From 6cc76f6bbef0778d31841af4d63ead60210fc49d Mon Sep 17 00:00:00 2001
From: ghostloda <78798447@qq.con>
Date: Thu, 15 May 2025 10:34:43 +0800
Subject: [PATCH] fix: Update Prometheus query for Ascend NPU device health
 status

Signed-off-by: ghostloda <78798447@qq.com>
---
 server/internal/provider/ascend/provider.go | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/server/internal/provider/ascend/provider.go b/server/internal/provider/ascend/provider.go
index 6c84d83a..efe8fbed 100644
--- a/server/internal/provider/ascend/provider.go
+++ b/server/internal/provider/ascend/provider.go
@@ -43,7 +43,7 @@ type DeviceMeta struct {
 
 func (a *Ascend) GetDevicesFromPrometheus(node *corev1.Node) map[string]*util.DeviceInfo {
 	device := make(map[string]*util.DeviceInfo)
-	queryString := fmt.Sprintf("npu_chip_info_health_status{node=\"%s\"}", node.Name)
+	queryString := fmt.Sprintf("npu_chip_info_health_status{} * on(namespace, pod) group_left(node) kube_pod_info{node=\"%s\"}", node.Name)
 	vs, err := a.prom.Query(context.Background(), queryString)
 	if err != nil {
 		a.log.Warnf("query %s failed", queryString)