Add locking for interval sources. This reduces the chance of repeated documents from long running email downloads.

Signed-off-by: Roberto Rosario <roberto.rosario.gonzalez@gmail.com>

HISTORY.rst

@@ -147,6 +147,7 @@
 - Update rendering of the readonly multiselect widget to conform to Django's updated field class interface.
 - Add warning when using SQLite as the database backend.
 - Use Mailgun's flanker library to process the email sources.
+- Add locking for interval sources. This reduces the chance of repeated documents from long running email downloads.
 
 2.7.3 (2017-09-11)
 ==================

docs/releases/3.0.rst

@@ -466,6 +466,7 @@ Other changes worth mentioning
 - Make error messages persistent and increase the timeout of warning to 10 seconds.
 - Improve rendering of the details form.
 - Update rendering of the readonly multiselect widget to conform to Django's updated field class interface.
+- Add locking for interval sources. This reduces the chance of repeated documents from long running email downloads.
 
 
 Removals

mayan/apps/sources/literals.py

@@ -59,6 +59,7 @@ SOURCE_CHOICES = (
     (SOURCE_CHOICE_EMAIL_IMAP, _('IMAP email')),
 )
 
+DEFAULT_SOURCE_LOCK_EXPIRE = 600
 DEFAULT_INTERVAL = 600
 DEFAULT_METADATA_ATTACHMENT_NAME = 'metadata.yaml'
 DEFAULT_POP3_TIMEOUT = 60

mayan/apps/sources/models.py

@@ -551,10 +551,6 @@ class EmailBaseModel(IntervalBaseModel):
                     }
                 )
 
-    # TODO: Add lock to avoid running more than once concurrent same document
-    # download
-    # TODO: Use message ID for lock
-
     @staticmethod
     def process_message(source, message_text):
         counter = 1

mayan/apps/sources/tasks.py

@@ -10,8 +10,12 @@ from django.utils.translation import ugettext_lazy as _
 from mayan.celery import app
 
 from common.compressed_files import CompressedFile, NotACompressedFile
+from lock_manager import LockError
+from lock_manager.runtime import locking_backend
 
-from .literals import DEFAULT_SOURCE_TASK_RETRY_DELAY
+from .literals import (
+    DEFAULT_SOURCE_LOCK_EXPIRE, DEFAULT_SOURCE_TASK_RETRY_DELAY
+)
 
 logger = logging.getLogger(__name__)
 
@@ -22,10 +26,19 @@ def task_check_interval_source(source_id):
         app_label='sources', model_name='Source'
     )
 
-    source = Source.objects.get_subclass(pk=source_id)
+    lock_id = 'task_check_interval_source-%d' % source_id
-    if source.enabled:
+    try:
+        logger.debug('trying to acquire lock: %s', lock_id)
+        lock = locking_backend.acquire_lock(lock_id, DEFAULT_SOURCE_LOCK_EXPIRE)
+    except LockError:
+        logger.debug('unable to obtain lock: %s' % lock_id)
+    else:
+        logger.debug('acquired lock: %s', lock_id)
+
         try:
-            source.check_source()
+            source = Source.objects.get_subclass(pk=source_id)
+            if source.enabled:
+                source.check_source()
         except Exception as exception:
             logger.error('Error processing source: %s; %s', source, exception)
             source.logs.create(
@@ -33,6 +46,8 @@ def task_check_interval_source(source_id):
             )
         else:
             source.logs.all().delete()
+        finally:
+            lock.release()
 
 
 @app.task(bind=True, default_retry_delay=DEFAULT_SOURCE_TASK_RETRY_DELAY, ignore_result=True)

mayan/apps/sources/tests/test_models.py

@@ -23,71 +23,6 @@ from .literals import (
 )
 
