parsl 2025.8.4__py3-none-any.whl → 2025.11.10__py3-none-any.whl

parsl/executors/high_throughput/interchange.py

@@ -23,7 +23,6 @@ from parsl.monitoring.radios.base import MonitoringRadioSender
 from parsl.monitoring.radios.zmq import ZMQRadioSender
 from parsl.process_loggers import wrap_with_logs
 from parsl.serialize import serialize as serialize_object
-from parsl.utils import setproctitle
 from parsl.version import VERSION as PARSL_VERSION
 
 PKL_HEARTBEAT_CODE = pickle.dumps((2 ** 32) - 1)
@@ -56,6 +55,7 @@ class Interchange:
                  cert_dir: Optional[str],
                  manager_selector: ManagerSelector,
                  run_id: str,
+                 _check_python_mismatch: bool,
                  ) -> None:
         """
         Parameters
@@ -99,6 +99,11 @@ class Interchange:
 
         cert_dir : str | None
             Path to the certificate directory.
+
+        _check_python_mismatch : bool
+            If True, the interchange and worker managers must run the same version of
+            Python. Running different versions can cause inter-process communication
+            errors, so proceed with caution.
         """
         self.cert_dir = cert_dir
         self.logdir = logdir
@@ -126,15 +131,13 @@ class Interchange:
         logger.info("Connected to client")
 
         self.run_id = run_id
+        self._check_python_mismatch = _check_python_mismatch
 
         self.hub_address = hub_address
         self.hub_zmq_port = hub_zmq_port
 
         self.pending_task_queue: SortedList[Any] = SortedList(key=lambda tup: (tup[0], tup[1]))
 
-        # count of tasks that have been received from the submit side
-        self.task_counter = 0
-
         # count of tasks that have been sent out to worker pools
         self.count = 0
 
@@ -157,6 +160,7 @@ class Interchange:
         logger.info(f"Bound to port {worker_port} for incoming worker connections")
 
         self._ready_managers: Dict[bytes, ManagerRecord] = {}
+        self._logged_manager_count_token: object = None
         self.connected_block_history: List[str] = []
 
         self.heartbeat_threshold = heartbeat_threshold
@@ -213,7 +217,7 @@ class Interchange:
 
         reply: Any  # the type of reply depends on the command_req received (aka this needs dependent types...)
 
-        if self.command_channel in self.socks and self.socks[self.command_channel] == zmq.POLLIN:
+        if self.socks.get(self.command_channel) == zmq.POLLIN:
             logger.debug("entering command_server section")
 
             command_req = self.command_channel.recv_pyobj()
@@ -222,35 +226,29 @@ class Interchange:
                 reply = self.connected_block_history
 
             elif command_req == "WORKERS":
-                num_workers = 0
-                for manager in self._ready_managers.values():
-                    num_workers += manager['worker_count']
-                reply = num_workers
+                reply = sum(m['worker_count'] for m in self._ready_managers.values())
 
             elif command_req == "MANAGERS":
                 reply = []
-                for manager_id in self._ready_managers:
-                    m = self._ready_managers[manager_id]
-                    idle_since = m['idle_since']
-                    if idle_since is not None:
-                        idle_duration = time.time() - idle_since
-                    else:
-                        idle_duration = 0.0
-                    resp = {'manager': manager_id.decode('utf-8'),
-                            'block_id': m['block_id'],
-                            'worker_count': m['worker_count'],
-                            'tasks': len(m['tasks']),
-                            'idle_duration': idle_duration,
-                            'active': m['active'],
-                            'parsl_version': m['parsl_version'],
-                            'python_version': m['python_version'],
-                            'draining': m['draining']}
+                now = time.time()
+                for manager_id, m in self._ready_managers.items():
+                    idle_duration = now - (m['idle_since'] or now)
+                    resp = {
+                        'manager': manager_id.decode('utf-8'),
+                        'block_id': m['block_id'],
+                        'worker_count': m['worker_count'],
+                        'tasks': len(m['tasks']),
+                        'idle_duration': idle_duration,
+                        'active': m['active'],
+                        'parsl_version': m['parsl_version'],
+                        'python_version': m['python_version'],
+                        'draining': m['draining']
+                    }
                     reply.append(resp)
 
             elif command_req == "MANAGERS_PACKAGES":
                 reply = {}
-                for manager_id in self._ready_managers:
-                    m = self._ready_managers[manager_id]
+                for manager_id, m in self._ready_managers.items():
                     manager_id_str = manager_id.decode('utf-8')
                     reply[manager_id_str] = m["packages"]
 
