Release (#1057)

* chore(deps): update dependencies

* chore(firefox): upgrade to FF 118.0.2

* docs(version): add changelog and bump version

* chore(formatting): fix package.json formatting

CHANGELOG.md

@@ -1,5 +1,11 @@
 # Changelog
 
+## v0.25.0 - 2023-10-13
+
+Bump to Firefox 118.0.2
+Introduce StorageWatchdog #1056 (Thanks @gridl0ck for contributing this)
+Upgrade Docker image to Ubuntu 22.04 #1055
+
 ## v0.24.0 - 2023-09-05
 
 Bump to Firefox 117

Extension/package.json

@@ -9,34 +9,34 @@
     "start": "Start is required for the manual_test.py to run"
   },
   "devDependencies": {
-    "@babel/cli": "^7.22.15",
-    "@babel/core": "^7.22.15",
+    "@babel/cli": "^7.23.0",
+    "@babel/core": "^7.23.0",
     "@babel/eslint-parser": "^7.22.15",
-    "@babel/preset-env": "^7.22.15",
+    "@babel/preset-env": "^7.23.0",
     "@types/download": "^8.0.2",
-    "@types/firefox-webext-browser": "^111.0.1",
-    "@typescript-eslint/eslint-plugin": "^6.6.0",
-    "@typescript-eslint/parser": "^6.6.0",
+    "@types/firefox-webext-browser": "^111.0.2",
+    "@typescript-eslint/eslint-plugin": "^6.7.5",
+    "@typescript-eslint/parser": "^6.7.5",
     "ajv": "^8.12.0",
     "body-parser": "^1.20.2",
     "download": "^8.0.0",
-    "eslint": "^8.48.0",
+    "eslint": "^8.51.0",
     "eslint-config-prettier": "^8.0.0",
     "eslint-plugin-html": "^7.1.0",
     "eslint-plugin-import": "^2.28.1",
-    "eslint-plugin-jsdoc": "^46.5.1",
+    "eslint-plugin-jsdoc": "^46.8.2",
     "eslint-plugin-json": "^3.1.0",
-    "eslint-plugin-mozilla": "^3.1.0",
+    "eslint-plugin-mozilla": "^3.2.0",
     "eslint-plugin-no-unsanitized": "^4.0.2",
     "eslint-plugin-prettier": "^5.0.0",
     "eslint-plugin-unicorn": "^48.0.1",
     "express": "^4.18.2",
     "prettier": "^3.0.3",
     "safe-compare": "^1.1.4",
-    "ts-loader": "^9.4.4",
+    "ts-loader": "^9.5.0",
     "typedoc": "^0.25.1",
     "typescript": "^5.2.2",
-    "web-ext": "^7.6.2",
+    "web-ext": "^7.8.0",
     "webpack": "^5.88.2",
     "webpack-cli": "^5.1.4"
   },

Extension/src/loggingdb.ts

@@ -76,7 +76,7 @@ export const open = async function (
     );
     console.log("StorageController started?", rv);
   }
-
+  storageController.send(JSON.stringify(`Browser-${crawlID}`));
   // Listen for incoming urls as visit ids
   listeningSocket = new socket.ListeningSocket(listeningSocketCallback);
   console.log("Starting socket listening for incoming connections.");

