The only tool you need to know what is happening and how to fix it.
Go alertingexception-trackerissue-trackinglogsmetricsmonitoringobservability-platformopen-telemetryotelsession-replaytelemetrytraces
fix: setup skill fix for js and backend
ddux597a01986888
diff --git a/skills/traceway-setup/SKILL.md b/skills/traceway-setup/SKILL.mdindex bacfd1fc..4b125d0a 100644--- a/skills/traceway-setup/SKILL.md+++ b/skills/traceway-setup/SKILL.md@@ -151,6 +151,8 @@ span?.setStatus({ code: SpanStatusCode.ERROR, message: error.message });(Go: `span.RecordError(err, trace.WithStackTrace(true))`; the stack trace option is what produces the `exception.stacktrace` attribute.)+**Return HTTP 500 when an exception happens.** A request that throws or panics must respond with `500` and set the span status to `ERROR` — never let it fall through to a `200`/`2xx`. The common trap is a recover/panic middleware that reports the exception but lets the response writer keep its default `200`: Traceway then records the endpoint transaction as a *success*, so the Endpoints page looks healthy while Issues fills up, and the exception becomes an unlinked island instead of correlating to a failed request (and, with distributed tracing, to the frontend call that triggered it). Always pair "an exception was recorded" with "respond `500` and set span status `ERROR`". Genuine client errors that are NOT bugs — validation `422`, auth `401`, not-found `404` — keep their real 4xx status and should not be recorded as exceptions in the first place.+### If the user explicitly asks for the Traceway Go SDKOnly on explicit request, instead of OTel: `go get go.tracewayapp.com/tracewaygin` (or `tracewaychi`, `tracewayfiber`, `tracewayfasthttp`, `tracewayhttp`), add the middleware with connection string `<project-token>@https://<instance>/api/report`. Trade-off: the Go SDK natively emits the built-in system metric names (`cpu.used_pcnt`, `mem.used`, `go.*`) that populate the dashboard's built-in charts; on the OTel path those charts stay empty and host metrics come from the Traceway OTel Agent (Step 5).diff --git a/skills/traceway-setup/frontend-js.md b/skills/traceway-setup/frontend-js.mdindex 39a4a14c..6e0481fa 100644--- a/skills/traceway-setup/frontend-js.md+++ b/skills/traceway-setup/frontend-js.md@@ -198,7 +198,7 @@ clearAttributes();### Distributed tracing-The SDK instruments `fetch` (and XHR in the jQuery SDK) to propagate a `traceway-trace-id` header on same-origin requests, linking frontend errors to the backend requests that caused them. No configuration. For Axios:+The SDK instruments `fetch` (and XHR in the jQuery SDK) to propagate a `traceway-trace-id` header on same-origin requests, and the backend echoes the same id back on the response. Errors the SDK captures on its own (uncaught exceptions, unhandled rejections, `captureHttpServerErrors`) link to the originating backend request automatically. For Axios, register the interceptor:```javascriptimport { createAxiosInterceptor } from "@tracewayapp/frontend";@@ -206,6 +206,29 @@ import { createAxiosInterceptor } from "@tracewayapp/frontend";api.interceptors.request.use(createAxiosInterceptor());```+**Manual captures in a `fetch`/API wrapper MUST pass the trace id explicitly — this is the #1 reason frontend issues show up unlinked to the backend.** The SDK holds the active distributed-trace id only *for the duration of the request* and clears it the instant the fetch settles. A `captureException` call in your `catch` or `if (!res.ok)` branch runs *after* `await fetch`, by which point the active id is already `null`, so the exception is stored with `distributedTraceId: null` and never connects to the backend trace — even though the request header was sent correctly. Read the id the backend echoed on the response and pass it through as the third argument:++```javascript+import { captureExceptionWithAttributes } from "@tracewayapp/frontend";++async function request(path, options) {+ const res = await fetch(path, options);+ if (!res.ok) {+ const err = new Error(`Request to ${path} failed (${res.status})`);+ const distributedTraceId = res.headers.get("traceway-trace-id") || undefined;+ captureExceptionWithAttributes(+ err,+ { path, method: options?.method || "GET", status: String(res.status) },+ distributedTraceId ? { distributedTraceId } : undefined+ );+ throw err;+ }+ return res.json();+}+```++The same applies to the React `useTraceway().captureExceptionWithAttributes`, the Vue/Svelte equivalents, and any other place you capture by hand after awaiting a request. When the frontend and backend are on different origins, the backend must also add `traceway-trace-id` to `Access-Control-Expose-Headers` or the browser cannot read it off the response.+## Bundler Plugin (debug IDs)`@tracewayapp/bundler-plugin` embeds a deterministic 128-bit debug ID into each emitted chunk and its `.map` (ECMA-426 debug ID format, interoperable with Sentry tooling). The backend then matches stack frames to the exact map from the same build, immune to filename collisions and concurrent deploys. Without it, matching falls back to filename (fine with content-hashed bundle names, broken with stable names like `app.js`).