Aws Rds
AWS RDS relational database service for managed databases. Use when provisioning databases, configuring backups, managing replicas, troubleshooting connectivity, or optimizing performance.
This skill is reviewed through Azure/git-ape PRs and maintained by its listed author. It is not bundled or activated with core. See the Skill Registry and use /git-ape-skills install aws-rds for explicit project or personal installation. Review is not a guarantee of safety.
Details
| Property | Value |
|---|---|
| Skill Directory | .github/community-skills/aws-rds/ |
| Author | dawright22 |
| Maturity | experimental |
| Available with Core | ❌ No — explicit installation required |
| Usage | /aws-rds Database engine, feature, or issue to look up (e.g. 'PostgreSQL Multi-AZ', 'read replica', 'IAM authentication', 'Performance Insights') |
Documentation
AWS RDS
Amazon Relational Database Service (RDS) provides managed relational databases including MySQL, PostgreSQL, MariaDB, Oracle, SQL Server, and Aurora. RDS handles provisioning, patching, backups, and failover.
Table of Contents
Core Concepts
DB Instance Classes
| Category | Example | Use Case |
|---|---|---|
| Standard | db.m6g.large | General purpose |
| Memory Optimized | db.r6g.large | High memory workloads |
| Burstable | db.t3.medium | Variable workloads, dev/test |
Storage Types
| Type | IOPS | Use Case |
|---|---|---|
| gp3 | 3,000-16,000 | Most workloads |
| io1/io2 | Up to 256,000 | High-performance OLTP |
| magnetic | N/A | Legacy, avoid |
Multi-AZ Deployments
- Multi-AZ Instance: Synchronous standby in different AZ
- Multi-AZ Cluster: One writer, two reader instances (Aurora-like)
Read Replicas
Asynchronous copies for read scaling. Can be cross-region.
Common Patterns
Create a PostgreSQL Instance
AWS CLI:
# Create DB subnet group
aws rds create-db-subnet-group \
--db-subnet-group-name my-db-subnet-group \
--db-subnet-group-description "Private subnets for RDS" \
--subnet-ids subnet-12345678 subnet-87654321
# Create security group (allow PostgreSQL from app)
aws ec2 create-security-group \
--group-name rds-postgres-sg \
--description "RDS PostgreSQL access" \
--vpc-id vpc-12345678
aws ec2 authorize-security-group-ingress \
--group-id sg-rds12345 \
--protocol tcp \
--port 5432 \
--source-group sg-app12345
# Create RDS instance
aws rds create-db-instance \
--db-instance-identifier my-postgres \
--db-instance-class db.t3.medium \
--engine postgres \
--engine-version 16.1 \
--master-username admin \
--master-user-password 'SecurePassword123!' \
--allocated-storage 100 \
--storage-type gp3 \
--db-subnet-group-name my-db-subnet-group \
--vpc-security-group-ids sg-rds12345 \
--multi-az \
--backup-retention-period 7 \
--storage-encrypted \
--no-publicly-accessible
boto3:
import boto3
rds = boto3.client('rds')
response = rds.create_db_instance(
DBInstanceIdentifier='my-postgres',
DBInstanceClass='db.t3.medium',
Engine='postgres',
EngineVersion='16.1',
MasterUsername='admin',
MasterUserPassword='SecurePassword123!',
AllocatedStorage=100,
StorageType='gp3',
DBSubnetGroupName='my-db-subnet-group',
VpcSecurityGroupIds=['sg-rds12345'],
MultiAZ=True,
BackupRetentionPeriod=7,
StorageEncrypted=True,
PubliclyAccessible=False
)
Create Read Replica
aws rds create-db-instance-read-replica \
--db-instance-identifier my-postgres-replica \
--source-db-instance-identifier my-postgres \
--db-instance-class db.t3.medium \
--availability-zone us-east-1b
Take a Snapshot
aws rds create-db-snapshot \
--db-snapshot-identifier my-postgres-snapshot-2024-01-15 \
--db-instance-identifier my-postgres
Restore from Snapshot
aws rds restore-db-instance-from-db-snapshot \
--db-instance-identifier my-postgres-restored \
--db-snapshot-identifier my-postgres-snapshot-2024-01-15 \
--db-instance-class db.t3.medium \
--db-subnet-group-name my-db-subnet-group \
--vpc-security-group-ids sg-rds12345
Point-in-Time Recovery
aws rds restore-db-instance-to-point-in-time \
--source-db-instance-identifier my-postgres \
--target-db-instance-identifier my-postgres-pitr \
