You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@beam.apache.org by "Kenneth Knowles (Jira)" <ji...@apache.org> on 2022/01/24 20:26:00 UTC
[jira] [Updated] (BEAM-13721) ZeroDivisionError if source bundle smaller than 1mb
[ https://issues.apache.org/jira/browse/BEAM-13721?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kenneth Knowles updated BEAM-13721:
-----------------------------------
Description:
Hi,
I built a (GCP) DataFlow + apache-beam (version 2.24.0) pipeline, using python.
The pipeline's stages are:
1. reading from BigQuery
2. running a custom function (using the ParDo while passing it a class inheriting 'beam.DoFn'), that transforms the read data.
3. writing the transformed data to BigQuery
The pipeline works fine when stage 1 is querying a small amount of data, but when it is querying data from the last six months (lots of data), I am getting this error:
{code:python}
apache_beam.runners.dataflow.dataflow_runner.DataflowRuntimeException: Dataflow pipeline failed. State: FAILED, Error:
Traceback (most recent call last):
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/batchworker.py", line 649, in do_work
work_executor.execute()
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/executor.py", line 179, in execute
op.start()
File "dataflow_worker/native_operations.py", line 38, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 39, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 44, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 48, in dataflow_worker.native_operations.NativeReadOperation.start
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/workercustomsources.py", line 69, in _{_}iter{_}_
self._source.start_position, self._source.stop_position)
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 78, in get_range_tracker
start_position, stop_position, self._source_bundles)
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 131, in _{_}init{_}_
self._compute_cumulative_weights(source_bundles[start[0]:last]) + [1] *
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 154, in _compute_cumulative_weights
running_total.append(max(min_diff, min(1, running_total[-1] + w / total)))
ZeroDivisionError: float division by zero
{code}
I saw this issue in your repository:
https://issues.apache.org/jira/browse/BEAM-10004
which is referencing the same problem I have, but even though this fix is already implemented in my version of apache-beam (2.24.0), I still get this error.
Can you please guide me in fixing this issue?
Is it something I am doing wrong or is this your bug?
Thank you in advance and have a good one!
Inbar Dekel
was:
Hi,
I built a (GCP) DataFlow + apache-beam (version 2.24.0) pipeline, using python.
The pipeline's stages are:
1. reading from BigQuery
2. running a custom function (using the ParDo while passing it a class inheriting 'beam.DoFn'), that transforms the read data.
3. writing the transformed data to BigQuery
The pipeline works fine when stage 1 is querying a small amount of data, but when it is querying data from the last six months (lots of data), I am getting this error:
apache_beam.runners.dataflow.dataflow_runner.DataflowRuntimeException: Dataflow pipeline failed. State: FAILED, Error:
Traceback (most recent call last):
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/batchworker.py", line 649, in do_work
work_executor.execute()
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/executor.py", line 179, in execute
op.start()
File "dataflow_worker/native_operations.py", line 38, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 39, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 44, in dataflow_worker.native_operations.NativeReadOperation.start
File "dataflow_worker/native_operations.py", line 48, in dataflow_worker.native_operations.NativeReadOperation.start
File "/usr/local/lib/python3.7/site-packages/dataflow_worker/workercustomsources.py", line 69, in __iter__
self._source.start_position, self._source.stop_position)
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 78, in get_range_tracker
start_position, stop_position, self._source_bundles)
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 131, in __init__
self._compute_cumulative_weights(source_bundles[start[0]:last]) + [1] *
File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 154, in _compute_cumulative_weights
running_total.append(max(min_diff, min(1, running_total[-1] + w / total)))
ZeroDivisionError: float division by zero
I saw this issue in your repository:
https://issues.apache.org/jira/browse/BEAM-10004
which is referencing the same problem I have, but even though this fix is already implemented in my version of apache-beam (2.24.0), I still get this error.
Can you please guide me in fixing this issue?
Is it something I am doing wrong or is this your bug?
Thank you in advance and have a good one!
Inbar Dekel
> ZeroDivisionError if source bundle smaller than 1mb
> ---------------------------------------------------
>
> Key: BEAM-13721
> URL: https://issues.apache.org/jira/browse/BEAM-13721
> Project: Beam
> Issue Type: Bug
> Components: io-py-gcp
> Affects Versions: 2.24.0
> Reporter: Inbar Dekel
> Priority: P2
>
> Hi,
> I built a (GCP) DataFlow + apache-beam (version 2.24.0) pipeline, using python.
> The pipeline's stages are:
> 1. reading from BigQuery
> 2. running a custom function (using the ParDo while passing it a class inheriting 'beam.DoFn'), that transforms the read data.
> 3. writing the transformed data to BigQuery
>
> The pipeline works fine when stage 1 is querying a small amount of data, but when it is querying data from the last six months (lots of data), I am getting this error:
>
> {code:python}
> apache_beam.runners.dataflow.dataflow_runner.DataflowRuntimeException: Dataflow pipeline failed. State: FAILED, Error:
> Traceback (most recent call last):
> File "/usr/local/lib/python3.7/site-packages/dataflow_worker/batchworker.py", line 649, in do_work
> work_executor.execute()
> File "/usr/local/lib/python3.7/site-packages/dataflow_worker/executor.py", line 179, in execute
> op.start()
> File "dataflow_worker/native_operations.py", line 38, in dataflow_worker.native_operations.NativeReadOperation.start
> File "dataflow_worker/native_operations.py", line 39, in dataflow_worker.native_operations.NativeReadOperation.start
> File "dataflow_worker/native_operations.py", line 44, in dataflow_worker.native_operations.NativeReadOperation.start
> File "dataflow_worker/native_operations.py", line 48, in dataflow_worker.native_operations.NativeReadOperation.start
> File "/usr/local/lib/python3.7/site-packages/dataflow_worker/workercustomsources.py", line 69, in _{_}iter{_}_
> self._source.start_position, self._source.stop_position)
> File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 78, in get_range_tracker
> start_position, stop_position, self._source_bundles)
> File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 131, in _{_}init{_}_
> self._compute_cumulative_weights(source_bundles[start[0]:last]) + [1] *
> File "/usr/local/lib/python3.7/site-packages/apache_beam/io/concat_source.py", line 154, in _compute_cumulative_weights
> running_total.append(max(min_diff, min(1, running_total[-1] + w / total)))
> ZeroDivisionError: float division by zero
> {code}
>
> I saw this issue in your repository:
>
> https://issues.apache.org/jira/browse/BEAM-10004
>
> which is referencing the same problem I have, but even though this fix is already implemented in my version of apache-beam (2.24.0), I still get this error.
>
> Can you please guide me in fixing this issue?
> Is it something I am doing wrong or is this your bug?
>
> Thank you in advance and have a good one!
>
> Inbar Dekel
--
This message was sent by Atlassian Jira
(v8.20.1#820001)