1
0

Spring gRPC on Spring Boot 4: four call types, exception mapping, and a symptom-first troubleshooting suite

Companion code for the ankurm.com guide. Verified on Spring Boot 4.1.0, spring-grpc 1.1.0,
grpc-java 1.80.0, protobuf-java 4.34.2, JDK 25.0.3. results-full.txt is unedited mvn test
output: 11 tests, 0 failures. No installs needed - protoc and the gRPC codegen plugin resolve
as Maven artifacts, and every test uses the in-process transport.

_01_basics       all four call types and how their failure modes differ: iterator semantics on
                 server streaming, the half-close that client streaming hangs without, and the
                 independence of the two streams in bidi.
_02_troubleshooting
                 four failures reproduced then fixed - the 4 MB message limit and which side
                 enforces it, the absent default deadline, cancellation that never interrupts a
                 thread, and errors that arrive as UNKNOWN with no description.
_03_exceptions   @GrpcAdvice / @GrpcExceptionHandler: domain exception to NOT_FOUND with
                 trailers, validation to INVALID_ARGUMENT, and a catch-all that returns a
                 deliberate INTERNAL without leaking the original message across the boundary.

docs/01          symptom-first troubleshooting index, each entry marked [tested] or [documented]
docs/02          complete spring.grpc.* property reference, and the split between the Boot 4.1
                 integration (org.springframework.boot, version 4.1.0, owns properties and
                 auto-configuration) and the Spring gRPC project (org.springframework.grpc,
                 version 1.1.0, owns the programming model). The older standalone
                 spring-grpc-spring-boot-starter stops at 1.0.3 and is what most search results
                 describe.

Findings worth the commit message
  - The in-process transport CANNOT enforce message size limits: it passes messages by
    reference and never serialises them. A 4 MB + 1 KB message goes through cleanly in tests
    and fails in production with RESOURCE_EXHAUSTED. The test asserts this rather than
    pretending otherwise. Same blind spot covers compression, TLS, keepalive and LB.
  - Two property names cost real time while writing this:
    spring.grpc.server.inprocess.name (not in-process) and
    spring.grpc.client.channel.<n>.target (singular channel, and target not address).
    The second failure surfaces as UnknownHostException on the channel NAME.
  - gRPC still has no default deadline, and cancellation only sets a Context flag.
This commit is contained in:
2026-07-31 22:53:13 +05:30
commit 8547981634
17 changed files with 1654 additions and 0 deletions