@@ -316,6 +314,7 @@ class Interchange:
             self.process_manager_socket_message(interesting_managers, monitoring_radio, kill_event)
             self.expire_bad_managers(interesting_managers, monitoring_radio)
             self.expire_drained_managers(interesting_managers, monitoring_radio)
+            self.log_manager_counts(interesting_managers)
             self.process_tasks_to_send(interesting_managers, monitoring_radio)
 
         self.zmq_context.destroy()
@@ -327,20 +326,20 @@ class Interchange:
         """Process incoming task message(s).
         """
 
-        if self.task_incoming in self.socks and self.socks[self.task_incoming] == zmq.POLLIN:
+        if self.socks.get(self.task_incoming) == zmq.POLLIN:
             logger.debug("start task_incoming section")
             msg = self.task_incoming.recv_pyobj()
 
             # Process priority, higher number = lower priority
-            resource_spec = msg.get('resource_spec', {})
+            task_id = msg['task_id']
+            resource_spec = msg['context'].get('resource_spec', {})
             priority = resource_spec.get('priority', float('inf'))
-            queue_entry = (-priority, -self.task_counter, msg)
+            queue_entry = (-priority, -task_id, msg)
 
-            logger.debug("putting message onto pending_task_queue")
+            logger.debug("Putting task %s onto pending_task_queue", task_id)
 
             self.pending_task_queue.add(queue_entry)
-            self.task_counter += 1
-            logger.debug(f"Fetched {self.task_counter} tasks so far")
+            logger.debug("Put task %s onto pending_task_queue", task_id)
 
     def process_manager_socket_message(
         self,
@@ -360,9 +359,10 @@ class Interchange:
             mtype = meta['type']
         except Exception as e:
             logger.warning(
-                f'Failed to read manager message ([{type(e).__name__}] {e})'
+                'Failed to read manager message; ignoring message'
+                f' (Exception: [{type(e).__name__}] {e})'
             )
-            logger.debug('Message:\n   %r\n', msg_parts, exc_info=e)
+            logger.debug('Raw message bytes:\n   %r\n', msg_parts, exc_info=e)
             return
 
         logger.debug(
@@ -402,7 +402,9 @@ class Interchange:
             logger.info(f'Registration info for manager {manager_id!r}: {meta}')
             self._send_monitoring_info(monitoring_radio, new_rec)
 
-            if (mgr_minor_py, mgr_parsl_v) != (ix_minor_py, ix_parsl_v):
+            python_mismatch: bool = ix_minor_py != mgr_minor_py
+            parsl_mismatch: bool = ix_parsl_v != mgr_parsl_v
+            if parsl_mismatch or (self._check_python_mismatch and python_mismatch):
                 kill_event.set()
                 vm_exc = VersionMismatch(
                     f"py.v={ix_minor_py} parsl.v={ix_parsl_v}",
@@ -523,15 +525,24 @@ class Interchange:
                 m['active'] = False
                 self._send_monitoring_info(monitoring_radio, m)
 
+    def log_manager_counts(self, interesting_managers: Set[bytes]) -> None:
+        count_interesting = len(interesting_managers)
+        count_ready = len(self._ready_managers)
+
+        new_logged_manager_count_token = (count_interesting, count_ready)
+
+        if self._logged_manager_count_token != new_logged_manager_count_token:
+
+            logger.debug(
+                "Managers count (interesting/total): %d/%d",
+                count_interesting,
+                count_ready
+            )
+            self._logged_manager_count_token = new_logged_manager_count_token
+
     def process_tasks_to_send(self, interesting_managers: Set[bytes], monitoring_radio: Optional[MonitoringRadioSender]) -> None:
         # Check if there are tasks that could be sent to managers
 
