如何在Java gRPC客户端运行时抑制UnknownHostException异常？

问题场景 #

当使用Java gRPC阻塞存根向无法解析的主机发起请求时，代码示例如下：

ManagedChannel channel = ManagedChannelBuilder.forAddress("SomeUnknownHost", 12345)
                                                .usePlaintext()
                                                .build();
MyServiceGrpc.MyServiceBlockingStub clientStub;
clientStub = MyServiceGrpc.newBlockingStub(channel);
clientStub.increment(Message.newBuilder()
                            .setNumber(1)
                            .build());

运行时异常 #

上述代码会抛出UnknownHostException，具体异常信息如下：

Mar 03, 2024 2:43:02 A.M. io.grpc.internal.ManagedChannelImpl$NameResolverListener handleErrorInSyncContext
WARNING: [Channel<1>: (SomeUnknownHost:12345)] Failed to resolve name. status=Status{code=UNAVAILABLE, description=Unable to resolve host UnknownHostException, cause=java.lang.RuntimeException: java.net.UnknownHostException: foobar: nodename nor servname provided, or not known
    at io.grpc.internal.DnsNameResolver.resolveAddresses(DnsNameResolver.java:223)
    at io.grpc.internal.DnsNameResolver.doResolve(DnsNameResolver.java:282)
    at io.grpc.internal.DnsNameResolver$Resolve.run(DnsNameResolver.java:318)
    at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
    at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
    at java.base/java.lang.Thread.run(Thread.java:1589)
Caused by: java.net.UnknownHostException: SomeUnknownHost: nodename nor servname provided, or not known
    at java.base/java.net.Inet6AddressImpl.lookupAllHostAddr(Native Method)
    at java.base/java.net.Inet6AddressImpl.lookupAllHostAddr(Inet6AddressImpl.java:52)
    at java.base/java.net.InetAddress$PlatformResolver.lookupByName(InetAddress.java:1059)
    at java.base/java.net.InetAddress.getAddressesFromNameService(InetAddress.java:1673)
    at java.base/java.net.InetAddress$NameServiceAddresses.get(InetAddress.java:1003)
    at java.base/java.net.InetAddress.getAllByName0(InetAddress.java:1663)
    at java.base/java.net.InetAddress.getAllByName(InetAddress.java:1528)
    at io.grpc.internal.DnsNameResolver$JdkAddressResolver.resolveAddress(DnsNameResolver.java:632)
    at io.grpc.internal.DnsNameResolver.resolveAddresses(DnsNameResolver.java:219)
    ... 5 more

现有尝试方案 #

曾参考以下思路：

在gRPC中，调用默认是“快速失败”的，这意味着一旦发生网络故障，调用会被视为失败。你可以在存根上覆盖此行为

尝试通过在客户端存根中启用Wait-for-Ready调用选项来修改默认行为，修改后的存根代码如下：

public static final class MyServiceBlockingStub extends io.grpc.stub.AbstractStub<MyServiceBlockingStub> {
    private MyServiceBlockingStub(io.grpc.Channel channel) {
      super(channel);
    }

    private MyServiceBlockingStub(io.grpc.Channel channel,
        io.grpc.CallOptions callOptions) {
      super(channel, callOptions.withWaitForReady());
    }

    @java.lang.Override
    protected MyServiceBlockingStub build(io.grpc.Channel channel,
        io.grpc.CallOptions callOptions) {
      return new MyServiceBlockingStub(channel, callOptions.withWaitForReady());
    }

    public Message increment(Message request) {
      System.out.println("WAIT-FOR-READY IS ON: " + getCallOptions().withWaitForReady().isWaitForReady()); // TRUE
      return blockingUnaryCall(
          getChannel(), getIncrementMethod(), getCallOptions().withWaitForReady(), request);
    }
}

但修改后运行时仍会抛出UnknownHostException异常。

可行解决方法 #

1. 正确启用Wait-for-Ready模式 #

修改存根类的方式容易出错，更简洁可靠的方式是在创建存根时直接调用withWaitForReady()方法，确保CallOptions正确配置：

MyServiceGrpc.MyServiceBlockingStub clientStub = MyServiceGrpc.newBlockingStub(channel)
    .withWaitForReady();

启用该模式后，gRPC客户端不会在域名解析失败时立即抛出异常，而是会持续尝试建立连接，直到达到通道的超时时间（默认无硬性超时，可通过ManagedChannelBuilder的keepAlive相关参数调整）。

2. 配置域名解析重试策略 #

若需要更精细地控制DNS解析的重试逻辑，可以通过配置通道的默认服务参数实现。例如，设置重试策略让客户端在遇到UNAVAILABLE状态（包含DNS解析失败）时自动重试：

String serviceConfig = "{\"retryPolicy\":{\"maxAttempts\":5,\"initialBackoff\":\"0.1s\",\"maxBackoff\":\"1s\",\"backoffMultiplier\":2,\"retryableStatusCodes\":[\"UNAVAILABLE\"]}}";
ManagedChannel channel = ManagedChannelBuilder.forAddress("SomeUnknownHost", 12345)
        .usePlaintext()
        .defaultServiceConfig(serviceConfig)
        .build();

该配置会让客户端最多尝试5次，重试间隔按倍数递增。

3. 捕获并处理异常 #

如果只是需要抑制异常抛出，而非持续重试，可以在调用gRPC方法时捕获StatusRuntimeException，并根据状态码进行处理：

try {
    clientStub.increment(Message.newBuilder().setNumber(1).build());
} catch (StatusRuntimeException e) {
    if (e.getStatus().getCode() == Status.Code.UNAVAILABLE) {
        // 忽略DNS解析失败异常，或执行自定义降级逻辑
        System.out.println("域名解析失败，已忽略该异常");
    } else {
        // 处理其他异常
        throw e;
    }
}

内容的提问来源于stack exchange，提问作者nawK