188
docs/01-troubleshooting.md Normal file
View File

@@ -0,0 +1,188 @@
# Troubleshooting Spring gRPC on Boot 4
Symptom-first. Each entry is something whose error message points somewhere other than its cause.
Entries marked **[tested]** are reproduced in
[`HardToDiagnoseTest`](../src/test/java/com/ankurm/grpc/_02_troubleshooting/HardToDiagnoseTest.java);
entries marked **[documented]** need a real network and are stated rather than asserted, because a
test claiming to prove them over in-process transport would be lying.
---
## `UNAVAILABLE: Unable to resolve host <name>` — **[tested]**
**Cause, 90% of the time: a typo in a property name**, not a networking problem.
An unmatched channel name is passed straight to the DNS resolver as a target. Boot does not warn
that the channel is unconfigured. The exact names, both of which are easy to get wrong:
```properties
spring.grpc.server.inprocess.name=orders-test # NOT "in-process"
spring.grpc.client.channel.orders.target=static://... # SINGULAR "channel"; "target", NOT "address"
```
Also check `spring.grpc.client.inprocess.enabled=true` if you are using the in-process transport
from the client side.
## `IllegalStateException: No grpc channel factory found that supports target : <name>`
`GrpcChannelFactory` is a composite that asks each registered factory `supports(target)` **before**
named-channel indirection is applied. Passing a bare logical name to `createChannel()` can therefore
fail even though the name is correctly configured. In application code, inject the stub (via
`@ImportGrpcClients`) and let Boot resolve the target; in tests, pass the full target.
## `RESOURCE_EXHAUSTED: gRPC message exceeds maximum size 4194304` — **[tested]**
The default limit is **4 MB, per message, on the receiver**, and the two sides are configured
separately.
```properties
# large RESPONSE -> the CLIENT is the receiver
spring.grpc.client.channel.<name>.inbound.message.max-size=16MB
# large REQUEST -> the SERVER is the receiver
spring.grpc.server.inbound.message.max-size=16MB
```
Per call: `stub.withMaxInboundMessageSize(16 * 1024 * 1024)`.
Half of all "I raised the limit and it still fails" reports are the asymmetry: raising it on the
server does nothing for a large response.
> **The testing trap.** The in-process transport passes messages **by reference and never
> serialises them**, so it *cannot* enforce size limits. A payload-size regression passes every
> in-process test you have and fails in production. The same applies to compression and anything
> else depending on the wire format. This is demonstrated — the test asserts that a 4 MB + 1 KB
> message goes through in-process without complaint.
Prefer streaming to large unary messages: the limit is per message, so 1,000 small messages are
fine. Do not raise the limit globally — a large limit turns a malformed request into an OOM.
## Calls that hang forever — **[tested]**
**gRPC has no default deadline.** A call without one waits indefinitely. Nothing warns you, and in
testing the server always responds quickly, so it surfaces the day something downstream is slow —
as a thread pool that fills up with no errors in the log.
```java
stub.withDeadlineAfter(2, TimeUnit.SECONDS).getOrder(request);
```
Per channel: `spring.grpc.client.channel.<name>.default.deadline=2s`, or register a
`DefaultDeadlineSetupClientInterceptor`.
Deadlines are **absolute and propagate**: if A calls B with 2s remaining, B sees 2s, not a fresh 2s.
Set the deadline at the edge; do not re-set it at every hop, or a deep chain multiplies its budget.
## The server keeps working after the client gave up — **[tested]**
gRPC **does not interrupt your thread** on cancellation. It sets a flag on the `Context`.
```java
for (Item item : items) {
if (Context.current().isCancelled()) return; // no onCompleted/onError -- stream is closed
responseObserver.onNext(convert(item));
}
```
For blocking work handed to another thread, propagate with `Context.current().wrap(runnable)`
otherwise the flag is invisible there. Calling `onNext`/`onCompleted` after cancellation throws
`IllegalStateException`.
## Every error is `UNKNOWN` with no message — **[tested]**
Any exception escaping a handler becomes `UNKNOWN` with a **null description** — deliberately, since
