Skip to content

Flaky test_error_during_startup[--no-nanny] #6762

@gjoseph92

Description

@gjoseph92

Identical to #6761

____________________ test_error_during_startup[--no-nanny] _____________________
ConnectionRefusedError: [Errno 61] Connection refused
The above exception was the direct cause of the following exception:
addr = 'tcp://127.0.0.1:50397', timeout = 5, deserialize = True
handshake_overrides = None
connection_args = {'extra_conn_args': {}, 'require_encryption': False, 'ssl_context': None}
scheme = 'tcp', loc = '127.0.0.1:50397'
backend = <distributed.comm.tcp.TCPBackend object at 0x12a9c7ca0>
connector = <distributed.comm.tcp.TCPConnector object at 0x135280820>
comm = None, time_left = <function connect.<locals>.time_left at 0x135e89120>
backoff_base = 0.01
asyncdefconnect(
        addr, timeout=None, deserialize=True, handshake_overrides=None, **connection_args
    ):
"""
    Connect to the given address (a URI such as ``tcp://127.0.0.1:1234``)
    and yield a ``Comm`` object.  If the connection attempt fails, it is
    retried until the *timeout* is expired.
    """
if timeout isNone:
            timeout = dask.config.get("distributed.comm.timeouts.connect")
        timeout = parse_timedelta(timeout, default="seconds")
        scheme, loc = parse_address(addr)
        backend = registry.get_backend(scheme)
        connector = backend.get_connector()
        comm = None
        start = time()
deftime_left():
            deadline = start + timeout
returnmax(0, deadline - time())
        backoff_base = 0.01
        attempt = 0
# Prefer multiple small attempts than one long attempt. This should protect
# primarily from DNS race conditions
# gh3104, gh4176, gh4167
        intermediate_cap = timeout / 5
        active_exception = None
while time_left() > 0:
try:
>               comm = await asyncio.wait_for(
                    connector.connect(loc, deserialize=deserialize, **connection_args),
                    timeout=min(intermediate_cap, time_left()),
                )
distributed/comm/core.py:291: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
fut = <Task finished name='Task-2297' coro=<BaseTCPConnector.connect() done, defined at /Users/runner/work/distributed/distr...'in <distributed.comm.tcp.TCPConnector object at 0x135280820>: ConnectionRefusedError: [Errno 61] Connection refused')>
timeout = 1.6936261653900146
asyncdefwait_for(fut, timeout):
"""Wait for the single Future or coroutine to complete, with timeout.
    Coroutine will be wrapped in Task.
    Returns result of the Future or coroutine.  When a timeout occurs,
    it cancels the task and raises TimeoutError.  To avoid the task
    cancellation, wrap it in shield().
    If the wait is cancelled, the task is also cancelled.
    This function is a coroutine.
    """
        loop = events.get_running_loop()
if timeout isNone:
returnawait fut
if timeout <= 0:
            fut = ensure_future(fut, loop=loop)
if fut.done():
return fut.result()
await _cancel_and_wait(fut, loop=loop)
try:
return fut.result()
except exceptions.CancelledError as exc:
raise exceptions.TimeoutError() fromexc
        waiter = loop.create_future()
        timeout_handle = loop.call_later(timeout, _release_waiter, waiter)
        cb = functools.partial(_release_waiter, waiter)
        fut = ensure_future(fut, loop=loop)
        fut.add_done_callback(cb)
try:
# wait until the future completes or the timeout
try:
await waiter
except exceptions.CancelledError:
if fut.done():
return fut.result()
else:
                    fut.remove_done_callback(cb)
# We must ensure that the task is not running
# after wait_for() returns.
# See https://bugs.python.org/issue32751
await _cancel_and_wait(fut, loop=loop)
raise
if fut.done():
>               return fut.result()
../../../miniconda3/envs/dask-distributed/lib/python3.10/asyncio/tasks.py:445: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
self = <distributed.comm.tcp.TCPConnector object at 0x135280820>
address = '127.0.0.1:50397', deserialize = True
connection_args = {'extra_conn_args': {}, 'require_encryption': False, 'ssl_context': None}
ip = '127.0.0.1', port = 50397, kwargs = {}
asyncdefconnect(self, address, deserialize=True, **connection_args):
self._check_encryption(address, connection_args)
        ip, port = parse_host_port(address)
        kwargs = self._get_connect_args(**connection_args)
try:
            stream = awaitself.client.connect(
                ip, port, max_buffer_size=MAX_BUFFER_SIZE, **kwargs
            )
# Under certain circumstances tornado will have a closed connnection with an
# error and not raise a StreamClosedError.
#
# This occurs with tornado 5.x and openssl 1.1+
if stream.closed() and stream.error:
raise StreamClosedError(stream.error)
except StreamClosedError as e:
# The socket connect() call failed
>           convert_stream_closed_error(self, e)
distributed/comm/tcp.py:461: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
obj = <distributed.comm.tcp.TCPConnector object at 0x135280820>
exc = ConnectionRefusedError(61, 'Connection refused')