-        logger.debug(
-            "Managers count (interesting/total): %d/%d",
-            len(interesting_managers),
-            len(self._ready_managers)
-        )
-
         if interesting_managers and self.pending_task_queue:
             shuffled_managers = self.manager_selector.sort_managers(self._ready_managers, interesting_managers)
 
@@ -624,6 +635,8 @@ def start_file_logger(filename: str, level: int = logging.DEBUG, format_string:
 
 
 if __name__ == "__main__":
+    from parsl.utils import setproctitle
+
     setproctitle("parsl: HTEX interchange")
 
     config = pickle.load(sys.stdin.buffer)

parsl/executors/high_throughput/mpi_executor.py

@@ -16,7 +16,6 @@ from parsl.executors.status_handling import BlockProviderExecutor
 from parsl.jobs.states import JobStatus
 from parsl.launchers import SimpleLauncher
 from parsl.monitoring.radios.base import RadioConfig
-from parsl.providers import LocalProvider
 from parsl.providers.base import ExecutionProvider
 
 
@@ -47,7 +46,7 @@ class MPIExecutor(HighThroughputExecutor):
     @typeguard.typechecked
     def __init__(self,
                  label: str = 'MPIExecutor',
-                 provider: ExecutionProvider = LocalProvider(),
+                 provider: Optional[ExecutionProvider] = None,
                  launch_cmd: Optional[str] = None,
                  interchange_launch_cmd: Optional[str] = None,
                  address: Optional[str] = None,

parsl/executors/high_throughput/mpi_resource_management.py

@@ -9,7 +9,6 @@ from enum import Enum
 from typing import Dict, List, Optional
 
 from parsl.multiprocessing import SpawnContext
-from parsl.serialize import pack_res_spec_apply_message, unpack_res_spec_apply_message
 
 logger = logging.getLogger(__name__)
 
@@ -146,11 +145,11 @@ class MPITaskScheduler(TaskScheduler):
         )
         acquired_nodes = []
         with self._free_node_counter.get_lock():
-            if num_nodes <= self._free_node_counter.value:  # type: ignore[attr-defined]
-                self._free_node_counter.value -= num_nodes  # type: ignore[attr-defined]
+            if num_nodes <= self._free_node_counter.value:
+                self._free_node_counter.value -= num_nodes
             else:
                 raise MPINodesUnavailable(
-                    requested=num_nodes, available=self._free_node_counter.value  # type: ignore[attr-defined]
+                    requested=num_nodes, available=self._free_node_counter.value
                 )
 
             for i in range(num_nodes):
@@ -163,17 +162,14 @@ class MPITaskScheduler(TaskScheduler):
         for node in nodes:
             self.nodes_q.put(node)
         with self._free_node_counter.get_lock():
-            self._free_node_counter.value += len(nodes)  # type: ignore[attr-defined]
+            self._free_node_counter.value += len(nodes)
 
     def put_task(self, task_package: dict):
         """Schedule task if resources are available otherwise backlog the task"""
-        user_ns = locals()
-        user_ns.update({"__builtins__": __builtins__})
-        _f, _args, _kwargs, resource_spec = unpack_res_spec_apply_message(task_package["buffer"])
+        resource_spec = task_package.get("context", {}).get("resource_spec", {})
 
