Spring boot starter for gRPC framework.
- 1. Features
- 2. Setup
- 3. Usage
- 4. Showcase
- 4.1. Service implementation
- 4.2. Interceptors support
- 4.3. Distributed tracing support (Spring Cloud Sleuth integration)
- 4.4. GRPC server metrics (Micrometer.io integration)
- 4.5. Spring Boot Validation support
- 4.6. Spring cloud stream support
- 4.7. Spring security support
- 4.8. Transport Security (TLS)
- 4.9. Custom gRPC Server Configuration
- 5. Events
- 6. Reactive API support
- 7. Error handling
- 8. Implementing message validation
- 9. GRPC response observer and Spring @Transactional caveats
- 10. Spring Security Integration
- 11. Health check
- 12. Spring actuator support
- 13. Consul Integration
- 14. Eureka Integration
- 15. License
Autoconfigures and runs the embedded gRPC server with @GRpcService-enabled beans as part of spring-boot application (short video)
Gradle
users are advised to apply the plugin :
plugins {
id "io.github.lognet.grpc-spring-boot" version '5.1.5'
}
io.github.lognet.grpc-spring-boot gradle plugin dramatically simplifies the project setup.
repositories {
mavenCentral()
// maven { url "https://oss.sonatype.org/content/repositories/snapshots" } // for snapshot builds
}
dependencies {
implementation 'io.github.lognet:grpc-spring-boot-starter:5.1.5'
}
By default, starter pulls io.grpc:grpc-netty-shaded
as transitive dependency, if you are forced to use pure grpc-netty
dependency:
implementation ('io.github.lognet:grpc-spring-boot-starter:5.1.5') {
exclude group: 'io.grpc', module: 'grpc-netty-shaded'
}
implementation 'io.grpc:grpc-netty:1.58.0' // (1)
-
Make sure to pull the version that matches the release.
Both libraries' presence on classpath is also supported with grpc.netty-server.on-collision-prefer-shaded-netty
property.
If you are using Spring Boot Dependency Management plugin, it might pull not the same version as the version this started was compiled against, causing binary incompatibility issue.
In this case you’ll need to forcibly and explicitly set the grpc
version to use (see version matrix here ):
configurations.all {
resolutionStrategy.eachDependency { details ->
if ("io.grpc".equalsIgnoreCase(details.requested.group)) {
details.useVersion "1.58.0"
}
}
}
ℹ️
|
The release notes with compatibility matrix can be found here |
Follow this guide to generate stub and server interface(s) from your .proto
file(s).
If you are stack with maven - use this link.
-
Annotate your server interface implementation(s) with
@org.lognet.springboot.grpc.GRpcService
-
Optionally configure the server port in your
application.yml/properties
. Default port is6565
.
grpc:
port: 6565
ℹ️
|
A random port can be defined by setting the port to 0 .The actual port being used can then be retrieved by using @LocalRunningGrpcPort annotation on int field which will inject the running port (explicitly configured or randomly selected)
|
-
Optionally enable server reflection (see https://github.com/grpc/grpc-java/blob/master/documentation/server-reflection-tutorial.md)
grpc:
enableReflection: true
-
Optionally set the startup phase order (defaults to
Integer.MAX_VALUE
).
grpc:
start-up-phase: XXX
-
Optionally set the number of seconds to wait for preexisting calls to finish during graceful server shutdown. New calls will be rejected during this time. A negative value is equivalent to an infinite grace period. Default value is
0
(means don’t wait).
grpc:
shutdownGrace: 30
-
Netty-specific server properties can be specified under
grpc.netty-server
prefix.
By configuring one of thegrpc.netty-server.xxxx
values you are implicitly setting transport to be Netty-based.
grpc:
netty-server:
keep-alive-time: 30s (1)
max-inbound-message-size: 10MB (2)
primary-listen-address: 10.10.15.23:0 (3)
additional-listen-addresses:
- 192.168.0.100:6767 (4)
on-collision-prefer-shaded-netty: false (5)
-
Duration
type properties can be configured with string value format described here. -
DataSize
type properties can be configured with string value described here -
Exposed on external network IP with custom port.
