I’ve setup a Postgres RDS instance and am attempting to launch a service through ECS that connects to this instance. The service is a Scala service deployed as a Dockerfile to ECR. When ECS attempts to spin up a task, CloudWatch shows the following logs:

[ZScheduler-Worker-1] DEBUG c.z.hikari.util.DriverDataSource - Loaded driver with class name org.postgresql.Driver for jdbcUrl=jdbc:postgresql://{endpoint}.us-east-1.rds.amazonaws.com:5432/{database}
[ZScheduler-Worker-1] INFO  com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Added connection org.postgresql.jdbc.PgConnection@23d97170
[ZScheduler-Worker-1] INFO  com.zaxxer.hikari.HikariDataSource - Hikari-PostgreSQL - Start completed.

[ZScheduler-Worker-1] INFO [36mlevel=INFO[0m [37mthread=zio-fiber-6[0m [36mmessage="Datasource configured at jdbc:postgresql://{endpoint}.us-east-1.rds.amazonaws.com:5432/{database}"[0m
[Hikari-PostgreSQL housekeeper] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Before cleanup stats (total=1, active=0, idle=1, waiting=0)
[Hikari-PostgreSQL housekeeper] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After cleanup  stats (total=1, active=0, idle=1, waiting=0)
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Added connection org.postgresql.jdbc.PgConnection@19b9a3ff
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After adding stats (total=2, active=0, idle=2, waiting=0)
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Added connection org.postgresql.jdbc.PgConnection@5cbb2eac
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After adding stats (total=3, active=0, idle=3, waiting=0)
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Added connection org.postgresql.jdbc.PgConnection@61ead5c5
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After adding stats (total=4, active=0, idle=4, waiting=0)
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Added connection org.postgresql.jdbc.PgConnection@61762b3d
[Hikari-PostgreSQL connection adder] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After adding stats (total=5, active=0, idle=5, waiting=0)
[ZScheduler-Worker-0] INFO  com.zaxxer.hikari.HikariDataSource - Hikari-PostgreSQL - Shutdown initiated...
[ZScheduler-Worker-0] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - Before shutdown stats (total=5, active=0, idle=5, waiting=0)
[Hikari-PostgreSQL connection closer] DEBUG com.zaxxer.hikari.pool.PoolBase - Hikari-PostgreSQL - Closing connection org.postgresql.jdbc.PgConnection@23d97170: (connection evicted)
[Hikari-PostgreSQL connection closer] DEBUG com.zaxxer.hikari.pool.PoolBase - Hikari-PostgreSQL - Closing connection org.postgresql.jdbc.PgConnection@19b9a3ff: (connection evicted)
[Hikari-PostgreSQL connection closer] DEBUG com.zaxxer.hikari.pool.PoolBase - Hikari-PostgreSQL - Closing connection org.postgresql.jdbc.PgConnection@5cbb2eac: (connection evicted)
[Hikari-PostgreSQL connection closer] DEBUG com.zaxxer.hikari.pool.PoolBase - Hikari-PostgreSQL - Closing connection org.postgresql.jdbc.PgConnection@61ead5c5: (connection evicted)
[Hikari-PostgreSQL connection closer] DEBUG com.zaxxer.hikari.pool.PoolBase - Hikari-PostgreSQL - Closing connection org.postgresql.jdbc.PgConnection@61762b3d: (connection evicted)
[ZScheduler-Worker-0] DEBUG com.zaxxer.hikari.pool.HikariPool - Hikari-PostgreSQL - After shutdown stats (total=0, active=0, idle=0, waiting=0)
[ZScheduler-Worker-0] INFO  com.zaxxer.hikari.HikariDataSource - Hikari-PostgreSQL - Shutdown completed.

I first thought this had to do with the connection pool not being configured correctly, or potentially security groups. However, when running this code locally I can connect with the instance just fine (it’s set to be publicly accessible to check this). When building the image locally and running that, it gives me the same error as the ECS task does (see above), so it’s likely a Docker issue? I’ve already tried some suggestions passing in a dns flag when running the image, but that does not work.