com.hazelcast.cp.internal

Class RaftService

java.lang.Object

com.hazelcast.cp.internal.RaftService

All Implemented Interfaces:

CPSubsystemManagementService, SnapshotAwareService<MetadataRaftGroupSnapshot>, RaftNodeLifecycleAwareService, GracefulShutdownAwareService, ManagedService, MembershipAwareService, PreJoinAwareService

public class RaftService
extends Object
implements ManagedService, SnapshotAwareService<MetadataRaftGroupSnapshot>, GracefulShutdownAwareService, MembershipAwareService, CPSubsystemManagementService, PreJoinAwareService, RaftNodeLifecycleAwareService

Contains RaftNode instances that run the Raft consensus algorithm for the created CP groups. Also implements CP subsystem management methods.

Field Summary

Fields

Modifier and Type	Field and Description
static String	SERVICE_NAME

Constructor Summary

Constructors

Constructor and Description
RaftService(NodeEngine nodeEngine)

Method Summary

Modifier and Type	Method and Description
boolean	awaitUntilDiscoveryCompleted(long timeout, TimeUnit timeUnit) Blocks until CP discovery process is completed, or the timeout occurs, or the current thread is interrupted, whichever happens first.
RaftGroupId	createRaftGroupForProxy(String name)
InternalCompletableFuture<RaftGroupId>	createRaftGroupForProxyAsync(String name)
void	createRaftNode(CPGroupId groupId, Collection<CPMemberInfo> members)
void	destroyRaftNode(CPGroupId groupId)
ICompletableFuture<Void>	forceDestroyCPGroup(String groupName) this method is idempotent
ICompletableFuture<Collection<CPGroupId>>	getAllCPGroupIds()
Collection<RaftNode>	getAllRaftNodes()
CPSubsystemConfig	getConfig()
ICompletableFuture<CPGroup>	getCPGroup(CPGroupId groupId)
ICompletableFuture<CPGroup>	getCPGroup(String name) Returns the active CP group with the given name.
ICompletableFuture<Collection<CPGroupId>>	getCPGroupIds() Returns all active CP group ids.
Collection<CPGroupId>	getCPGroupIdsLocally()
CPGroupInfo	getCPGroupLocally(CPGroupId groupId)
ICompletableFuture<Collection<CPMember>>	getCPMembers() Returns the current list of CP members
static String	getGroupNameForProxy(String name)
RaftInvocationManager	getInvocationManager()
CPMemberInfo	getLocalCPMember() Returns the local CP member if this Hazelcast member is part of the CP subsystem, returns null otherwise.
RaftGroupId	getMetadataGroupId()
MetadataRaftGroupManager	getMetadataGroupManager()
static String	getObjectNameForProxy(String name)
RaftNode	getOrInitRaftNode(CPGroupId groupId)
Operation	getPreJoinOperation() An operation to be executed on the joining member before it is set as joined.
RaftNode	getRaftNode(CPGroupId groupId)
void	handleActiveCPMembers(RaftGroupId latestMetadataGroupId, long membersCommitIndex, Collection<CPMemberInfo> members)
void	handleAppendEntries(CPGroupId groupId, AppendRequest request, CPMember target)
void	handleAppendResponse(CPGroupId groupId, AppendFailureResponse response, CPMember target)
void	handleAppendResponse(CPGroupId groupId, AppendSuccessResponse response, CPMember target)
void	handlePreVoteRequest(CPGroupId groupId, PreVoteRequest request, CPMember target)
void	handlePreVoteResponse(CPGroupId groupId, PreVoteResponse response, CPMember target)
void	handleSnapshot(CPGroupId groupId, InstallSnapshot request, CPMember target)
void	handleVoteRequest(CPGroupId groupId, VoteRequest request, CPMember target)
void	handleVoteResponse(CPGroupId groupId, VoteResponse response, CPMember target)
void	init(NodeEngine nodeEngine, Properties properties) Initializes this service.
boolean	isDiscoveryCompleted() Returns whether CP discovery process is completed or not.
boolean	isRaftGroupDestroyed(CPGroupId groupId)
void	memberAdded(MembershipServiceEvent event) Invoked when a new member is added to the cluster.
void	memberAttributeChanged(MemberAttributeServiceEvent event) Invoked when a member attribute is changed.
void	memberRemoved(MembershipServiceEvent event) Invoked when an existing member leaves the cluster.
void	onRaftGroupDestroyed(CPGroupId groupId) Called on the thread of the Raft group when the given Raft node is destroyed, either gracefully or forcefully.
void	onRaftNodeSteppedDown(CPGroupId groupId) Called on the thread of the Raft group when the given Raft node is stepped down, either because it is shutting down, or it could not be added to the Raft group
boolean	onShutdown(long timeout, TimeUnit unit) A hook method that's called during graceful shutdown to provide safety for data managed by this service.
ICompletableFuture<Void>	promoteToCPMember() Promotes the local Hazelcast member to a CP member.
ICompletableFuture<Void>	removeCPMember(String cpMemberUuid) Removes the given unreachable CP member from the active CP members list and all CP groups it belongs to.
void	reset() Resets this service back to its initial state.
ICompletableFuture<Void>	restart() Wipes and resets the whole CP subsystem and initializes it as if the Hazelcast cluster is starting up initially.
void	restartLocal(long seed)
void	restoreSnapshot(CPGroupId groupId, long commitIndex, MetadataRaftGroupSnapshot snapshot) Restores the snapshot for specified CPGroupId.
void	shutdown(boolean terminate) Shuts down this service.
void	stepDownRaftNode(CPGroupId groupId)
MetadataRaftGroupSnapshot	takeSnapshot(CPGroupId groupId, long commitIndex) Creates a snapshot for specified CPGroupId.
static String	withoutDefaultGroupName(String name)

