fix(vps): report stable creation progress

server_provisioning.py

@@ -786,13 +786,16 @@ def create_vps_server(config: Dict, user_id: str, server_id: str, namespace: str
     memory_request = f"{ram_gb * 0.5}Gi"
 
     if config['image'] == 'windows':
-        return create_windows_vps(config, user_id, server_id, namespace, image_config, ports, storage_size, cpu_cores, memory_limit, memory_request)
+        return create_windows_vps(
+            config, user_id, server_id, namespace, image_config, ports,
+            storage_size, cpu_cores, memory_limit, memory_request, progress_callback
+        )
 
     # Create DataVolume
     data_volume_name = f"{server_id}-rootdisk"
 
     if progress_callback:
-        progress_callback('Initializing', 0, 'Preparing VPS creation...')
+        progress_callback('Preparing storage', 3, 'Preparing the VPS system disk.')
 
     data_volume_body = {
         'apiVersion': 'cdi.kubevirt.io/v1beta1',
@@ -978,11 +981,13 @@ final_message: "AnchorHost VPS ready! User: {image_config['user']}, Password: {i
     if dv_created and k8s_manager.connected:
         logger.info("⏳ Waiting for DataVolume to be ready...")
         if progress_callback:
-            progress_callback('Pending', 0, 'Creating DataVolume...')
+            progress_callback('Allocating storage', 5, 'Allocating storage for the system image.')
 
         max_wait = 600
-        wait_interval = 5  # Check more frequently for better progress updates
+        wait_interval = 5
         elapsed = 0
+        latest_progress = 12
+        image_ready = False
 
         while elapsed < max_wait:
             try:
@@ -998,67 +1003,75 @@ final_message: "AnchorHost VPS ready! User: {image_config['user']}, Password: {i
                 phase = status.get('phase', '')
                 progress = status.get('progress', '')
 
-                # Extract progress percentage if available
+                # CDI reports the actual image download percentage.
-                progress_percent = 0
+                progress_percent = None
                 if progress:
-                    # Progress format is usually "XX.X%" or "XX.XX%"
                     try:
                         progress_percent = float(progress.replace('%', ''))
                     except (ValueError, AttributeError):
                         pass
 
-                # Map phase to progress percentage if not available
+                phase_progress_map = {
-                if progress_percent == 0:
+                    'Pending': 5,
-                    phase_progress_map = {
+                    'PVCBound': 9,
-                        'Pending': 5,
+                    'ImportScheduled': 12,
-                        'PVCBound': 10,
+                    'Succeeded': 84
-                        'ImportScheduled': 15,
-                        'ImportInProgress': 50,
-                        'Succeeded': 100
-                    }
-                    progress_percent = phase_progress_map.get(phase, 0)
-
-                # Create progress message
-                phase_messages = {
-                    'Pending': 'Waiting for storage to be allocated...',
-                    'PVCBound': 'Storage allocated, preparing import...',
-                    'ImportScheduled': 'Import scheduled, starting download...',
-                    'ImportInProgress': f'Downloading image... ({progress if progress else "in progress"})',
-                    'Succeeded': 'Image download completed!',
-                    'Failed': 'Image download failed'
                 }
-                message = phase_messages.get(phase, f'Processing... ({phase})')
+                if phase == 'ImportInProgress' and progress_percent is not None:
+                    progress_percent = 12 + max(0, min(100, progress_percent)) * 0.72
+                else:
+                    progress_percent = phase_progress_map.get(phase, latest_progress)
+                latest_progress = max(latest_progress, int(progress_percent))
+
+                phase_messages = {
+                    'Pending': ('Allocating storage', 'Waiting for the storage volume.'),
+                    'PVCBound': ('Preparing image download', 'Storage is ready. Preparing the system image.'),
+                    'ImportScheduled': ('Starting image download', 'The system image download has been scheduled.'),
+                    'ImportInProgress': ('Downloading system image', f'Downloading the system image ({progress or "in progress"}).'),
+                    'Succeeded': ('System image ready', 'The system image is ready.')
+                }
+                progress_phase, message = phase_messages.get(
+                    phase, ('Preparing system image', 'Waiting for the system image importer.')
+                )
 
