Cluster Standby
* {@link HeartbeatRequestDecoder#TEMPLATE_ID}
* (null)
*
* Standby snapshot notifications from Cluster Standby
* {@link StandbySnapshotDecoder#TEMPLATE_ID}
* (null)
*
*
*
* @param authorisationServiceSupplier {@link AuthorisationServiceSupplier} to use for the consensus module.
* @return this for a fluent API.
*/
public Context authorisationServiceSupplier(final AuthorisationServiceSupplier authorisationServiceSupplier)
{
this.authorisationServiceSupplier = authorisationServiceSupplier;
return this;
}
/**
* Should an {@link AgentInvoker} be used for running the {@link ConsensusModule} rather than run it on
* a thread with a {@link AgentRunner}.
*
* @param useAgentInvoker use {@link AgentInvoker} be used for running the {@link ConsensusModule}?
* @return this for a fluent API.
*/
public ConsensusModule.Context useAgentInvoker(final boolean useAgentInvoker)
{
this.useAgentInvoker = useAgentInvoker;
return this;
}
/**
* Should an {@link AgentInvoker} be used for running the {@link ConsensusModule} rather than run it on
* a thread with a {@link AgentRunner}.
*
* @return true if the {@link ConsensusModule} will be run with an {@link AgentInvoker} otherwise false.
*/
public boolean useAgentInvoker()
{
return useAgentInvoker;
}
/**
* Set the {@link ShutdownSignalBarrier} that can be used to shut down a consensus module.
*
* @param barrier that can be used to shut down a consensus module.
* @return this for a fluent API.
*/
public Context shutdownSignalBarrier(final ShutdownSignalBarrier barrier)
{
shutdownSignalBarrier = barrier;
return this;
}
/**
* Get the {@link ShutdownSignalBarrier} that can be used to shut down a consensus module.
*
* @return the {@link ShutdownSignalBarrier} that can be used to shut down a consensus module.
*/
public ShutdownSignalBarrier shutdownSignalBarrier()
{
return shutdownSignalBarrier;
}
/**
* Set the {@link Runnable} that is called when the {@link ConsensusModule} processes a termination action.
*
* @param terminationHook that can be used to terminate a consensus module.
* @return this for a fluent API.
*/
public Context terminationHook(final Runnable terminationHook)
{
this.terminationHook = terminationHook;
return this;
}
/**
* Get the {@link Runnable} that is called when the {@link ConsensusModule} processes a termination action.
*
* The default action is to call signal on the {@link #shutdownSignalBarrier()}.
*
* @return the {@link Runnable} that can be used to terminate a consensus module.
*/
public Runnable terminationHook()
{
return terminationHook;
}
/**
* Set the {@link ClusterMarkFile} in use.
*
* @param markFile to use.
* @return this for a fluent API.
*/
public Context clusterMarkFile(final ClusterMarkFile markFile)
{
this.markFile = markFile;
return this;
}
/**
* The {@link ClusterMarkFile} in use.
*
* @return {@link ClusterMarkFile} in use.
*/
public ClusterMarkFile clusterMarkFile()
{
return markFile;
}
Context nodeStateFile(final NodeStateFile nodeStateFile)
{
this.nodeStateFile = nodeStateFile;
return this;
}
/**
* The {@link NodeStateFile} that is used by the consensus module.
*
* @return node state file.
*/
public NodeStateFile nodeStateFile()
{
return nodeStateFile;
}
/**
* Set the error buffer length in bytes to use.
*
* @param errorBufferLength in bytes to use.
* @return this for a fluent API.
* @see Configuration#ERROR_BUFFER_LENGTH_PROP_NAME
*/
public Context errorBufferLength(final int errorBufferLength)
{
this.errorBufferLength = errorBufferLength;
return this;
}
/**
* The error buffer length in bytes.
*
* @return error buffer length in bytes.
* @see Configuration#ERROR_BUFFER_LENGTH_PROP_NAME
*/
@Config
public int errorBufferLength()
{
return errorBufferLength;
}
/**
* Set the {@link DistinctErrorLog} in use.
*
* @param errorLog to use.
* @return this for a fluent API.
*/
public Context errorLog(final DistinctErrorLog errorLog)
{
this.errorLog = errorLog;
return this;
}
/**
* The {@link DistinctErrorLog} in use.
*
* @return {@link DistinctErrorLog} in use.
*/
public DistinctErrorLog errorLog()
{
return errorLog;
}
/**
* The source of random values for timeouts used in elections.
*
* @param random source of random values for timeouts used in elections.
* @return this for a fluent API.
*/
public Context random(final Random random)
{
this.random = random;
return this;
}
/**
* The source of random values for timeouts used in elections.
