workers.securityworker: small fixes

This change adjusts our batch size to coerce to integer after all
floating point math in order to get a more accurate end result. In
addition, we handle the scenario when there are no longer any images in
the database to be scanned when finding the min id.

workers/securityworker.py

@@ -1,7 +1,7 @@
 import logging.config
 import time
 
-from math import floor, log10
+from math import log10
 
 import features
 
@@ -29,19 +29,16 @@ class SecurityWorker(Worker):
   def __init__(self):
     super(SecurityWorker, self).__init__()
     validator = SecurityConfigValidator(app.config)
-    if validator.valid():
-      self._target_version = app.config.get('SECURITY_SCANNER_ENGINE_VERSION_TARGET', 3)
-      self._analyzer = LayerAnalyzer(app.config, secscan_api)
-
-      # Get the ID of the first image we want to analyze.
-      self._min_id = app.config.get('SECURITY_SCANNER_INDEXING_MIN_ID')
-      if self._min_id is None:
-        self._min_id = get_min_id_for_sec_scan(self._target_version)
-
-      interval = app.config.get('SECURITY_SCANNER_INDEXING_INTERVAL', DEFAULT_INDEXING_INTERVAL)
-      self.add_operation(self._index_images, interval)
-    else:
+    if not validator.valid():
       logger.warning('Failed to validate security scan configuration')
+      return
+
+    self._target_version = app.config.get('SECURITY_SCANNER_ENGINE_VERSION_TARGET', 3)
+    self._analyzer = LayerAnalyzer(app.config, secscan_api)
+    self._min_id = None
+
+    interval = app.config.get('SECURITY_SCANNER_INDEXING_INTERVAL', DEFAULT_INDEXING_INTERVAL)
+    self.add_operation(self._index_images, interval)
 
   def _index_images(self):
     def batch_query():
@@ -53,10 +50,14 @@ class SecurityWorker(Worker):
     if max_id is None:
       return
 
+    if self.min_id is None or self.min_id > max_id:
+      logger.info('Could not find any available images for scanning.')
+      return
+
     max_unscanned_images_gauge.Set(max_id)
 
     # 4^log10(total) gives us a scalable batch size into the billions.
-    batch_size = 4 ** int(floor(log10(max(10, max_id - self._min_id))))
+    batch_size = int(4 ** log10(max(10, max_id - self.min_id)))
 
     with UseThenDisconnect(app.config):
       to_scan_generator = yield_random_entries(
@@ -64,7 +65,7 @@ class SecurityWorker(Worker):
         get_image_pk_field(),
         batch_size,
         max_id,
-        self._min_id,
+        self.min_id,
       )
       for candidate, abt, num_remaining in to_scan_generator:
         try:
@@ -77,7 +78,25 @@ class SecurityWorker(Worker):
 
     # If we reach this point, we analyzed every images up to max_id, next time the worker runs,
     # we want to start from the next image.
-    self._min_id = max_id + 1
+    self.min_id = max_id + 1
+
+
+  @property
+  def min_id(self):
+    """ If it hasn't already been determined, finds the ID of the first image to be analyzed.
+        First checks the config, then the database, and returns None if there are no images
+        available for scanning.
+    """
+    if self._min_id is None:
+      self._min_id = app.config.get('SECURITY_SCANNER_INDEXING_MIN_ID')
+      if self._min_id is None:
+        self._min_id = get_min_id_for_sec_scan(self._target_version)
+    return self._min_id
+
+  @min_id.setter
+  def min_id(self, new_min_id):
+    self._min_id = new_min_id
+
 
 if __name__ == '__main__':
   app.register_blueprint(v2_bp, url_prefix='/v2')