Methods inherited from class java.lang.Object

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

Field Detail

SERVICE_NAME

public static final String SERVICE_NAME

See Also:

Constant Field Values

Constructor Detail

RaftService

public RaftService(NodeEngine nodeEngine)

Method Detail

init

public void init(NodeEngine nodeEngine,
                 Properties properties)

Description copied from interface: ManagedService

Initializes this service.

Specified by:

init in interface ManagedService

Parameters:

nodeEngine - the NodeEngine that this service belongs to

properties - the Properties (can be used to pass settings to the service)

reset

public void reset()

Description copied from interface: ManagedService

Resets this service back to its initial state.

TODO: what is the purpose of reset

Specified by:

reset in interface ManagedService

shutdown

public void shutdown(boolean terminate)

Description copied from interface: ManagedService

Shuts down this service.

TODO: what is the purpose of the terminate variable

Specified by:

shutdown in interface ManagedService

Parameters:

terminate - true to shut down this service

takeSnapshot

public MetadataRaftGroupSnapshot takeSnapshot(CPGroupId groupId,
                                              long commitIndex)

Description copied from interface: SnapshotAwareService

Creates a snapshot for specified CPGroupId. The snapshot object returned from this method will be stored among the Raft log and it can be sent to other Raft nodes anytime. Therefore, this method must not return a snapshot object that can mutate after the takeSnapshot() call.

Specified by:

takeSnapshot in interface SnapshotAwareService<MetadataRaftGroupSnapshot>

Parameters:

groupId - CPGroupId which is snapshot requested for

commitIndex - commitIndex of the taken snapshot

Returns:

snapshot for specified CPGroupId.

restoreSnapshot

public void restoreSnapshot(CPGroupId groupId,
                            long commitIndex,
                            MetadataRaftGroupSnapshot snapshot)

Description copied from interface: SnapshotAwareService

Restores the snapshot for specified CPGroupId.

Specified by:

restoreSnapshot in interface SnapshotAwareService<MetadataRaftGroupSnapshot>

Parameters:

groupId - CPGroupId of the snapshot to be restored

commitIndex - commitIndex of the restored snapshot

snapshot - snapshot for specified CPGroupId

getAllCPGroupIds

public ICompletableFuture<Collection<CPGroupId>> getAllCPGroupIds()

getCPGroupIds

public ICompletableFuture<Collection<CPGroupId>> getCPGroupIds()

Description copied from interface: CPSubsystemManagementService

Returns all active CP group ids.

Specified by:

getCPGroupIds in interface CPSubsystemManagementService

getCPGroup

public ICompletableFuture<CPGroup> getCPGroup(CPGroupId groupId)

getCPGroup

public ICompletableFuture<CPGroup> getCPGroup(String name)

Description copied from interface: CPSubsystemManagementService

Returns the active CP group with the given name. There can be at most one active CP group with a given name.

Specified by:

getCPGroup in interface CPSubsystemManagementService

restart

public ICompletableFuture<Void> restart()

Description copied from interface: CPSubsystemManagementService

Wipes and resets the whole CP subsystem and initializes it as if the Hazelcast cluster is starting up initially. This method must be used only when the Metadata CP group loses its majority and cannot make progress anymore.

After this method is called, all CP state and data are wiped and the CP members start with empty state.

This method can be invoked only from the Hazelcast master member.

This method must not be called while there are membership changes in the cluster. Before calling this method, please make sure that there is no new member joining and all existing Hazelcast members have seen the same member list.

Use with caution: This method is NOT idempotent and multiple invocations can break the whole system! After calling this API, you must observe the system to see if the restart process is successfully completed or failed before making another call.

