Browse AWS detectors for leftover storage, orphaned resources, and stale infrastructure that can quietly turn into recurring spend.
Use this hub to find leftover AWS storage, stale infrastructure, and orphaned network resources before they keep adding quiet monthly spend.
Why this matters
These are the AWS waste patterns most likely to hide in ordinary storage, networking, and operations work.
Storage cleanup often lags behind teardown, migration, and rollback work, leaving detached EBS volumes and incomplete S3 uploads behind after the workload changed.
Public IP reservations often survive cutovers and stale environments because they feel safer to leave in place than to verify and remove.
Failed exports, artifact uploads, and ingestion jobs can leave multipart upload parts behind unless abort cleanup is configured.
How to use this page
Stay inside AWS while still moving quickly from a specific issue to the broader waste pattern behind it.
AWS category guides
These pages connect related AWS detectors into broader waste-reduction motions for storage and infrastructure cleanup.
Review stale AWS infrastructure spend across unassociated Elastic IPs, idle load balancers, quiet databases, and the surrounding ownership drift that keeps old environments hanging around.
Category guide
Idle and Underused Resources
Reduce AWS logging cost by fixing missing retention and self-targeted access logging before quiet log storage turns into recurring waste.
Category guide
Logging Cost Optimization
Reduce AWS storage costs by finding detached EBS volumes and abandoned uploads after workloads are gone.
Category guide
Storage Cost Optimization
Published AWS detectors
These detector pages cover the concrete AWS signals that support the broader cleanup themes above.
CloudWatch log groups with no retention policy default to infinite retention, which lets stored logs accumulate indefinitely and turn routine operational logging into background storage cost.
Detector ID
aws-cw-logs-no-retention
Application, network, and classic load balancers can keep billing after traffic disappears or targets are removed. This detector surfaces ELB resources with no recent traffic or no registered backends so teams can review obvious post-cutover and post-teardown drift.
Potential savings
$18 to $450 / month
$216 to $5,400 / year
Detector ID
aws-elb-idle
RDS instances with near-zero activity over a 14-day window, plus already-stopped instances that still warrant retained-footprint review.
Potential savings
$30 to $1,200 / month
$360 to $14,400 / year
Detector ID
aws-rds-idle
S3 access logging loop cost starts when a bucket writes access logs back to itself instead of a dedicated logging bucket. The issue usually appears when access logging is enabled quickly and the destination bucket choice is never revisited.
Detector ID
aws-s3-logging-loop
S3 incomplete multipart uploads create storage cost when failed or abandoned uploads leave billable parts behind. This usually happens after artifact, export, or ingestion jobs stop before cleanup runs.
Potential savings
$25 to $600 / month
$300 to $7,200 / year
Detector ID
aws-s3-incomplete-multipart-uploads
Unattached EBS volumes keep billing after an EC2 instance is deleted, replaced, or scaled down. The cost is easy to miss because the storage survives even when the workload is gone.
Potential savings
$40 to $2,200 / month
$480 to $26,400 / year
Detector ID
aws-ebs-unattached
Allocated Elastic IPs keep billing when they are no longer associated with an instance or network interface. The waste is easy to miss because the address can remain reserved after cutovers, teardown, or failed cleanup.
Potential savings
$4 to $24 / month
$48 to $288 / year
Detector ID
aws-eip-unassociated
Many AWS cost leaks are not outages. They are resources that still appear valid enough to ignore, so they survive normal operational review even while they keep charging.
Start with the detector if you already suspect a specific resource pattern. Start with the category guide if you know the cleanup theme but want the full set of related checks.
No. They also cover quieter storage-governance problems, such as incomplete upload cleanup and stale teardown residue, that can compound spend without looking like classic idleness.
Early Access
Cloud Waste Hunter is being built to connect detector findings into one review flow with savings estimates, cleanup context, and practical next steps.
