fix

Signed-off-by: Kuromesi <[email protected]>

Author: Kuromesi

pkg/epp/backend/vllm/metrics.go

@@ -49,7 +49,15 @@ const (
 	KvCacheMaxTokenCapacityMetricName = "vllm:gpu_cache_max_token_capacity"
 )
 
-type PodMetricsClientImpl struct{}
+type PodMetricsClientImpl struct {
+	scrapePort int32
+	scrapePath string
+}
+
+func (p *PodMetricsClientImpl) UpdateScrapeOptions(port int32, path string) {
+	p.scrapePort = port
+	p.scrapePath = path
+}
 
 // FetchMetrics fetches metrics from a given pod.
 func (p *PodMetricsClientImpl) FetchMetrics(
@@ -62,7 +70,7 @@ func (p *PodMetricsClientImpl) FetchMetrics(
 
 	// Currently the metrics endpoint is hard-coded, which works with vLLM.
 	// TODO(https://github.com/kubernetes-sigs/gateway-api-inference-extension/issues/16): Consume this from InferencePool config.
-	url := existing.Address + ":" + strconv.Itoa(int(port)) + "/metrics"
+	url := "http://" + existing.Address + ":" + strconv.Itoa(int(port)) + "/metrics"
 
 	req, err := http.NewRequestWithContext(ctx, http.MethodGet, url, nil)
 	if err != nil {