*
* @return source of random values for timeouts used in elections.
*/
public Random random()
{
return random;
}
/**
* Provides an {@link TimerService} supplier for the idle strategy for the agent duty cycle.
*
* @param timerServiceSupplier supplier for the idle strategy for the agent duty cycle.
* @return this for a fluent API.
*/
public Context timerServiceSupplier(final TimerServiceSupplier timerServiceSupplier)
{
this.timerServiceSupplier = timerServiceSupplier;
return this;
}
/**
* Supplier of the {@link TimerService} instances.
*
* @return supplier of dynamically created {@link TimerService} instances.
*/
@Config
public TimerServiceSupplier timerServiceSupplier()
{
return timerServiceSupplier;
}
/**
* Register a ConsensusModuleExtension to extend the behaviour of the
* consensus module instead of using ClusteredServices.
*
* @param consensusModuleExtension supplier for consensus module extension
* @return this for a fluent API.
*/
public Context consensusModuleExtension(final ConsensusModuleExtension consensusModuleExtension)
{
this.consensusModuleExtension = consensusModuleExtension;
return this;
}
/**
* Registered consensus module extension.
*
* @return Registered consensus module extension or null.
*/
public ConsensusModuleExtension consensusModuleExtension()
{
return consensusModuleExtension;
}
/**
* Deprecated for removal.
*
* @return {@code null}.
*/
@Deprecated
public NameResolver nameResolver()
{
return null;
}
/**
* Deprecated for removal.
*
* @param ignore as deprecated.
* @return this for fluent API.
*/
@Deprecated
public Context nameResolver(final NameResolver ignore)
{
return this;
}
/**
* Indicate whether this node should accept snapshots from standby nodes
*
* @return true if this node should accept snapshots from standby nodes, false
* otherwise.
* @see Configuration#CLUSTER_ACCEPT_STANDBY_SNAPSHOTS_PROP_NAME
* @see Configuration#acceptStandbySnapshots()
*/
@Config(id = "CLUSTER_ACCEPT_STANDBY_SNAPSHOTS")
public boolean acceptStandbySnapshots()
{
return acceptStandbySnapshots;
}
/**
* Indicate whether this node should accept snapshots from standby nodes
*
* @param acceptStandbySnapshots true if this node should accept snapshots from standby nodes,
* false otherwise.
* @return this for a fluent API.
* @see Configuration#CLUSTER_ACCEPT_STANDBY_SNAPSHOTS_PROP_NAME
* @see Configuration#acceptStandbySnapshots()
*/
public ConsensusModule.Context acceptStandbySnapshots(final boolean acceptStandbySnapshots)
{
this.acceptStandbySnapshots = acceptStandbySnapshots;
return this;
}
/**
* Get the counter used to track standby snapshots accepted by this node.
*
* @return the counter for standby snapshots.
*/
public Counter standbySnapshotCounter()
{
return standbySnapshotCounter;
}
/**
* Set the counter used to track standby snapshots accepted by this node.
*
* @param standbySnapshotCounter the counter for standby snapshots.
* @return this for a fluentAPI.
*/
public Context standbySnapshotCounter(final Counter standbySnapshotCounter)
{
this.standbySnapshotCounter = standbySnapshotCounter;
return this;
}
/**
* Get the counter used to track the number of elections on this node.
*
* @return the counter for elections.
* @since 1.44.0
*/
public Counter electionCounter()
{
return electionCounter;
}
/**
* Set the counter used to track the number of elections on this node.
*
* @param electionCounter the counter for elections.
* @return this for a fluentAPI.
* @since 1.44.0
*/
public Context electionCounter(final Counter electionCounter)
{
this.electionCounter = electionCounter;
return this;
}
/**
* Get the counter used to track the leadership term id.
*
* @return the counter containing the leadership term id.
* @since 1.44.0
*/
public Counter leadershipTermIdCounter()
{
return leadershipTermId;
}
/**
* Set the counter used to track the number of elections on this node.
*
* @param leadershipTermId the counter for elections.
* @return this for a fluentAPI.
* @since 1.44.0
*/
public Context leadershipTermIdCounter(final Counter leadershipTermId)
{
this.leadershipTermId = leadershipTermId;
return this;
}
/**
* Delete the cluster directory.
*/
public void deleteDirectory()
{
if (null != clusterDir)
{
IoUtil.delete(clusterDir, false);
}
}
/**
* Close the context and free applicable resources.
*
* If {@link #ownsAeronClient()} is true then the {@link #aeron()} client will be closed.
