We have not seen the issue with Python 3.6 on 2.16+ after applying this
patch. 🎉
Thanks!
On Thu, Nov 21, 2019 at 4:41 PM Thomas Weise wrote:
> We are currently verifying the patch. Will report back tomorrow.
>
> On Thu, Nov 21, 2019 at 8:40 AM Valentyn Tymofieiev
> wrote:
>
>> That would be he
We are currently verifying the patch. Will report back tomorrow.
On Thu, Nov 21, 2019 at 8:40 AM Valentyn Tymofieiev
wrote:
> That would be helpful, thanks a lot! It should be a straightforward patch.
> Also, thanks Guenther, for sharing your investigation on
> https://bugs.python.org/issue34572
That would be helpful, thanks a lot! It should be a straightforward patch.
Also, thanks Guenther, for sharing your investigation on
https://bugs.python.org/issue34572, it was very helpful.
On Thu, Nov 21, 2019 at 8:25 AM Thomas Weise wrote:
> Valentyn, thanks a lot for following up on this.
>
>
Valentyn, thanks a lot for following up on this.
If the change can be cherry picked in isolation, we should be able to
verify this soon (with 2.16).
On Thu, Nov 21, 2019 at 8:12 AM Valentyn Tymofieiev
wrote:
> To close the loop here: To my knowledge this issue affects all Python 3
> users of P
To close the loop here: To my knowledge this issue affects all Python 3
users of Portable Flink/Spark runners, and Dataflow Python Streaming users,
including users on Python 3.7.3 and newer versions.
The issue is addressed on Beam master, and we have a cherry-pick out for
Beam 2.17.0.
Workaround
I also opened https://issues.apache.org/jira/browse/BEAM-8651 to track this
issue and any recommendation for the users that will come out of it.
On Thu, Nov 7, 2019 at 6:25 PM Valentyn Tymofieiev
wrote:
> I think we have heard of this issue from the same source:
>
> This looks exactly like a ra
I think we have heard of this issue from the same source:
This looks exactly like a race condition that we've encountered on Python
> 3.7.1: There's a bug in some older 3.7.x releases that breaks the
> thread-safety of the unpickler, as concurrent unpickle threads can access a
> module before it
Thanks Valentyn,
Aggregation_transform.py doesn't have any transformation method which
extends beam.DoFn. We are using plain python method which we passed in
beam.Map(). I am not sure how to get the dump of serialized_fn. Can you
please let me the process?
I also heard that some people ran into
+user@, bcc: dev@
https://issues.apache.org/jira/browse/BEAM-6158 may be contributing to this
issue, although we saw instances of this bug in exactly opposite scenarios
- when pipeline was defined *in one file*, but not in multiple files.
Could you try replacing instances of super() in aggregation
Hi All,
We have noticed a weird intermittent issue on Python3 but we don't run into
this issue on python2. Sometimes when we are trying to submit the pipeline,
we get AttributeError (Check the stack trace below). we have
double-checked and we do find the attribute/methods are present in the
right
10 matches
Mail list logo