-                # Send progress update
                 if progress_callback:
-                    progress_callback(phase, progress_percent, message)
+                    progress_callback(progress_phase, latest_progress, message)
 
                 if phase == 'Succeeded':
-                    logger.info("✅ DataVolume ready!")
+                    logger.info("VPS system image is ready")
+                    image_ready = True
                     break
                 elif phase == 'Failed':
                     error_message = status.get('conditions', [{}])[0].get('message', 'Unknown error') if status.get('conditions') else 'Unknown error'
-                    logger.error(f"❌ DataVolume failed: {error_message}")
+                    logger.error(f"VPS system image import failed: {error_message}")
                     if progress_callback:
-                        progress_callback('Failed', 0, f'Image download failed: {error_message}')
+                        progress_callback('Failed', latest_progress, f'System image download failed: {error_message}')
-                    break
+                    raise RuntimeError(f'System image download failed: {error_message}')
 
             except Exception as e:
-                logger.error(f"⚠ Error checking DataVolume: {e}")
+                if isinstance(e, RuntimeError):
+                    raise
+                logger.warning(f"Could not read VPS image import status: {e}")
                 if progress_callback:
-                    progress_callback('Error', 0, f'Error checking progress: {str(e)}')
+                    progress_callback(
+                        'Downloading system image', latest_progress,
+                        'Waiting for the image download status to update.'
+                    )
 
+            if image_ready:
+                break
             time.sleep(wait_interval)
             elapsed += wait_interval
 
         if elapsed >= max_wait:
-            logger.warning(f"⚠ DataVolume wait timeout after {max_wait}s")
+            logger.warning(f"VPS image import did not finish within {max_wait}s")
             if progress_callback:
-                progress_callback('Timeout', 0, 'Image download timed out, proceeding anyway...')
+                progress_callback('Failed', latest_progress, 'The system image download timed out.')
+            raise TimeoutError('System image download timed out; the virtual machine was not started.')
 
     # Create VM
     if progress_callback:
-        progress_callback('CreatingVM', 90, 'Creating virtual machine...')
+        progress_callback('Starting virtual machine', 88, 'Creating the virtual machine resource.')
 
     try:
         custom_objects_api.create_namespaced_custom_object(
@@ -1068,15 +1081,16 @@ final_message: "AnchorHost VPS ready! User: {image_config['user']}, Password: {i
             plural="virtualmachines",
             body=vm_body
         )
-        logger.info(f"✅ Created VirtualMachine {server_id}")
+        logger.info(f"Created VirtualMachine {server_id}")
-        if progress_callback:
-            progress_callback('Completed', 100, 'VPS created successfully!')
     except Exception as e:
         logger.error(f"❌ Failed to create VM: {e}")
         if progress_callback:
             progress_callback('Error', 0, f'Failed to create VM: {str(e)}')
         raise Exception(f"Failed to create VM: {str(e)}")
 
+    if progress_callback:
+        progress_callback('Configuring network', 92, 'Setting up the VPS network and ports.')
+
     # Create Service with FIXED port handling
     if ports:
         service_ports = []
@@ -1123,17 +1137,14 @@ final_message: "AnchorHost VPS ready! User: {image_config['user']}, Password: {i
             ports.extend(actual_ports)
 
         except Exception as e:
-            logger.error(f"⚠ Failed to create service: {e}")
+            logger.error(f"Failed to create VPS network service: {e}")
-
+            if progress_callback:
-    # Background status sync
+                progress_callback('Failed', 92, 'Could not create the VPS network service.')
-    if k8s_manager.connected:
+            raise RuntimeError(f'Failed to create VPS network service: {e}') from e
-        threading.Thread(
-            target=wait_and_sync_vm_status,
-            args=(server_id, user_id, namespace),
-            daemon=True
-        ).start()
 
     # Create server record
+    if progress_callback:
+        progress_callback('Registering VPS', 97, 'Saving the VPS to your server list.')
     server_data = {
         'id': server_id,
         'name': config['name'],
@@ -1151,7 +1162,18 @@ final_message: "AnchorHost VPS ready! User: {image_config['user']}, Password: {i
         'k8s_name': server_id
     }
 