*/
public void close()
{
CloseHelper.close(countedErrorHandler, recordingLog);
CloseHelper.close(countedErrorHandler, nodeStateFile);
CloseHelper.close(countedErrorHandler, markFile);
if (errorHandler instanceof AutoCloseable)
{
CloseHelper.quietClose((AutoCloseable)errorHandler); // Ignore error to ensure the rest will be closed
}
if (ownsAeronClient)
{
CloseHelper.close(aeron);
}
else if (!aeron.isClosed())
{
CloseHelper.closeAll(
timedOutClientCounter,
clusterControlToggle,
snapshotCounter,
moduleStateCounter,
electionStateCounter,
clusterNodeRoleCounter,
commitPosition);
}
}
Context logPublisher(final LogPublisher logPublisher)
{
this.logPublisher = logPublisher;
return this;
}
LogPublisher logPublisher()
{
return logPublisher;
}
Context egressPublisher(final EgressPublisher egressPublisher)
{
this.egressPublisher = egressPublisher;
return this;
}
EgressPublisher egressPublisher()
{
return egressPublisher;
}
boolean isLogMdc()
{
return isLogMdc;
}
/**
* Start up the consensus module using pre-supplied state skipping the recovery process. Internal use only.
*
* @param bootstrapState to initialize the consensus module.
* @return this for a fluent API.
*/
ConsensusModule.Context bootstrapState(final ConsensusModuleStateExport bootstrapState)
{
this.boostrapState = bootstrapState;
return this;
}
ConsensusModuleStateExport boostrapState()
{
return boostrapState;
}
private void concludeMarkFile()
{
final String aeronDirectory = aeron.context().aeronDirectoryName();
final String authenticatorClassName = authenticatorSupplier.getClass().getName();
ClusterMarkFile.checkHeaderLength(
aeronDirectory,
controlChannel(),
ingressChannel,
null,
authenticatorClassName);
markFile.encoder()
.archiveStreamId(archiveContext.controlRequestStreamId())
.serviceStreamId(serviceStreamId)
.consensusModuleStreamId(consensusModuleStreamId)
.ingressStreamId(ingressStreamId)
.memberId(clusterMemberId)
.serviceId(SERVICE_ID)
.clusterId(clusterId)
.aeronDirectory(aeronDirectory)
.controlChannel(controlChannel)
.ingressChannel(ingressChannel)
.serviceName(null)
.authenticator(authenticatorClassName)
.servicesClusterDir(clusterServicesDirectoryName);
markFile.updateActivityTimestamp(epochClock.time());
markFile.signalReady();
markFile.force();
}
private TimerServiceSupplier getTimerServiceSupplierFromSystemProperty()
{
final String timeServiceClassName = Configuration.timerServiceSupplier();
if (WheelTimerServiceSupplier.class.getName().equals(timeServiceClassName))
{
return new WheelTimerServiceSupplier(
clusterClock.timeUnit(),
0,
findNextPositivePowerOfTwo(
clusterClock.timeUnit().convert(wheelTickResolutionNs, TimeUnit.NANOSECONDS)),
ticksPerWheel);
}
else if (PriorityHeapTimerServiceSupplier.class.getName().equals(timeServiceClassName))
{
return new PriorityHeapTimerServiceSupplier();
}
throw new ClusterException("invalid TimerServiceSupplier: " + timeServiceClassName);
}
private void validateLogChannel()
{
final ChannelUri logChannelUri = ChannelUri.parse(logChannel);
if (logChannelUri.containsKey(INITIAL_TERM_ID_PARAM_NAME))
{
throw new ConfigurationException("logChannel must not contain: " + INITIAL_TERM_ID_PARAM_NAME);
}
if (logChannelUri.containsKey(TERM_ID_PARAM_NAME))
{
throw new ConfigurationException("logChannel must not contain: " + TERM_ID_PARAM_NAME);
}
if (logChannelUri.containsKey(TERM_OFFSET_PARAM_NAME))
{
throw new ConfigurationException("logChannel must not contain: " + TERM_OFFSET_PARAM_NAME);
}
}
/**
* {@inheritDoc}
*/
public String toString()
{
return "ConsensusModule.Context" +
"\n{" +
"\n isConcluded=" + isConcluded() +
"\n ownsAeronClient=" + ownsAeronClient +
"\n aeronDirectoryName='" + aeronDirectoryName + '\'' +
"\n aeron=" + aeron +
