README.md

1. Usage in NiFi

You can use NiFi processor (nifi-cql-nar-*.nar).

• ✅ PutCQL as NiFi v2 processor with controller, where inputs are FlowFiles

2. PutCQL (NiFi processor)

2.1 Preconditions for usage

You have to do these steps (only one-time action):

1. You need the NAR file 'nifi-cql-nar-*.nar'
   • the last version is here
2. Import the *.nar file to the NiFi lib directory
   • expected location in Linux e.g. '/opt/nifi/current-nifi/lib'
3. STOP NiFi
4. START NiFi

2.2 Add controller and setup properties

CQLControllerService setting (key items):

• IP Addresses:
  • IP addresses of CQL engine with comma delimiter e.g. '10.129.53.10, 10.129.53.11, 10.129.53.12'
• Port:
  • Port for communication with CQL engine (default is 9042)
• Username:
  • Username for login to CQL
• Password:
  • Password for login to CQL
• Local Data Center:
  • Name of local data center in CQL typically e.g. 'dc1' or 'datacenter1', etc.
• Connection Timeout:
  • 900 (in seconds)
• Request Timeout:
  • 60 (in seconds)
• Consistency Level:
  • Default consistency level, e.g. LOCAL_ONE, LOCAL_QUORUM, etc.

2.3 Add processor and setup properties

Input

• FlowFile with CSV content for import (content see), where the CSV content (with header) is based on 'keyspace.table' definition in CQL (from data types point of view).

Output

• cql.count (FlowFile attribute)
  • The amount of write rows to CQL.

PutCQL setting (key items):

• Write Consistency Level:
  • E.g. LOCAL_ONE, LOCAL_QUORUM, etc.
• Table:
  • Schema and table name in CQL for write/put a data (expected format 'keyspace.table') e.g. 'cqlschema.cqltable'
  • The data types defined in the table will be used for value conversions from the CSV file.
• Batch Size:
  • Size of batch for write to CQL (default is 200)
• Dry Run:
  • The simulation of write to CQL (default is false)