AWS Non-Production Multi-AZ RDS Detector
Uncover development, testing, and staging RDS databases configured with redundant Multi-AZ standbys that double monthly compute costs.
1. Detection Scope & Operational Boundaries
Every ArthaOps detector enforces strict epistemic honesty. We clearly separate what can be mathematically proven via agentless telemetry from what requires contextual human judgement.
What This Detector Identifies
- Amazon RDS PostgreSQL, MySQL, and MariaDB instances configured with Multi-AZ = true in non-production accounts or environments.
- Database instances bearing environment tags such as dev, staging, test, qa, or sandbox.
- Instances where standby replica infrastructure doubles hourly instance and storage rates without serving live customer traffic.
What It Cannot Establish Alone
- Whether the non-production cluster is actively conducting automated failover testing, disaster recovery drills, or chaos engineering experiments.
- Whether contractual client SLAs or compliance certifications mandate identical Multi-AZ configuration in staging environments.
2. Telemetry & Evidence Requirements
ArthaOps operates strictly via Plane 1 (Read-Only Discovery). The following CloudFormation / IAM permissions and CloudWatch metrics are evaluated:
Verification Protocol: Tested against synthetic PostgreSQL and MySQL multi-AZ database fixtures.
Validation Context: Validated in development and QA environments where staging instances accidentally inherited production high-availability configurations.
3. Reproducible AWS CLI Verification Recipe
List all RDS database instances in ap-south-1 that have Multi-AZ enabled and evaluate instance identifiers and tags.
# List non-production Multi-AZ database instances in ap-south-1
aws rds describe-db-instances \
--region ap-south-1 \
--query "DBInstances[?MultiAZ==`true` && (contains(DBInstanceIdentifier, 'dev') || contains(DBInstanceIdentifier, 'staging') || contains(DBInstanceIdentifier, 'test') || contains(DBInstanceIdentifier, 'qa'))].{Identifier:DBInstanceIdentifier,Class:DBInstanceClass,Engine:Engine,MultiAZ:MultiAZ,AllocatedStorage:AllocatedStorage}" \
--output table---------------------------------------------------------------------------------------------
| DescribeDBInstances |
+------------------+---------------+-------------+----------+-------------------------------+
| AllocatedStorage | Class | Engine | MultiAZ | Identifier |
+------------------+---------------+-------------+----------+-------------------------------+
| 200 | db.m6g.xlarge | postgres | True | staging-orders-db-instance |
+------------------+---------------+-------------+----------+-------------------------------+Output Interpretation: Multi-AZ synchronously provisions a hot standby replica in an adjacent availability zone. In staging and dev environments, this configuration doubles the monthly bill with zero production protection.
4. Transparent Economic Model
Every cost metric published by ArthaOps references published cloud pricing APIs. We decouple gross provisioned spend from net reclaimable waste:
- 730 operational hours per standard calendar month.
- Multi-AZ doubles both compute instance hours and provisioned storage costs.
- Non-production environments do not require synchronous cross-AZ failover outside planned testing windows.
- Provisioned IOPS rates on io1/io2 and backup retention charges beyond 1 day are excluded from base savings calculation.
5. False-Positive Scenarios & Operational Risks
Never delete or terminate cloud resources blindly. These validated architectural scenarios must be reviewed before initiating remediation:
Documented False-Positive Scenarios
- Pre-Production Load & Failover Testing: Staging environments running automated chaos tests or load benchmarking must replicate production Multi-AZ behavior.
- Regulatory Mirroring: Specific financial services mandates require pre-production topologies to exactly mirror production infrastructure configurations.
Blast Radius & Remediation Risks
- Brief Maintenance Window: Converting an RDS instance from Multi-AZ to Single-AZ deletes the secondary standby replica. This causes zero primary downtime, but subsequent failover capabilities are unavailable until re-enabled.
- Mitigation: Capture a manual DB snapshot prior to modifying the instance configuration.
6. Remediation Procedure & Rollback Runbook
Follow this ordered sequence of verified AWS CLI actions to remediate the resource with complete rollback availability:
Capture Manual Pre-Modification DB Snapshot
Create an explicit manual snapshot before modifying database topology.
aws rds create-db-snapshot --region ap-south-1 --db-instance-identifier staging-orders-db-instance --db-snapshot-identifier staging-orders-pre-single-az-backupWait for Snapshot Completion
Confirm snapshot state reaches available.
aws rds wait db-snapshot-completed --region ap-south-1 --db-snapshot-identifier staging-orders-pre-single-az-backupConvert Instance to Single-AZ
Modify instance to remove the Multi-AZ standby replica. The standby replica is decommissioned without primary interruption.
aws rds modify-db-instance --region ap-south-1 --db-instance-identifier staging-orders-db-instance --no-multi-az --apply-immediatelyForensic Rollback Procedure
To re-enable Multi-AZ at any time, execute `aws rds modify-db-instance --region ap-south-1 --db-instance-identifier <id> --multi-az --apply-immediately`. AWS will synchronously provision and sync a new secondary standby replica within 15–30 minutes.
ArthaOps Assurance: ArthaOps requires snapshot confirmation before executing RDS configuration changes and verifies non-production tag fencing.
Converting to Single-AZ immediately deletes the standby replica. In the rare event of a primary host failure prior to re-enabling Multi-AZ, recovery must be executed via snapshot restore.
How This Appears in the ArthaOps Control Plane
db.m6g.xlarge PostgreSQL instance running with Multi-AZ standby in staging environment with zero external failover testing.
Test Your Infrastructure in 180 Seconds
Explore our zero-setup interactive sample report covering real Mumbai findings, or launch a 100% read-only agentless scan across your AWS environment.