Specified by:

restart in interface CPSubsystemManagementService

restartLocal

public void restartLocal(long seed)

promoteToCPMember

public ICompletableFuture<Void> promoteToCPMember()

Description copied from interface: CPSubsystemManagementService

Promotes the local Hazelcast member to a CP member.

This method is idempotent. If the local member is already in the active CP members list, then this method has no effect. When the current member is promoted to a CP member, its member UUID is assigned as CP member UUID.

Once the returned Future object is completed, the promoted CP member has been added to the CP groups that have missing members, i.e., whose size is smaller than CPSubsystemConfig.getGroupSize().

If the local member is currently being removed from the active CP members list, then the returned Future object will throw IllegalArgumentException.

If there is an ongoing membership change in the CP subsystem when this method is invoked, then the returned Future object throws IllegalStateException

If the CP subsystem initial discovery process has not completed when this method is invoked, then the returned Future object throws IllegalStateException

Specified by:

promoteToCPMember in interface CPSubsystemManagementService

removeCPMember

public ICompletableFuture<Void> removeCPMember(String cpMemberUuid)

Description copied from interface: CPSubsystemManagementService

Removes the given unreachable CP member from the active CP members list and all CP groups it belongs to. If any other active CP member is available, it will replace the removed CP member in its CP groups. Otherwise, CP groups which the removed CP member is a member of will shrink and their majority values will be recalculated.

Before removing a CP member from the CP subsystem, please make sure that it is declared as unreachable by Hazelcast's failure detector and removed from Hazelcast's member list. The behavior is undefined when a running CP member is removed from the CP subsystem.

Specified by:

removeCPMember in interface CPSubsystemManagementService

forceDestroyCPGroup

public ICompletableFuture<Void> forceDestroyCPGroup(String groupName)

this method is idempotent

Specified by:

forceDestroyCPGroup in interface CPSubsystemManagementService

getCPMembers

public ICompletableFuture<Collection<CPMember>> getCPMembers()

Description copied from interface: CPSubsystemManagementService

Returns the current list of CP members

Specified by:

getCPMembers in interface CPSubsystemManagementService

isDiscoveryCompleted

public boolean isDiscoveryCompleted()

Description copied from interface: CPSubsystemManagementService

Returns whether CP discovery process is completed or not.

Specified by:

isDiscoveryCompleted in interface CPSubsystemManagementService

Returns:

true if CP discovery completed, false otherwise

See Also:

CPSubsystemManagementService.awaitUntilDiscoveryCompleted(long, TimeUnit)

awaitUntilDiscoveryCompleted

public boolean awaitUntilDiscoveryCompleted(long timeout,
                                            TimeUnit timeUnit)
                                     throws InterruptedException

Description copied from interface: CPSubsystemManagementService

Blocks until CP discovery process is completed, or the timeout occurs, or the current thread is interrupted, whichever happens first.

Specified by:

awaitUntilDiscoveryCompleted in interface CPSubsystemManagementService

Parameters:

timeout - maximum time to wait

timeUnit - time unit of the timeout

Returns:

true if CP discovery completed, false otherwise

Throws:

InterruptedException - if interrupted while waiting

See Also:

CPSubsystemManagementService.isDiscoveryCompleted()

onShutdown

public boolean onShutdown(long timeout,
                          TimeUnit unit)

Description copied from interface: GracefulShutdownAwareService

A hook method that's called during graceful shutdown to provide safety for data managed by this service. Shutdown process is blocked until this method returns or shutdown timeouts. If this method does not return in time, then it's considered as failed to gracefully shutdown.

Specified by:

onShutdown in interface GracefulShutdownAwareService

Parameters:

timeout - timeout for graceful shutdown

unit - time unit

Returns:

true if graceful shutdown is successful, false otherwise

getPreJoinOperation

public Operation getPreJoinOperation()

Description copied from interface: PreJoinAwareService

An operation to be executed on the joining member before it is set as joined. As is the case with PostJoinAwareService.getPostJoinOperation()s, no partition locks, no key-based locks, no service level locks, no database interaction are allowed. Additionally, a pre-join operation is executed while the cluster lock is being held on the joining member, so it is important that the operation finishes quickly and does not interact with other locks. The Operation.getPartitionId() method should return a negative value. This means that the operations should not implement PartitionAwareOperation.

Pre join operations should return response, which may also be a null response.

Specified by:

getPreJoinOperation in interface PreJoinAwareService

Returns:

an operation to be executed on joining member before it is set as joined. Can be null.

memberAdded

public void memberAdded(MembershipServiceEvent event)

Description copied from interface: MembershipAwareService

Invoked when a new member is added to the cluster. Should be non-blocking.

Specified by:

memberAdded in interface MembershipAwareService

Parameters:

event - the event for a new member added to the cluster

memberRemoved

public void memberRemoved(MembershipServiceEvent event)

Description copied from interface: MembershipAwareService

Invoked when an existing member leaves the cluster. Should be non-blocking.

