Replication Equality Requirements:
- The same logical structure
- The same physical structure of the databases. All Storage Areas except after-image extents
- The same OpenEdge version and bit-version of the databases.
- The same Database block (dblocksize) and Before-Image (BI) block sizes.
- Same BI cluster size.
- If large file support is enabled, it must be enabled on both databases.
- The values for the following database startup parameters: Lock Table Entries (-L), Number of Users (-n), Maximum Servers (-Mn) and -maxxxids for JTA enabled databases [ since OpenEdge 11.3 ], so the transaction tables are at least equally sized in order that AI transaction notes to be successfully applied from source to target. Since OpenEdge 11.7 when these values for the target are less than the source, the agent will not start. This is no longer a requirement since OpenEdge 12.2: Why is error 15571 on Replication enabled target databases important?
- Both the source and the target machines must have the same endian ordering, to assure the way multiple byte integers are stored in memory is the same - either by MSB (most-significant byte) or LSB (least-significant byte). Systems storing by MSB are called Big Endian, and those storing by LSB are called Little Endian. The term endianess is used in general to describe when binary files are portable between platforms; those platforms with the same endianess may use binary data transparently. Typically, UNIX machines and Windows machines use different endian ordering for storage. Therefore, a Windows source database can be replicated to another Windows machine, but not to an HPUX machine. For further clarification refer to Article: Is Replication using OpenEdge Replication and Replication Plus supported cross Platforms?
Although it is not an equality requirement to have the same Buffer Pool (-B), APW/ BIW or other performance related parameters and processes, it is recommended that these are set at reasonable values and page writers are started on Enterprise databases. When the target database can't write the data fast enough and that becomes the bottleneck, it can cause the Database Service Manager Queue (or sometimes referred to as the -pica buffer/queue or RPLS-Q) to fill up on the source database that results in production performance issues and database freezes.
