Skip to main content

Tiered Cloud Backup to S3 Standard, Wasabi, and Glacier Deep Archive with RcloneView

· 4 min read
Tayson
Tayson
Senior Engineer

Keep recent restores fast and long-term retention cheap: hot in S3 Standard, warm in Wasabi/R2, and archive in Glacier Deep Archivewith RcloneView schedules and lifecycle-friendly flags.

A single storage class rarely fits every file. Design a tiered pipeline: copy fresh data to S3 Standard for rapid access, mirror to a low-cost warm tier (Wasabi/R2) for geo redundancy, and push monthly snapshots to Glacier Deep Archive for compliance retention. RcloneView layers a GUI over rclone so you can schedule Syncs, verify with Compare, and keep lifecycle rules intactno shell scripts.

Relevant docs

RcloneView app preview

Manage & Sync All Clouds in One Place

RcloneView is a cross-platform GUI for rclone. Compare folders, transfer or sync files, and automate multi-cloud workflows with a clean, visual interface.

  • One-click jobs: Copy · Sync · Compare
  • Schedulers & history for reliable automation
  • Works with Google Drive, OneDrive, Dropbox, S3, WebDAV, SFTP and more
WindowsmacOSLinux
Get Started Free →

Free core features. Plus automations available.

Why tiered backup with RcloneView​

  • Recovery speed: Recent files stay in S3 Standard for low-latency pulls.
  • Cost control: Warm copy in Wasabi/R2; deep archive in Glacier for cents per TB-month.
  • Resilience: Multi-cloud and multi-region reduce single-provider risk.
  • No scripting: Visual jobs, schedules, and logs instead of cron/YAML.
  • Proof: Compare jobs surface drift before you need to restore.

Architecture at a glance​

[Primary (NAS/Drive/OneDrive)] --(Hourly Sync)--> [S3 Standard hot copy]
                                         \
                                          --(Daily Sync)--> [Wasabi/R2 warm copy]
                                          --(Monthly Copy)--> [Glacier Deep Archive]
  • Hot: S3 Standard (fast restore).
  • Warm: Wasabi or R2 (cheap + egress-friendly for restores).
  • Cold: Glacier Deep Archive for 90-365 day retention.

Prerequisites​

  • Three remotes configured in RcloneView (e.g., s3:hot, wasabi:warm, s3:archive).
  • Versioning on hot/warm buckets; lifecycle rules on s3:archive for retention.
  • IAM/API permissions: list/read/write + multipart; Glacier restore permissions for cold tier.

Step 1: Build the hot and warm Sync jobs​

  1. Create Sync job (Primary -> S3 hot).
  2. Add performance flags in Advanced Options:
--s3-chunk-size=64M --s3-upload-concurrency=8 --checksum --checkers=8 --transfers=8
  1. Create a second Sync job (Primary -> Wasabi/R2 warm) using similar flags; add filters to skip temp/cache.
  2. Save both jobs.

Step 2: Add the cold archive Copy job​

  1. Create a Copy job (S3 hot -> Glacier archive bucket). Use Copy, not Sync, to avoid deletions on archive.
  2. In Advanced Options, keep lifecycle-friendly flags minimal:
--s3-storage-class=DEEP_ARCHIVE --s3-chunk-size=64M --s3-upload-concurrency=4 --checkers=4
  1. Ensure the archive bucket has lifecycle to expire old versions after compliance windows.

Step 3: Schedule everything​

  • Hot job: Hourly cadence with retries/backoff.
  • Warm job: Daily (nightly) to reduce egress impact.
  • Cold job: Monthly Copy (first Sunday) for compliance snapshots.
  • Add logging and notifications on all jobs.
  • Add a weekly Compare (hot vs warm) to catch drift early.
Configure the job scheduler in RcloneView

Step 4: Verify integrity​

  • Run Compare with --checksum between hot and warm after the first full sync.
  • For the archive, spot-check restores: issue a Glacier retrieval for a test prefix and confirm hashes locally.
  • Keep a small canary file (e.g., canary.txt) in every tier and check its presence in Compare reports.

Step 5: Restore playbook​

  • Fast restores: pull from warm (Wasabi/R2) or hot (S3 Standard) depending on egress/location.
  • Deep restores: initiate Glacier restore for the needed prefix; then Copy back to warm/hot.
  • Use Mount in RcloneView for browsing before bulk copy to avoid restoring the wrong folders.
--dry-run --checksum --no-traverse  # validate before destructive moves

Tuning tips​

  • Latency/egress sensitive: lower --transfers during business hours; raise off-hours.
  • Glacier costs: bundle archives monthly; avoid frequent small uploads to reduce PUT and retrieval requests.
  • Security: pair hot/warm with Object Lock (see immutable guide) to block ransomware deletes.
  • Naming: prefix snapshots with date folders (e.g., archive/2025-11/) to simplify lifecycle and restores.

Troubleshooting checklist​

  • Restore delays: Glacier retrieval can take hoursplan RPO/RTO accordingly.
  • AccessDenied on archive: ensure the IAM role allows glacier:InitiateJob/restore for the bucket.
  • Drift found: rerun hot/warm Sync; if archive missing objects, re-Copy the delta from hot.
  • Costs spiking: enable lifecycle expiration after retention; cap concurrency during peak egress windows.

Quick-start templates​

Job: Hot Sync (Primary -> S3 Standard)
Flags: --checksum --checkers=8 --transfers=8 --s3-chunk-size=64M --s3-upload-concurrency=8
Schedule: Hourly at :10, 2 retries (10m backoff)

Job: Warm Sync (Primary -> Wasabi)
Flags: --checksum --checkers=8 --transfers=6 --s3-chunk-size=64M --s3-upload-concurrency=6
Schedule: Nightly at 01:00, 2 retries

Job: Cold Copy (S3 Standard -> Glacier Deep Archive)
Flags: --s3-storage-class=DEEP_ARCHIVE --checkers=4 --transfers=2 --s3-chunk-size=64M --s3-upload-concurrency=4
Schedule: First Sunday monthly at 02:00
Verification: Weekly Compare (S3 hot vs Wasabi warm) with --checksum
Canary: keep /canary.txt in all tiers for health checks

Tier once, schedule always, and keep both costs and RPO under control with RcloneView.