Before we look at the details of the lookup processor, we will explain the basics of Apache NiFi. There are four important terms to know:
FlowFile : A FlowFile is essentially the data. It moves in the NiFi UI and can be analyzed there. If new data is read, a corresponding FlowFile is created. Once this has reached its destination and the data has been written again (e.g. in a database or a file), the FlowFile is deleted. FlowFiles have attributes. These are key-value pairs that are stored in memory and the content, which is not stored in memory. Content is the content, while attributes only represent metadata.
Processor : Processors argentina telegram screening act as "black boxes" in NiFi and usually perform exactly one task. The SplitJson processor, for example, splits a JSON array. Processors can access the attributes and content of a FlowFile and edit both. In the current NiFi version 1.17.0 there are more than 300 processors for a wide variety of tasks.
Queue/Connection : A data flow is created when several processors are linked together. This is done via connections or queues. These also serve as queues if a processor is slower and cannot process the incoming flow files quickly enough.
Controller service : Controller services act as daemons in the background and provide a specific resource that can then be used by the other processors or controller services.
For example, a database controller service, such as the DBCPConnectionPool service, provides a database connection. This is configured in a central location and several processors can access it. In total, there are over 100 different controller services that provide cloud connections, email functionality, database connections and many other functions.