You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@beam.apache.org by "Brian Hulette (Jira)" <ji...@apache.org> on 2022/05/04 22:35:00 UTC
[jira] [Commented] (BEAM-14410) FnRunnerTest with non-trivial (order 1000 elements) input flakes in non-cython environment
[ https://issues.apache.org/jira/browse/BEAM-14410?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17531973#comment-17531973 ]
Brian Hulette commented on BEAM-14410:
--------------------------------------
It looks like using np.int64 instances as elements is also part of the problem. When I run the test with just list(range(5000)) I wasn't able to reproduce the flake.
> FnRunnerTest with non-trivial (order 1000 elements) input flakes in non-cython environment
> ------------------------------------------------------------------------------------------
>
> Key: BEAM-14410
> URL: https://issues.apache.org/jira/browse/BEAM-14410
> Project: Beam
> Issue Type: Bug
> Components: sdk-py-core
> Reporter: Brian Hulette
> Priority: P2
> Labels: flake
> Time Spent: 20m
> Remaining Estimate: 0h
>
> See [https://github.com/apache/beam/pull/17384#issuecomment-1117418018]
>
> {code:java}
> apache_beam/runners/portability/fn_api_runner/fn_runner_test.py:141:
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
> apache_beam/pipeline.py:596: in __exit__
> self.result = self.run()
> apache_beam/pipeline.py:573: in run
> return self.runner.run_pipeline(self, self._options)
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:200: in run_pipeline
> pipeline.to_runner_api(default_environment=self._default_environment))
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:208: in run_via_runner_api
> return self.run_stages(stage_context, stages)
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:409: in run_stages
> runner_execution_context, bundle_context_manager, bundle_input)
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:742: in _execute_bundle
> bundle_manager))
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:966: in _run_bundle
> data_input, data_output, input_timers, expected_timer_output)
> apache_beam/runners/portability/fn_api_runner/fn_runner.py:1309: in process_bundle
> output.transform_id).append(output.data)
> apache_beam/runners/portability/fn_api_runner/execution.py:242: in append
> windowed_key_value = coder_impl.decode_from_stream(input_stream, True)
> apache_beam/coders/coder_impl.py:1446: in decode_from_stream
> value = self._value_coder.decode_from_stream(in_stream, nested)
> apache_beam/coders/coder_impl.py:994: in decode_from_stream
> c in enumerate(self._coder_impls)
> apache_beam/coders/coder_impl.py:994: in <listcomp>
> c in enumerate(self._coder_impls)
> apache_beam/coders/coder_impl.py:617: in decode_from_stream
> return in_stream.read_all(nested)
> apache_beam/coders/slow_stream.py:137: in read_all
> return self.read(self.read_var_int64() if nested else self.size())
> apache_beam/coders/slow_stream.py:148: in read_var_int64
> byte = self.read_byte()
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ self = <apache_beam.coders.slow_stream.InputStream object at 0x7f8433d63190> def read_byte(self):
> # type: () -> int
> self.pos += 1
> > return self.data[self.pos - 1]
> E IndexError: index out of rangeapache_beam/coders/slow_stream.py:142: IndexError{code}
--
This message was sent by Atlassian Jira
(v8.20.7#820007)