Timeout should not trigger a retry by default

Author: honzakral

Changelog.rst

@@ -3,6 +3,12 @@
 Changelog
 =========
 
+1.3.0 (dev)
+-----------
+
+ * Timeout now doesn't trigger a retry by default (can be overriden by setting
+ `retry_on_timeout=True`)
+
 1.2.0 (2014-08-03)
 ------------------
 

elasticsearch/connection/http_urllib3.py

@@ -3,7 +3,7 @@
 from urllib3.exceptions import ReadTimeoutError
 
 from .base import Connection
-from ..exceptions import ConnectionError
+from ..exceptions import ConnectionError, ConnectionTimeout
 from ..compat import urlencode
 
 class Urllib3HttpConnection(Connection):
@@ -33,7 +33,7 @@ def __init__(self, host='localhost', port=9200, http_auth=None, use_ssl=False, m
  def perform_request(self, method, url, params=None, body=None, timeout=None, ignore=()):
  url = self.url_prefix + url
  if params:
- url = '%s?%s' % (url, urlencode(params or {}))
+ url = '%s?%s' % (url, urlencode(params))
  full_url = self.host + url
 
  start = time.time()
@@ -52,7 +52,7 @@ def perform_request(self, method, url, params=None, body=None, timeout=None, ign
  raw_data = response.data.decode('utf-8')
  except ReadTimeoutError as e:
  self.log_request_fail(method, full_url, body, time.time() - start, exception=e)
- raise ConnectionError('TIMEOUT', str(e), e)
+ raise ConnectionTimeout('TIMEOUT', str(e), e)
  except Exception as e:
  self.log_request_fail(method, full_url, body, time.time() - start, exception=e)
  raise ConnectionError('N/A', str(e), e)

elasticsearch/transport.py

@@ -4,7 +4,8 @@
 from .connection import Urllib3HttpConnection
 from .connection_pool import ConnectionPool
 from .serializer import JSONSerializer, Deserializer, DEFAULT_SERIALIZERS
-from .exceptions import ConnectionError, TransportError, SerializationError
+from .exceptions import ConnectionError, TransportError, SerializationError, \
+ ConnectionTimeout
 
 # get ip/port from "inet[wind/127.0.0.1:9200]"
 ADDRESS_RE = re.compile(r'/(?P<host>[\.:0-9a-f]*):(?P<port>[0-9]+)\]?$')
@@ -45,7 +46,7 @@ def __init__(self, hosts, connection_class=Urllib3HttpConnection,
  sniff_on_start=False, sniffer_timeout=None, sniff_timeout=.1,
  sniff_on_connection_fail=False, serializer=JSONSerializer(), serializers=None,
  default_mimetype='application/json', max_retries=3,
- send_get_body_as='GET', **kwargs):
+ retry_on_timeout=False, send_get_body_as='GET', **kwargs):
  """
  :arg hosts: list of dictionaries, each containing keyword arguments to
  create a `connection_class` instance
@@ -67,6 +68,8 @@ def __init__(self, hosts, connection_class=Urllib3HttpConnection,
  :arg default_mimetype: when no mimetype is specified by the server
  response assume this mimetype, defaults to `'application/json'`
  :arg max_retries: maximum number of retries before an exception is propagated
+ :arg retry_on_timeout: should timeout trigger a retry on different
+ node? (default `False`)
  :arg send_get_body_as: for GET requests with body this option allows
  you to specify an alternate way of execution for environments that
  don't support passing bodies with GET requests. If you set this to
@@ -89,6 +92,7 @@ def __init__(self, hosts, connection_class=Urllib3HttpConnection,
  self.deserializer = Deserializer(_serializers, default_mimetype)
 
  self.max_retries = max_retries
+ self.retry_on_timeout = retry_on_timeout
  self.send_get_body_as = send_get_body_as
 
  # data serializer
@@ -282,6 +286,16 @@ def perform_request(self, method, url, params=None, body=None):
 
  try:
  status, headers, data = connection.perform_request(method, url, params, body, ignore=ignore, timeout=timeout)
+ except ConnectionTimeout:
+ if self.retry_on_timeout:
+ # only mark as dead if we are retrying
+ self.mark_dead(connection)
+ # raise exception on last retry
+ if attempt == self.max_retries:
+ raise
+ else:
+ raise
+
  except ConnectionError:
  self.mark_dead(connection)