defconvert_stream_closed_error(obj, exc):
"""
    Re-raise StreamClosedError as CommClosedError.
    """
if exc.real_error isnotNone:
# The stream was closed because of an underlying OS error
            exc = exc.real_error
ifisinstance(exc, ssl.SSLError):
if"UNKNOWN_CA"in exc.reason:
raise FatalCommClosedError(f"in {obj}: {exc.__class__.__name__}: {exc}")
>           raise CommClosedError(f"in {obj}: {exc.__class__.__name__}: {exc}") fromexc
E           distributed.comm.core.CommClosedError: in <distributed.comm.tcp.TCPConnector object at 0x135280820>: ConnectionRefusedError: [Errno 61] Connection refused
distributed/comm/tcp.py:142: CommClosedError
The above exception was the direct cause of the following exception:
monkeypatch = <_pytest.monkeypatch.MonkeyPatch object at 0x136fbe500>
nanny = '--no-nanny'
loop = <tornado.platform.asyncio.AsyncIOLoop object at 0x13520a530>
@pytest.mark.parametrize("nanny", ["--nanny", "--no-nanny"])
deftest_error_during_startup(monkeypatch, nanny, loop):
# see https://github.com/dask/distributed/issues/6320
        scheduler_port = open_port()
        scheduler_addr = f"tcp://127.0.0.1:{scheduler_port}"
        monkeypatch.setenv("DASK_SCHEDULER_ADDRESS", scheduler_addr)
with popen(
            [
"dask-scheduler",
f"--port={scheduler_port}",
"--dashboard-address=:0",
            ],
        ):
>           with Client(scheduler_addr, loop=loop) as c:
distributed/cli/tests/test_dask_worker.py:719: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
distributed/client.py:940: in __init__
self.start(timeout=timeout)
distributed/client.py:1098: in start
    sync(self.loop, self._start, **kwargs)
distributed/utils.py:405: in sync
raise exc.with_traceback(tb)
distributed/utils.py:378: in f
    result = yield future
../../../miniconda3/envs/dask-distributed/lib/python3.10/site-packages/tornado/gen.py:762: in run
    value = future.result()
distributed/client.py:1178: in _start
awaitself._ensure_connected(timeout=timeout)
distributed/client.py:1241: in _ensure_connected
    comm = await connect(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
addr = 'tcp://127.0.0.1:50397', timeout = 5, deserialize = True
handshake_overrides = None
connection_args = {'extra_conn_args': {}, 'require_encryption': False, 'ssl_context': None}
scheme = 'tcp', loc = '127.0.0.1:50397'
backend = <distributed.comm.tcp.TCPBackend object at 0x12a9c7ca0>
connector = <distributed.comm.tcp.TCPConnector object at 0x135280820>
comm = None, time_left = <function connect.<locals>.time_left at 0x135e89120>
backoff_base = 0.01
asyncdefconnect(
        addr, timeout=None, deserialize=True, handshake_overrides=None, **connection_args
    ):
"""
    Connect to the given address (a URI such as ``tcp://127.0.0.1:1234``)
    and yield a ``Comm`` object.  If the connection attempt fails, it is
    retried until the *timeout* is expired.
    """
if timeout isNone:
            timeout = dask.config.get("distributed.comm.timeouts.connect")
        timeout = parse_timedelta(timeout, default="seconds")
        scheme, loc = parse_address(addr)
        backend = registry.get_backend(scheme)
        connector = backend.get_connector()
        comm = None
        start = time()
deftime_left():
            deadline = start + timeout
returnmax(0, deadline - time())
        backoff_base = 0.01
        attempt = 0
# Prefer multiple small attempts than one long attempt. This should protect
# primarily from DNS race conditions
# gh3104, gh4176, gh4167
        intermediate_cap = timeout / 5
        active_exception = None
while time_left() > 0:
try:
                comm = await asyncio.wait_for(
                    connector.connect(loc, deserialize=deserialize, **connection_args),
                    timeout=min(intermediate_cap, time_left()),
                )
break
except FatalCommClosedError:
raise
# Note: CommClosed inherits from OSError
except (asyncio.TimeoutError, OSError) as exc:
                active_exception = exc
# As descibed above, the intermediate timeout is used to distributed
# initial, bulk connect attempts homogeneously. In particular with
# the jitter upon retries we should not be worred about overloading
# any more DNS servers
                intermediate_cap = timeout
# FullJitter see https://aws.amazon.com/blogs/architecture/exponential-backoff-and-jitter/
                upper_cap = min(time_left(), backoff_base * (2**attempt))
                backoff = random.uniform(0, upper_cap)
                attempt += 1
                logger.debug(
"Could not connect to %s, waiting for %s before retrying", loc, backoff
                )
await asyncio.sleep(backoff)
else:
>           raiseOSError(
f"Timed out trying to connect to {addr} after {timeout} s"
            ) fromactive_exception
E           OSError: Timed out trying to connect to tcp://127.0.0.1:50397 after 5 s
distributed/comm/core.py:317: OSError

Metadata

Metadata

Assignees

No one assigned

    Labels

    flaky testIntermittent failures on CI.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions