Testcontainers fail health checks when running with Kiri

[realonbebeto]

Hello @libracoder,

I've noticed that my Testcontainers tests run successfully with both OrbStack and Docker Desktop, but fail with Kiri due to container health checks reporting an unhealthy status.

For context, here's how I'm configuring my PostgreSQL container using testcontainers-rs:

/// A wrapper around a Postgres container for use in tests.
pub struct PostgresContainer {
    #[allow(dead_code)]
    container: ContainerAsync<GenericImage>,
    pub db_port: u16,
}

impl PostgresContainer {
    pub async fn start() -> anyhow::Result<Self> {
        let container = GenericImage::new(POSTGRES_IMAGE, POSTGRES_TAG)
            .with_exposed_port(POSTGRES_PORT.tcp())
            .with_wait_for(WaitFor::message_on_stdout(
                "database system is ready to accept connections",
            ))
            .with_env_var("POSTGRES_DB", POSTGRES_DB)
            .with_env_var("POSTGRES_USER", POSTGRES_USER)
            .with_env_var("POSTGRES_PASSWORD", POSTGRES_PASSWORD)
            .with_container_name(format!("postgres-{}", Uuid::now_v7()))
            .start()
            .await
            .expect("postgres failed to start");

        let db_port = container.get_host_port_ipv4(5432).await?;

        Ok(Self { container, db_port })
    }
}

The same configuration works without issues on OrbStack and Docker Desktop.

[libracoder]

Hello @realonbebeto sorry for all the troubles.
I will reporduce this and fix accrodingly and make a new realease with other fixes by tommorow morning.