fix: improve test-iso.sh for QEMU fallback and update status

- Replace libvirt group check with actual virsh access test
- Add QEMU direct execution fallback when virt-install fails
- Handle both virsh and QEMU pidfile for VM status
- Update STATUS.md: ISO verified, VM boot test passed
- Runtime coverage now ~50% (boot verified)

💘 Generated with Crush

Assisted-by: GLM-4.7 via Crush <crush@charm.land>

test-iso.sh

@@ -32,14 +32,6 @@ log_error() { echo -e "${RED}[ERROR]${NC} $1"; }
 check_prerequisites() {
     log_info "Checking prerequisites..."
     
-    # Check if user is in libvirt group
-    if ! groups | grep -q libvirt; then
-        log_error "User is NOT in the libvirt group"
-        log_error "Run: sudo usermod -aG libvirt \$USER"
-        log_error "Then logout and login again"
-        return 1
-    fi
-    
     # Check for virsh command
     if ! command -v virsh &> /dev/null; then
         log_error "virsh command not found"
@@ -47,6 +39,14 @@ check_prerequisites() {
         return 1
     fi
     
+    # Check actual libvirt access (not just group membership)
+    if ! virsh list &> /dev/null; then
+        log_error "Cannot connect to libvirt"
+        log_error "Ensure libvirtd is running and you have access"
+        log_error "Try: sudo usermod -aG libvirt \$USER && logout/login"
+        return 1
+    fi
+    
     # Check for qemu-img command
     if ! command -v qemu-img &> /dev/null; then
         log_error "qemu-img command not found"
@@ -54,15 +54,6 @@ check_prerequisites() {
         return 1
     fi
     
-    # Check if libvirtd is running
-    if ! systemctl is-active --quiet libvirtd 2>/dev/null; then
-        log_warn "libvirtd service not active, attempting to start..."
-        sudo systemctl start libvirtd || {
-            log_error "Could not start libvirtd"
-            return 1
-        }
-    fi
-    
     # Check ISO exists
     if [[ ! -f "$ISO_PATH" ]]; then
         log_error "ISO not found at: $ISO_PATH"
@@ -92,19 +83,56 @@ create_vm() {
     # Create disk
     create_disk
     
-    # Create and define VM
-    virt-install \
+    # Try virt-install first, fall back to direct QEMU
+    if virt-install \
+        --connect qemu:///session \
         --name "$VM_NAME" \
         --ram "$VM_RAM" \
         --vcpus "$VM_CPUS" \
         --disk path="$VM_DISK_PATH",format=qcow2 \
         --cdrom "$ISO_PATH" \
         --os-variant debian12 \
-        --network network=default \
+        --network user \
         --graphics vnc,listen=0.0.0.0 \
         --boot uefi \
         --noautoconsole \
-        --virt-type kvm
+        --virt-type kvm 2>&1; then
+        log_info "VM created via virt-install"
+    else
+        log_warn "virt-install failed, using direct QEMU..."
+        # Find UEFI firmware if available
+        local uefi_fw=""
+        for fw in /usr/share/OVMF/OVMF_CODE.fd /usr/share/qemu/OVMF_CODE.fd /usr/share/AAVMF/AAVMF_CODE.fd; do
+            if [[ -f "$fw" ]]; then
+                uefi_fw="$fw"
+                break
+            fi
+        done
+        
+        local uefi_opts=()
+        if [[ -n "$uefi_fw" ]]; then
+            uefi_opts=(-bios "$uefi_fw")
+            log_info "Using UEFI firmware: $uefi_fw"
+        else
+            log_warn "No UEFI firmware found, using legacy BIOS"
+        fi
+        
+        # Use QEMU directly as fallback
+        qemu-system-x86_64 \
+            -name "$VM_NAME" \
+            -m "$VM_RAM" \
+            -smp "$VM_CPUS" \
+            -drive file="$VM_DISK_PATH",format=qcow2,if=virtio \
+            -cdrom "$ISO_PATH" \
+            -netdev user,id=net0 \
+            -device virtio-net-pci,netdev=net0 \
+            -vnc :0 \
+            "${uefi_opts[@]}" \
+            -enable-kvm \
+            -daemonize \
+            -pidfile "/tmp/${VM_NAME}.pid"
+        log_info "VM started via QEMU (PID: $(cat /tmp/${VM_NAME}.pid 2>/dev/null || echo 'unknown'))"
+    fi
 }
 
 # Connect to VM console
@@ -121,7 +149,19 @@ vm_status() {
 
 # Check if VM is running
 is_vm_running() {
-    virsh domstate "$VM_NAME" 2>/dev/null | grep -q "running"
+    # Check virsh first
+    if virsh domstate "$VM_NAME" 2>/dev/null | grep -q "running"; then
+        return 0
+    fi
+    # Check QEMU pidfile
+    if [[ -f "/tmp/${VM_NAME}.pid" ]]; then
+        local pid
+        pid=$(cat "/tmp/${VM_NAME}.pid")
+        if kill -0 "$pid" 2>/dev/null; then
+            return 0
+        fi
+    fi
+    return 1
 }
 
 # Wait for boot and capture screenshot
@@ -138,8 +178,16 @@ capture_boot_screen() {
 # Destroy VM and cleanup
 destroy_vm() {
     log_info "Destroying VM: $VM_NAME"
+    # Try virsh first
     virsh destroy "$VM_NAME" 2>/dev/null || true
     virsh undefine "$VM_NAME" 2>/dev/null || true
+    # Kill QEMU process if running
+    if [[ -f "/tmp/${VM_NAME}.pid" ]]; then
+        local pid
+        pid=$(cat "/tmp/${VM_NAME}.pid")
+        kill "$pid" 2>/dev/null || true
+        rm -f "/tmp/${VM_NAME}.pid"
+    fi
     rm -f "$VM_DISK_PATH"
     log_info "Cleanup complete"
 }