-    return create_server_record(server_data)
+    server = create_server_record(server_data)
+    if not server:
+        raise RuntimeError('VPS resources were created, but the server record could not be saved.')
+
+    if k8s_manager.connected:
+        threading.Thread(
+            target=wait_and_sync_vm_status,
+            args=(server_id, user_id, namespace),
+            daemon=True
+        ).start()
+
+    return server
 
 def wait_and_sync_vm_status(server_id: str, user_id: str, namespace: str):
     """Background VM status sync"""
@@ -1240,9 +1262,12 @@ def wait_and_sync_pod_status(server_id: str, user_id: str, namespace: str):
         sync_server_status_from_k8s(server_id, user_id)
 
 def create_windows_vps(config: Dict, user_id: str, server_id: str, namespace: str, image_config: Dict,
-                      ports: List, storage_size: str, cpu_cores: int, memory_limit: str, memory_request: str) -> Dict:
+                      ports: List, storage_size: str, cpu_cores: int, memory_limit: str, memory_request: str,
+                      progress_callback=None) -> Dict:
     """Create Windows VPS"""
     pvc_name = f"{server_id}-rootdisk"
+    if progress_callback:
+        progress_callback('Preparing storage', 5, 'Preparing storage for the Windows VPS.')
     if not k8s_manager.create_pvc(namespace, pvc_name, storage_size):
         raise Exception("Failed to create PVC")
 
@@ -1281,6 +1306,8 @@ def create_windows_vps(config: Dict, user_id: str, server_id: str, namespace: st
         }
     }
 
+    if progress_callback:
+        progress_callback('Starting virtual machine', 88, 'Creating the Windows virtual machine resource.')
     custom_objects_api.create_namespaced_custom_object(
         group="kubevirt.io",
         version="v1",
@@ -1289,6 +1316,9 @@ def create_windows_vps(config: Dict, user_id: str, server_id: str, namespace: st
         body=vm_body
     )
 
+    if progress_callback:
+        progress_callback('Configuring network', 92, 'Setting up the Windows VPS network and ports.')
+
     # Create Service with FIXED port handling
     if ports:
         service_ports = []
@@ -1334,7 +1364,10 @@ def create_windows_vps(config: Dict, user_id: str, server_id: str, namespace: st
             ports.extend(actual_ports)
 
         except Exception as e:
-            logger.error(f"⚠ Failed to create service: {e}")
+            logger.error(f"Failed to create Windows VPS network service: {e}")
+            if progress_callback:
+                progress_callback('Failed', 92, 'Could not create the VPS network service.')
+            raise RuntimeError(f'Failed to create Windows VPS network service: {e}') from e
 
     server_data = {
         'id': server_id,
@@ -1353,7 +1386,20 @@ def create_windows_vps(config: Dict, user_id: str, server_id: str, namespace: st
         'k8s_name': server_id
     }
 
-    return create_server_record(server_data)
+    if progress_callback:
+        progress_callback('Registering VPS', 97, 'Saving the VPS to your server list.')
+    server = create_server_record(server_data)
+    if not server:
+        raise RuntimeError('VPS resources were created, but the server record could not be saved.')
+
+    if k8s_manager.connected:
+        threading.Thread(
+            target=wait_and_sync_vm_status,
+            args=(server_id, user_id, namespace),
+            daemon=True
+        ).start()
+
+    return server
 
 def create_website_server(config: Dict, user_id: str, server_id: str, namespace: str) -> Dict:
     """Create Website Server with FIXED port handling"""

