I'm setting up logical replication between Cloud SQL Postgres (as a replica) and AWS RDS Postgres (source). I'm using guideline provided by GCP -> https://cloud.google.com/sql/docs/postgres/replication/configure-replication-from-external.
I was creating a source representation instance using request:
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
--header 'Content-Type: application/json' \
--data @./source.json \
-X POST \
https://sqladmin.googleapis.com/sql/v1beta4/projects/${project_id}/instances
where source.json looks like this:
{
"name": "aws-boa-source",
"region": "us-east4",
"databaseVersion": "POSTGRES_14",
"onPremisesConfiguration": {
"hostPort": "{aws_rds_endpoint}:{aws_rds_port}",
"username": "test",
"password": "test!"
}
}
Also, I'm creating a CloudSQL Postgres replica using such request:
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
--header 'Content-Type: application/json' \
--data '{"settings":{"tier":"db-custom-2-7680","dataDiskSizeGb":"100","ipConfiguration":{"privateNetwork":"projects/${project_id}/global/networks/${network}"},"availabilityType":"ZONAL"},"masterInstanceName":"aws-boa-source","region":"us-east4","databaseVersion":"POSTGRES_14","name":"aws-boa-replica"}' \
-X POST \
https://sqladmin.googleapis.com/sql/v1beta4/projects/${project_id}/instances
When I'm checking whether the setup is configured properly using this request:
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
--header 'Content-Type: application/json' \
--data '{
"syncMode": "EXTERNAL_SYNC_MODE_UNSPECIFIED"
}' \
-X POST \
https://sqladmin.googleapis.com/sql/v1beta4/projects/${project_id}/instances/aws-boa-source/verifyExternalSyncSettings
I'm receiving such error:
{
"error": {
"code": 500,
"message": "Internal error encountered.",
"errors": [
{
"message": "Internal error encountered.",
"domain": "global",
"reason": "backendError"
}
],
"status": "INTERNAL"
}
}
Additionally, my AWS RDS Postgres instance is publicly accessible and I'm able to connect to to it from my local PC.
What is more curious, I was able to set up replication once using the same configuration. But then I destroyed the whole infrastructure and I wanted to setup it again. Now I'm encountered such problems.
Does anybody know why I'm receiving such error?