-
-
Notifications
You must be signed in to change notification settings - Fork 719
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Python 3.10 #5952
Python 3.10 #5952
Conversation
Unit Test Results 18 files + 18 18 suites +18 9h 25m 41s ⏱️ + 9h 25m 41s For more details on these failures, see this check. Results for commit b0308ac. ± Comparison against base commit b872d45. ♻️ This comment has been updated with latest results. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for working on this Thomas! 😄
Had a few questions below 🙂
cb7548b
to
b3a5b38
Compare
2423e95
to
090f5ee
Compare
In 3.10, the _internal_ `_loop` attribute of several asyncio classes has been removed.
by interacting with the ac.condition we bind it to the loop on py3.10
ERROR asyncio:base_events.py:1729 Exception in callback BaseAsyncIOLoop._handle_events(18, 1) handle: <Handle BaseAsyncIOLoop._handle_events(18, 1)> Traceback (most recent call last): File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/asyncio/events.py", line 80, in _run self._context.run(self._callback, *self._args) File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/site-packages/tornado/platform/asyncio.py", line 189, in _handle_events handler_func(fileobj, events) File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/site-packages/tornado/netutil.py", line 276, in accept_handler callback(connection, address) File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/site-packages/tornado/tcpserver.py", line 288, in _handle_connection connection = ssl_wrap_socket( File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/site-packages/tornado/netutil.py", line 615, in ssl_wrap_socket return context.wrap_socket(socket, server_hostname=server_hostname, **kwargs) File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/ssl.py", line 512, in wrap_socket return self.sslsocket_class._create( File "/home/graingert/miniconda3/envs/dask-distributed-310/lib/python3.10/ssl.py", line 1061, in _create self._sslobj = self._context._wrap_socket( ssl.SSLError: Cannot create a server socket with a PROTOCOL_TLS_CLIENT context (_ssl.c:794)
…her_semaphore_loop skip reason
Seeing 2 test failures. One with The former seems to be due to a missing file and the latter due to contention around a file. Not seeing other issues around these tests on the issue tracker (so don't know if they are known to be flaky). Though wouldn't be surprised if these are due to sporadic test failures. Did see PR ( #5994 ) modernizing Idk if these should be blockers, but ok going ahead without them (if others agree) to get Python 3.10 included. |
test_restart_sync_no_center can be nuked. That's happening at #5994 . Taking a look at the test_restart report now. |
For test_restart I think it's not related to this PR. That being said, I think that folks in this PR might be able to help diagnose what's going on. I think that that failing test (and others) might boil down to this question: #6001 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Some observations:
- Pytorch is no longer tested on Python 3.10.
- The default SSL Context purpose has changed from
SERVER_AUTH
toCLIENT_AUTH
- Do the docs build OK?
Yeah agree it is probably unrelated.
Maybe just wanted to avoid holding up Python 3.10 addition over this test failure, but if others want to debug and fix it that is fine as well |
I don't think that this PR should block on that test
…On Fri, Mar 25, 2022 at 11:29 AM jakirkham ***@***.***> wrote:
For test_restart I think it's not related to this PR.
Yeah agree it is probably unrelated.
That being said, I think that folks in this PR might be able to help
diagnose what's going on. I think that that failing test (and others) might
boil down to this question: #6001
<#6001>
Maybe just wanted to avoid holding up Python 3.10 addition over this test
failure, but if others want to debug and fix it that is fine as well
—
Reply to this email directly, view it on GitHub
<#5952 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACKZTBY5UE5NUDM4PI6BLDVBXSXFANCNFSM5Q4TV4TQ>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
@fjetter did you have more thoughts here or is this good from your perspective? 🙂 |
The environment.yaml still includes too many dependencies. All the "only tested here" dependencies should be removed since they are tested in the py3.9 build |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
remove "# Only tested here" deps
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm also wondering why we are testing
- paramiko
- scikit-learn/scipy
- prometheus
- netcdf4
- h5py
- ...
in this build but we can clean up these build files in a follow up PR.
Co-authored-by: Florian Jetter <fjetter@users.noreply.github.com>
combines #5778 and #5353
pre-commit run --all-files