leaking exception text across a service boundary is an information-disclosure risk.
```java
Metadata trailers = new Metadata();
trailers.put(REASON_KEY, "ORDER_LOCKED");
responseObserver.onError(new StatusRuntimeException(
Status.FAILED_PRECONDITION.withDescription("order is locked"), trailers));
```
Choose the code carefully — it is the contract that retry policies, circuit breakers and dashboards
key off:
| Do **not** retry | Retry may help |
|---|---|
| `INVALID_ARGUMENT`, `NOT_FOUND`, `ALREADY_EXISTS`, `FAILED_PRECONDITION`, `PERMISSION_DENIED`, `UNAUTHENTICATED` | `UNAVAILABLE`, `DEADLINE_EXCEEDED`, `RESOURCE_EXHAUSTED`, `ABORTED` |
Getting this wrong makes non-retryable failures look retryable and turns one bad request into a
storm.
## `IllegalStateException: call already closed` / `half-closed`
You called `onNext` or `onCompleted` after the stream ended. Usual causes:
- responding after cancellation or deadline expiry (check `Context.current().isCancelled()` first);
- calling `onCompleted()` twice;
- calling `onError()` and then `onCompleted()``onError` is terminal;
- two threads writing to one `StreamObserver`. **`StreamObserver` is not thread-safe.** Concurrent
`onNext` corrupts the stream, and the symptom is usually a deserialization error on the *far*
side, pointing nowhere near the bug.
## A client-streaming call never completes
You forgot the half-close. `requestObserver.onCompleted()` is what tells the server "no more
requests"; without it the server's `onCompleted` never fires and the call hangs until the deadline —
or forever, if you did not set one.
## Retries do nothing — **[documented]**
gRPC retries are configured through the **service config**, not a client API, and are off by
default:
```properties
spring.grpc.client.channel.orders.service-config.methodConfig[0].retryPolicy.maxAttempts=4
spring.grpc.client.channel.orders.service-config.methodConfig[0].retryPolicy.initialBackoff=0.5s
spring.grpc.client.channel.orders.service-config.methodConfig[0].retryPolicy.retryableStatusCodes[0]=UNAVAILABLE
```
Three reasons they silently do nothing:
1. The status you are failing with is not in `retryableStatusCodes`.
2. The response was already partially delivered — gRPC will not retry a committed stream.
3. Retry is disabled on the channel (`enableRetry()` / the corresponding property).
Only make idempotent methods retryable. gRPC has no idea whether your RPC is safe to repeat.
## Connections drop every N minutes through a proxy — **[documented]**
Load balancers and ingress controllers close idle HTTP/2 connections. gRPC reuses one long-lived
connection, so an idle stream is not an idle *connection* — the proxy disagrees.
```properties
spring.grpc.client.channel.<name>.keepalive.time=30s
spring.grpc.client.channel.<name>.keepalive.timeout=5s
spring.grpc.server.keepalive.connection.max-idle-time=...
spring.grpc.server.keepalive.connection.max-age=...
```
Keep the client's keepalive interval **shorter** than the proxy's idle timeout. Note the server can
reject keepalives that are too frequent (`ENHANCE_YOUR_CALM` / `too_many_pings`), which looks like a
random disconnect — so both ends need to agree.
## All traffic goes to one backend — **[documented]**
The default load-balancing policy is `pick_first`, not `round_robin`. With a DNS target resolving to
several addresses, a gRPC client picks one connection and keeps it. This is correct behaviour and
almost never what people expect behind a headless Kubernetes service.
Set `round_robin` via service config, and remember that DNS re-resolution happens on connection
failure, not on a timer — so scaling up does not redistribute existing connections.
---
## Test transport limitations, collected
The in-process transport is excellent for testing protocol behaviour and misleading for anything
wire-related. It **does** exercise interceptors, metadata, status codes, deadlines, cancellation and
flow control. It does **not** exercise:
- message size limits (no serialization — proven in the test suite),
- compression,
- TLS and credentials negotiation,
- keepalive, GOAWAY, idle timeouts,
- name resolution and load balancing,
- HTTP/2 framing and header limits.
Anything in that second list needs a real port, and belongs in a smaller, slower test tier.

