Trying to come up with an exhaustive list of *possible* ways the DSI (ASE2.DB2) could go down would take all day to document.
Common issues (some duplicates of what Avinash has posted)
- add login in PDS; add user in PDB; DSI attempts to add user in RDB and fails because login was not added to the RDS
- maintenance user (login used by DSI to perform operations in RDB) doesn't have necessary permissions to perform various operations (eg, no write permissions on a table), and DSI goes down
- replicating a table with no unique indexes can lead to row count mismatch errors when DELETEs/UPDATEs are applied in the RDB, and DSI goes down
- any time the PDB and RDB get out of sync (data-wise, schema-wise)
- cross-database RI constraint failure/error
- RDB becomes unavailable (RDB log fills, RDB is offline, RDS is down)
- hitting any ASE error that is configured in the repserver as 'STOP REPLICATION'