Backing Up and Recovering a Database on the Headquarters Node or Regional Node

This topic describes how to backup and recover a database of the headquarters or regional node.

To back up and recover a database on the headquarters node or regional node

1. Make sure you do regular backups and maintenance tasks on the headquarters database and on the regional databases.

   It is strongly recommended that you regularly do this work.

2. If a failure occurs, then do the following work:

   • If a failure occurs on a regional database, and if you cannot recover it to the exact point of failure, then you must reextract the regional database on the Siebel Server at the parent node.

   • If transactions are lost on the headquarters node, then you must do the following:

     • Reextract all child nodes.

     • Reextract the users who synchronize with the regional server whose regional database failed.

If you can recover the regional node to the exact point of failure, then the S_DOCK_STATUS values of the headquarters and regional nodes are synchronized. Replication Manager synchronizes the following transactions after recovery to the point of failure:

• Transactions that Replication Manager had not synchronized at the time of failure

• Transactions that Replication Manager created on the headquarters node and the regional nodes after the point of failure