SocketAddress
type properties string value format:-
host:port
(ifport
value is less than 1, uses random value) -
host:
(uses default grpc port,6565
)
-
-
Exposed on internal network IP as well with predefined port
6767
. -
In case you have both
shaded
andpure
netty libraries in dependencies, pick theNettyServerBuilder
type that should be created. This is the type that will be passed toGRpcServerBuilderConfigurer
(see Custom gRPC Server Configuration), defaults totrue
(i.e.io.grpc.netty.shaded.io.grpc.netty.NettyServerBuilder
;io.grpc.netty.NettyServerBuilder
iffalse
)
The starter supports also the in-process server
, which should be used for testing purposes :
grpc:
enabled: false (1)
inProcessServerName: myTestServer (2)
-
Disables the default server (
NettyServer
). -
Enables the
in-process
server.
ℹ️
|
If you enable both the NettyServer and in-process server, they will both share the same instance of HealthStatusManager and GRpcServerBuilderConfigurer (see Custom gRPC Server Configuration).
|
In the grpc-spring-boot-starter-demo
project you can find fully functional examples with integration tests.
The service definition from .proto
file looks like this :
service Greeter {
rpc SayHello ( HelloRequest) returns ( HelloReply) {}
}
Note the generated io.grpc.examples.GreeterGrpc.GreeterImplBase
class that extends io.grpc.BindableService
.
All you need to do is to annotate your service implementation with @org.lognet.springboot.grpc.GRpcService
@GRpcService
public static class GreeterService extends GreeterGrpc.GreeterImplBase{
@Override
public void sayHello(GreeterOuterClass.HelloRequest request, StreamObserver<GreeterOuterClass.HelloReply> responseObserver) {
final GreeterOuterClass.HelloReply.Builder replyBuilder = GreeterOuterClass.HelloReply.newBuilder().setMessage("Hello " + request.getName());
responseObserver.onNext(replyBuilder.build());
responseObserver.onCompleted();
}
}
The starter supports the registration of two kinds of interceptors: Global and Per Service.
In both cases the interceptor has to implement io.grpc.ServerInterceptor
interface.
-
Per service
@GRpcService(interceptors = { LogInterceptor.class })
public class GreeterService extends GreeterGrpc.GreeterImplBase{
// ommited
}
LogInterceptor
will be instantiated via spring factory if there is bean of type LogInterceptor
, or via no-args constructor otherwise.
-
Global
@GRpcGlobalInterceptor
public class MyInterceptor implements ServerInterceptor{
// ommited
}
The annotation on java config factory method is also supported :
@Configuration
public class MyConfig{
@Bean
@GRpcGlobalInterceptor
public ServerInterceptor globalInterceptor(){
return new ServerInterceptor(){
@Override
public <ReqT, RespT> ServerCall.Listener<ReqT> interceptCall(ServerCall<ReqT, RespT> call, Metadata headers, ServerCallHandler<ReqT, RespT> next) {
// your logic here
return next.startCall(call, headers);
}
};
}
}
The particular service also has the opportunity to disable the global interceptors :
@GRpcService(applyGlobalInterceptors = false)
public class GreeterService extends GreeterGrpc.GreeterImplBase{
// ommited
}
Global interceptors can be ordered using Spring’s @Ordered
or @Priority
annotations.
Following Spring’s ordering semantics, lower order values have higher priority and will be executed first in the interceptor chain.
@GRpcGlobalInterceptor
@Order(10)
public class A implements ServerInterceptor{
// will be called before B
}
@GRpcGlobalInterceptor
@Order(20)
public class B implements ServerInterceptor{
// will be called after A
}
The starter uses built-in interceptors to implement error handling, Spring Security
, Validation
and Metrics
integration.
Their order can also be controlled by below properties :
-
grpc.recovery.interceptor-order
(error handling interceptor order, defaults toOrdered.HIGHEST_PRECEDENCE
) -
grpc.security.auth.interceptor-order
( defaults toOrdered.HIGHEST_PRECEDENCE+1
) -
grpc.validation.interceptor-order
( defaults toOrdered.HIGHEST_PRECEDENCE+10
) -
grpc.metrics.interceptor-order
( defaults toOrdered.HIGHEST_PRECEDENCE+20
)
This gives you the ability to set up the desired order of built-in and your custom interceptors.
Keep on reading !!! There is more
The way grpc interceptor works is that it intercepts the call and returns the server call listener, which in turn can intercept the request message as well, before forwarding it to the actual service call handler :
By setting grpc.security.auth.fail-fast
property to false
all downstream interceptors as well as all upstream interceptors (On_Message) will still be executed in case of authentication/authorization failure
Assuming interceptor_2
is securityInterceptor
:
This started is natively supported by spring-cloud-sleuth
project.
Please continue to sleuth grpc integration.
By including org.springframework.boot:spring-boot-starter-actuator
dependency,
the starter will collect gRPC server metrics , broken down by
-
method
- gRPC service method FQN (Fully Qualified Name) -
result
- Response status code -
address
- server local address (if you exposed additional listen addresses, withgrpc.netty-server.additional-listen-addresses
property)
After configuring the exporter of your choice,
you should see the timer
named grpc.server.calls
.
By defining GRpcMetricsTagsContributor
bean in your application context, you can add custom tags to the grpc.server.calls
timer.
You can also use RequestAwareGRpcMetricsTagsContributor
bean to tag unary and streaming calls.
Demo is here
💡
|
Keep the dispersion low not to blow up the cardinality of the metric. |
RequestAwareGRpcMetricsTagsContributor
can be still executed for failed authentication if metric
interceptor has higher precedence than security
interceptor and grpc.security.auth.fail-fast
set to false
.
This case is covered by this test.
💡
|
Make sure to read Interceptors ordering chapter. |
Make sure to include below dependencies :
implementation "org.springframework.boot:spring-boot-starter-actuator"
implementation "io.micrometer:micrometer-registry-prometheus"
implementation 'org.springframework.boot:spring-boot-starter-web'
Configuration :
management:
metrics:
export:
prometheus:
enabled: true
endpoints:
web:
exposure:
include: "*"
Standard /actuator/metrics
and /actuator/prometheus
endpoints will render grpc.server.calls
metrics (see demo here).
ℹ️
|
GRPC scrapping proposal |
The starter can be autoconfigured to validate request/response gRPC service messages. Please continue to Implementing message validation for configuration details.
The starter internally defines the bean of type java.util.function.Consumer
which is being considered for function registry when spring-cloud-stream
is on classpath, which is undesirable (spring-cloud-stream
auto-registers the channel if you have exactly one Consumer/Supplier/Function bean in the application context, so you already have one if you use this starter together with spring-cloud-stream
).
According to this, it is recommended to use spring.cloud.function.definition
property in production ready applications and not to rely on the auto-discovery.
Please refer to GRPC Kafka Stream demo, the essential part is this line.
The starter provides built-in support for authenticating and authorizing users leveraging integration with Spring Security framework.
Please refer to the sections on Spring Security Integration for details on supported authentication providers and configuration options.
The transport security can be configured using root certificate together with its private key path:
grpc:
security:
cert-chain: classpath:cert/server-cert.pem
private-key: file:../grpc-spring-boot-starter-demo/src/test/resources/cert/server-key.pem
The value of both properties is in form supported by ResourceEditor.
The client side should be configured accordingly :
((NettyChannelBuilder)channelBuilder)
.useTransportSecurity()
.sslContext(GrpcSslContexts.forClient().trustManager(certChain).build());
This starter will pull the io.netty:netty-tcnative-boringssl-static
dependency by default to support SSL.
If you need another SSL/TLS support, please exclude this dependency and follow Security Guide.
ℹ️
|
If the more detailed tuning is needed for security setup, please use custom configurer described in Custom gRPC Server Configuration |
To intercept the io.grpc.ServerBuilder
instance used to build the io.grpc.Server
, you can add bean that inherits from org.lognet.springboot.grpc.GRpcServerBuilderConfigurer
to your context and override the configure
method.
Multiple configurers are also supported.
By the time of invocation of configure
method, all discovered services, including theirs interceptors, had been added to the passed builder.
In your implementation of configure
method, you can add your custom configuration:
@Component
public class MyGRpcServerBuilderConfigurer extends GRpcServerBuilderConfigurer{
@Override
public void configure(ServerBuilder<?> serverBuilder){
serverBuilder
.executor(YOUR EXECUTOR INSTANCE)
.useTransportSecurity(YOUR TRANSPORT SECURITY SETTINGS);
((NettyServerBuilder)serverBuilder)// cast to NettyServerBuilder (which is the default server) for further customization
.sslContext(GrpcSslContexts // security fine tuning
.forServer(...)
.trustManager(...)
.build())
.maxConnectionAge(...)
