Naming Convention – Data Archive Connections

Challenge

During a Data Archive project many connections must be set up to perform the archive and restore functions of the Data Archive product. The correct connection must be selected for the project to work accurately as intended. If a Data Archive source or target is chosen for a Data Restore project, the project will fail. The same holds true if a Restore source or target is chosen for a Data Archive project. The number of connections in the development or testing phase of a project can be numerous and if a naming convention isn’t followed the chance of choosing the wrong connection increases. Production typically has fewer connections to choose from, but like the lower environments, if the wrong one is chosen the job will fail. Since it is not always obvious that the problem stems from an improperly chosen connection, debugging can be time consuming.

Description

In the Production environment a Data Archive Source connection, a Data Archive Target connection, a Data Restore Source connection and a Data Restore Target connection will be defined. In the Test and Development phases of the project, many environments may be used and each one will require the same connections as the Production environment. At a minimum, there will be a Dev environment, but usually there is also a Test environment and there could be UAT, SIT, or even more. Each of the connections should be named in a manner to eliminate confusion and make choosing the wrong connection less likely.

Development, Test, UAT, SIT …

It is common practice to share the same archive database for all the different source environments and to use a separate target schema for each source and to have only one Repository for all the connections to be defined. To eliminate confusion in this type of environment, the database or service name should be included in the connection name along with the type of connection. In the case of the target, a mention of the corresponding source should also be included. The following example of connection names from a repository with Dev, Test, UAT, and SIT may be used as a guide. For the example below there are 4 sources, DEV, TEST, UAT, and SIT and one Target database DEVARC with a separate schema for each source.

• DEV Archive Source
• DEVARC DEV Archive Target
• DEVARC DEV Restore Source
• DEV Restore Target
• TEST Archive Source
• DEVARC TEST Archive Target
• DEVARC TEST Restore Source
• TEST Restore Target
• UAT Archive Source
• DEVARC UAT Archive Target
• DEVARC UAT Restore Source
• UAT Restore Target
• SIT Archive Source
• DEVARC SIT Archive Target
• DEVARC SIT Restore Source
• SIT Restore Target

Production

In the Production environment there is generally one source and one target and for the restore connections. The source is the data archive target and the target is the data archive source. The database name should still be included in the connection names and the type of connection should also be included. However, the associated source or target name does not need to be included in the archive target or the restore source since there is only one. In the example below there is one source PROD and one target PRODARC.

• PROD Archive Source
• PRODARC Archive Target
• PRODARC Restore Source
• PROD Restore Target

The chance of choosing the wrong connection for a specific project will be reduced and the likelihood of a successful project will be greatly increased if the naming conventions demonstrated above are followed.