feat: add scripts for testing and automating Rancher upgrades and chart checks

scripts/upgrade_rancher_step.py

@@ -0,0 +1,67 @@
+import paramiko
+import sys
+import time
+
+sys.stdout.reconfigure(encoding='utf-8')
+
+target_version = sys.argv[1] if len(sys.argv) > 1 else "2.7.10"
+key_path = r"C:\Users\Snarf\.ssh\id_ed25519_snarfnet"
+master_ip = "140.44.4.11"
+
+print(f"==================================================")
+print(f"Starting Rancher upgrade to version {target_version}")
+print(f"==================================================")
+
+client = paramiko.SSHClient()
+client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+
+def run_cmd(cmd, check=True):
+    print(f"\n>>> Running: {cmd}")
+    stdin, stdout, stderr = client.exec_command(f"KUBECONFIG=/etc/rancher/k3s/k3s.yaml {cmd}")
+    out = stdout.read().decode('utf-8')
+    err = stderr.read().decode('utf-8')
+    exit_status = stdout.channel.recv_exit_status()
+    if out:
+        print(out.strip())
+    if err:
+        print(f"STDERR: {err.strip()}")
+    if check and exit_status != 0:
+        raise RuntimeError(f"Command failed with exit code {exit_status}")
+    return out, err, exit_status
+
+try:
+    client.connect(master_ip, username="root", key_filename=key_path, timeout=15)
+    
+    # 1. Take snapshot
+    snap_name = f"pre-rancher-{target_version}-{int(time.time())}"
+    print(f"\n--- Taking etcd snapshot: {snap_name} ---")
+    run_cmd(f"k3s etcd-snapshot save --name {snap_name}")
+    
+    # 2. Perform Helm upgrade
+    print(f"\n--- Upgrading Rancher chart to {target_version} ---")
+    upgrade_cmd = (
+        f"helm upgrade rancher rancher-stable/rancher "
+        f"--namespace cattle-system "
+        f"--version {target_version} "
+        f"--reuse-values"
+    )
+    run_cmd(upgrade_cmd)
+    
+    # 3. Wait for deployment rollout
+    print("\n--- Waiting for deployment/rancher rollout status ---")
+    run_cmd("kubectl -n cattle-system rollout status deployment/rancher --timeout=360s")
+    
+    # 4. Check cattle-system pods
+    print("\n--- Checking cattle-system pods ---")
+    run_cmd("kubectl -n cattle-system get pods -o wide")
+    
+    # 5. Check ingress / HTTP response
+    print("\n--- Checking Rancher endpoint ---")
+    run_cmd("curl -k -I -s https://aerodrome.snarfnet.net | head -n 10", check=False)
+    
+    client.close()
+    print(f"\nSUCCESS: Rancher upgraded to {target_version} successfully!")
+
+except Exception as ex:
+    print(f"\nFAILED: {ex}")
+    sys.exit(1)