Database Connection String Security#

Your database connection string is the keys to your kingdom. A leaked connection string means full access to your data — reads, writes, deletes, everything.

Yet developers still hardcode credentials in config files, commit them to Git, and share them over Slack. Let's fix that.

The Problem with Traditional Connection Strings#

A typical connection string looks like this:

postgresql://admin:SuperSecret123@db.example.com:5432/production

Everything an attacker needs is right there: username, password, host, port, and database name. If this string leaks through logs, error messages, or source control, your database is compromised.

Common leak vectors:

• Hardcoded in application config committed to Git
• Logged in plaintext by ORMs or connection pools
• Shared in Slack/email between team members
• Visible in CI/CD pipeline logs
• Stored in .env files without proper .gitignore

Connection Encryption with SSL/TLS#

Why Encryption Matters#

Without SSL/TLS, database traffic travels in plaintext. Anyone on the network path (cloud VPC notwithstanding) can sniff queries and results.

# Unencrypted — vulnerable to sniffing
Client → [SELECT * FROM users WHERE email='...'] → Database
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
         Anyone on the network can read this

# Encrypted with TLS
Client → [encrypted bytes] → Database
         Only client and server can decrypt

Configuring SSL/TLS for PostgreSQL#

# Connection with SSL required
connection:
  host: db.example.com
  port: 5432
  sslmode: verify-full
  sslrootcert: /certs/ca.pem
  sslcert: /certs/client-cert.pem
  sslkey: /certs/client-key.pem

SSL modes ranked by security:

Always use verify-full in production. Anything less is vulnerable to man-in-the-middle attacks.

Configuring SSL/TLS for MySQL#

[client]
ssl-mode = VERIFY_IDENTITY
ssl-ca = /certs/ca.pem
ssl-cert = /certs/client-cert.pem
ssl-key = /certs/client-key.pem

MongoDB TLS Configuration#

const client = new MongoClient(uri, {
  tls: true,
  tlsCAFile: '/certs/ca.pem',
  tlsCertificateKeyFile: '/certs/client.pem',
  tlsAllowInvalidCertificates: false,  // never set true in production
  tlsAllowInvalidHostnames: false       // never set true in production
});

IAM Authentication: Eliminate Passwords Entirely#

IAM-based authentication replaces static passwords with cloud identity tokens. No passwords to rotate, no credentials to leak.

AWS RDS IAM Authentication#

import boto3

def get_rds_auth_token(host, port, user, region):
    client = boto3.client('rds', region_name=region)
    token = client.generate_db_auth_token(
        DBHostname=host,
        Port=port,
        DBUsername=user,
        Region=region
    )
    return token

# Token is valid for 15 minutes — auto-rotated
token = get_rds_auth_token('mydb.abc123.us-east-1.rds.amazonaws.com', 5432, 'app_user', 'us-east-1')

GCP Cloud SQL IAM Authentication#

from google.cloud.sql.connector import Connector

connector = Connector()

def get_connection():
    return connector.connect(
        "project:region:instance",
        "pg8000",
        user="service-account@project.iam.gserviceaccount.com",
        db="production",
        enable_iam_auth=True
    )

Azure AD Authentication for Azure SQL#

# Connection string using Azure AD
Server=myserver.database.windows.net;
Database=mydb;
Authentication=Active Directory Default;
Encrypt=True;

IAM auth benefits:

• No static passwords to manage or rotate
• Credentials tied to cloud identity (service account, role)
• Automatic expiration (15-minute tokens for AWS)
• Centralized access control through IAM policies
• Full audit trail in cloud provider logs

Short-Lived Credentials with HashiCorp Vault#

Static credentials are a liability. Vault generates temporary database credentials on demand.