-        nodes_needed = resource_spec.get("num_nodes")
-        tid = task_package["task_id"]
-        if nodes_needed:
+        if nodes_needed := resource_spec.get("num_nodes"):
+            tid = task_package["task_id"]
             try:
                 allocated_nodes = self._get_nodes(nodes_needed)
             except MPINodesUnavailable:
@@ -183,9 +179,6 @@ class MPITaskScheduler(TaskScheduler):
             else:
                 resource_spec["MPI_NODELIST"] = ",".join(allocated_nodes)
                 self._map_tasks_to_nodes[tid] = allocated_nodes
-                buffer = pack_res_spec_apply_message(_f, _args, _kwargs, resource_spec)
-                task_package["buffer"] = buffer
-                task_package["resource_spec"] = resource_spec
 
         self.pending_task_q.put(task_package)
 

parsl/executors/high_throughput/process_worker_pool.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 
 import argparse
+import importlib
 import logging
 import math
 import multiprocessing
@@ -17,7 +18,7 @@ from importlib.metadata import distributions
 from multiprocessing.context import SpawnProcess
 from multiprocessing.managers import DictProxy
 from multiprocessing.sharedctypes import Synchronized
-from typing import Dict, List, Optional, Sequence
+from typing import Callable, Dict, List, Optional, Sequence
 
 import psutil
 import zmq
@@ -348,7 +349,7 @@ class Manager:
 
             logger.debug(
                 'ready workers: %d, pending tasks: %d',
-                self.ready_worker_count.value,  # type: ignore[attr-defined]
+                self.ready_worker_count.value,
                 pending_task_count,
             )
 
@@ -373,10 +374,12 @@ class Manager:
             if socks.get(ix_sock) == zmq.POLLIN:
                 pkl_msg = ix_sock.recv()
                 tasks = pickle.loads(pkl_msg)
+                del pkl_msg
+
                 last_interchange_contact = time.time()
 
                 if tasks == HEARTBEAT_CODE:
-                    logger.debug("Got heartbeat from interchange")
+                    logger.debug("Got heartbeat response from interchange")
                 elif tasks == DRAINED_CODE:
                     logger.info("Got fully drained message from interchange - setting kill flag")
                     self._stop_event.set()
@@ -454,6 +457,7 @@ class Manager:
                                               'exception': serialize(RemoteExceptionWrapper(*sys.exc_info()))}
                             pkl_package = pickle.dumps(result_package)
                             self.pending_result_queue.put(pkl_package)
+                            del pkl_package
                     except KeyError:
                         logger.info("Worker {} was not busy when it died".format(worker_id))
 
@@ -603,6 +607,10 @@ def update_resource_spec_env_vars(mpi_launcher: str, resource_spec: Dict, node_i
 
 
 def _init_mpi_env(mpi_launcher: str, resource_spec: Dict):
+    for varname in resource_spec:
+        envname = "PARSL_" + str(varname).upper()
+        os.environ[envname] = str(resource_spec[varname])
+
     node_list = resource_spec.get("MPI_NODELIST")
     if node_list is None:
         return
@@ -753,8 +761,8 @@ def worker(
             worker_enqueued = True
 
         try:
-            # The worker will receive {'task_id':<tid>, 'buffer':<buf>}
             req = task_queue.get(timeout=task_queue_timeout)
+            # req is {'task_id':<tid>, 'buffer':<buf>, 'resource_spec':<dict>}
         except queue.Empty:
             continue
 
@@ -766,17 +774,33 @@ def worker(
             ready_worker_count.value -= 1
         worker_enqueued = False
 
-        _init_mpi_env(mpi_launcher=mpi_launcher, resource_spec=req["resource_spec"])
+        ctxt = req["context"]
+        res_spec = ctxt.get("resource_spec", {})
+
+        _init_mpi_env(mpi_launcher=mpi_launcher, resource_spec=res_spec)
+
+        exec_func: Callable = execute_task
+        exec_args = ()
+        exec_kwargs = {}
 
         try:
-            result = execute_task(req['buffer'])
+            if task_executor := ctxt.get("task_executor", None):
+                mod_name, _, fn_name = task_executor["f"].rpartition(".")
+                exec_mod = importlib.import_module(mod_name)
+                exec_func = getattr(exec_mod, fn_name)
+
+                exec_args = task_executor.get("a", ())
+                exec_kwargs = task_executor.get("k", {})
+
+            result = exec_func(req['buffer'], *exec_args, **exec_kwargs)
             serialized_result = serialize(result, buffer_threshold=1000000)
         except Exception as e:
             logger.info('Caught an exception: {}'.format(e))
             result_package = {'type': 'result', 'task_id': tid, 'exception': serialize(RemoteExceptionWrapper(*sys.exc_info()))}
         else:
             result_package = {'type': 'result', 'task_id': tid, 'result': serialized_result}
-            # logger.debug("Result: {}".format(result))
+            del serialized_result
+        del req
 