websocket_service.py

@@ -48,6 +48,17 @@ import websockets
 
 __all__ = ['DateTimeEncoder', 'attach_to_game_console', 'ensure_node_shell_pod', 'execute_minecraft_save_all', 'handle_2fa_disable', 'handle_2fa_enable', 'handle_2fa_login', 'handle_2fa_setup', 'handle_add_credits', 'handle_authenticate', 'handle_change_server_version', 'handle_check_demo_eligibility', 'handle_convert_demo_to_paid', 'handle_copy_path', 'handle_create_backup', 'handle_create_custom_domain', 'handle_create_directory', 'handle_create_domain', 'handle_create_server', 'handle_create_ticket', 'handle_delete_backup', 'handle_delete_custom_domain', 'handle_delete_domain', 'handle_delete_path', 'handle_delete_server', 'handle_download_file', 'handle_exec_command', 'handle_expose_port', 'handle_game_console_command', 'handle_game_console_stream', 'handle_generate_sftp', 'handle_get_backups', 'handle_get_credits', 'handle_get_minecraft_settings', 'handle_get_server_creation_progress', 'handle_get_server_logs', 'handle_get_server_metrics', 'handle_get_servers', 'handle_get_ticket', 'handle_get_ticket_stats', 'handle_get_tickets', 'handle_link_domain', 'handle_list_custom_domains', 'handle_list_domains', 'handle_list_files', 'handle_login', 'handle_move_path', 'handle_node_terminal_connect', 'handle_node_terminal_disconnect', 'handle_node_terminal_input', 'handle_pack_zip', 'handle_read_file', 'handle_register', 'handle_remove_port', 'handle_restart_server', 'handle_restore_backup', 'handle_send_ticket_message', 'handle_start_chunked_upload', 'handle_start_server', 'handle_stop_server', 'handle_sync_server_status', 'handle_unlink_domain', 'handle_unpack_zip', 'handle_update_minecraft_settings', 'handle_update_ticket_status', 'handle_upload_file', 'handle_upload_file_chunk', 'handle_verify_custom_domain', 'handle_vps_terminal_connect', 'handle_vps_terminal_disconnect', 'handle_vps_terminal_input', 'handle_websocket', 'handle_write_file', 'notify_admins', 'notify_ticket_participants', 'parse_ls_output', 'send_message', 'send_verification_email', 'start_websocket_server', 'stream_game_console_output', 'stream_metrics', 'sync_user_domains_with_dns']
 
+_server_creation_progress_by_user: Dict[str, Dict] = {}
+_server_creation_progress_lock = threading.Lock()
+
+
+def _set_server_creation_progress(user_id: str, progress: Dict) -> Dict:
+    """Keep the latest creation status available after a WebSocket reconnect."""
+    snapshot = {**progress, 'timestamp': time.time()}
+    with _server_creation_progress_lock:
+        _server_creation_progress_by_user[str(user_id)] = snapshot
+    return snapshot
+
 
 
 async def handle_create_server(client_id, data):
@@ -58,6 +69,14 @@ async def handle_create_server(client_id, data):
         return
 
     user_id = client_info['user_id']
+    server_type = data.get('serverType', '')
+    _set_server_creation_progress(user_id, {
+        'phase': 'Preparing',
+        'progress': 1,
+        'message': 'Checking capacity and preparing your server.',
+        'serverType': server_type,
+        'state': 'active'
+    })
 
     try:
         # Check if this is a demo server request
@@ -67,6 +86,18 @@ async def handle_create_server(client_id, data):
         server_name = data.get('name', '').strip()
         is_valid, error_msg = validate_server_name(server_name)
         if not is_valid:
+            failed = _set_server_creation_progress(user_id, {
+                'phase': 'Failed',
+                'progress': 1,
+                'message': error_msg,
+                'serverType': server_type,
+                'state': 'failed'
+            })
+            await send_message(client_id, {'type': 'server_creation_progress', **failed})
+            await send_message(client_id, {
+                'type': 'server_creation_failed',
+                'error': error_msg
+            })
             await send_message(client_id, {
                 'type': 'error',
                 'error': error_msg
@@ -80,7 +111,15 @@ async def handle_create_server(client_id, data):
         capacity_check = validate_server_capacity(data)
 
         if not capacity_check['success']:
+            failed = _set_server_creation_progress(user_id, {
+                'phase': 'Capacity check',
+                'progress': 1,
+                'message': capacity_check['message'],
+                'serverType': server_type,
+                'state': 'failed'
+            })
             # Send detailed capacity error
+            await send_message(client_id, {'type': 'server_creation_progress', **failed})
             await send_message(client_id, {
                 'type': 'capacity_error',
                 'error': capacity_check['message'],
@@ -90,43 +129,48 @@ async def handle_create_server(client_id, data):
             })
             return
 
-        # Capacity check passed, proceed with server creation
+        # Capacity check passed, proceed with server creation.
-        # Create progress callback that sends websocket messages
         loop = asyncio.get_event_loop()
 
         def progress_callback(phase, progress, message):
-            """Thread-safe progress callback that sends websocket messages"""
+            """Store and forward progress updates from the provisioning thread."""
             try:
-                # Store progress in client info for reconnection recovery
+                with _server_creation_progress_lock:
-                if client_id in connected_clients:
+                    previous = dict(_server_creation_progress_by_user.get(str(user_id), {}))
-                    connected_clients[client_id]['server_creation_progress'] = {
+                state = 'failed' if phase in ('Failed', 'Error', 'Timeout') else 'active'
-                        'phase': phase,
+                snapshot = _set_server_creation_progress(user_id, {
-                        'progress': progress,
+                    'phase': phase,
-                        'message': message,
+                    'progress': max(int(previous.get('progress', 0)), max(0, min(99, int(progress)))),
-                        'timestamp': time.time()
+                    'message': message,
-                    }
+                    'serverType': server_type,
+                    'state': state
+                })
 