.maxConnectionAgeGrace(...);
}
};
}
@Component
public class MyCustomCompressionGRpcServerBuilderConfigurer extends GRpcServerBuilderConfigurer{
@Override
public void configure(ServerBuilder<?> serverBuilder){
serverBuilder
.compressorRegistry(YOUR COMPRESSION REGISTRY)
.decompressorRegistry(YOUR DECOMPRESSION REGISTRY) ;
}
};
}
ℹ️
|
If you enable both NettyServer and in-process servers, the configure method will be invoked on the same instance of configurer.If you need to differentiate between the passed serverBuilder s, you can check the type.This is the current limitation. |
GRpcServerInitializedEvent
is published upon server startup, you can consume it using regular spring API.
Starting from version 5.1.0
, spring-boot-starter-gradle-plugin
integrates SalesForce’s reactive-grpc protoc plugin :
import org.lognet.springboot.grpc.gradle.ReactiveFeature
plugins {
id "io.github.lognet.grpc-spring-boot"
}
grpcSpringBoot {
reactiveFeature.set(ReactiveFeature.REACTOR) // or ReactiveFeature.RX
}
Here are the tests and reactive grpc sample service.
The starter registers the GRpcExceptionHandlerInterceptor
which is responsible to propagate the service-thrown exception to the error handlers.
The error handling method could be registered by having @GRpcServiceAdvice
annotated bean with methods annotated with @GRpcExceptionHandler
annotations.
These are considered as global
error handlers and the method with exception type parameter the nearest by the type hierarchy to the thrown exception is invoked.
The signature of the error handler has to follow the below pattern:
Return type | Parameter 1 | Parameter 2 |
---|---|---|
io.grpc.Status |
any |
GRpcExceptionScope |
@GRpcServiceAdvice
class MyHandler1{
@GRpcExceptionHandler
public Status handle (MyCustomExcpetion exc, GRpcExceptionScope scope){
}
@GRpcExceptionHandler
public Status handle (IllegalArgumentException exc, GRpcExceptionScope scope){
}
}
@GRpcServiceAdvice
class MyHandler2 {
@GRpcExceptionHandler
public Status anotherHandler (NullPointerException npe,GRpcExceptionScope scope){
}
}
You can have as many advice
beans and handler methods as you want as long as they don’t interfere with each other and don’t create handled exception type ambiguity.
The grpc
service bean is also discovered for error handlers, having the higher precedence than global error handling methods discovered in @GRpcServiceAdvice
beans. The service-level error handling methods are considered private
and invoked only when the exception is thrown by this service:
class SomeException extends Exception{
}
class SomeRuntimeException extends RuntimeException{
}
@GRpcService
public class HelloService extends GreeterGrpc.GreeterImplBase{
@Override
public void sayHello(GreeterOuterClass.HelloRequest request, StreamObserver<GreeterOuterClass.HelloReply> responseObserver) {
...
throw new GRpcRuntimeExceptionWrapper(new SomeException()) ; // (1)
//or
throw new GRpcRuntimeExceptionWrapper(new SomeException(), "myHint");// (2)
//or
throw new SomeRuntimeException(); //(3)
}
@GRpcExceptionHandler
public Status privateHandler (SomeException npe,GRpcExceptionScope scope){
// INVOKED when thrown from HelloService service
String myHint = scope.getHintAs(String.class); // (4)
scope.getResponseHeaders().put(Metadata.Key.of("custom", Metadata.ASCII_STRING_MARSHALLER), "Value");// (5)
}
@GRpcExceptionHandler
public Status privateHandler (SomeRuntimeException npe,GRpcExceptionScope scope){
// INVOKED when thrown from HelloService service
}
}
@GRpcServiceAdvice
class MyHandler {
@GRpcExceptionHandler
public Status anotherHandler (SomeException npe,GRpcExceptionScope scope){
// NOT INVOKED when thrown from HelloService service
}
@GRpcExceptionHandler
public Status anotherHandler (SomeRuntimeException npe,GRpcExceptionScope scope){
// NOT INVOKED when thrown from HelloService service
}
}
-
Because the nature of
grpc
service API that doesn’t allow throwing checked exception, the special runtime exception type is provided to wrap the checked exception. It’s then getting unwrapped when looking for the handler method. -
When throwing the
GRpcRuntimeExceptionWrapper
exception, you can also pass thehint
object which is then accessible from thescope
object inhandler
method. -
Runtime exception can be thrown as-is and doesn’t need to be wrapped.
