address review comments

hazelcast/metrics.py

@@ -93,7 +93,7 @@ class ProbeUnit(object):
  ENUM = 6
  """0..n, ordinal of an enum."""
 
- # New enum values should be should be converted to a tag to ensure backward
+ # New enum values should be converted to a tag to ensure backward
  # compatibility during compressing the metrics. see ProbeUnit#newUnit
  # handling in Java for that. We don't implement this functionality yet in
  # the Python client, as we don't use such enum members.
@@ -112,29 +112,29 @@ class ValueType(object):
 class MetricsCompressor(object):
  """Compresses metrics into a ``bytearray`` blob.
 
-  The compressor uses dictionary based delta compression and deflates
-  the resulting ``bytearray`` by using ``zlib.compress()``. This
-  compressor doesn't use the textual representation of the
-  :class:`MetricDescriptor` hence it is agnostic to the order of the
-  tags in that representation.
-
-  Adding metrics by calling :func:`add_long` or :func:`add_double`
-  builds a dictionary by mapping all words found in the passed
-  :class:`MetricDescriptor`s to ``int``s and these ``int``s will be
-  written to the resulting ``bytearray`` blob. Before these ``int``s
-  are written, the current :class:`MetricDescriptor` is compared to the
-  previous one and only the fields of the descriptor that are different
-  from the previous will be written to the metrics blob.
-
-  When the blob is retrieved from this compressor (when the metrics
-  collection cycle finishes) the dictionary is stored in the dictionary
-  blob. The compressor iterates over the words stored in the dictionary
-  in ascending order and writes them to the blob by skipping the first
-  N characters that are equal to the first N character of the previously
-  written word, hence using delta compression here too.
-
-  After both the metrics and the dictionary blob is constructed, they
-  are copied into a final blob in the following structure:
+ The compressor uses dictionary based delta compression and deflates
+ the resulting ``bytearray`` by using ``zlib.compress()``. This
+ compressor doesn't use the textual representation of the
+ :class:`MetricDescriptor` hence it is agnostic to the order of the
+ tags in that representation.
+
+ Adding metrics by calling :func:`add_long` or :func:`add_double`
+ builds a dictionary by mapping all words found in the passed
+ :class:`MetricDescriptor`s to ``int``s and these ``int``s will be
+ written to the resulting ``bytearray`` blob. Before these ``int``s
+ are written, the current :class:`MetricDescriptor` is compared to the
+ previous one and only the fields of the descriptor that are different
+ from the previous will be written to the metrics blob.
+
+ When the blob is retrieved from this compressor (when the metrics
+ collection cycle finishes) the dictionary is stored in the dictionary
+ blob. The compressor iterates over the words stored in the dictionary
+ in ascending order and writes them to the blob by skipping the first
+ N characters that are equal to the first N character of the previously
+ written word, hence using delta compression here too.
+
+ After both the metrics and the dictionary blob is constructed, they
+ are copied into a final blob in the following structure:
 
  +--------------------------------+--------------------+
  | Compressor version | 2 bytes (short) |
@@ -213,7 +213,7 @@ def _write_descriptor(self, descriptor):
  else:
  self._metrics_buf.write_byte(descriptor.unit)
 
- # include excludedTargets and tags bytes for compatibility purposes
+ # include excludedTargets and tags bits for compatibility purposes
  if mask & _MASK_EXCLUDED_TARGETS == 0:
  self._metrics_buf.write_byte(0)
 
@@ -245,7 +245,7 @@ def _calculate_descriptor_mask(self, descriptor):
  if descriptor.unit == last_descriptor.unit:
  mask |= _MASK_UNIT
 
- # include excludedTargets and tags bytes for compatibility purposes
+ # include excludedTargets and tags bits for compatibility purposes
  mask |= _MASK_EXCLUDED_TARGETS
  mask |= _MASK_TAG_COUNT
 

hazelcast/statistics.py

@@ -1,4 +1,5 @@
 import logging
+import os
 
 from hazelcast.core import CLIENT_TYPE
 from hazelcast.invocation import Invocation
@@ -21,7 +22,6 @@
 _KEY_VALUE_SEPARATOR = "="
 _EMPTY_ATTRIBUTE_VALUE = ""
 
-_DEFAULT_PROBE_VALUE = 0
 _NEAR_CACHE_DESCRIPTOR_PREFIX = "nearcache"
 _NEAR_CACHE_DESCRIPTOR_DISCRIMINATOR = "name"
 
@@ -94,7 +94,7 @@ def _register_gauges(self):
  )
  self._register_system_gauge(
  "os.systemLoadAverage",
- lambda: psutil.cpu_percent(),
+ lambda: os.getloadavg()[0],
  ValueType.DOUBLE,
  )
  self._register_system_gauge(
@@ -124,6 +124,7 @@ def _register_gauges(self):
  )
 
  def _register_system_gauge(self, gauge_name, gauge_fn, value_type=ValueType.LONG):
+ # Try a gauge function read, we will register it if it succeeds.
  try:
  gauge_fn()
  self._registered_system_gauges[gauge_name] = (gauge_fn, value_type)
@@ -133,6 +134,7 @@ def _register_system_gauge(self, gauge_name, gauge_fn, value_type=ValueType.LONG
  )
 
  def _register_process_gauge(self, gauge_name, gauge_fn, value_type=ValueType.LONG):
+ # Try a gauge function read, we will register it if it succeeds.
  try:
  process = psutil.Process()
  gauge_fn(process)