NiFi ExecuteFlumeSource error - "unable to load source type"

[#1]

Hi guys,

We have Flume setup on the Hadoop cluster and it is working. I am trying to use the Flume data as source in NiFi, I set up the required properties in NiFi's ExecuteFlumeSource processor, but it gives "unable to load source type" error.

Any thoughts on what's causing the error?

Thank you.

[#2]

@Raj B Check the value of the SourceType field in the ExecuteFlumeSource. The type should be valid source type, i.e. avro, thrift, netcat, etc.

Check out the @Timothy Spann article on using flume with NiFi:

https://community.hortonworks.com/articles/48271/using-apache-flume-sources-and-sinks-with-apache-n....

[#3]

@cduby I was following the example in the NiFi documentation; looking at the example it didn't seem there were a set of standard source types; @Timothy Spann's article didn't say anything about the source types being preset few either; do you know what source type to use if my incoming data is in HL7 format.

[#4 · Accepted solution]

@Raj B Here are the flume source types. In the documentation for each source type you will see the type Property defines the string to use in the SourceType field.

https://flume.apache.org/FlumeUserGuide.html#flume-sources

For example if look at JMS Source, the type property specifies the component type name is jms.

HL7 is a data format but the source type for flume is a protocol that describes how the data is delivered. For example HL7 format data could be delivered using Kafka.

If the HL7 data is already delivered using flume, check the config files to see what the source type value is.

[#5]

thanks @cduby, I see what you're saying. Let me check what I need to use for source type in my case.