You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@nifi.apache.org by "Mark Payne (Jira)" <ji...@apache.org> on 2020/11/17 15:42:00 UTC

[jira] [Created] (NIFI-8018) Introduce a Flow Sequence ID to both flow.xml.gz and Provenance events

Mark Payne created NIFI-8018:
--------------------------------

             Summary: Introduce a Flow Sequence ID to both flow.xml.gz and Provenance events
                 Key: NIFI-8018
                 URL: https://issues.apache.org/jira/browse/NIFI-8018
             Project: Apache NiFi
          Issue Type: Improvement
          Components: Core Framework
            Reporter: Mark Payne


By default, every time a user makes a change to a NiFi dataflow, we write a backup of the flow to the archive directory first. We should introduce a new 'flowSequenceId' that is a one-up counter every time that a flow is updated. This should be stored in the flow.xml.gz. That will cause it to also be present in the archived flows.

Additionally, ensure that when a flow is inherited from cluster, that we update the internal state to match the sequence id from the cluster's flow.

Then, when generating provenance events, we should include the flow sequence id.

By doing this, it means that whenever a user views a particular Provenance Event, they will have the unique Sequence ID of the flow. That flow can then be retrieved from the archive in order to ensure that any unexpected behavior that is shown in the Provenance Events can be correlated to the flow as it was at the time, rather than looking at the flow as it exists currently.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)