[
https://issues.apache.org/jira/browse/AIRFLOW-6529?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17095556#comment-17095556
]
ASF GitHub Bot commented on AIRFLOW-6529:
-----------------------------------------
jhtimmins commented on pull request #7128:
URL: https://github.com/apache/airflow/pull/7128#issuecomment-621294026
@sarutak @potiuk @ashb do any of you know what the status of this is? It
looks like all requested changes have been made. Won't solve all problems
related to adding Python 3.8 support, but gets at least part of the way there.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Serialization error occurs when the scheduler tries to run on macOS.
> --------------------------------------------------------------------
>
> Key: AIRFLOW-6529
> URL: https://issues.apache.org/jira/browse/AIRFLOW-6529
> Project: Apache Airflow
> Issue Type: Bug
> Components: scheduler
> Affects Versions: 1.10.8
> Environment: macOS
> Python 3.8
> multiprocessing with spawn mode
> Reporter: Kousuke Saruta
> Assignee: Kousuke Saruta
> Priority: Major
>
> When we try to run the scheduler on macOS, we will get a serialization error
> like as follows.
> {code}
> ____________ _____________
> ____ |__( )_________ __/__ /________ __
> ____ /| |_ /__ ___/_ /_ __ /_ __ \_ | /| / /
> ___ ___ | / _ / _ __/ _ / / /_/ /_ |/ |/ /
> _/_/ |_/_/ /_/ /_/ /_/ \____/____/|__/
> [2020-01-10 19:54:41,974] {executor_loader.py:59} INFO - Using executor
> SequentialExecutor
> [2020-01-10 19:54:41,983] {scheduler_job.py:1462} INFO - Starting the
> scheduler
> [2020-01-10 19:54:41,984] {scheduler_job.py:1469} INFO - Processing each file
> at most -1 times
> [2020-01-10 19:54:41,984] {scheduler_job.py:1472} INFO - Searching for files
> in /Users/sarutak/airflow/dags
> [2020-01-10 19:54:42,025] {scheduler_job.py:1474} INFO - There are 27 files
> in /Users/sarutak/airflow/dags
> [2020-01-10 19:54:42,025] {scheduler_job.py:1527} INFO - Resetting orphaned
> tasks for active dag runs
> [2020-01-10 19:54:42,059] {scheduler_job.py:1500} ERROR - Exception when
> executing execute_helper
> Traceback (most recent call last):
> File
> "/Users/sarutak/work/oss/airflow-env/master-python3.8.1/lib/python3.8/site-packages/airflow/jobs/scheduler_job.py",
> line 1498, in _execute
> self._execute_helper()
> File
> "/Users/sarutak/work/oss/airflow-env/master-python3.8.1/lib/python3.8/site-packages/airflow/jobs/scheduler_job.py",
> line 1531, in _execute_helper
> self.processor_agent.start()
> File
> "/Users/sarutak/work/oss/airflow-env/master-python3.8.1/lib/python3.8/site-packages/airflow/utils/dag_processing.py",
> line 348, in start
> self._process.start()
> File "/opt/python/3.8.1/lib/python3.8/multiprocessing/process.py", line
> 121, in start
> self._popen = self._Popen(self)
> File "/opt/python/3.8.1/lib/python3.8/multiprocessing/context.py", line
> 224, in _Popen
> return _default_context.get_context().Process._Popen(process_obj)
> File "/opt/python/3.8.1/lib/python3.8/multiprocessing/context.py", line
> 283, in _Popen
> return Popen(process_obj)
> File
> "/opt/python/3.8.1/lib/python3.8/multiprocessing/popen_spawn_posix.py", line
> 32, in __init__
> super().__init__(process_obj)
> File "/opt/python/3.8.1/lib/python3.8/multiprocessing/popen_fork.py", line
> 19, in __init__
> self._launch(process_obj)
> File
> "/opt/python/3.8.1/lib/python3.8/multiprocessing/popen_spawn_posix.py", line
> 47, in _launch
> reduction.dump(process_obj, fp)
> File "/opt/python/3.8.1/lib/python3.8/multiprocessing/reduction.py", line
> 60, in dump
> ForkingPickler(file, protocol).dump(obj)
> AttributeError: Can't pickle local object
> 'SchedulerJob._execute.<locals>.processor_factory'
> {code}
> The reason is scheduler try to run subprocesses using multiprocessing with
> spawn mode and the mode tries to pickle objects. In this case,
> `processor_factory` inner method is tried to be pickled.
> Actually, as of Python 3.8, spawn mode is the default mode in macOS.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
