Hello everyone, Our team has been using Nifi Registry with Postgres 13 as the metadata backend on Openshift (Kubernetes). We have deployed 1.15.3 recently and wanted to test upgrading the deployment to 1.16.0. This did not work as planned and the nifi registry pod failed to boot. The pertinent part of the error log seems to be the following:
Migration checksum mismatch for migration version 6 -> Applied to database : -260833722 -> Resolved locally : -1218242517. Either revert the changes to the migration, or run repair to update the schema history. Migration checksum mismatch for migration version 8 -> Applied to database : -2057645592 -> Resolved locally : -1726468007. Either revert the changes to the migration, or run repair to update the schema history. Need more flexibility with validation rules? Learn more: https://rd.gt/3AbJUZE We checked the database and found flyway_schema_history table that seems to hold these checksum values for the two files: - V6__AddFlowPersistence.sql - V8__AddUserGroupPolicy.sql We tried changing the recorded values, but changing that did not help. I gathered two information from the release notes and migration notes for 1.16.0: * - Security: Aggressively update libraries such as Jetty, Netty, Logging related libs (log4j, slf4j, etc..), and moved to latest H2 Database to eliminate any worries about a vulnerability in the older line. Included a super slick mechanism to upgrade from the old database to the new one without the user having to worry about it. * - This release removed support for MySQL 5.6 /5.7 and Postgres 9.x for NiFi Registry's database, supported databases are now H2, Postgres 10.x - 14.x and MySQL 8.x What ways do I have to resolve the error? Cheers, Tony Sim