         logger.info("Completed executor task {}".format(tid))
         try:
@@ -788,6 +812,7 @@ def worker(
                                         })
 
         result_queue.put(pkl_package)
+        del pkl_package, result_package
         tasks_in_progress.pop(worker_id)
         logger.info("All processing finished for executor task {}".format(tid))
 

parsl/executors/high_throughput/zmq_pipes.py

@@ -74,51 +74,37 @@ class CommandClient:
 
         reply = '__PARSL_ZMQ_PIPES_MAGIC__'
         with self._lock:
-            for _ in range(max_retries):
-                try:
-                    logger.debug("Sending command client command")
-
-                    if timeout_s is not None:
-                        remaining_time_s = start_time_s + timeout_s - time.monotonic()
-                        poll_result = self.zmq_socket.poll(timeout=remaining_time_s * 1000, flags=zmq.POLLOUT)
-                        if poll_result == zmq.POLLOUT:
-                            pass  # this is OK, so continue
-                        elif poll_result == 0:
-                            raise CommandClientTimeoutError("Waiting for command channel to be ready for a command")
-                        else:
-                            raise InternalConsistencyError(f"ZMQ poll returned unexpected value: {poll_result}")
-
-                    self.zmq_socket.send_pyobj(message, copy=True)
-
-                    if timeout_s is not None:
-                        logger.debug("Polling for command client response or timeout")
-                        remaining_time_s = start_time_s + timeout_s - time.monotonic()
-                        poll_result = self.zmq_socket.poll(timeout=remaining_time_s * 1000, flags=zmq.POLLIN)
-                        if poll_result == zmq.POLLIN:
-                            pass  # this is OK, so continue
-                        elif poll_result == 0:
-                            logger.error("Command timed-out - command client is now bad forever")
-                            self.ok = False
-                            raise CommandClientTimeoutError("Waiting for a reply from command channel")
-                        else:
-                            raise InternalConsistencyError(f"ZMQ poll returned unexpected value: {poll_result}")
-
-                    logger.debug("Receiving command client response")
-                    reply = self.zmq_socket.recv_pyobj()
-                    logger.debug("Received command client response")
-                except zmq.ZMQError:
-                    logger.exception("Potential ZMQ REQ-REP deadlock caught")
-                    logger.info("Trying to reestablish context")
-                    self.zmq_context.recreate()
-                    self.create_socket_and_bind()
+            logger.debug("Sending command client command")
+
+            if timeout_s is not None:
+                remaining_time_s = start_time_s + timeout_s - time.monotonic()
+                poll_result = self.zmq_socket.poll(timeout=remaining_time_s * 1000, flags=zmq.POLLOUT)
+                if poll_result == zmq.POLLOUT:
+                    pass  # this is OK, so continue
+                elif poll_result == 0:
+                    raise CommandClientTimeoutError("Waiting for command channel to be ready for a command")
                 else:
-                    break
-
-        if reply == '__PARSL_ZMQ_PIPES_MAGIC__':
-            logger.error("Command channel run retries exhausted. Unable to run command")
-            raise Exception("Command Channel retries exhausted")
+                    raise InternalConsistencyError(f"ZMQ poll returned unexpected value: {poll_result}")
+
+            self.zmq_socket.send_pyobj(message, copy=True)
+
+            if timeout_s is not None:
+                logger.debug("Polling for command client response or timeout")
+                remaining_time_s = start_time_s + timeout_s - time.monotonic()
+                poll_result = self.zmq_socket.poll(timeout=remaining_time_s * 1000, flags=zmq.POLLIN)
+                if poll_result == zmq.POLLIN:
+                    pass  # this is OK, so continue
+                elif poll_result == 0:
+                    logger.error("Command timed-out - command client is now bad forever")
+                    self.ok = False
+                    raise CommandClientTimeoutError("Waiting for a reply from command channel")
+                else:
+                    raise InternalConsistencyError(f"ZMQ poll returned unexpected value: {poll_result}")
 
