From 5db4afb76660901a48f885325091f1830898852e Mon Sep 17 00:00:00 2001
From: Robert Jacobson <Robert.C.Jacobson@nasa.gov>
Date: Thu, 4 Sep 2025 16:54:04 -0400
Subject: [PATCH] verify pod is in Running state before attempting port-forward

---
 containers/kubernetes/kubernetes.sh | 51 +++++++++++++++++++++++++++--
 1 file changed, 49 insertions(+), 2 deletions(-)

diff --git a/containers/kubernetes/kubernetes.sh b/containers/kubernetes/kubernetes.sh
index 781ac48a..ae43f0e0 100755
--- a/containers/kubernetes/kubernetes.sh
+++ b/containers/kubernetes/kubernetes.sh
@@ -5,6 +5,9 @@ REALM=${OU}-${ENVIRO}-${CONTEXT}
 
 MISSION=m01
 SPACECRAFT=sc01
+# Timeouts for starting the cluster
+TIMEOUT_SECONDS=300 # 5 minutes
+SLEEP_INTERVAL=15    # Wait 15 seconds between checks
 
 # resource names, to be used with context, cluster, namespace, pods, and services
 RESOURCE_NAME=${REALM}-${MISSION}
@@ -47,8 +50,50 @@ kubectl create deployment ${K8S_DEPLOYMENT} --context ${K8S_CONTEXT} --namespace
   --image=${IMAGE_URI} \
   --replicas=1
 
-echo "Sleeping for 120 seconds to ensure that pod(s) are up and running"
-sleep 120
+echo "Waiting for pod in namespace ${K8S_NAMESPACE} to be in a 'Running' state..."
+
+start_time=$(date +%s)
+while true; do
+    # Use kubectl with jq to find the pod and get its status and name
+    # We select the pod whose name starts with our prefix
+    pod_info=$(kubectl get pods -n "$K8S_NAMESPACE" -o json 2>/dev/null | \
+               jq -r '.items[] | select(.metadata.name | startswith("'$K8S_NAMESPACE'")) | {name: .metadata.name, status: .status.phase}')
+
+    # Check if a pod was found and get its name and status
+    if [ -n "$pod_info" ]; then
+        POD_NAME=$(echo "$pod_info" | jq -r '.name')
+        pod_status=$(echo "$pod_info" | jq -r '.status')
+    else
+        POD_NAME=""
+        pod_status=""
+    fi
+
+    # Check if the pod is in the "Running" status
+    if [ "$pod_status" == "Running" ]; then
+        echo "Success: Pod '$POD_NAME' is now running."
+        break # Exit the loop
+    fi
+
+    # Check for timeout
+    current_time=$(date +%s)
+    elapsed_time=$((current_time - start_time))
+    if [ "$elapsed_time" -ge "$TIMEOUT_SECONDS" ]; then
+        if [ -z "$POD_NAME" ]; then
+            echo "Error: Timeout reached ($TIMEOUT_SECONDS seconds). No pod with prefix '$POD_NAME_PREFIX' was found."
+        else
+            echo "Error: Timeout reached ($TIMEOUT_SECONDS seconds). Pod '$POD_NAME' status is '$pod_status'."
+        fi
+        exit 1
+    fi
+
+    # Report current status and wait before the next check
+    if [ -n "$POD_NAME" ]; then
+        echo "Current status for '$POD_NAME': '$pod_status'. Waiting..."
+    else
+        echo "No pod with prefix '$POD_NAME_PREFIX' found yet. Waiting..."
+    fi
+    sleep "$SLEEP_INTERVAL"
+done
 
 kubectl scale deployment ${K8S_DEPLOYMENT} --context ${K8S_CONTEXT} --namespace ${K8S_NAMESPACE} --replicas=1
 
@@ -65,3 +110,5 @@ kubectl expose deployment ${K8S_DEPLOYMENT} --context ${K8S_CONTEXT} --namespace
 # port-forwarding
 kubectl port-forward service/${K8S_SERVICE} --context ${K8S_CONTEXT} --namespace ${K8S_NAMESPACE} \
   ${APP_EXTERNAL_PORT}:${APP_INTERNAL_PORT} &
+
+echo "open browser to http://localhost:${APP_EXTERNAL_PORT}/vnc.html and click on 'Connect'"