Services
How to implement, deploy, access, and re-deploy GridGain services, including deployment strategies, interceptors, and the IgniteServices framework.
Overview
A service is a piece of functionality that can be deployed to a GridGain cluster and execute specific operations. You can have multiple instances of a service on one or multiple nodes.
GridGain services have the following features:
Load balancing
In all cases, other than singleton service deployment, GridGain automatically makes sure that approximately equal number of services are deployed on each node within the cluster. Whenever cluster topology changes, GridGain re-evaluates service deployments and may re-deploy an already deployed service to another node for better load balancing.
Fault tolerance
GridGain guarantees that services are continuously available, and are deployed according to the specified configuration, regardless of any topology changes or node crashes.
Hot Redeployment
You can use GridGain's DeploymentSpi configuration to re-deploy services without restarting the cluster. See Re-deploying Services.
GridGain services can be used as a backbone of a micro-services based solution or application. Learn more about this use case from the following series of articles:
Refer to a service example implementation in the Apache Ignite examples.
Implementing a Service
A service implements the Service interface. The Service interface has three methods:
init(): this method is called by Ignite before the service is deployed (and before theexecute()method is called)execute(): starts execution of the servicecancel(): cancels service execution
In .NET, implement the IService interface, which declares the same three methods as Init(), Execute(), and Cancel(). Each method receives an IServiceContext object that exposes the service name, the execution ID, and the cancellation flag.
Injecting the Service Context
In Java, a service gets its ServiceContext through the @ServiceContextResource annotation (org.apache.ignite.resources.ServiceContextResource). Put the annotation on a ServiceContext field or on a setter method of the service implementation. GridGain injects the context before it calls the init() method.
The service context exposes the service name, the execution ID, the cancellation flag, and the context of the current service call:
unsupported
unsupported
Service Call Context
A caller can pass a set of attributes to a service method in a service call context. Build the context with ServiceCallContext.builder(), and pass it when you get the service proxy. The put() method of the builder accepts a String or a byte[] value.
Inside the service, ServiceContext.currentCallContext() returns the call context of the method call that the current thread executes. Read the attributes with the attribute(name) and binaryAttribute(name) methods. If the caller did not pass a call context, currentCallContext() returns null.
The Java thin client can also pass a call context to a service. See Executing Ignite Services.
Deploying Services
You can deploy your service either programmatically at runtime, or by providing a service configuration as part of the node configuration. In the latter case, the service is deployed when the cluster starts.
Deploying Services at Runtime
You can deploy services at runtime via the instance of IgniteServices, which can be obtained from an instance of Ignite by calling the Ignite.services() method.
The IgniteServices interface has a number of methods for deploying services:
deploy(ServiceConfiguration)deploys a service defined by a given configuration.deployNodeSingleton(...)ensures that an instance of the service is running on each server node.deployClusterSingleton(...)deploys a single instance of the service per cluster. If the cluster node on which the service is deployed stops, GridGain automatically redeploys the service on another node.deployKeyAffinitySingleton(...)deploys a single instance of the service on the primary node for a given cache key.deployMultiple(...)deploys the given number of instances of the service.
This is an example of cluster singleton deployment:
In .NET, the equivalent interface is IServices, obtained by calling IIgnite.GetServices().
unsupported
And here is how to deploy a cluster singleton using ServiceConfiguration:
unsupported
Deploying Services at Node Startup
You can specify your service as part of the node configuration and start the service together with the node. If your service is a node singleton, the service is started on each node of the cluster. If the service is a cluster singleton, it is started in the first cluster node, and is redeployed to one of the other nodes if the first node terminates. The service must be available on the classpath of each node.
Below is an example of configuring a cluster singleton service:
unsupported
unsupported
Deploying to a Subset of Nodes
When you obtain the IgniteServices interface by calling ignite.services(), the IgniteServices instance is associated with all server nodes. It means that GridGain chooses where to deploy the service from the set of all server nodes. You can change the set of nodes considered for service deployment by using various approaches describe below.
Cluster Singleton
A cluster singleton is a deployment strategy where there is only one instance of the service in the cluster, and GridGain guarantees that the instance is always available. In case the cluster node on which the service is deployed crashes or stops, GridGain automatically redeploys the instance to another node.
ClusterGroup
You can use the ClusterGroup interface to deploy services to a subset of nodes. If the service is a node singleton, the service is deployed on all nodes from the subset. If the service is a cluster singleton, it is deployed on one of the nodes from the subset.
unsupported
Node Filter
You can use node attributes to define the subset of nodes meant for service deployment. This is achieved by using a node filter. A node filter is an IgnitePredicate<ClusterNode> that GridGain calls for each node associated with the IgniteService interface. If the predicate returns true for a given node, the node is included.
Here is an example of a node filter. The filter includes the server nodes that have the "west.coast.node" attribute.
In .NET, a node filter implements the IClusterNodeFilter interface, whose Invoke() method returns true for the nodes to include. The assembly that contains the filter class must be loaded on all nodes.
unsupported
Deploy the service using the node filter:
unsupported
Cache Key
Affinity-based deployment allows you to deploy a service to the primary node for a specific key in a specific cache. Refer to the Affinity Colocation section for details. For an affinity-base deployment, specify the desired cache and key in the service configuration. The cache does not have to contain the key. The node is determined by the affinity function. If the cluster topology changes in a way that the key is re-assigned to another node, the service is redeployed to that node as well.
unsupported
Accessing Services
You can access the service at runtime via a service proxy. Proxies can be either sticky or non-sticky. A sticky proxy always connects to the same cluster node to access a remotely deployed service. A non-sticky proxy load-balances remote service invocations among all cluster nodes on which the service is deployed.
