Previously we were exposing internal cluster BrokerId's to the client.
However these need to be rewritten to the BrokerId's used by the shotover instances.
This PR introduces the required message rewriting logic.
The isr_nodes (in-sync-replica nodes) can follow the same logic as replica_nodes, so the logic is just copied over from there.
Previously we were exposing internal cluster BrokerId's to the client. However these need to be rewritten to the BrokerId's used by the shotover instances. This PR introduces the required message rewriting logic.
The
isr_nodes
(in-sync-replica nodes) can follow the same logic asreplica_nodes
, so the logic is just copied over from there.