go network_error ai_generated true

rpc error: code = Unavailable desc = last connection error: connection error: desc = "transport: Error while dialing: dial tcp: lookup grpc.example.com on 10.0.0.2:53: no such host"

ID: go/grpc-unavailable-subchannel-transient-failure

Also available as: JSON · Markdown · 中文
80%Fix Rate
87%Confidence
0Evidence
2025-03-04First Seen

Version Compatibility

VersionStatusIntroducedDeprecatedNotes
1.30+ active — — —

Root Cause

DNS resolution failed for the gRPC target. Common in Kubernetes when the service name is wrong, the DNS server is unreachable, or the pod's resolv.conf is misconfigured.

generic

中文

gRPC 目标的 DNS 解析失败。常见于 Kubernetes 中服务名错误、DNS 服务器不可达，或 Pod 的 resolv.conf 配置错误时。

Workarounds

  1. 92% success 
    // Use <service>.<namespace>.svc.cluster.local
conn, _ := grpc.Dial("users.default.svc.cluster.local:50051", grpc.WithTransportCredentials(insecure.NewCredentials()))
  2. 88% success 
    conn, _ := grpc.Dial("dns:///users.default.svc.cluster.local:50051",
  grpc.WithTransportCredentials(insecure.NewCredentials()),
  grpc.WithDefaultServiceConfig(`{"loadBalancingConfig":[{"round_robin":{}}]}`))

Dead Ends

Common approaches that don't work:

  1. 75% fail

    Breaks when the backend pod is rescheduled; loses load balancing across replicas

  2. 90% fail

    DNS NXDOMAIN returns immediately; a longer timeout does not make the name resolve