The following code snippet obtains a non-sticky proxy to the service and calls a service method:
The .NET thin client can also call services. See Calling Services for details, including how to invoke a service asynchronously.
unsupported
Service Call Interceptors
You can intercept calls to service methods with one or more service call interceptors. An interceptor lets you add middleware logic — such as security checks, auditing, logging, or metrics — that applies to all business methods of a service without changing the service implementation.
A service call interceptor implements the ServiceCallInterceptor interface, which has a single method:
The interceptor receives the method name, the method arguments, and the service context. Call next.call() to pass control to the next interceptor in the chain; the last interceptor in the chain invokes the service method itself.
You register interceptors on the ServiceConfiguration object with the setInterceptors() method. When you specify multiple interceptors, they run in the order in which you pass them.
unsupported
unsupported
Un-deploying Services
To undeploy a service, use the IgniteServices.cancel(serviceName) or IgniteServices.cancelAll() methods.
In .NET, the corresponding methods are IServices.Cancel(name) and IServices.CancelAll().
unsupported
Re-deploying Services
If you want to update the implementation of a service without stopping the cluster, you can do it if you use the GridGain's DeploymentSPI configuration.
Use the following procedure to redeploy the service:
Update the JAR file(s) in the location where the service is stored (pointed to by your
UriDeploymentSpi.uriListproperty). GridGain will reload the new classes after the configured update period.Add the service implementation to the classpass of a client node and start the client.
Call the
Ignite.services().cancel()method on the client node to stop the service.Deploy the service from the client node.
Stop the client node.
In this way, you don't have to stop the server nodes, so you don't interrupt the operation of your cluster.
Service Events
The services generate the following events:
EVT_SERVICE_METHOD_EXECUTION_STARTED
Is raised right before a service method execution starts.
The node where the service method execution is failed.
EVT_SERVICE_METHOD_EXECUTION_FINISHED
Is raised right after a service method execution finishes.
The node where the service method execution is failed.
EVT_SERVICE_METHOD_EXECUTION_FAILED
Is raised in case a service method execution fails.
The node where the service method execution is failed.
New IgniteServices Framework
Starting from version 8.8.1, GridGain offers a newer framework for IgniteServices. We recommend to switch to this framework because it's better optimized and results in lower load on your cluster. To move to the new framework, turn the IGNITE_EVENT_DRIVEN_SERVICE_PROCESSOR_ENABLED JVM property to true.
Enabling the New Framework (with Downtime)
All nodes in the cluster (both server and client) must have the same IGNITE_EVENT_DRIVEN_SERVICE_PROCESSOR_ENABLED value. Therefore, the move to true will require a full restart of the cluster and the clients, which will result in downtime.
To move to the new implementation:
Stop all nodes.
Change
IGNITE_EVENT_DRIVEN_SERVICE_PROCESSOR_ENABLEDtotrueon all nodes.Restart all nodes.
Enabling the New Framework While Upgrading (with Downtime)
In GridGain 8.9.1 and above, IGNITE_EVENT_DRIVEN_SERVICE_PROCESSOR_ENABLED=true by default. When upgrading to one of the GridGain 8.9.x versions, you will automatically move to the new service framework. However, this might also require downtime.
To move to the new framework while upgrading to a new GridGain version with downtime:
Stop all nodes.
Upgrade to 8.9.x.
Restart all nodes.
Avoiding Downtime While Upgrading
If downtime is not an option in your environment, you can upgrade from GridGain 8.8.x to 8.9.x without enabling the new IgniteServices framework, using the Rolling Upgrade.
Upgrading Server Nodes
To upgrade server nodes:
Make sure Rolling Upgrades are enabled.
Stop one of the server nodes.
Upgrade the version and set
IGNITE_EVENT_DRIVEN_SERVICE_PROCESSOR_ENABLED=falsefor the above server node.Restart the above server node.
Repeat steps (2)-(4) for the remaining server nodes.
Upgrading Nodes with Thick Clients
To upgrade nodes with a thick client:
Stop the client node.
Upgrade to the new version.
(Optionally, starting from 8.9.4) Add to the configuration file
IGNITE_SERVICES_SET_REMOTE_FILTER_ON_START=true.Restart the node.
Repeat the above steps on the remaining nodes.
Setting IGNITE_SERVICES_SET_REMOTE_FILTER_ON_START=true on client nodes initiates optimizations that reduce load on the cluster (compared to the legacy framework). To set the above option to true, all server nodes must run GridGain 8.8.39 or above, or 8.9.4 or above. You can set IGNITE_SERVICES_SET_REMOTE_FILTER_ON_START=true on nodes while upgrading these nodes to the above GridGain versions.
Enabling Optimizations with the Legacy Services Framework
To switch on optimizations for the legacy framework while upgrading to a new GridGain version (without downtime):
Make sure that all server nodes have been upgraded to 8.8.39+ or 8.9.4+.
Stop a client node.
Upgrade the client node to the same version the server nodes had been upgraded to.
Set
IGNITE_SERVICES_SET_REMOTE_FILTER_ON_START=trueon the client node.Restart the client node.
If your entire cluster (server and client nodes) is already on GridGain 8.8.39+ or 8.9.4+, all you need to do to switch on the optimizations is to set IGNITE_SERVICES_SET_REMOTE_FILTER_ON_START=true when reloading your thick client.
This page is: Copyright © 2026 MariaDB. All rights reserved.
Last updated
Was this helpful?

