Add stop-lab script

README.md

@@ -43,6 +43,11 @@ After a host reboot (VMs already exist, just need to come back up):
 ./start-lab.sh   # re-arms libvirt/network/NAT state and starts the 3 VMs
 ```
 
+To pause the lab without destroying it (e.g. to free RAM/CPU):
+```bash
+./stop-lab.sh    # gracefully shuts down all 3 VMs; ./start-lab.sh brings them back
+```
+
 ## Access
 - Web UI: <https://10.10.10.11:8006> and <https://10.10.10.12:8006>
 - Login: `root` / value of `ROOT_PASSWORD` in `lab.env`

stop-lab.sh

@@ -0,0 +1,59 @@
+#!/usr/bin/env bash
+# Gracefully shut down the lab's VMs without destroying anything — libvirt
+# domains, the network, and storage pools are left defined so ./start-lab.sh
+# can bring the lab back up later. Use ./teardown.sh instead to remove it all.
+#
+# Shuts down the PVE nodes first, then the storage VM last, so nothing is
+# still relying on its NFS/iSCSI when it disappears (mirrors the storage-first
+# order start-lab.sh uses on the way up).
+set -uo pipefail
+source "$(dirname "$0")/lab.env"
+
+sudo -v   # prompt for the sudo password once, up front
+
+request_shutdown() {
+  local name="$1" state
+  state=$(sudo virsh domstate "$name" 2>/dev/null || echo "missing")
+  case "$state" in
+    "shut off") echo "    $name already shut off" ;;
+    missing)    echo "    $name is not defined in libvirt; skipping" ;;
+    *)          echo "    sending ACPI shutdown to $name (state: $state)"; sudo virsh shutdown "$name" >/dev/null 2>&1 ;;
+  esac
+}
+
+wait_shutoff() {
+  local name="$1" tries=0
+  while true; do
+    local state
+    state=$(sudo virsh domstate "$name" 2>/dev/null || echo "missing")
+    [[ "$state" == "shut off" || "$state" == "missing" ]] && { echo "    $name is down"; return 0; }
+    sleep 5
+    tries=$((tries+1))
+    if (( tries % 6 == 0 )); then echo "    still waiting for $name to power off... (${tries}0s)"; fi
+    if (( tries > 36 )); then return 1; fi   # ~3 minutes
+  done
+}
+
+force_off_if_needed() {
+  local name="$1"
+  if ! wait_shutoff "$name"; then
+    echo "    WARN: $name didn't shut down cleanly after 3m; forcing off"
+    sudo virsh destroy "$name" >/dev/null 2>&1 || true
+  fi
+}
+
+echo "==> Shutting down PVE nodes"
+request_shutdown "$PVE1_NAME"
+request_shutdown "$PVE2_NAME"
+echo "==> Waiting for PVE nodes to power off"
+force_off_if_needed "$PVE1_NAME"
+force_off_if_needed "$PVE2_NAME"
+
+echo "==> Shutting down storage VM (after the PVE nodes, so nothing still depends on its NFS/iSCSI)"
+request_shutdown "$STORE_NAME"
+echo "==> Waiting for $STORE_NAME to power off"
+force_off_if_needed "$STORE_NAME"
+
+echo
+echo "Lab stopped. VMs, network, and storage pools are still defined (nothing was destroyed)."
+echo "Run ./start-lab.sh to bring it back up."