"\n deleteDirOnStart=" + deleteDirOnStart +
"\n clusterDirectoryName='" + clusterDirectoryName + '\'' +
"\n clusterDir=" + clusterDir +
"\n recordingLog=" + recordingLog +
"\n markFile=" + markFile +
"\n fileSyncLevel=" + fileSyncLevel +
"\n appVersion=" + appVersion +
"\n clusterId=" + clusterId +
"\n clusterMemberId=" + clusterMemberId +
"\n appointedLeaderId=" + appointedLeaderId +
"\n clusterMembers='" + clusterMembers + '\'' +
"\n clusterConsensusEndpoints='" + clusterConsensusEndpoints + '\'' +
"\n clusterMembersIgnoreSnapshot=" + clusterMembersIgnoreSnapshot +
"\n ingressChannel='" + ingressChannel + '\'' +
"\n ingressStreamId=" + ingressStreamId +
"\n ingressFragmentLimit=" + ingressFragmentLimit +
"\n logChannel='" + logChannel + '\'' +
"\n logStreamId=" + logStreamId +
"\n memberEndpoints='" + memberEndpoints + '\'' +
"\n replayChannel='" + replayChannel + '\'' +
"\n replayStreamId=" + replayStreamId +
"\n controlChannel='" + controlChannel + '\'' +
"\n consensusModuleStreamId=" + consensusModuleStreamId +
"\n serviceStreamId=" + serviceStreamId +
"\n snapshotChannel='" + snapshotChannel + '\'' +
"\n snapshotStreamId=" + snapshotStreamId +
"\n consensusChannel='" + consensusChannel + '\'' +
"\n consensusStreamId=" + consensusStreamId +
"\n replicationChannel='" + replicationChannel + '\'' +
"\n logFragmentLimit=" + logFragmentLimit +
"\n serviceCount=" + serviceCount +
"\n errorBufferLength=" + errorBufferLength +
"\n maxConcurrentSessions=" + maxConcurrentSessions +
"\n ticksPerWheel=" + ticksPerWheel +
"\n wheelTickResolutionNs=" + wheelTickResolutionNs +
"\n timerServiceSupplier=" + timerServiceSupplier +
"\n sessionTimeoutNs=" + sessionTimeoutNs +
"\n leaderHeartbeatTimeoutNs=" + leaderHeartbeatTimeoutNs +
"\n leaderHeartbeatIntervalNs=" + leaderHeartbeatIntervalNs +
"\n startupCanvassTimeoutNs=" + startupCanvassTimeoutNs +
"\n electionTimeoutNs=" + electionTimeoutNs +
"\n electionStatusIntervalNs=" + electionStatusIntervalNs +
"\n terminationTimeoutNs=" + terminationTimeoutNs +
"\n threadFactory=" + threadFactory +
"\n idleStrategySupplier=" + idleStrategySupplier +
"\n clusterClock=" + clusterClock +
"\n epochClock=" + epochClock +
"\n random=" + random +
"\n errorLog=" + errorLog +
"\n errorHandler=" + errorHandler +
"\n errorCounter=" + errorCounter +
"\n countedErrorHandler=" + countedErrorHandler +
"\n moduleStateCounter=" + moduleStateCounter +
"\n electionStateCounter=" + electionStateCounter +
"\n clusterNodeRoleCounter=" + clusterNodeRoleCounter +
"\n commitPosition=" + commitPosition +
"\n controlToggle=" + clusterControlToggle +
"\n nodeControlToggle=" + nodeControlToggle +
"\n snapshotCounter=" + snapshotCounter +
"\n timedOutClientCounter=" + timedOutClientCounter +
"\n standbySnapshotCounter=" + standbySnapshotCounter +
"\n electionCounter=" + electionCounter +
"\n leadershipTermId=" + leadershipTermId +
"\n shutdownSignalBarrier=" + shutdownSignalBarrier +
"\n terminationHook=" + terminationHook +
"\n archiveContext=" + archiveContext +
"\n authenticatorSupplier=" + authenticatorSupplier +
"\n authorisationServiceSupplier=" + authorisationServiceSupplier +
"\n logPublisher=" + logPublisher +
"\n egressPublisher=" + egressPublisher +
"\n isLogMdc=" + isLogMdc +
"\n useAgentInvoker=" + useAgentInvoker +
"\n cycleThresholdNs=" + cycleThresholdNs +
"\n dutyCycleTracker=" + dutyCycleTracker +
"\n totalSnapshotDurationThresholdNs=" + totalSnapshotDurationThresholdNs +
"\n totalSnapshotDurationTracker=" + totalSnapshotDurationTracker +
"\n acceptStandbySnapshots=" + acceptStandbySnapshots +
"\n boostrapState=" + boostrapState +
"\n}";
}
}
/**
* {@inheritDoc}
*/
public String toString()
{
return "ConsensusModule{" +
"conductor=" + conductor +
'}';
}
}