VERSION

@@ -1 +1 @@
-0.24.0
+0.25.0

environment.yaml

@@ -9,42 +9,42 @@ dependencies:
 - dill=0.3.7
 - dill=0.3.7
 - easyprocess=1.1
-- gcsfs=2023.9.1
+- gcsfs=2023.9.2
 - geckodriver=0.33.0
-- ipython=8.15.0
+- ipython=8.16.1
 - isort=5.12.0
 - leveldb=1.23
 - multiprocess=0.70.15
-- mypy=1.5.1
-- nodejs=20.7.0
-- pandas=2.1.0
+- mypy=1.6.0
+- nodejs=20.8.0
+- pandas=2.1.1
 - pillow=10.0.1
 - pip=23.2.1
+- plyvel=1.5.0
 - pre-commit=3.4.0
 - psutil=5.9.5
 - pyarrow=13.0.0
 - pytest-asyncio=0.21.1
 - pytest-cov=4.1.0
 - pytest=7.4.2
-- python=3.11.5
-- pyvirtualdisplay=3.0
+- python=3.12.0
+- pyvirtualdisplay=2.2
 - recommonmark=0.7.1
-- redis-py=5.0.0
-- s3fs=2023.9.1
-- selenium=4.12.0
-- sentry-sdk=1.31.0
+- redis-py=5.0.1
+- s3fs=2023.9.2
+- selenium=4.13.0
+- sentry-sdk=1.32.0
 - sphinx-markdown-tables=0.0.17
 - sphinx=7.2.6
 - tabulate=0.9.0
 - tblib=2.0.0
 - wget=1.20.3
 - pip:
-  - dataclasses-json==0.6.0
+  - dataclasses-json==0.6.1
   - domain-utils==0.7.1
-  - jsonschema==4.19.0
-  - plyvel==1.5.0
+  - jsonschema==4.19.1
   - tranco==0.6
-  - types-pyyaml==6.0.12.11
-  - types-redis==4.6.0.6
+  - types-pyyaml==6.0.12.12
+  - types-redis==4.6.0.7
   - types-tabulate==0.9.0.3
 name: openwpm

openwpm/browser_manager.py

@@ -549,15 +549,17 @@ def kill_browser_manager(self):
         if self.browser_manager is not None and self.browser_manager.pid is not None:
             self.logger.debug(
                 "BROWSER %i: Attempting to kill BrowserManager with pid %i. "
-                "Browser PID: %s"
-                % (self.browser_id, self.browser_manager.pid, self.geckodriver_pid)
+                "Browser PID: %s",
+                self.browser_id,
+                self.browser_manager.pid,
+                self.geckodriver_pid,
             )
             try:
                 os.kill(self.browser_manager.pid, signal.SIGKILL)
             except OSError:
                 self.logger.debug(
-                    "BROWSER %i: Browser manager process does "
-                    "not exist" % self.browser_id
+                    "BROWSER %i: Browser manager process does not exist",
+                    self.browser_id,
                 )
                 pass
 
@@ -566,13 +568,14 @@ def kill_browser_manager(self):
                 os.kill(self.display_pid, signal.SIGKILL)
             except OSError:
                 self.logger.debug(
-                    "BROWSER %i: Display process does not exit" % self.browser_id
+                    "BROWSER %i: Display process does not exit", self.browser_id
                 )
                 pass
             except TypeError:
                 self.logger.error(
-                    "BROWSER %i: PID may not be the correct "
-                    "type %s" % (self.browser_id, str(self.display_pid))
+                    "BROWSER %i: PID may not be the correct " "type %s",
+                    self.browser_id,
+                    str(self.display_pid),
                 )
         if self.display_port is not None:  # xvfb display lock
             # lockfile = "/tmp/.X%s-lock" % self.display_port
@@ -584,8 +587,9 @@ def kill_browser_manager(self):
                 os.remove(lockfile)
             except OSError:
                 self.logger.debug(
-                    "BROWSER %i: Screen lockfile (%s) already "
-                    "removed" % (self.browser_id, lockfile)
+                    "BROWSER %i: Screen lockfile (%s) already removed",
+                    self.browser_id,
+                    lockfile,
                 )
                 pass
 

openwpm/deploy_browsers/deploy_firefox.py

@@ -78,7 +78,7 @@ def deploy_firefox(
     display_port = None
     display = None
     if display_mode == "headless":
-        fo.headless = True
+        fo.add_argument("--headless")
         fo.add_argument("--width={}".format(DEFAULT_SCREEN_RES[0]))
         fo.add_argument("--height={}".format(DEFAULT_SCREEN_RES[1]))
     if display_mode == "xvfb":

openwpm/storage/storage_controller.py

@@ -72,11 +72,11 @@ def __init__(
         self._shutdown_flag = False
         self._relaxed = False
         self.logger = logging.getLogger("openwpm")
-        self.store_record_tasks: DefaultDict[VisitId, List[Task[None]]] = defaultdict(
+        self.store_record_tasks: DefaultDict[VisitId, list[Task[None]]] = defaultdict(
             list
         )
         """Contains all store_record tasks for a given visit_id"""
-        self.finalize_tasks: List[Tuple[VisitId, Optional[Task[None]], bool]] = []
+        self.finalize_tasks: list[tuple[VisitId, Optional[Task[None]], bool]] = []
         """Contains all information required for update_completion_queue to work
             Tuple structure is: VisitId, optional completion token, success
         """
@@ -97,18 +97,21 @@ async def _handler(
             self.logger.error(
                 "An exception occurred while processing records", exc_info=e
             )
+        writer.close()
+        await writer.wait_closed()
 