-                # Use run_coroutine_threadsafe to send async message from executor thread
                 future = asyncio.run_coroutine_threadsafe(
                     send_message(client_id, {
                         'type': 'server_creation_progress',
-                        'phase': phase,
+                        **snapshot
-                        'progress': progress,
-                        'message': message
                     }),
                     loop
                 )
-                # Don't wait for completion to avoid blocking
+                future.result(timeout=2)
-                # The future will complete asynchronously
             except Exception as e:
                 logger.error(f"Error sending progress update: {e}")
 
-        # Run blocking create_server in executor with progress callback
+        # Run blocking provisioning outside the event loop.
         server = await loop.run_in_executor(None, create_server, data, user_id, progress_callback)
 
-        # Clear server creation progress on completion
+        completed = _set_server_creation_progress(user_id, {
-        if client_id in connected_clients:
+            'phase': 'Provisioned',
-            connected_clients[client_id].pop('server_creation_progress', None)
+            'progress': 100,
+            'message': 'Server resources are ready. The VPS may take a little longer to boot.',
+            'serverType': server_type,
+            'state': 'complete'
+        })
+        await send_message(client_id, {
+            'type': 'server_creation_progress',
+            **completed
+        })
 
         # Get user credits
         user = get_user_by_id(user_id)
@@ -145,10 +189,22 @@ async def handle_create_server(client_id, data):
         })
     except Exception as e:
         logger.error(f"Error creating server: {e}", exc_info=True)
-
+        previous = _server_creation_progress_by_user.get(str(user_id), {})
-        # Clear server creation progress on error
+        failed = _set_server_creation_progress(user_id, {
-        if client_id in connected_clients:
+            'phase': 'Failed',
-            connected_clients[client_id].pop('server_creation_progress', None)
+            'progress': previous.get('progress', 0),
+            'message': str(e),
+            'serverType': data.get('serverType', ''),
+            'state': 'failed'
+        })
+        await send_message(client_id, {
+            'type': 'server_creation_progress',
+            **failed
+        })
+        await send_message(client_id, {
+            'type': 'server_creation_failed',
+            'error': str(e)
+        })
 
         # Check if it's a capacity-related error
         error_msg = str(e)
@@ -1269,25 +1325,25 @@ async def handle_get_server_creation_progress(client_id, data):
         await send_message(client_id, {'type': 'error', 'error': 'Not authenticated'})
         return
 
-    # Get stored progress if available
+    user_id = str(client_info['user_id'])
-    progress = client_info.get('server_creation_progress')
+    with _server_creation_progress_lock:
+        progress = _server_creation_progress_by_user.get(user_id)
     if progress:
-        # Check if progress is recent (within last 5 minutes)
+        expiry = 1800 if progress.get('state') == 'active' else 600
-        if time.time() - progress.get('timestamp', 0) < 300:
+        if time.time() - progress.get('timestamp', 0) < expiry:
             await send_message(client_id, {
                 'type': 'server_creation_progress',
-                'phase': progress.get('phase', ''),
+                **progress
-                'progress': progress.get('progress', 0),
-                'message': progress.get('message', 'Creating server...')
             })
         else:
-            # Progress is stale, clear it
+            with _server_creation_progress_lock:
-            client_info.pop('server_creation_progress', None)
+                _server_creation_progress_by_user.pop(user_id, None)
             await send_message(client_id, {
                 'type': 'server_creation_progress',
                 'phase': 'Unknown',
                 'progress': 0,
-                'message': 'No active server creation found'
+                'message': 'No recent server creation status is available.',
+                'state': 'unknown'
             })
     else:
         # No progress stored
@@ -1295,7 +1351,8 @@ async def handle_get_server_creation_progress(client_id, data):
             'type': 'server_creation_progress',
             'phase': 'Unknown',
             'progress': 0,
-            'message': 'No active server creation found'
+            'message': 'No recent server creation status is available.',
+            'state': 'unknown'
         })
 
 async def handle_create_domain(client_id, data):