classClusterSingletonManager extends Actor with FSM[State, Data]
Manages singleton actor instance among all cluster nodes or a group
of nodes tagged with a specific role. At most one singleton instance
is running at any point in time.
The ClusterSingletonManager is supposed to be started on all nodes,
or all nodes with specified role, in the cluster with actorOf.
The actual singleton is started on the oldest node by creating a child
actor from the supplied singletonProps.
The singleton actor is always running on the oldest member, which can
be determined by akka.cluster.Member#isOlderThan.
This can change when removing members. A graceful hand over can normally
be performed when current oldest node is leaving the cluster. Be aware that
there is a short time period when there is no active singleton during the
hand-over process.
The cluster failure detector will notice when oldest node
becomes unreachable due to things like JVM crash, hard shut down,
or network failure. When the crashed node has been removed (via down) from the
cluster then a new oldest node will take over and a new singleton actor is
created. For these failure scenarios there will not be a graceful hand-over,
but more than one active singletons is prevented by all reasonable means. Some
corner cases are eventually resolved by configurable timeouts.
You access the singleton actor with actorSelection using the names you have
specified when creating the ClusterSingletonManager. You can subscribe to
akka.cluster.ClusterEvent.MemberEvent and sort the members by age
(akka.cluster.ClusterEvent.Member#isOlderThan) to keep track of oldest member.
Alternatively the singleton actor may broadcast its existence when it is started.
Use factory method to create the
akka.actor.Props for the actor.
Arguments
singletonPropsakka.actor.Props of the singleton actor instance.
singletonName The actor name of the child singleton actor.
terminationMessage When handing over to a new oldest node
this terminationMessage is sent to the singleton actor to tell
it to finish its work, close resources, and stop.
The hand-over to the new oldest node is completed when the
singleton actor is terminated.
Note that akka.actor.PoisonPill is a perfectly fine
terminationMessage if you only need to stop the actor.
role Singleton among the nodes tagged with specified role.
If the role is not specified it's a singleton among all nodes in
the cluster.
maxHandOverRetries When a node is becoming oldest it sends
hand-over request to previous oldest. This is retried with the
retryInterval until the previous oldest confirms that the hand
over has started, or this maxHandOverRetries limit has been
reached. If the retry limit is reached it takes the decision to be
the new oldest if previous oldest is unknown (typically removed),
otherwise it initiates a new round by throwing
akka.contrib.pattern.ClusterSingletonManagerIsStuck and expecting
restart with fresh state. For a cluster with many members you might
need to increase this retry limit because it takes longer time to
propagate changes across all nodes.
maxTakeOverRetries When a oldest node is not oldest any more
it sends take over request to the new oldest to initiate the normal
hand-over process. This is especially useful when new node joins and becomes
oldest immediately, without knowing who was previous oldest. This is retried
with the retryInterval until this retry limit has been reached. If the retry
limit is reached it initiates a new round by throwing
akka.contrib.pattern.ClusterSingletonManagerIsStuck and expecting
restart with fresh state. This will also cause the singleton actor to be
stopped. maxTakeOverRetries must be less than maxHandOverRetries to
ensure that new oldest doesn't start singleton actor before previous is
stopped for certain corner cases.
Manages singleton actor instance among all cluster nodes or a group of nodes tagged with a specific role. At most one singleton instance is running at any point in time.
The ClusterSingletonManager is supposed to be started on all nodes, or all nodes with specified role, in the cluster with
actorOf
. The actual singleton is started on the oldest node by creating a child actor from the suppliedsingletonProps
.The singleton actor is always running on the oldest member, which can be determined by akka.cluster.Member#isOlderThan. This can change when removing members. A graceful hand over can normally be performed when current oldest node is leaving the cluster. Be aware that there is a short time period when there is no active singleton during the hand-over process.
The cluster failure detector will notice when oldest node becomes unreachable due to things like JVM crash, hard shut down, or network failure. When the crashed node has been removed (via down) from the cluster then a new oldest node will take over and a new singleton actor is created. For these failure scenarios there will not be a graceful hand-over, but more than one active singletons is prevented by all reasonable means. Some corner cases are eventually resolved by configurable timeouts.
You access the singleton actor with
actorSelection
using the names you have specified when creating the ClusterSingletonManager. You can subscribe to akka.cluster.ClusterEvent.MemberEvent and sort the members by age (akka.cluster.ClusterEvent.Member#isOlderThan) to keep track of oldest member. Alternatively the singleton actor may broadcast its existence when it is started.Use factory method to create the akka.actor.Props for the actor.
Arguments
singletonProps akka.actor.Props of the singleton actor instance.
singletonName The actor name of the child singleton actor.
terminationMessage When handing over to a new oldest node this
terminationMessage
is sent to the singleton actor to tell it to finish its work, close resources, and stop. The hand-over to the new oldest node is completed when the singleton actor is terminated. Note that akka.actor.PoisonPill is a perfectly fineterminationMessage
if you only need to stop the actor.role Singleton among the nodes tagged with specified role. If the role is not specified it's a singleton among all nodes in the cluster.
maxHandOverRetries When a node is becoming oldest it sends hand-over request to previous oldest. This is retried with the
retryInterval
until the previous oldest confirms that the hand over has started, or thismaxHandOverRetries
limit has been reached. If the retry limit is reached it takes the decision to be the new oldest if previous oldest is unknown (typically removed), otherwise it initiates a new round by throwing akka.contrib.pattern.ClusterSingletonManagerIsStuck and expecting restart with fresh state. For a cluster with many members you might need to increase this retry limit because it takes longer time to propagate changes across all nodes.maxTakeOverRetries When a oldest node is not oldest any more it sends take over request to the new oldest to initiate the normal hand-over process. This is especially useful when new node joins and becomes oldest immediately, without knowing who was previous oldest. This is retried with the
retryInterval
until this retry limit has been reached. If the retry limit is reached it initiates a new round by throwing akka.contrib.pattern.ClusterSingletonManagerIsStuck and expecting restart with fresh state. This will also cause the singleton actor to be stopped.maxTakeOverRetries
must be less thanmaxHandOverRetries
to ensure that new oldest doesn't start singleton actor before previous is stopped for certain corner cases.