Member since
03-26-2023
3
Posts
0
Kudos Received
0
Solutions
03-27-2023
10:40 AM
1 Kudo
@ManishR NiFi offers many components (processors, controller services, reporting tasks, etc) that can be used to construct a flow based program on the NiFi canvas (Referred to as a NiFi dataflow). While this list of default available components may be different depending on the release of NiFi being used, NiFi has embedded documentation found under help within the NiFi UI that shows all components available in that installed release. Apache NiFi also publishes the same info for the most current released version here: https://nifi.apache.org/docs/nifi-docs/ Selecting a component from the documentation with open a description of the component and all list configurable properties. Building a dataflow on the NiFi canvas consist of dragging and dropping new component processors to the canvas. You can then drag connection between these components to construct your end-to-end dataflow. There are 100s of component processors available out of the box and even more that you can download and add to your NiFi from the apache community. Once a dataflow is built and configured, starting those components would result in the creation of FlowFile (for testing, you can add a GenerateFlowFile processor that generates a FlowFile rather then ingesting content from an external source like the local file system, kafka, DB, etc. As each component executes against a FlowFile, that FlowFile is routed to one of the available relationships the particular processor offers. These relationships would be assigned to one of the connection exiting the processor and connecting to another downstream processor. The following Apache NiFi docs explain how to build a dtaflow: https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#building-dataflow This covers how to search for a component in yoru dataflow(s): https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#search Then when it comes to looking at the detailed lineage of an individual FlowFile, you can use NiFi's data Provenance for that. Individual processor components generate provenance events as they execute on FlowFile (create, route, drop, etc...). You can look at the entire lineage from create to drop of a FlowFile (assuming you configure NiFi provenance with enough storage to store all the lineage). BY default NiFI is configured to only use 10GB for Provenance and only store Provenance for 24 hours, but this can be configured in the nifi.properties file. You can write click on NiFi processor component in your dataflow and Select data provenance from the pop-up context menu. This will open a provenance search query result set that show FlowFile that traversed the component. You can select one and even expand the lineage of that select component. The lineage of a FlowFile will show all events associated to that FlowFile created by the processor components that FlowFile traversed. This covers how to use NiFi's Data Provenance: https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#data_provenance If you found that the provided solution(s) assisted you with your query, please take a moment to login and click Accept as Solution below each response that helped. Thank you, Matt
... View more
03-27-2023
10:40 AM
1 Kudo
@ManishR NiFi offers many components (processors, controller services, reporting tasks, etc) that can be used to construct a flow based program on the NiFi canvas (Referred to as a NiFi dataflow). While this list of default available components may be different depending on the release of NiFi being used, NiFi has embedded documentation found under help within the NiFi UI that shows all components available in that installed release. Apache NiFi also publishes the same info for the most current released version here: https://nifi.apache.org/docs/nifi-docs/ Selecting a component from the documentation with open a description of the component and all list configurable properties. Building a dataflow on the NiFi canvas consist of dragging and dropping new component processors to the canvas. You can then drag connection between these components to construct your end-to-end dataflow. There are 100s of component processors available out of the box and even more that you can download and add to your NiFi from the apache community. Once a dataflow is built and configured, starting those components would result in the creation of FlowFile (for testing, you can add a GenerateFlowFile processor that generates a FlowFile rather then ingesting content from an external source like the local file system, kafka, DB, etc. As each component executes against a FlowFile, that FlowFile is routed to one of the available relationships the particular processor offers. These relationships would be assigned to one of the connection exiting the processor and connecting to another downstream processor. The following Apache NiFi docs explain how to build a dtaflow: https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#building-dataflow This covers how to search for a component in yoru dataflow(s): https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#search Then when it comes to looking at the detailed lineage of an individual FlowFile, you can use NiFi's data Provenance for that. Individual processor components generate provenance events as they execute on FlowFile (create, route, drop, etc...). You can look at the entire lineage from create to drop of a FlowFile (assuming you configure NiFi provenance with enough storage to store all the lineage). BY default NiFI is configured to only use 10GB for Provenance and only store Provenance for 24 hours, but this can be configured in the nifi.properties file. You can write click on NiFi processor component in your dataflow and Select data provenance from the pop-up context menu. This will open a provenance search query result set that show FlowFile that traversed the component. You can select one and even expand the lineage of that select component. The lineage of a FlowFile will show all events associated to that FlowFile created by the processor components that FlowFile traversed. This covers how to use NiFi's Data Provenance: https://nifi.apache.org/docs/nifi-docs/html/user-guide.html#data_provenance If you found that the provided solution(s) assisted you with your query, please take a moment to login and click Accept as Solution below each response that helped. Thank you, Matt
... View more