     async def handler(
         self, reader: asyncio.StreamReader, _: asyncio.StreamWriter
     ) -> None:
         """Created for every new connection to the Server"""
-        self.logger.debug("Initializing new handler")
+        client_name = await get_message_from_reader(reader)
+        self.logger.info(f"Initializing new handler for {client_name}")
         while True:
             try:
                 record: Tuple[str, Any] = await get_message_from_reader(reader)
             except IncompleteReadError:
                 self.logger.info(
-                    "Terminating handler, because the underlying socket closed"
+                    f"Terminating handler for {client_name}, because the underlying socket closed"
                 )
                 break
             if len(record) != 2:
@@ -248,6 +251,7 @@ async def update_status_queue(self) -> NoReturn:
             )
 
     async def shutdown(self, completion_queue_task: Task[None]) -> None:
+        self.logger.info("Entering self.shutdown")
         completion_tokens = {}
         visit_ids = list(self.store_record_tasks.keys())
         for visit_id in visit_ids:
@@ -261,6 +265,7 @@ async def shutdown(self, completion_queue_task: Task[None]) -> None:
             self.completion_queue.put((visit_id, False))
 
         await self.structured_storage.shutdown()
+        self.logger.info("structured_storage is shut down")
 
         if self.unstructured_storage is not None:
             await self.unstructured_storage.flush_cache()
@@ -342,13 +347,21 @@ async def _run(self) -> None:
         update_completion_queue = asyncio.create_task(
             self.update_completion_queue(), name="CompletionQueueFeeder"
         )
-        # Blocks until we should shutdown
+        # Blocks until we should shut down
         await self.should_shutdown()
-
+        self.logger.info(f"Closing Server")
         server.close()
+        self.logger.info("Closed Server")
+        self.logger.info("Cancelling status_queue_update")
         status_queue_update.cancel()
+        self.logger.info("Cancelled status_queue_update")
+        self.logger.info("Cancelling timeout_check")
         timeout_check.cancel()
+        self.logger.info("Cancelled timeout_check")
+        self.logger.info("Starting wait_closed")
         await server.wait_closed()
+        self.logger.info("Completed wait_closed")
+
         await self.shutdown(update_completion_queue)
 
     def run(self) -> None:
@@ -359,10 +372,11 @@ def run(self) -> None:
 class DataSocket:
     """Wrapper around ClientSocket to make sending records to the StorageController more convenient"""
 
-    def __init__(self, listener_address: Tuple[str, int]) -> None:
+    def __init__(self, listener_address: Tuple[str, int], client_name: str) -> None:
         self.socket = ClientSocket(serialization="dill")
         self.socket.connect(*listener_address)
         self.logger = logging.getLogger("openwpm")
+        self.socket.send(client_name)
 
     def store_record(
         self, table_name: TableName, visit_id: VisitId, data: Dict[str, Any]
@@ -443,7 +457,7 @@ def save_configuration(
         browser_version: str,
     ) -> None:
         assert self.listener_address is not None
-        sock = DataSocket(self.listener_address)
+        sock = DataSocket(self.listener_address, "StorageControllerHandle")
         task_id = random.getrandbits(32)
         sock.store_record(
             TableName("task"),
@@ -467,6 +481,7 @@ def save_configuration(
                 },
             )
         sock.finalize_visit_id(INVALID_VISIT_ID, success=True)
+        sock.close()
 
     def launch(self) -> None:
         """Starts the storage controller"""

openwpm/task_manager.py

@@ -291,7 +291,9 @@ def _launch_storage_controller(
         )
         assert self.manager_params.storage_controller_address is not None
         # open connection to storage controller for saving crawl details
-        self.sock = DataSocket(self.manager_params.storage_controller_address)
+        self.sock = DataSocket(
+            self.manager_params.storage_controller_address, "TaskManager"
+        )
 
     def _shutdown_manager(
         self, during_init: bool = False, relaxed: bool = True