-
Obtain the hint object.
-
Send custom headers to the client.
Authentication failure is propagated via AuthenticationException
and authorization failure - via AccessDeniedException
.
Validation failure is propagated via ConstraintViolationException
: for failed request - with Status.INVALID_ARGUMENT
as a hint , and for failed response - with Status.FAILED_PRECONDITION
as a hint.
The demo is here
Thanks to Bean Validation configuration support via XML deployment descriptor , it’s possible to
provide the constraints for generated classes via XML instead of instrumenting the generated messages with custom protoc
compiler.
-
Add
org.springframework.boot:spring-boot-starter-validation
dependency to your project. -
Create
META-INF/validation.xml
and constraints declarations file(s). (IntelliJ IDEA has great auto-complete support for authorizing bean validation constraints xml files )
See also samples fromHibernate
validator documentation
You can find demo configuration and corresponding tests here
Note, that both request
and response
messages are being validated.
If your gRPC method uses the same request and response message type, you can use org.lognet.springboot.grpc.validation.group.RequestMessage
and
org.lognet.springboot.grpc.validation.group.ResponseMessage
validation groups to apply different validation logic :
...
<getter name="someField">
<!--should be empty for request message-->
<constraint annotation="javax.validation.constraints.Size">
<groups>
<value>org.lognet.springboot.grpc.validation.group.RequestMessage</value> (1)
</groups>
<element name="min">0</element>
<element name="max">0</element>
</constraint>
<!--should NOT be empty for response message-->
<constraint annotation="javax.validation.constraints.NotEmpty">
<groups>
<value>org.lognet.springboot.grpc.validation.group.ResponseMessage</value> (2)
</groups>
</constraint>
</getter>
...
-
Apply this constraint only for
request
message -
Apply this constraint only for
response
message
Note also custom cross-field constraint and its usage :
<bean class="io.grpc.examples.GreeterOuterClass$Person">
<class>
<constraint annotation="org.lognet.springboot.grpc.demo.PersonConstraint"/>
</class>
<!-- ... -->
</bean>
As described in Interceptors ordering chapter, you can give validation
interceptor the higher precedence than security
interceptor and set grpc.security.auth.fail-fast
property to false
.
In this scenario, if call is both unauthenticated and invalid, the client will get Status.INVALID_ARGUMENT
instead of Status.PERMISSION_DENIED/Status.UNAUTHENTICATED
response status.
Demo is here
While it’s still possible to have your rpc methods annotated with @Transactional
(with spring.aop.proxy-target-class=true
if it’s not enabled by default), chances are to get unpredictable behaviour. Consider below grpc method implementation :
@GRpcService
class MyGrpcService extends ...{
@Autowired
private MyJpaRepository repo;
@Transactional //(1)
public void rpcCall(Req request, StreamOvserver<Res> observer) {
Res response = // Database operations via repo
observer.onNext(response); //(2)
observer.onCompleted();
}//(3)
}
-
The method is annotated as
@Transactional
, Spring will commit the transaction at some time after methods returns -
Response is returned to the caller
-
Methods returns, transaction eventually committed.
Theoretically, and as you can see - practically, there is small time-span when client (if the network latency is minimal, and your grpc server encouraged context switch right after <2>) can try to access the database via another grpc call before the transaction is committed.
The solution to overcome this situation is to externalize the transactional logic into separate service class :
@Service
class MyService{
@Autowired
private MyJpaRepository repo;
@Transactional //(1)
public Res doTransactionalWork(){
// Database operations via repo
return result;
}//(2)
}
@GRpcService
class MyGrpcService extends ...{
@Autowired
private MyService myService;
public void rpcCall(Req request, StreamOvserver<Res> observer) {
Res response = myService.doTransactionalWork();
observer.onNext(response); //(3)
observer.onCompleted();
}
}
-
Service method is transactional
-
Transaction is eventually committed.
-
Reply after transaction is committed.
By following this approach you also decouple the transport layer and business logic that now can be tested separately.
Scheme | Dependencies |
---|---|
Basic |
|
Bearer |
|
Custom |
|
GRPC security configuration follows the same principals and APIs as Spring WEB security configuration, it’s enabled by default if you have org.springframework.security:spring-security-config
dependency in your classpath.
