logger adjust

daemons/execute_monitor_container.go

@@ -178,37 +178,43 @@ func (cm *ContainerMonitor) watchJob(clientset *kubernetes.Clientset, execID str
 	}
 
 	if podName == "" {
-		l.Error().Msg("No pod found for job after 60s")
+		cm.failExec(execID, l, "No pod found for job after 60s")
 		return
 	}
 
 	l.Info().Str("pod", podName).Msg("Pod found for job")
 
 	// Wait for the pod to be Running or terminal (up to 120s)
+	podReady := false
 	for i := 0; i < 120; i++ {
 		pod, err := clientset.CoreV1().Pods(ns).Get(context.Background(), podName, metav1.GetOptions{})
 		if err != nil {
-			l.Error().Err(err).Str("pod", podName).Msg("Failed to get pod status")
+			cm.failExec(execID, l, "Failed to get pod status: "+err.Error())
 			return
 		}
 		phase := pod.Status.Phase
 		if phase == corev1.PodRunning || phase == corev1.PodSucceeded || phase == corev1.PodFailed {
 			l.Info().Str("pod", podName).Str("phase", string(phase)).Msg("Pod phase")
+			podReady = true
 			break
 		}
 		time.Sleep(time.Second)
 	}
+	if !podReady {
+		cm.failExec(execID, l, "Pod never reached a running/terminal phase after 120s")
+		return
+	}
 
 	// Stream pod logs
 	req := clientset.CoreV1().Pods(ns).GetLogs(podName, &corev1.PodLogOptions{Follow: true})
 	stream, err := req.Stream(context.Background())
 	if err != nil {
-		l.Error().Err(err).Str("pod", podName).Msg("Failed to stream pod logs")
-	} else {
-		defer stream.Close()
-		l.Info().Str("pod", podName).Msg("Streaming pod logs")
-		logExecution(stream, l)
+		cm.failExec(execID, l, "Failed to stream pod logs: "+err.Error())
+		return
 	}
+	defer stream.Close()
+	l.Info().Str("pod", podName).Msg("Streaming pod logs")
+	logExecution(stream, l)
 
 	// Log final job status
 	job, err := clientset.BatchV1().Jobs(ns).Get(context.Background(), jobName, metav1.GetOptions{})