# How Vault dynamic secrets work
App → Vault: "I need database credentials"
Vault → Database: CREATE ROLE temp_user WITH PASSWORD 'random' VALID UNTIL '15min'
Vault → App: {username: "temp_user", password: "random", ttl: "15m"}

# After TTL expires
Vault → Database: DROP ROLE temp_user

Vault Database Secrets Engine Configuration#

# Enable the database secrets engine
vault secrets enable database

# Configure PostgreSQL connection
vault write database/config/production \
  plugin_name=postgresql-database-plugin \
  connection_url="postgresql://{{username}}:{{password}}@db.example.com:5432/production" \
  allowed_roles="app-role" \
  username="vault_admin" \
  password="admin_password"

# Create a role with a creation statement
vault write database/roles/app-role \
  db_name=production \
  creation_statements="CREATE ROLE \"{{name}}\" WITH LOGIN PASSWORD '{{password}}' VALID UNTIL '{{expiration}}'; GRANT SELECT, INSERT, UPDATE ON ALL TABLES IN SCHEMA public TO \"{{name}}\";" \
  default_ttl="1h" \
  max_ttl="24h"

Application Usage#

import hvac

client = hvac.Client(url='https://vault.example.com:8200')
creds = client.secrets.database.generate_credentials('app-role')

db_config = {
    'host': 'db.example.com',
    'user': creds['data']['username'],      # v-app-role-abc123
    'password': creds['data']['password'],  # randomly generated
    'database': 'production'
}
# Credentials auto-expire after TTL

Secrets Management for Connection Strings#

Never Hardcode — Use a Secrets Manager#

# BAD — hardcoded in environment
DATABASE_URL=postgresql://admin:secret@db.example.com:5432/prod

# GOOD — reference to secret
DATABASE_URL=vault:database/creds/app-role
DATABASE_URL=aws-sm:prod/database/connection
DATABASE_URL=gcp-sm:projects/123/secrets/db-connection/versions/latest

Kubernetes Secrets with External Secrets Operator#

apiVersion: external-secrets.io/v1beta1
kind: ExternalSecret
metadata:
  name: database-credentials
spec:
  refreshInterval: 15m
  secretStoreRef:
    name: vault-backend
    kind: ClusterSecretStore
  target:
    name: db-credentials
  data:
    - secretKey: username
      remoteRef:
        key: database/creds/app-role
        property: username
    - secretKey: password
      remoteRef:
        key: database/creds/app-role
        property: password

Secret Rotation Strategy#

# Rotation timeline
Day 0:  Generate new credentials (Credential B)
Day 0:  Update application to accept both A and B
Day 1:  Switch primary to Credential B
Day 3:  Verify no traffic using Credential A
Day 5:  Revoke Credential A

# Automated rotation with AWS Secrets Manager
aws secretsmanager rotate-secret \
  --secret-id prod/database/credentials \
  --rotation-lambda-arn arn:aws:lambda:us-east-1:123:function:rotate-db-creds \
  --rotation-rules AutomaticallyAfterDays=30

Audit Logging Database Connections#

Every connection to your database should be logged and monitored.

PostgreSQL Connection Audit Logging#

-- Enable connection logging
ALTER SYSTEM SET log_connections = on;
ALTER SYSTEM SET log_disconnections = on;
ALTER SYSTEM SET log_line_prefix = '%t [%p]: user=%u,db=%d,app=%a,client=%h ';

-- Log all DDL statements
ALTER SYSTEM SET log_statement = 'ddl';

What to Monitor#

# Alert on these patterns
- Connection from unknown IP addresses
- Connections using the root/admin account
- Unusual connection volume spikes
- Failed authentication attempts (brute force)
- Connections outside business hours (if applicable)
- Connections from deprecated credential sets

Centralized Audit with pgAudit#

-- Install pgAudit extension
CREATE EXTENSION pgaudit;

-- Log all reads and writes
ALTER SYSTEM SET pgaudit.log = 'read, write, ddl';

-- Log role-specific activity
ALTER ROLE app_user SET pgaudit.log = 'write, ddl';

Security Checklist#

Common Mistakes#

1. Using sslmode=require instead of verify-full — require encrypts but does not verify the server identity
2. Sharing one credential across all services — if one service is compromised, all services are compromised
3. Long-lived credentials without rotation — credentials should expire, not live forever
4. Logging connection strings in error messages — sanitize all logs to strip credentials
5. Storing secrets in environment variables without encryption — env vars are visible in process listings and crash dumps

Summary#

---

Audit your database security at codelit.io — 413 architecture articles and growing. Generate architecture diagrams, run security checks, and visualize your infrastructure.