You can use @Secured
annotation on services/methods to protect your endpoints, or by using API and overriding defaults (which precesses @Secured
annotation ):
@Configuration
class MySecurityCfg extends GrpcSecurityConfigurerAdapter {
@Override
public void configure(GrpcSecurity builder) throws Exception {
MethodsDescriptor<?,?> adminMethods = MyServiceGrpc.getSomeMethod();
builder
.authorizeRequests()
.methods(adminMethods).hasAnyRole("admin")
.anyMethodExcluding(adminMethods).hasAnyRole("user")
.withSecuredAnnotation();(1)
}
}
-
or combine
API
with@Secured
annotations.
This default configuration secures GRPC methods/services annotated with org.springframework.security.access.annotation.@Secured
annotation.
Leaving value of the annotation empty (@Secured({})
) means : authenticate
only, no authorization will be performed.
If JwtDecoder
bean exists in your context, it will also register JwtAuthenticationProvider
to handle the validation of authentication claim.
BasicAuthSchemeSelector
and BearerTokenAuthSchemeSelector
are also automatically registered to support authentication with username/password and bearer token.
By setting grpc.security.auth.enabled
to false
, GRPC security can be turned-off.
Customization of GRPC security configuration is done by extending GrpcSecurityConfigurerAdapter
(Various configuration examples and test scenarios are here.)
@Configuration
public class GrpcSecurityConfiguration extends GrpcSecurityConfigurerAdapter {
@Autowired
private JwtDecoder jwtDecoder;
@Override
public void configure(GrpcSecurity builder) throws Exception {
builder.authorizeRequests()(1)
.methods(GreeterGrpc.getSayHelloMethod()).hasAnyAuthority("SCOPE_profile")(2)
.and()
.authenticationProvider(JwtAuthProviderFactory.withAuthorities(jwtDecoder));(3)
}
}
-
Get hold of authorization configuration object
-
MethodDefinition
ofsayHello
method is allowed for authenticated users withSCOPE_profile
authority. -
Use
JwtAuthenticationProvider
to validate user claim (BEARER
token) against resource server configured withspring.security.oauth2.resourceserver.jwt.issuer-uri
property.
One is possible to plug in your own bespoke authentication provider by implementing AuthenticationSchemeSelector
interface.
@Configuration
public class GrpcSecurityConfiguration extends GrpcSecurityConfigurerAdapter {
@Override
public void configure(GrpcSecurity builder) throws Exception {
builder.authorizeRequests()
.anyMethod().authenticated()//(1)
.and()
.authenticationSchemeSelector(new AuthenticationSchemeSelector() { //(2)
@Override
public Optional<Authentication> getAuthScheme(CharSequence authorization) {
return new MyAuthenticationObject();// (3)
}
})
.authenticationProvider(new AuthenticationProvider() {// (4)
@Override
public Authentication authenticate(Authentication authentication) throws AuthenticationException {
MyAuthenticationObject myAuth= (MyAuthenticationObject)authentication;
//validate myAuth
return MyValidatedAuthenticationObject(withAuthorities);//(5)
}
@Override
public boolean supports(Class<?> authentication) {
return MyAuthenticationObject.class.isInstance(authentication);
}
});
}
}
-
Secure all services methods.
-
Register your own
AuthenticationSchemeSelector
. -
Based on provided authorization header - return
Authentication
object as a claim (not authenticated yet) -
Register your own
AuthenticationProvider
that supports validation ofMyAuthenticationObject
-
Validate provided
authentication
and return validated and authenticatedAuthentication
object
AuthenticationSchemeSelector
can also be registered by defining Spring bean in your application context:
@Bean
public AuthenticationSchemeSelector myCustomSchemeSelector(){
return authHeader->{
// your logic here
};
}
Client side configuration support section explains how to pass custom authorization scheme and claim from GRPC client.
Starting from version 4.5.9
you can also use standard @PreAuthorize
and @PostAuthorize
annotations on grpc service methods and grpc service types.
