feat(observability): OpenTelemetry 全链路追踪(Jaeger + NATS 跨总线传播)
新增 sundynix-shared/otelx:otelx.Init(ctx,服务名) 注册 W3C 传播器 + OTLP/HTTP 批量导出(默认 localhost:4318,docker 里的 Jaeger);OTEL_SDK_DISABLED=true 关导出。 Jaeger 不在线 / 导出器构建失败都不阻断启动(可观测性是增益而非依赖)。 - NATS 跨进程传播(无现成中间件):bus/trace.go 的 natsHeaderCarrier + inject/extract, PublishTask/CallTool 注入 traceparent,ConsumeTasks/ServeTool 抽出续上 → 链路跨总线连成一棵树。 - 埋点:gateway 挂 otelgin(HTTP server span,链路根);dispatcher task.execute → node.<kind>(每节点,nctx 下传使工具/LLM 挂到节点下)→ llm.stream/llm.generate; bus 自动出 tool.call(client)↔tool.serve(server) 成对跨服务 span。 - docker-compose 加 jaeger all-in-one(UI :16686,OTLP :4318)。 - 依赖修复:otlptracehttp 触发 genproto 单体(旧)vs 拆分模块 ambiguous import(milvus 拉旧版), pin genproto 至后拆分版(go.work 工作区全局生效)。 - production_readiness.md 1.1 更新为「已实现」。 验证:真实 input→retriever→agent 任务在 Jaeger 出 14 span / 3 服务的完整树, 跨 NATS(publish→consume)、跨服务(tool.call→tool.serve)均连通, 瓶颈 kb_search 692ms、llm 1597ms 一眼可见;四模块 build+vet+test 全绿。 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -13,8 +13,11 @@ import (
|
||||
"github.com/cloudwego/eino-ext/components/model/openai"
|
||||
"github.com/cloudwego/eino/components/model"
|
||||
"github.com/cloudwego/eino/schema"
|
||||
"go.opentelemetry.io/otel/attribute"
|
||||
"go.opentelemetry.io/otel/trace"
|
||||
|
||||
"github.com/sundynix/sundynix-shared/contract"
|
||||
"github.com/sundynix/sundynix-shared/otelx"
|
||||
)
|
||||
|
||||
// requestTimeout 是单次推理请求的上限。
|
||||
@@ -102,8 +105,15 @@ func (p *Pool) ChatStream(ctx context.Context, msgs []ChatMessage, onToken func(
|
||||
if cm == nil {
|
||||
return fmt.Errorf("no model configured")
|
||||
}
|
||||
ctx, span := otelx.Tracer().Start(ctx, "llm.stream",
|
||||
trace.WithAttributes(
|
||||
attribute.String("llm.model", p.ModelName()),
|
||||
attribute.Int("llm.messages", len(msgs)),
|
||||
))
|
||||
defer span.End()
|
||||
sr, err := cm.Stream(ctx, toSchema(msgs))
|
||||
if err != nil {
|
||||
span.RecordError(err)
|
||||
return fmt.Errorf("llm stream: %w", err)
|
||||
}
|
||||
defer sr.Close()
|
||||
@@ -128,8 +138,15 @@ func (p *Pool) Chat(ctx context.Context, msgs []ChatMessage) (string, error) {
|
||||
if cm == nil {
|
||||
return "", fmt.Errorf("no model configured")
|
||||
}
|
||||
ctx, span := otelx.Tracer().Start(ctx, "llm.generate",
|
||||
trace.WithAttributes(
|
||||
attribute.String("llm.model", p.ModelName()),
|
||||
attribute.Int("llm.messages", len(msgs)),
|
||||
))
|
||||
defer span.End()
|
||||
out, err := cm.Generate(ctx, toSchema(msgs))
|
||||
if err != nil {
|
||||
span.RecordError(err)
|
||||
return "", fmt.Errorf("llm generate: %w", err)
|
||||
}
|
||||
return out.Content, nil
|
||||
|
||||
Reference in New Issue
Block a user