We are investigating the possibility to use transactional replication for an archive / reporting database in order to offload the production database. If we use the same software/client for accessing the archive, the client will update some tables in the archive/consumer like access logs i.e. primary keys/identifiers will not match the production database (which is not an application problem). The log function only adds rows to a table. Is this possible or will the transactional replication fail? Other issues?
↧