Call Type | Input object ref | Output object ref | Sample |
---|---|---|---|
Unary |
By parameter name |
|
@Override
@PreAuthorize("#person.age<12")
@PostAuthorize("returnObject.description.length()>0")
public void unary(Person person, StreamObserver<Assignment> responseObserver) {
} |
Input stream, |
|
|
@Override
@PreAuthorize("#p0.getAge()<12")
@PostAuthorize("returnObject.description.length()>0")
public StreamObserver<Person> inStream(StreamObserver<Assignment> responseObserver) {
} |
Single request, |
By parameter name |
|
@Override
@PreAuthorize("#person.age<12")
@PostAuthorize("returnObject.description.length()>0")
public void outStream(Person person, StreamObserver<Assignment> responseObserver) {
} |
Bidi stream |
|
|
@Override
@PreAuthorize("#p0.age<12")
@PostAuthorize("returnObject.description.length()>0")
public StreamObserver<Person> bidiStream(StreamObserver<Assignment> responseObserver) {
} |
To obtain Authentication
object in the implementation of secured method, please use below snippet
final Authentication auth = GrpcSecurity.AUTHENTICATION_CONTEXT_KEY.get();
Starting from 4.5.6
, the Authentication
object can also be obtained via standard Spring API :
final Authentication auth = SecurityContextHolder.getContext().getAuthentication();
By adding io.github.lognet:grpc-client-spring-boot-starter
dependency to your java grpc client application you can easily configure per-channel or per-call credentials :
- Per-channel
-
class MyClient{ public void doWork(){ final AuthClientInterceptor clientInterceptor = new AuthClientInterceptor((1) AuthHeader.builder() .bearer() .binaryFormat(true)(3) .tokenSupplier(this::generateToken)(4) ); Channel authenticatedChannel = ClientInterceptors.intercept( ManagedChannelBuilder.forAddress("host", 6565).build(), clientInterceptor (2) ); // use authenticatedChannel to invoke GRPC service } private ByteBuffer generateToken(){ (4) // generate bearer token against your resource server } }
-
Create client interceptor
-
Intercept channel
-
Turn the binary format on/off:
-
When
true
, the authentication header is sent withAuthorization-bin
key using binary marshaller. -
When
false
, the authentication header is sent withAuthorization
key using ASCII marshaller.
-
-
Provide token generator function (Please refer to for example.)
-
- Per-call
-
class MyClient{ public void doWork(){ AuthCallCredentials callCredentials = new AuthCallCredentials( (1) AuthHeader.builder().basic("user","pwd".getBytes()) ); final SecuredGreeterGrpc.SecuredGreeterBlockingStub securedFutureStub = SecuredGreeterGrpc.newBlockingStub(ManagedChannelBuilder.forAddress("host", 6565));(2) final String reply = securedFutureStub .withCallCredentials(callCredentials)(3) .sayAuthHello(Empty.getDefaultInstance()).getMessage(); } }
-
Create call credentials with basic scheme
-
Create service stub
-
Attach call credentials to the call
AuthHeader
could also be built with bespoke authorization scheme :AuthHeader .builder() .authScheme("myCustomAuthScheme") .tokenSupplier(()->generateMyCustomToken())
-
The starter registers the default implementation of HealthServiceImpl.
You can provide you own by registering ManagedHealthStatusService bean in your application context.
If you have org.springframework.boot:spring-boot-starter-actuator
and org.springframework.boot:spring-boot-starter-web
in the classpath, the starter will expose:
-
grpc
health indicator under/actuator/health
endpoint. -
/actuator/grpc
endpoint.
Starting from version 3.3.0
, the starter will auto-register the running grpc server in Consul registry if org.springframework.cloud:spring-cloud-starter-consul-discovery
is in classpath and
spring.cloud.service-registry.auto-registration.enabled
is NOT set to false
.
The registered service name will be prefixed with grpc-
,i.e. grpc-${spring.application.name}
to not interfere with standard registered web-service name if you choose to run both embedded Grpc
and Web
servers.
ConsulDiscoveryProperties
are bound from configuration properties prefixed by spring.cloud.consul.discovery
and then the values are overwritten by grpc.consul.discovery
prefixed properties (if set). This allows you to have separate consul discovery configuration for rest
and grpc
services if you choose to expose both from your application.
spring:
cloud:
consul:
discovery:
metadata:
myKey: myValue (1)
tags:
- myWebTag (2)
grpc:
consul:
discovery:
tags:
- myGrpcTag (3)
-
Both
rest
andgrpc
services are registered with metadatamyKey=myValue
-
Rest services are registered with
myWebTag
-
Grpc services are registered with
myGrpcTag
Setting spring.cloud.consul.discovery.register-health-check
(or grpc.consul.discovery.register-health-check
) to true
will register GRPC health check service with Consul.
