I'm trying to use tenant tags with the kafka low level consumer to split consuming v consumed partitions across servers. However the offline servers don't seem to be getting any segments. Are there additional steps needed to get this to work. Am also using upsert functionality.

Consumed is still part of RT table. Offline is what is pushed from offline ingestion flow

Are you trying to use this? https://docs.pinot.apache.org/operators/operating-pinot/tuning/realtime#moving-completed-segments-to-different-hosts

yes I was following these docs, trying to get better query performance by separating consuming from consumed with low level kafka consumer.

Do the docs help answer your questions?

Look for logs in the controller by SegmentRelocator class and TableRebalancer class

<http://LOGGER.info|LOGGER.info>("Relocating COMPLETED segments for table: {}", tableNameWithType);
 <http://LOGGER.info|LOGGER.info>(
        "Start rebalancing table: {} with dryRun: {}, reassignInstances: {}, includeConsuming: {}, bootstrap: {}, "
            + "downtime: {}, minReplicasToKeepUpForNoDowntime: {}, enableStrictReplicaGroup: {}, bestEfforts: {}",

Will have a look. Does this rebalncer work without any additional setup e.g. minion workers?

Yes, it's a periodic task on the controller called SegmentRelocator. It should run every hour by default. You can check in the controller looks at startup, it should print each periodic task and the frequency