- All the target databases must be pre-created (empty preferred) for all engine migrations.
- The database username and password should be the same for all the DBs in a single cluster.
- Setup your cloud DB server on AWS RDS, see Amazon Relational Database Service Documentation for details.
- Create a Scalearc cluster with source (on-prem) DB server and target (RDS) DB server with the source being in “Read + Write” mode while target being in “Standby, No Traffic” mode. See Creating Clusters for details. If you wish to use our Scalearc server deployment, then you can from here ScaleArc v2020.0.develop-549 (credentials?)
- Configure the source DB server in accordance with DMS prerequisites. Some of them are to enable
wal_level (pgsql), wal_logs, bin_log, max_connections, max_packet_size, etc (mysql) for better migration. These settings become important in the case of large multi-db migrations. Make sure to set these before running the migrations. Some examples of this are given in the troubleshooting guide. See Sources for data migration in the AWS Database Migration Service documentation for details.
- Make sure the same username and password is used for source and target DB, if not create a user in the source and/or target DB. This is also a prerequisite for Scalearc cluster creation.
- Make note of the list of source database names that you wish to migrate along with their sizes in GB. Your DBA can find this information for you, typically from
information_schema, pg_catalog, etc.
- To avoid connection errors, please do the following points before starting the migration.
- Make a note of source and target, host, port, username, password, engine, database names, and their sizes in GB.
- Engines supported:
-
MYSQL
Versions supported: 5.5, 5.6, 5.7
Link to support and limitations doc can be found here.
-
MSSQL
Versions supported: 2005, 2008, 2008R2, 2012, 2014, 2016, 2017, 2019
Link to support and limitations doc can be found here.
-
PGSQL
Versions supported: 10.x, 11.x
Link to support and limitations doc can be found here.
|
