micro/go-micro · error

stream API request failed: %w

Error message

stream API request failed: %w

What it means

Wraps the transport error from http.DefaultClient.Do in the OpenAI provider's Stream call. The request failed at the network layer before a status code was obtained: DNS failure, connection refused, TLS error, or context cancellation. The original error is preserved with %w.

Source

Thrown at ai/openai/openai.go:224

	if p.opts.Effort != "" {
		apiReq["reasoning_effort"] = p.opts.Effort
	}
	reqBody, err := json.Marshal(apiReq)
	if err != nil {
		return nil, fmt.Errorf("failed to marshal stream request: %w", err)
	}
	apiURL := strings.TrimRight(p.opts.BaseURL, "/") + "/v1/chat/completions"
	httpReq, err := http.NewRequestWithContext(ctx, http.MethodPost, apiURL, bytes.NewReader(reqBody))
	if err != nil {
		return nil, fmt.Errorf("failed to create stream request: %w", err)
	}
	httpReq.Header.Set("Content-Type", "application/json")
	httpReq.Header.Set("Accept", "text/event-stream")
	httpReq.Header.Set("Authorization", "Bearer "+p.opts.APIKey)

	httpResp, err := http.DefaultClient.Do(httpReq)
	if err != nil {
		return nil, fmt.Errorf("stream API request failed: %w", err)
	}
	if httpResp.StatusCode != http.StatusOK {
		defer httpResp.Body.Close()
		respBody, _ := io.ReadAll(httpResp.Body)
		return nil, fmt.Errorf("stream API error (%s): %s", httpResp.Status, string(respBody))
	}
	return &openAIStream{body: httpResp.Body, scanner: bufio.NewScanner(httpResp.Body)}, nil
}

type openAIStream struct {
	body    io.ReadCloser
	scanner *bufio.Scanner
	closed  bool
}

func (s *openAIStream) Recv() (*ai.Response, error) {
	for s.scanner.Scan() {
		line := strings.TrimSpace(s.scanner.Text())

View on GitHub (pinned to 24529f1404)

Solutions

1. Verify reachability: curl -i $BASEURL/v1/chat/completions from the same host
2. Correct opts.BaseURL (scheme, host, port)
3. Increase the context timeout or fix premature cancellation of ctx
4. Configure HTTPS_PROXY / trust the corporate CA if behind a TLS-intercepting proxy

Example fix

// before
ctx, cancel := context.WithTimeout(context.Background(), 500*time.Millisecond)
// after
ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)

// cheap connectivity check to the endpoint host before streaming
conn, err := net.DialTimeout("tcp", host+":443", 3*time.Second)
if err != nil {
    return fmt.Errorf("endpoint unreachable: %w", err)
}
conn.Close()

_, err := provider.Stream(ctx, req)
if err != nil {
    if errors.Is(err, context.DeadlineExceeded) {
        // retry with a longer timeout
    } else if isNetworkError(err) { // errors.As(err, &net.Error)
        // exponential backoff retry
    }
    return err
}

When it happens

Trigger: Calling Stream() when the OpenAI-compatible endpoint is unreachable: wrong BaseURL host/port, no internet/DNS, TLS interception with untrusted certs, or the passed ctx is cancelled or times out before the response.

Common situations: Typo'd BaseURL (e.g. api.openai.com misspelled); running in an air-gapped environment or behind a corporate proxy that strips TLS; context deadline too short for slow network; firewall blocking outbound 443.

Understand the failure class

Background: ECONNREFUSED and "connection refused" / "could not connect to server" errors: what they mean and how to fix them — this error's family across 44 libraries.

Related errors

• stream API request failed: %w
• failed to create stream request: %w
• stream API error (%s): %s
• API request failed: %w
• API error: nil response

AI-assisted analysis of micro/go-micro@24529f1404 (2026-09-01). Data as JSON: /api/errors/fba1826e476a965b. Report an issue: GitHub.