[jira] [Commented] (BEAM-186) The streaming WindowedWordCount example is broken on Dataflow

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/BEAM-186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15233186#comment-15233186 ] 

ASF GitHub Bot commented on BEAM-186:
-------------------------------------

GitHub user herohde opened a pull request:

    https://github.com/apache/incubator-beam/pull/156

    [BEAM-186] Fix pubsub injector for streaming examples

    


You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/herohde/incubator-beam master

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/incubator-beam/pull/156.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #156
    
----
commit 2ea17a4f1959838ef955abbe27b4b914f4169006
Author: Henning Rohde <he...@google.com>
Date:   2016-04-09T00:06:18Z

    [BEAM-186] Fix pubsub injector for streaming examples

----


> The streaming WindowedWordCount example is broken on Dataflow
> -------------------------------------------------------------
>
>                 Key: BEAM-186
>                 URL: https://issues.apache.org/jira/browse/BEAM-186
>             Project: Beam
>          Issue Type: Bug
>          Components: examples-java
>            Reporter: Henning Korsholm Rohde
>            Assignee: Henning Korsholm Rohde
>            Priority: Minor
>
> The streaming version of WindowedWordCount example is broken on Dataflow. Currently, the injector cannot start up and emits the error "java.lang.RuntimeException: DataflowWorkerHarness should not be used for non-streaming pipelines". This in turn means that no messages are generated for the streaming pipeline to process.
> The problem is that DataflowExampleUtils uses a batch workflow to inject Google PubSub messages. This workflow reuses the real options and modifies them to be batch, setting isStreaming to false etc. However, with versioned containers, the option for workerHarnessContainerImage must now also be changed to use the batch image.



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