View File

@@ -0,0 +1,144 @@
# Properties reference, and which project owns what
Two things that cause disproportionate confusion: the exact property names, and the fact that
**"Spring gRPC" now means two different things with two different version numbers**.
---
## 1. Boot 4.1 integration vs. the Spring gRPC project
These are separate artifacts with separate versions, and both are on your classpath.
| | **Spring Boot gRPC integration** | **Spring gRPC** |
|---|---|---|
| Group | `org.springframework.boot` | `org.springframework.grpc` |
| Artifacts | `spring-boot-starter-grpc-server`, `spring-boot-starter-grpc-client`, `spring-boot-grpc-server`, `spring-boot-grpc-client`, `spring-boot-grpc-test` | `spring-grpc-core` |
| Version here | **4.1.0** (the Boot version) | **1.1.0** |
| Owns | auto-configuration, `spring.grpc.*` properties, starters, actuator/health, observation wiring | the programming model: `@GrpcService`, `@GrpcAdvice`, `@GrpcExceptionHandler`, `GrpcChannelFactory`, `@ImportGrpcClients`, interceptor infrastructure |
| Managed by | the Boot BOM directly | the Boot BOM, via the `spring-grpc.version` property |
Practical consequences:
- **Property names (`spring.grpc.*`) come from Boot 4.1**, defined in `GrpcServerProperties` /
`GrpcClientProperties` under `org.springframework.boot.grpc.*.autoconfigure`. Version them against
your Boot version.
- **Annotations and interfaces come from Spring gRPC 1.1.0**, in `org.springframework.grpc.*`.
Version them against `spring-grpc.version`.
- Boot 4.1 also pins `grpc-java` (**1.80.0** here) and `protobuf-java` (**4.34.2**). Both are
older than the newest releases on Maven Central — that is deliberate, and overriding them
independently is how you get `NoSuchMethodError` between grpc-java and its Netty shading.
### The older path, and why you will see it in search results
Before Boot 4, the community project shipped its own starter:
```xml
<!-- Boot 3 era -->
<dependency>
<groupId>org.springframework.grpc</groupId>
<artifactId>spring-grpc-spring-boot-starter</artifactId>
</dependency>
```
That artifact stops at **1.0.3** and is not what Boot 4 uses. On Boot 4 use the Boot starters and
let the BOM manage everything:
```xml
<!-- Boot 4 -->
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-grpc-server</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-grpc-client</artifactId>
</dependency>
```
Most blog posts and Stack Overflow answers you find will be describing the older artifact. The
programming model is largely the same; the dependency coordinates and the property names are not.
---
## 2. Server properties — exact names
Read from `spring-boot-grpc-server-4.1.0.jar`'s configuration metadata. This is the complete list.
| Property | Notes |
|---|---|
| `spring.grpc.server.enabled` | |
| `spring.grpc.server.port` | `-1` disables the network server (in-process only); `0` picks an ephemeral port |
| `spring.grpc.server.address` | |
| `spring.grpc.server.inprocess.name` | **`inprocess`, one word — not `in-process`** |
| `spring.grpc.server.inbound.message.max-size` | the 4 MB default, for incoming **requests** |
| `spring.grpc.server.inbound.metadata.max-size` | header size limit |
| `spring.grpc.server.shutdown.grace-period` | |
| `spring.grpc.server.keepalive.time` | |
| `spring.grpc.server.keepalive.timeout` | |
| `spring.grpc.server.keepalive.permit.time` | minimum client ping interval the server tolerates |
| `spring.grpc.server.keepalive.permit.without-calls` | |
| `spring.grpc.server.keepalive.connection.max-age` | |
| `spring.grpc.server.keepalive.connection.max-idle-time` | |
| `spring.grpc.server.keepalive.connection.grace-period` | |
| `spring.grpc.server.ssl.enabled` / `.bundle` / `.client-auth` / `.secure` | uses Boot SSL bundles |
| `spring.grpc.server.health.enabled` | standard gRPC health service |
| `spring.grpc.server.health.service` | |
| `spring.grpc.server.health.include-overall-health` | |
| `spring.grpc.server.health.status.mapping` / `.order` | actuator status → gRPC serving status |
| `spring.grpc.server.health.schedule.enabled` / `.delay` / `.period` | |
| `spring.grpc.server.health.services.validate-membership` | |
| `spring.grpc.server.reflection.enabled` | needed for `grpcurl` without local `.proto` files |
| `spring.grpc.server.observation.enabled` | Micrometer observations |
| `spring.grpc.server.netty.transport` | |
| `spring.grpc.server.netty.domain-socket-path` | |
| `spring.grpc.server.servlet.enabled` / `.validate-http2` | gRPC over the servlet container |
| `spring.grpc.server.security.csrf.enabled` | |
| `spring.grpc.server.factory.enabled` | |
## 3. Client properties — exact names
The client side is a **map keyed by channel name**:
```
spring.grpc.client.channel.<name>.<property>
```
**`channel`, singular.** `channels` is wrong and produces no error — see below.
| Property | Notes |
|---|---|
| `spring.grpc.client.channel.<name>.target` | **`target`, not `address`** |
| `spring.grpc.client.channel.<name>.user-agent` | |
| `spring.grpc.client.channel.<name>.bypass-certificate-validation` | test/dev only |
| `spring.grpc.client.channel.<name>.default.deadline` | **the default deadline gRPC otherwise lacks** |
| `spring.grpc.client.channel.<name>.default.load-balancing-policy` | e.g. `round_robin` (default is `pick_first`) |
| `spring.grpc.client.channel.<name>.inbound.message.max-size` | 4 MB default, for incoming **responses** |
| `spring.grpc.client.channel.<name>.inbound.metadata.max-size` | |
| `spring.grpc.client.channel.<name>.keepalive.time` / `.timeout` / `.without-calls` | |
| `spring.grpc.client.channel.<name>.idle.timeout` | |
| `spring.grpc.client.channel.<name>.ssl.enabled` / `.bundle` | |
| `spring.grpc.client.channel.<name>.health.enabled` / `.service-name` | client-side health checking |
| `spring.grpc.client.channel.<name>.service-config.*` | retry, load balancing, throttling |
| `spring.grpc.client.enabled` | |
| `spring.grpc.client.inprocess.enabled` | register the in-process channel factory |
| `spring.grpc.client.observation.enabled` | |
| `spring.grpc.client.channelfactory.enabled` | |
`service-config` maps to the gRPC service config record with `loadbalancing`, `method`,
`retrythrottling` and `healthcheck` sections.
## 4. The two mistakes, and their symptoms
| Wrong | Right | Symptom |
|---|---|---|
| `spring.grpc.server.in-process.name` | `spring.grpc.server.inprocess.name` | property silently ignored; server never binds in-process |
| `spring.grpc.client.channels.x.address` | `spring.grpc.client.channel.x.target` | `UNAVAILABLE: Unable to resolve host x` / `UnknownHostException: x` |
The second is the expensive one. An unmatched channel name is handed to the DNS resolver **as a
literal target**, so the error names your logical channel as if it were a hostname — and sends you
to investigate DNS, service discovery and networking instead of a typo. Both cost real time while
building this repository.
**Guard against it:** add `spring-boot-configuration-processor` and your IDE will flag unknown
`spring.grpc.*` keys. It will not catch a wrong value inside the `channel.<name>` map (the keys
there are arbitrary), but it catches everything else.