-        return reply
+            logger.debug("Receiving command client response")
+            reply = self.zmq_socket.recv_pyobj()
+            logger.debug("Received command client response")
+            return reply
 
     def close(self):
         self.zmq_socket.close()
@@ -150,30 +136,18 @@ class TasksOutgoing:
         self.port = self.zmq_socket.bind_to_random_port(tcp_url(ip_address),
                                                         min_port=port_range[0],
                                                         max_port=port_range[1])
-        self.poller = zmq.Poller()
-        self.poller.register(self.zmq_socket, zmq.POLLOUT)
 
     def put(self, message):
         """ This function needs to be fast at the same time aware of the possibility of
         ZMQ pipes overflowing.
 
-        The timeout increases slowly if contention is detected on ZMQ pipes.
         We could set copy=False and get slightly better latency but this results
         in ZMQ sockets reaching a broken state once there are ~10k tasks in flight.
         This issue can be magnified if each the serialized buffer itself is larger.
         """
-        timeout_ms = 1
-        while True:
-            socks = dict(self.poller.poll(timeout=timeout_ms))
-            if self.zmq_socket in socks and socks[self.zmq_socket] == zmq.POLLOUT:
-                # The copy option adds latency but reduces the risk of ZMQ overflow
-                logger.debug("Sending TasksOutgoing message")
-                self.zmq_socket.send_pyobj(message, copy=True)
-                logger.debug("Sent TasksOutgoing message")
-                return
-            else:
-                timeout_ms *= 2
-                logger.debug("Not sending due to non-ready zmq pipe, timeout: {} ms".format(timeout_ms))
+        logger.debug("Sending TasksOutgoing message")
+        self.zmq_socket.send_pyobj(message)
+        logger.debug("Sent TasksOutgoing message")
 
     def close(self):
         self.zmq_socket.close()
@@ -206,20 +180,15 @@ class ResultsIncoming:
         self.port = self.results_receiver.bind_to_random_port(tcp_url(ip_address),
                                                               min_port=port_range[0],
                                                               max_port=port_range[1])
-        self.poller = zmq.Poller()
-        self.poller.register(self.results_receiver, zmq.POLLIN)
 
     def get(self, timeout_ms=None):
         """Get a message from the queue, returning None if timeout expires
         without a message. timeout is measured in milliseconds.
         """
-        socks = dict(self.poller.poll(timeout=timeout_ms))
-        if self.results_receiver in socks and socks[self.results_receiver] == zmq.POLLIN:
-            m = self.results_receiver.recv_multipart()
-            logger.debug("Received ResultsIncoming message")
-            return m
-        else:
-            return None
+        if zmq.POLLIN == self.results_receiver.poll(timeout_ms, zmq.POLLIN):
+            logger.debug("Receiving ResultsIncoming multipart message")
+            return self.results_receiver.recv_multipart()
+        return None
 
     def close(self):
         self.results_receiver.close()

parsl/executors/radical/executor.py

@@ -20,7 +20,7 @@ from parsl.app.errors import BashExitFailure, RemoteExceptionWrapper
 from parsl.app.python import timeout
 from parsl.data_provider.files import File
 from parsl.executors.base import ParslExecutor
-from parsl.serialize import deserialize, pack_res_spec_apply_message
+from parsl.serialize import deserialize, pack_apply_message
 from parsl.serialize.errors import DeserializationError, SerializationError
 from parsl.utils import RepresentationMixin
 
