Retry on redis connection errors

Sometimes, connections get closed by the server. This change
adds retrying to the code.

Conflict:
requirements.txt

NOTE(hberaud): This conflict is due to the cap of tenacity during wallaby
to unblock the gate. This requirement cap won't be an issue for victoria
and for previous stable branches.

fe9c430455

Change-Id: Iaab5ce609c0dcf7085f5dd43efbd37eb4b88f17b
(cherry picked from commit a025b4d4ee)
(cherry picked from commit 62d5d17a5d)
(cherry picked from commit 1ddddbba73)

requirements.txt

@@ -7,7 +7,7 @@ six>=1.9.0 # MIT
 voluptuous>=0.8.9 # BSD License
 msgpack>=0.4.0 # Apache-2.0
 fasteners>=0.7 # Apache-2.0
-tenacity>=3.2.1  # Apache-2.0
+tenacity>=3.2.1,<7.0.0  # Apache-2.0
 futurist>=1.2.0 # Apache-2.0
 oslo.utils>=3.15.0 # Apache-2.0
 oslo.serialization>=1.10.0 # Apache-2.0

tooz/drivers/redis.py

@@ -14,7 +14,6 @@
 #    License for the specific language governing permissions and limitations
 #    under the License.
 
-import contextlib
 from distutils import version
 import functools
 import logging
@@ -38,12 +37,37 @@ from tooz import utils
 LOG = logging.getLogger(__name__)
 
 
-@contextlib.contextmanager
+def _handle_failures(func=None, n_tries=15):
-def _translate_failures():
+
-    """Translates common redis exceptions into tooz exceptions."""
+    """Translates common redis exceptions into tooz exceptions.
+
+    This also enables retrying on certain exceptions.
+
+    :param func: the function to act on
+    :param n_tries: the number of retries
+    """
+
+    if func is None:
+        return functools.partial(
+                _handle_failures,
+                n_tries=n_tries
+                )
+
+    @functools.wraps(func)
+    def wrapper(*args, **kwargs):
+        ntries = n_tries
+        while ntries > 1:
             try:
-        yield
+                return func(*args, **kwargs)
-    except (exceptions.ConnectionError, exceptions.TimeoutError) as e:
+            except exceptions.ConnectionError as e:
+                # retry ntries times and then raise a connection error
+                ntries -= 1
+                if ntries >= 1:
+                    utils.raise_with_cause(coordination.ToozConnectionError,
+                                           encodeutils.exception_to_unicode(e),
+                                           cause=e)
+
+            except (exceptions.TimeoutError) as e:
                 utils.raise_with_cause(coordination.ToozConnectionError,
                                        encodeutils.exception_to_unicode(e),
                                        cause=e)
@@ -51,6 +75,8 @@ def _translate_failures():
                 utils.raise_with_cause(tooz.ToozError,
                                        encodeutils.exception_to_unicode(e),
                                        cause=e)
+        return func(*args, **kwargs)
+    return wrapper
 
 
 class RedisLock(locking.Lock):
@@ -66,23 +92,23 @@ class RedisLock(locking.Lock):
         self._coord = coord
         self._client = client
 
+    @_handle_failures
     def is_still_owner(self):
-        with _translate_failures():
         lock_tok = self._lock.local.token
         if not lock_tok:
             return False
         owner_tok = self._client.get(self.name)
         return owner_tok == lock_tok
 
+    @_handle_failures
     def break_(self):
-        with _translate_failures():
         return bool(self._client.delete(self.name))
 
+    @_handle_failures
     def acquire(self, blocking=True, shared=False):
         if shared:
             raise tooz.NotImplemented
         blocking, timeout = utils.convert_blocking(blocking)
-        with _translate_failures():
         acquired = self._lock.acquire(
             blocking=blocking, blocking_timeout=timeout)
         if acquired:
@@ -90,9 +116,9 @@ class RedisLock(locking.Lock):
                 self._coord._acquired_locks.add(self)
         return acquired
 
+    @_handle_failures
     def release(self):
         with self._exclusive_access:
-            with _translate_failures():
             try:
                 self._lock.release()
             except exceptions.LockError as e:
@@ -102,10 +128,10 @@ class RedisLock(locking.Lock):
                 self._coord._acquired_locks.discard(self)
             return True
 
+    @_handle_failures
     def heartbeat(self):
         with self._exclusive_access:
             if self.acquired:
-                with _translate_failures():
                 self._lock.reacquire()
                 return True
         return False
@@ -436,6 +462,7 @@ return 1
             return master_client
         return redis.StrictRedis(**kwargs)
 
+    @_handle_failures
     def _start(self):
         super(RedisDriver, self)._start()
         try:
@@ -449,7 +476,6 @@ return 1
             # Ensure that the server is alive and not dead, this does not
             # ensure the server will always be alive, but does insure that it
             # at least is alive once...
-            with _translate_failures():
             self._server_info = self._client.info()
             # Validate we have a good enough redis version we are connected
             # to so that the basic set of features we support will actually
@@ -509,12 +535,13 @@ return 1
     def _decode_group_id(self, group_id):
         return utils.to_binary(group_id, encoding=self._encoding)
 
+    @_handle_failures
     def heartbeat(self):
-        with _translate_failures():
         beat_id = self._encode_beat_id(self._member_id)
         expiry_ms = max(0, int(self.membership_timeout * 1000.0))
         self._client.psetex(beat_id, time_ms=expiry_ms,
                             value=self.STILL_ALIVE)
+
         for lock in self._acquired_locks.copy():
             try:
                 lock.heartbeat()
@@ -523,6 +550,7 @@ return 1
                             exc_info=True)
         return min(self.lock_timeout, self.membership_timeout)
 
+    @_handle_failures
     def _stop(self):
         while self._acquired_locks:
             lock = self._acquired_locks.pop()
@@ -537,7 +565,6 @@ return 1
             try:
                 # NOTE(harlowja): this will delete nothing if the key doesn't
                 # exist in the first place, which is fine/expected/desired...
-                with _translate_failures():
                 self._client.delete(beat_id)
             except tooz.ToozError:
                 LOG.warning("Unable to delete heartbeat key '%s'", beat_id,
@@ -755,5 +782,4 @@ return 1
         return result
 
 
-RedisFutureResult = functools.partial(coordination.CoordinatorResult,
+RedisFutureResult = functools.partial(coordination.CoordinatorResult)
-                                      failure_translator=_translate_failures)