[
https://issues.apache.org/jira/browse/NIFI-647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14573149#comment-14573149
]
Aldrin Piri commented on NIFI-647:
----------------------------------
Code looks good and confirmed expected behavior of how FORKs are handled. +1
> Duplicating provenance reporting via session creates unexpected behavior
> ------------------------------------------------------------------------
>
> Key: NIFI-647
> URL: https://issues.apache.org/jira/browse/NIFI-647
> Project: Apache NiFi
> Issue Type: Bug
> Components: Core Framework
> Affects Versions: 0.0.2, 0.1.0
> Reporter: Aldrin Piri
> Assignee: Mark Payne
> Fix For: 0.2.0
>
> Attachments:
> 0001-NIFI-647-When-FORK-Event-emitted-by-processor-Framew.patch
>
>
> This is related to NIFI-643. While that processor bug was resolved, there are
> issues with how this is managed in the framework.
> The unexpected behavior is created via the following:
> https://github.com/apache/incubator-nifi/blob/d29a2d688e437bae42c12115768cdb038b7406c5/nifi/nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/SplitJson.java#L196
> Summarily, a series of files (splits) are created from the parent file which
> generates the associated FORK provenance events. The separate call to the
> provenance reporter at Line 196 also requests a FORK for each of the
> generated files. This manifests itself as being CREATE events for each of
> the FORKed files and causes an error:
> {code}
> Found cycle in graph. This indicates that multiple events were registered
> claiming to have generated the same FlowFile (UUID =
> 21b663c2-2406-49cb-b740-76dd64f54be8)
> {code}
> Ideally, such behavior would also be detected by the test framework.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
