You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@beam.apache.org by "Luke Cwik (JIRA)" <ji...@apache.org> on 2016/03/22 22:09:25 UTC

[jira] [Commented] (BEAM-142) BigQueryIO: don't unnecessarily initialize an ExecutorService to validate parameters

    [ https://issues.apache.org/jira/browse/BEAM-142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15207317#comment-15207317 ] 

Luke Cwik commented on BEAM-142:
--------------------------------

If you use the executor service from PipelineOptions, it handles the case where pipeline construction is occurring on AppEngine
https://github.com/apache/incubator-beam/blob/master/sdk/src/main/java/com/google/cloud/dataflow/sdk/options/GcsOptions.java#L60

> BigQueryIO: don't unnecessarily initialize an ExecutorService to validate parameters
> ------------------------------------------------------------------------------------
>
>                 Key: BEAM-142
>                 URL: https://issues.apache.org/jira/browse/BEAM-142
>             Project: Beam
>          Issue Type: Bug
>          Components: sdk-java-gcp
>            Reporter: Davor Bonaci
>            Priority: Minor
>
> See http://stackoverflow.com/questions/36142464/kicking-off-dataflow-jobs-with-app-engine-errors-with-a-securityexception-on-add
> From BigQueryIO$Write$Bound.verifyTableEmpty, we call BigQueryTableInserter, which unnecessarily sets up ExecutorService for this case. As a result, the pipeline cannot be run in the App Engine environment.
> We should fix this by avoid initialization of the ExecutorService.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)