Transfer Directories for Off-line Locations

Just as the Hub has a Transfer directory to which files are copied before transfer to a Satellite, an off-line Satellite must have a Transfer directory to which files containing the updates from the Satellite's primary databases are copied before being transferred to the Hub, and to which files are written when they arrive from the Hub, ready to update the off-line location's secondary databases.

Environment variables must be set at both the Hub and the Satellite to point to the Transfer directory at each location.

Example:

The project code is ABC.

Cambridge is the Hub, with the location identifier CAM.

There are two off-line Satellites:

• Sydney has the location identifier SYD;

• Perth has the location identifier PER.

The transfer directory at the Hub has a subdirectory for each location:

pathname\sydney

pathname\perth

Each subdirectory will have the normal project-related directory structure, abc000, abcmac, etc.

Note:
The transfer directory is used in the same way as the transfer directory for an on-line location when the location is first generated. You can delete the directories for an on-line location once the project has been installed there, but you should keep the directories for the off-line locations, as they will be used for transferring updates.

Note:
The Transfer directory is always required as an intermediate Transfer directory when transferring between the Hub and an offline satellite. This is required to avoid corrupting data in the System and Misc databases. TRANSFER is always a two-stage process in each direction:

• Data is copied from the project to another location using TRANSFER TO <loc>.

• Data is copied into the project from another location using TRANSFER FROM <loc>.