@@ -441,11 +441,7 @@ class RadicalPilotExecutor(ParslExecutor, RepresentationMixin):
 
     def _pack_and_apply_message(self, func, args, kwargs):
         try:
-            buffer = pack_res_spec_apply_message(func,
-                                                 args,
-                                                 kwargs,
-                                                 resource_specification={},
-                                                 buffer_threshold=1024 * 1024)
+            buffer = pack_apply_message(func, args, kwargs, buffer_threshold=1 << 20)
             task_func = rp.utils.serialize_bson(buffer)
         except TypeError:
             raise SerializationError(func.__name__)

parsl/executors/radical/rpex_worker.py

@@ -5,7 +5,7 @@ import radical.pilot as rp
 import parsl.app.errors as pe
 from parsl.app.bash import remote_side_bash_executor
 from parsl.executors.execute_task import execute_task
-from parsl.serialize import serialize, unpack_res_spec_apply_message
+from parsl.serialize import serialize, unpack_apply_message
 
 
 class ParslWorker:
@@ -33,7 +33,7 @@ class ParslWorker:
 
         try:
             buffer = rp.utils.deserialize_bson(task['description']['executable'])
-            func, args, kwargs, _resource_spec = unpack_res_spec_apply_message(buffer)
+            func, args, kwargs = unpack_apply_message(buffer)
             ret = remote_side_bash_executor(func, *args, **kwargs)
             exc = (None, None)
             val = None

parsl/executors/taskvine/executor.py

@@ -107,13 +107,17 @@ class TaskVineExecutor(BlockProviderExecutor, putils.RepresentationMixin):
                  function_exec_mode: Union[Literal['regular'], Literal['serverless']] = 'regular',
                  manager_config: TaskVineManagerConfig = TaskVineManagerConfig(),
                  factory_config: TaskVineFactoryConfig = TaskVineFactoryConfig(),
-                 provider: Optional[ExecutionProvider] = LocalProvider(init_blocks=1),
+                 provider: Optional[ExecutionProvider] = None,
                  storage_access: Optional[List[Staging]] = None,
                  remote_monitoring_radio: Optional[RadioConfig] = None):
 
         # Set worker launch option for this executor
         if worker_launch_method == 'factory' or worker_launch_method == 'manual':
             provider = None
+        elif worker_launch_method == 'provider' and provider is None:
+            # provider method chosen, but no explicit provider supplied to __init__
+            # so default to LocalProvider
+            provider = LocalProvider(init_blocks=1)
 
         # Initialize the parent class with the execution provider and block error handling enabled.
         # If provider is None, then no worker is launched via the provider method.

parsl/executors/threads.py

@@ -29,12 +29,15 @@ class ThreadPoolExecutor(ParslExecutor, RepresentationMixin):
 
     @typeguard.typechecked
     def __init__(self, label: str = 'threads', max_threads: Optional[int] = 2,
-                 thread_name_prefix: str = '', storage_access: Optional[List[Staging]] = None,
+                 thread_name_prefix: str | None = None, storage_access: Optional[List[Staging]] = None,
                  working_dir: Optional[str] = None, remote_monitoring_radio: Optional[RadioConfig] = None):
         ParslExecutor.__init__(self)
         self.label = label
         self.max_threads = max_threads
-        self.thread_name_prefix = thread_name_prefix
+        if thread_name_prefix is None:
+            self.thread_name_prefix = "ThreadPoolExecutor-" + label
+        else:
+            self.thread_name_prefix = thread_name_prefix
 
         # we allow storage_access to be None now, which means something else to [] now
         # None now means that a default storage access list will be used, while

parsl/jobs/states.py

@@ -10,7 +10,7 @@ class JobState(IntEnum):
     """Defines a set of states that a job can be in"""
 
     UNKNOWN = 0
-    """The batch provider is unable to determinate a state for this job"""
+    """The batch provider is unable to determine a state for this job"""
 
     PENDING = 1
     """"This job is in the batch queue but has not started running"""
@@ -40,7 +40,7 @@ class JobState(IntEnum):
     """This job is held/suspended in the batch system"""
 
     MISSING = 8
-    """This job has reached a terminal state without the resources(managers/workers)
+    """This job has reached a terminal state without the resources (managers/workers)
     launched in the job connecting back to the Executor. This state is set by HTEX
     when it is able to infer that the block failed to start workers for eg due to
     bad worker environment or network connectivity issues.