Specified by:

memberRemoved in interface MembershipAwareService

Parameters:

event - the event for an existing member leaving the cluster

memberAttributeChanged

public void memberAttributeChanged(MemberAttributeServiceEvent event)

Description copied from interface: MembershipAwareService

Invoked when a member attribute is changed.

Specified by:

memberAttributeChanged in interface MembershipAwareService

Parameters:

event - the event for a member attribute being changed

getCPGroupIdsLocally

public Collection<CPGroupId> getCPGroupIdsLocally()

getCPGroupLocally

public CPGroupInfo getCPGroupLocally(CPGroupId groupId)

getMetadataGroupManager

public MetadataRaftGroupManager getMetadataGroupManager()

getInvocationManager

public RaftInvocationManager getInvocationManager()

handlePreVoteRequest

public void handlePreVoteRequest(CPGroupId groupId,
                                 PreVoteRequest request,
                                 CPMember target)

handlePreVoteResponse

public void handlePreVoteResponse(CPGroupId groupId,
                                  PreVoteResponse response,
                                  CPMember target)

handleVoteRequest

public void handleVoteRequest(CPGroupId groupId,
                              VoteRequest request,
                              CPMember target)

handleVoteResponse

public void handleVoteResponse(CPGroupId groupId,
                               VoteResponse response,
                               CPMember target)

handleAppendEntries

public void handleAppendEntries(CPGroupId groupId,
                                AppendRequest request,
                                CPMember target)

handleAppendResponse

public void handleAppendResponse(CPGroupId groupId,
                                 AppendSuccessResponse response,
                                 CPMember target)

handleAppendResponse

public void handleAppendResponse(CPGroupId groupId,
                                 AppendFailureResponse response,
                                 CPMember target)

handleSnapshot

public void handleSnapshot(CPGroupId groupId,
                           InstallSnapshot request,
                           CPMember target)

getAllRaftNodes

public Collection<RaftNode> getAllRaftNodes()

getRaftNode

public RaftNode getRaftNode(CPGroupId groupId)

getOrInitRaftNode

public RaftNode getOrInitRaftNode(CPGroupId groupId)

isRaftGroupDestroyed

public boolean isRaftGroupDestroyed(CPGroupId groupId)

getConfig

public CPSubsystemConfig getConfig()

getLocalCPMember

public CPMemberInfo getLocalCPMember()

Description copied from interface: CPSubsystemManagementService

Returns the local CP member if this Hazelcast member is part of the CP subsystem, returns null otherwise.

This field is initialized when the local Hazelcast member is one of the first CPSubsystemConfig.getCPMemberCount() members in the cluster and the CP subsystem discovery process is completed.

This method fails with HazelcastException if the CP subsystem is not enabled.

Specified by:

getLocalCPMember in interface CPSubsystemManagementService

Returns:

local CP member if available, null otherwise

See Also:

CPSubsystemManagementService.isDiscoveryCompleted(), CPSubsystemManagementService.awaitUntilDiscoveryCompleted(long, TimeUnit)

createRaftNode

public void createRaftNode(CPGroupId groupId,
                           Collection<CPMemberInfo> members)

destroyRaftNode

public void destroyRaftNode(CPGroupId groupId)

stepDownRaftNode

public void stepDownRaftNode(CPGroupId groupId)

createRaftGroupForProxy

public RaftGroupId createRaftGroupForProxy(String name)

createRaftGroupForProxyAsync

public InternalCompletableFuture<RaftGroupId> createRaftGroupForProxyAsync(String name)

withoutDefaultGroupName

public static String withoutDefaultGroupName(String name)

getGroupNameForProxy

public static String getGroupNameForProxy(String name)

getObjectNameForProxy

public static String getObjectNameForProxy(String name)

getMetadataGroupId

public RaftGroupId getMetadataGroupId()

handleActiveCPMembers

public void handleActiveCPMembers(RaftGroupId latestMetadataGroupId,
                                  long membersCommitIndex,
                                  Collection<CPMemberInfo> members)

onRaftGroupDestroyed

public void onRaftGroupDestroyed(CPGroupId groupId)

Description copied from interface: RaftNodeLifecycleAwareService

Called on the thread of the Raft group when the given Raft node is destroyed, either gracefully or forcefully.

Specified by:

onRaftGroupDestroyed in interface RaftNodeLifecycleAwareService

onRaftNodeSteppedDown

public void onRaftNodeSteppedDown(CPGroupId groupId)

Description copied from interface: RaftNodeLifecycleAwareService

Called on the thread of the Raft group when the given Raft node is stepped down, either because it is shutting down, or it could not be added to the Raft group

Specified by:

onRaftNodeSteppedDown in interface RaftNodeLifecycleAwareService