DBD: Data Replication System - Initialize all Replicated Tables
1: The process of initialization
Before you start a Data Replication server process for the first time, you have the option of initializing (preloading) the contents of one or more tables by running the Replication Initialization utility. Note that even if you do not initialize a previously-unreplicated table, Data Replication will still proceed normally – but your table won't contain any historical data before this point in time.
Initializing the table guarantees that after this point in time, the replicated table will contain exactly the same data that DB Distributor currently contains.
The initialization process:
- Preloads the (empty) table with current DB Distributor data.
- Is not required; it just provides an easy way to load a Table for the first time.
- Allows you to initialize a single table or initialize all replicated tables.
- The process reads through the DB Distributor table and rewrites each record to trigger replication of data to the database.
Will take a while for large tables, but:
1. it doesn't require exclusive access to the table.
2. Replication will continue even while the initialization is in process.
2: Initializing one or more tables
You may wish to initialize just a single table (the usual case), or you may wish to initialize all tables at once (probably only needed the first time you set up Data Replication).
You can initialize one or more tables with the Replication Initialization utility as follows:
GUI: DB Distributor Optional Modules / Data Replication / Processing / Initialize Replication
CHUI: D/R menu, option P1
Target Code
Select the Target Code you wish to process data for or leave blank to select all Target Codes.
If you are already replicating data to other targets, this would only need to be done for a new Target code. Otherwise, all Targets will be reinitialized.
This will load the current DB Distributor values into the data source if needed. Not required, provided as an easy way to load the data.
Table Code
Select the Table Codes you wish to process data for or leave blank to select all Table Codes associated with the Target Code.
- Pick one table to process or all replicated tables.
- Process will read through the DB Distributor table and re-write each record to trigger the sending of data to the data source.
Will take a while for large tables.
– Does not require exclusive access to the table.
– Replication will occur even while this is in process.
Click “Proceed” to run the utility.
Note that when you later start a Data Replication server process, transactions for current real-time DB Distributor data will be interleaved with transactions from the table initialization you have just executed.
You must be patient. It isn't unusual for the Data Replication process to take a full day (24 hours) before it catches up with all the preloaded data and settles down to the normal, more sporadic daily replication activity.