 
-@override_settings(OCR_AUTO_OCR=False)
-class UploadDocumentTestCase(BaseTestCase):
-    """
-    Test creating documents
-    """
-
-    def setUp(self):
-        super(UploadDocumentTestCase, self).setUp()
-        self.document_type = DocumentType.objects.create(
-            label=TEST_DOCUMENT_TYPE_LABEL
-        )
-
-    def tearDown(self):
-        self.document_type.delete()
-        super(UploadDocumentTestCase, self).tearDown()
-
-    def test_issue_gh_163(self):
-        """
-        Non-ASCII chars in document name failing in upload via watch folder
-        gh-issue #163 https://github.com/mayan-edms/mayan-edms/issues/163
-        """
-
-        temporary_directory = mkdtemp()
-        shutil.copy(TEST_NON_ASCII_DOCUMENT_PATH, temporary_directory)
-
-        watch_folder = WatchFolderSource.objects.create(
-            document_type=self.document_type, folder_path=temporary_directory,
-            uncompress=SOURCE_UNCOMPRESS_CHOICE_Y
-        )
-        watch_folder.check_source()
-
-        self.assertEqual(Document.objects.count(), 1)
-
-        document = Document.objects.first()
-
-        self.assertEqual(document.exists(), True)
-        self.assertEqual(document.size, 17436)
-
-        self.assertEqual(document.file_mimetype, 'image/png')
-        self.assertEqual(document.file_mime_encoding, 'binary')
-        self.assertEqual(document.label, TEST_NON_ASCII_DOCUMENT_FILENAME)
-        self.assertEqual(document.page_count, 1)
-
-        # Test Non-ASCII named documents inside Non-ASCII named compressed file
-
-        shutil.copy(
-            TEST_NON_ASCII_COMPRESSED_DOCUMENT_PATH, temporary_directory
-        )
-
-        watch_folder.check_source()
-        document = Document.objects.all()[1]
-
-        self.assertEqual(Document.objects.count(), 2)
-
-        self.assertEqual(document.exists(), True)
-        self.assertEqual(document.size, 17436)
-
-        self.assertEqual(document.file_mimetype, 'image/png')
-        self.assertEqual(document.file_mime_encoding, 'binary')
-        self.assertEqual(document.label, TEST_NON_ASCII_DOCUMENT_FILENAME)
-        self.assertEqual(document.page_count, 1)
-
-        shutil.rmtree(temporary_directory)
-
-
 @override_settings(OCR_AUTO_OCR=False)
 class CompressedUploadsTestCase(BaseTestCase):
     def setUp(self):
@@ -189,3 +124,67 @@ class EmailFilenameDecodingTestCase(BaseTestCase):
                 '<Document: test-02.png>'
             ),
         )
+
+
+@override_settings(OCR_AUTO_OCR=False)
+class UploadDocumentTestCase(BaseTestCase):
+    """
+    Test creating documents
+    """
+    def setUp(self):
+        super(UploadDocumentTestCase, self).setUp()
+        self.document_type = DocumentType.objects.create(
+            label=TEST_DOCUMENT_TYPE_LABEL
+        )
+
+    def tearDown(self):
+        self.document_type.delete()
+        super(UploadDocumentTestCase, self).tearDown()
+
+    def test_issue_gh_163(self):
+        """
+        Non-ASCII chars in document name failing in upload via watch folder
+        gh-issue #163 https://github.com/mayan-edms/mayan-edms/issues/163
+        """
+
+        temporary_directory = mkdtemp()
+        shutil.copy(TEST_NON_ASCII_DOCUMENT_PATH, temporary_directory)
+
+        watch_folder = WatchFolderSource.objects.create(
+            document_type=self.document_type, folder_path=temporary_directory,
+            uncompress=SOURCE_UNCOMPRESS_CHOICE_Y
+        )
+        watch_folder.check_source()
+
+        self.assertEqual(Document.objects.count(), 1)
+
+        document = Document.objects.first()
+
+        self.assertEqual(document.exists(), True)
+        self.assertEqual(document.size, 17436)
+
+        self.assertEqual(document.file_mimetype, 'image/png')
+        self.assertEqual(document.file_mime_encoding, 'binary')
+        self.assertEqual(document.label, TEST_NON_ASCII_DOCUMENT_FILENAME)
+        self.assertEqual(document.page_count, 1)
+
+        # Test Non-ASCII named documents inside Non-ASCII named compressed file
+
+        shutil.copy(
+            TEST_NON_ASCII_COMPRESSED_DOCUMENT_PATH, temporary_directory
+        )
+
+        watch_folder.check_source()
+        document = Document.objects.all()[1]
+
+        self.assertEqual(Document.objects.count(), 2)
+
+        self.assertEqual(document.exists(), True)
+        self.assertEqual(document.size, 17436)
+
+        self.assertEqual(document.file_mimetype, 'image/png')
+        self.assertEqual(document.file_mime_encoding, 'binary')
+        self.assertEqual(document.label, TEST_NON_ASCII_DOCUMENT_FILENAME)
+        self.assertEqual(document.page_count, 1)
+
+        shutil.rmtree(temporary_directory)