Question

Difficulty: HardDatabase Migration and Schema Conversion using DMS and SCT

An infrastructure modernization project requires migrating a high-throughput, on-premises Oracle database to Amazon Aurora PostgreSQL-Compatible Edition. The database is in active use 24/7, requiring a replication strategy that minimizes downtime to under 1 hour. A Solutions Architect uses the AWS Schema Conversion Tool (SCT) to convert the database schema and applies it to the target Aurora DB cluster. Next, the architect configures an AWS Database Migration Service (DMS) replication instance and a replication task with the migration type set to "Migrate existing data and replicate ongoing changes".

The replication task successfully performs the full load of all tables. However, once the replication task transitions to the Change Data Capture (CDC) phase, the task status changes to "Running with errors", and updates made on the source database are not reflected on the target database. Reviewing the DMS task logs reveals that the replication instance is unable to read the transaction changes from the source database's redo logs.

Which sequence of steps must the Solutions Architect perform on the source Oracle database to enable AWS DMS to successfully read the transaction changes and resume CDC replication?

  1. Configure the source database to run in ARCHIVELOG mode, enable minimal supplemental logging at the database level, and enable supplemental logging for the specific tables or columns being replicated.Answer
  2. B
    Configure the source database to run in NOARCHIVELOG mode, set the initialization parameter log_buffer to a value greater than 512 MB, and enable force logging at the database level.
  3. C
    Enable binary logging by setting the initialization parameter log_bin to ON, configure the binlog format to ROW, and grant the REPLICATION CLIENT privilege to the DMS database user.
  4. D
    Install the AWS SCT extraction agent on the source database server, configure the agent to read from the Oracle Active Data Guard standby database, and set the DMS task parameter UseLogminerReader to true.

Answer

Configure the source database to run in ARCHIVELOG mode, enable minimal supplemental logging at the database level, and enable supplemental logging for the specific tables or columns being replicated.
The correct option correctly identifies the mandatory Oracle database configurations for AWS DMS CDC: ARCHIVELOG mode, database-level minimal supplemental logging, and table-level supplemental logging. ARCHIVELOG mode ensures transaction logs are not lost, while minimal and table-level supplemental logging ensure that update and delete operations record primary keys or full column values in the redo logs, which DMS requires to replicate changes correctly.

Step-by-Step Solution

1
Enable database archiving
The database is configured to run in ARCHIVELOG mode, preserving redo log files in the archive destination.
AWS DMS requires access to archived redo log files if the online redo logs are overwritten before DMS can process them during CDC.
2
Enable database-level minimal supplemental logging
The command 'ALTER DATABASE ADD SUPPLEMENTAL LOG DATA' is executed.
This is a prerequisite for Oracle to write additional information, such as ROWIDs and primary keys, to the redo logs, which is required by the log reader.
3
Enable table-level supplemental logging
Supplemental log groups are created on the target tables (e.g., 'ALTER TABLE table_name ADD SUPPLEMENTAL LOG DATA (ALL) COLUMNS' or PRIMARY KEY columns).
Table-level supplemental logging ensures that update and delete operations in the redo log contain enough identifying column data for DMS to construct the corresponding statements on the target database.

Key Concept

AWS DMS Change Data Capture (CDC) prerequisites for Oracle source databases
Rate this question