There are 4 supported registration modes :
-
SINGLE_SERVER_WITH_GLOBAL_CHECK
(default)
In this mode the running grpc server is registered as single service with singlegrpc
check with emptyserviceId
.
Please note that default implementation does nothing and simply returnsSERVING
status. You might want to provide your custom Health check implementation for this mode. -
SINGLE_SERVER_WITH_CHECK_PER_SERVICE
In this mode the running grpc server is registered as single service with check per each discoveredgrpc
service. -
STANDALONE_SERVICES
In this mode each discovered grpc service is registered as single service with single check. Each registered service is tagged by its own service name. -
NOOP
- no grpc services registered. This mode is useful if you serve bothrest
andgrpc
services in your application, but for some reason, onlyrest
services should be registered with Consul.
grpc:
consule:
registration-mode: SINGLE_SERVER_WITH_CHECK_PER_SERVICE
When building production-ready services, the advise is to have separate project for your service(s) gRPC API that holds only proto-generated classes both for server and client side usage.
You will then add this project as implementation
dependency to your gRPC client
and gRPC server
projects.
To integrate Eureka
simply follow the great guide from Spring.
Below are the essential parts of configurations for both server and client projects.
-
Add eureka starter as dependency of your server project together with generated classes from
proto
files:
dependencies {
implementation('org.springframework.cloud:spring-cloud-starter-eureka')
implementation project(":yourProject-api")
}
-
Configure gRPC server to register itself with Eureka.
bootstrap.yamlspring: application: name: my-service-name (1)
-
Eureka’s
ServiceId
by default is the spring application name, provide it before the service registers itself with Eureka.application.yamlgrpc: port: 6565 (1) eureka: instance: nonSecurePort: ${grpc.port} (2) client: serviceUrl: defaultZone: http://${eureka.host:localhost}:${eureka.port:8761}/eureka/ (3)
-
Specify the port number the gRPC is listening on.
-
Register the eureka service port to be the same as
grpc.port
so client will know where to send the requests to. -
Specify the registry URL, so the service will register itself with.
-
-
Expose the gRPC service as part of Spring Boot Application.
EurekaGrpcServiceApp.java@SpringBootApplication @EnableEurekaClient public class EurekaGrpcServiceApp { @GRpcService public static class GreeterService extends GreeterGrpc.GreeterImplBase { @Override public void sayHello(GreeterOuterClass.HelloRequest request, StreamObserver<GreeterOuterClass.HelloReply> responseObserver) { } } public static void main(String[] args) { SpringApplication.run(DemoApp.class,args); } }
-
Add eureka starter as dependency of your client project together with generated classes from
proto
files:
dependencies {
implementation('org.springframework.cloud:spring-cloud-starter-eureka')
implementation project(":yourProject-api")
}
-
Configure client to find the eureka service registry:
eureka:
client:
register-with-eureka: false (1)
service-url:
defaultZone: http://${eureka.host:localhost}:${eureka.port:8761}/eureka/ (2)
-
false
if this project is not meant to act as a service to another client. -
Specify the registry URL, so this client will know where to look up the required service.
@EnableEurekaClient
@SpringBootApplication
public class GreeterServiceConsumerApplication {
public static void main(String[] args) {
SpringApplication.run(GreeterServiceConsumerApplication.class, args);
}
}
-
Use EurekaClient to get the coordinates of gRPC service instance from Eureka and consume the service :
@EnableEurekaClient
@Component
public class GreeterServiceConsumer {
@Autowired
private EurekaClient client;
public void greet(String name) {
final InstanceInfo instanceInfo = client.getNextServerFromEureka("my-service-name", false);//(1)
final ManagedChannel channel = ManagedChannelBuilder.forAddress(instanceInfo.getIPAddr(), instanceInfo.getPort())
.usePlaintext()
.build(); //(2)
final GreeterServiceGrpc.GreeterServiceFutureStub stub = GreeterServiceGrpc.newFutureStub(channel); //(3)
stub.greet(name); //(4)
}
}
-
Get the information about the
my-service-name
instance. -
Build
channel
accordingly. -
Create stub using the
channel
. -
Invoke the service.