The err: block of __checkpoint_db_internal contains disaggregated storage specific logic: updating checkpoint metadata when the stable timestamp has advanced without other changes, and advancing to the next checkpoint after metadata tracking is turned off. This logic is unrelated to the general checkpoint teardown around it.

The aim of this ticket is to extract the disagg specific logic into a static helper, improving code readability and reducing complexity.

Definition of Done

• __checkpoint_disagg_put_and_advance exists as a static helper in __checkpoint_txn.c and is called from __checkpoint_db_internal
• Existing tests pass