--restore-time 2024-01-15T10:30:00Z \
--db-instance-class db.t3.medium
Connect with IAM Authentication
import boto3
import psycopg2
rds = boto3.client('rds')
# Generate auth token
token = rds.generate_db_auth_token(
DBHostname='my-postgres.abc123.us-east-1.rds.amazonaws.com',
Port=5432,
DBUsername='iam_user',
Region='us-east-1'
)
# Connect using token as password
conn = psycopg2.connect(
host='my-postgres.abc123.us-east-1.rds.amazonaws.com',
port=5432,
database='mydb',
user='iam_user',
password=token,
sslmode='require'
)
CLI Reference
Instance Management
| Command | Description |
|---|---|
aws rds create-db-instance | Create instance |
aws rds describe-db-instances | List instances |
aws rds modify-db-instance | Modify settings |
aws rds delete-db-instance | Delete instance |
aws rds reboot-db-instance | Reboot instance |
aws rds start-db-instance | Start stopped instance |
aws rds stop-db-instance | Stop instance |
Backups
| Command | Description |
|---|---|
aws rds create-db-snapshot | Manual snapshot |
aws rds describe-db-snapshots | List snapshots |
aws rds restore-db-instance-from-db-snapshot | Restore from snapshot |
aws rds restore-db-instance-to-point-in-time | Point-in-time restore |
aws rds copy-db-snapshot | Copy snapshot |
Replicas
| Command | Description |
|---|---|
aws rds create-db-instance-read-replica | Create read replica |
aws rds promote-read-replica | Promote to standalone |
Best Practices
Security
- Never make publicly accessible — use VPC and security groups
- Enable encryption at rest (KMS) and in transit (SSL)
- Use IAM authentication for application access
- Store credentials in Secrets Manager with rotation
- Use parameter groups to enforce SSL
# Enforce SSL in PostgreSQL
aws rds modify-db-parameter-group \
--db-parameter-group-name my-pg-params \
--parameters "ParameterName=rds.force_ssl,ParameterValue=1,ApplyMethod=pending-reboot"
Performance
- Right-size instances — monitor CPU, memory, IOPS
- Use gp3 for cost-effective performance
- Enable Performance Insights for query analysis
- Use read replicas for read scaling
- Optimize queries — check slow query log
High Availability
- Enable Multi-AZ for production
- Use Aurora for mission-critical workloads
- Configure appropriate backup retention
- Test failover periodically
- Monitor replication lag for replicas
Cost Optimization
- Use Reserved Instances for steady-state workloads
- Stop dev/test instances when not in use
- Delete old snapshots regularly
- Right-size instance classes
Troubleshooting
Cannot Connect
Causes:
- Security group not allowing access
- Instance not in VPC subnet
- SSL required but not used
- Wrong endpoint/port
Debug:
# Check security group
aws ec2 describe-security-groups --group-ids sg-rds12345
# Check instance status
aws rds describe-db-instances \
--db-instance-identifier my-postgres \
--query "DBInstances[0].{Status:DBInstanceStatus,Endpoint:Endpoint}"
# Test connectivity from EC2
nc -zv my-postgres.abc123.us-east-1.rds.amazonaws.com 5432
High CPU/Memory
# Enable Performance Insights
aws rds modify-db-instance \
--db-instance-identifier my-postgres \
--enable-performance-insights \
--performance-insights-retention-period 7
Storage Full
# Enable storage autoscaling
aws rds modify-db-instance \
--db-instance-identifier my-postgres \
--max-allocated-storage 500
# Set CloudWatch alarm
aws cloudwatch put-metric-alarm \
--alarm-name "RDS-Storage-Low" \
--metric-name FreeStorageSpace \
--namespace AWS/RDS \
--dimensions Name=DBInstanceIdentifier,Value=my-postgres \
--statistic Average \
--period 300 \
--threshold 10000000000 \
--comparison-operator LessThanThreshold \
--evaluation-periods 2 \
--alarm-actions arn:aws:sns:us-east-1:123456789012:alerts
Replication Lag
aws cloudwatch get-metric-statistics \
--namespace AWS/RDS \
--metric-name ReplicaLag \
--dimensions Name=DBInstanceIdentifier,Value=my-postgres-replica \
--start-time $(date -d '1 hour ago' -u +%Y-%m-%dT%H:%M:%SZ) \
--end-time $(date -u +%Y-%m-%dT%H:%M:%SZ) \
--period 60 \
--statistics Average