Why do these errors occur repeatedly and without end when attempting to use AWS DMS full-load migrations even thought test target endpoint succeeded?

The "rep_net_server_select: Server poll timeout. failed" errors you're seeing in your AWS DMS logs are related to network connectivity issues between the DMS replication instance and your source or target databases. These errors occur repeatedly because DMS is continuously trying to establish or maintain a connection but is encountering timeouts.

These network connectivity issues could be caused by several factors:

1. Security group configurations that don't properly allow traffic between the DMS replication instance and your databases
2. Network ACLs or firewall rules blocking necessary connections
3. VPC configuration issues preventing proper routing
4. Database server performance bottlenecks causing connection delays
5. Network latency or packet loss between the DMS instance and databases

To troubleshoot and resolve these errors:

1. Verify that security groups for both source and target databases allow inbound and outbound traffic from the DMS replication instance
2. Check that the databases are accessible from the VPC where DMS is running
3. Ensure the source and target databases allow access from the public IP address of the data migration
4. Review CloudWatch logs for more detailed error information
5. Test direct connectivity between the replication instance and your databases
6. Consider increasing the network timeout settings in your task configuration

If the issue persists after addressing these network connectivity concerns, you may need to examine your database configurations or consider reaching out to AWS Support for more detailed troubleshooting assistance.
Sources
Troubleshooting for homogeneous data migrations in AWS DMS - AWS Database Migration Service
AWS DMS implementation guide: Building resilient database migrations through testing, monitoring, and SOPs | AWS Database Blog