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)
This commit is contained in:
Matthias Runge 2021-03-05 11:24:25 +01:00 committed by Hervé Beraud
parent e5025587a6
commit 1ddddbba73
2 changed files with 79 additions and 53 deletions

View File

@ -6,7 +6,7 @@ stevedore>=1.16.0 # Apache-2.0
voluptuous>=0.8.9 # BSD License voluptuous>=0.8.9 # BSD License
msgpack>=0.4.0 # Apache-2.0 msgpack>=0.4.0 # Apache-2.0
fasteners>=0.7 # 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 futurist>=1.2.0 # Apache-2.0
oslo.utils>=3.15.0 # Apache-2.0 oslo.utils>=3.15.0 # Apache-2.0
oslo.serialization>=1.10.0 # Apache-2.0 oslo.serialization>=1.10.0 # Apache-2.0

View File

@ -14,7 +14,6 @@
# License for the specific language governing permissions and limitations # License for the specific language governing permissions and limitations
# under the License. # under the License.
import contextlib
from distutils import version from distutils import version
import functools import functools
import logging import logging
@ -35,12 +34,37 @@ from tooz import utils
LOG = logging.getLogger(__name__) 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: 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, utils.raise_with_cause(coordination.ToozConnectionError,
encodeutils.exception_to_unicode(e), encodeutils.exception_to_unicode(e),
cause=e) cause=e)
@ -48,6 +72,8 @@ def _translate_failures():
utils.raise_with_cause(tooz.ToozError, utils.raise_with_cause(tooz.ToozError,
encodeutils.exception_to_unicode(e), encodeutils.exception_to_unicode(e),
cause=e) cause=e)
return func(*args, **kwargs)
return wrapper
class RedisLock(locking.Lock): class RedisLock(locking.Lock):
@ -63,23 +89,23 @@ class RedisLock(locking.Lock):
self._coord = coord self._coord = coord
self._client = client self._client = client
@_handle_failures
def is_still_owner(self): def is_still_owner(self):
with _translate_failures():
lock_tok = self._lock.local.token lock_tok = self._lock.local.token
if not lock_tok: if not lock_tok:
return False return False
owner_tok = self._client.get(self.name) owner_tok = self._client.get(self.name)
return owner_tok == lock_tok return owner_tok == lock_tok
@_handle_failures
def break_(self): def break_(self):
with _translate_failures():
return bool(self._client.delete(self.name)) return bool(self._client.delete(self.name))
@_handle_failures
def acquire(self, blocking=True, shared=False): def acquire(self, blocking=True, shared=False):
if shared: if shared:
raise tooz.NotImplemented raise tooz.NotImplemented
blocking, timeout = utils.convert_blocking(blocking) blocking, timeout = utils.convert_blocking(blocking)
with _translate_failures():
acquired = self._lock.acquire( acquired = self._lock.acquire(
blocking=blocking, blocking_timeout=timeout) blocking=blocking, blocking_timeout=timeout)
if acquired: if acquired:
@ -87,9 +113,9 @@ class RedisLock(locking.Lock):
self._coord._acquired_locks.add(self) self._coord._acquired_locks.add(self)
return acquired return acquired
@_handle_failures
def release(self): def release(self):
with self._exclusive_access: with self._exclusive_access:
with _translate_failures():
try: try:
self._lock.release() self._lock.release()
except exceptions.LockError as e: except exceptions.LockError as e:
@ -99,10 +125,10 @@ class RedisLock(locking.Lock):
self._coord._acquired_locks.discard(self) self._coord._acquired_locks.discard(self)
return True return True
@_handle_failures
def heartbeat(self): def heartbeat(self):
with self._exclusive_access: with self._exclusive_access:
if self.acquired: if self.acquired:
with _translate_failures():
self._lock.reacquire() self._lock.reacquire()
return True return True
return False return False
@ -433,6 +459,7 @@ return 1
return master_client return master_client
return redis.StrictRedis(**kwargs) return redis.StrictRedis(**kwargs)
@_handle_failures
def _start(self): def _start(self):
super(RedisDriver, self)._start() super(RedisDriver, self)._start()
try: try:
@ -446,7 +473,6 @@ return 1
# Ensure that the server is alive and not dead, this does not # Ensure that the server is alive and not dead, this does not
# ensure the server will always be alive, but does insure that it # ensure the server will always be alive, but does insure that it
# at least is alive once... # at least is alive once...
with _translate_failures():
self._server_info = self._client.info() self._server_info = self._client.info()
# Validate we have a good enough redis version we are connected # Validate we have a good enough redis version we are connected
# to so that the basic set of features we support will actually # to so that the basic set of features we support will actually
@ -506,12 +532,13 @@ return 1
def _decode_group_id(self, group_id): def _decode_group_id(self, group_id):
return utils.to_binary(group_id, encoding=self._encoding) return utils.to_binary(group_id, encoding=self._encoding)
@_handle_failures
def heartbeat(self): def heartbeat(self):
with _translate_failures():
beat_id = self._encode_beat_id(self._member_id) beat_id = self._encode_beat_id(self._member_id)
expiry_ms = max(0, int(self.membership_timeout * 1000.0)) expiry_ms = max(0, int(self.membership_timeout * 1000.0))
self._client.psetex(beat_id, time_ms=expiry_ms, self._client.psetex(beat_id, time_ms=expiry_ms,
value=self.STILL_ALIVE) value=self.STILL_ALIVE)
for lock in self._acquired_locks.copy(): for lock in self._acquired_locks.copy():
try: try:
lock.heartbeat() lock.heartbeat()
@ -520,6 +547,7 @@ return 1
exc_info=True) exc_info=True)
return min(self.lock_timeout, self.membership_timeout) return min(self.lock_timeout, self.membership_timeout)
@_handle_failures
def _stop(self): def _stop(self):
while self._acquired_locks: while self._acquired_locks:
lock = self._acquired_locks.pop() lock = self._acquired_locks.pop()
@ -534,7 +562,6 @@ return 1
try: try:
# NOTE(harlowja): this will delete nothing if the key doesn't # NOTE(harlowja): this will delete nothing if the key doesn't
# exist in the first place, which is fine/expected/desired... # exist in the first place, which is fine/expected/desired...
with _translate_failures():
self._client.delete(beat_id) self._client.delete(beat_id)
except tooz.ToozError: except tooz.ToozError:
LOG.warning("Unable to delete heartbeat key '%s'", beat_id, LOG.warning("Unable to delete heartbeat key '%s'", beat_id,
@ -752,5 +779,4 @@ return 1
return result return result
RedisFutureResult = functools.partial(coordination.CoordinatorResult, RedisFutureResult = functools.partial(coordination.CoordinatorResult)
failure_translator=_translate_failures)