[
https://issues.apache.org/jira/browse/NIFI-4596?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tanmoy resolved NIFI-4596.
--------------------------
Resolution: Implemented
Since expression language is supported, I am now using the Atrribute names in
the Cache identifier.
> The FetchDistributedMapCache processor should read the flowfile attributes
> from the previous processor
> ------------------------------------------------------------------------------------------------------
>
> Key: NIFI-4596
> URL: https://issues.apache.org/jira/browse/NIFI-4596
> Project: Apache NiFi
> Issue Type: Improvement
> Components: Extensions
> Affects Versions: 1.2.0
> Environment: Nifi running on a 3-node cluster. OS - CentOS
> Reporter: Tanmoy
> Priority: Minor
> Labels: features, performance
> Original Estimate: 504h
> Remaining Estimate: 504h
>
> The FetchDistributedMapCache processor should read the flowfile attributes
> from the previous processor. In that case, we can configure which Cache Entry
> Identifier to read from. Right now, the processor accepts only static labels.
> Hence, it cannot be used in a framework where we would like to have the input
> source name in the Cache Identifier. The PutDistributedMapCache does not have
> any such restricition.
> Current usecase: I am reading my config from HDFS and each time a flow file
> appears in SFTP server, I ingest them using my framework. Processing time can
> be improved if I used the cache. But, the FetchDistributedMapCache is not
> configured to accept flowfile attributes as Cache Identifier. This limits its
> usage as Filenames vary everytime.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
