System Maintenance

System Maintenance is the server health view inside Aeroplane. It helps you catch disk pressure before deployments, backups, or database volumes run the server out of space.

What Aeroplane Measures

Aeroplane records:

• Root filesystem usage from df.
• Docker storage and reclaimable data from Docker system data.
• Aeroplane data directory usage.
• Build artifact usage.
• Database backup usage.
• APT cache usage.
• System journal usage.

Maintenance history keeps the last 48 points for disk usage, Docker reclaimable bytes, and build artifact usage.

Alerts

Aeroplane raises maintenance alerts when:

• Disk usage crosses warning or critical thresholds.
• Docker reclaimable data grows large.
• Build artifacts grow large.

The exact underlying numbers are visible in the maintenance panels so you can decide whether cleanup is enough or the server needs more capacity.

Safe Cleanup

Safe cleanup can run these cleanup targets:

• Stopped Docker containers.
• Unused Docker images.
• Docker build cache.
• APT cache.
• System journals down to about 100M.
• Aeroplane build artifacts older than 24 hours.

Safe cleanup avoids Docker volume pruning because old database data can live in unattached volumes.

Volume Cleanup

Clean volumes is separate and requires confirmation. It runs Docker volume prune and can delete old data from removed containers.

Do not use volume cleanup as a casual disk fix. Create backups first and make sure you do not need any detached database volumes.

When to Use Maintenance

• Before a large Railway import.
• Before exporting an Aeroplane migration bundle.
• Before enabling disk+r2 backups on many databases.
• After repeated failed builds.
• When deploys fail with Docker or disk space errors.

Maintenance does not replace server monitoring, but it gives the control plane a first-class view of the resources it consumes.