Synology Glacier Backup: How It Works

Synology Glacier backup lets you send data from a Synology NAS to Amazon S3 Glacier, a low-cost archival storage service, using the built-in Cloud Sync tool. You set up a connection to your AWS account, choose a folder on the NAS and a bucket in S3, and Cloud Sync handles the rest. This article explains what Glacier backup actually does, how to configure it, what it costs, and where it falls short.

What Amazon S3 Glacier Is

S3 Glacier is an archival storage tier from Amazon Web Services, designed for data you rarely access but must keep for months or years. It is much cheaper per gigabyte than standard S3 storage, but retrieval is slow and costs extra. Glacier is not a backup service by itself; it is a destination. You still need a tool on your side to copy files there, and Synology’s Cloud Sync is that tool for NAS owners.

Glacier has several retrieval options. Expedited can return data in minutes but costs the most. Standard takes 3-5 hours. Bulk can take up to 12 hours and is the cheapest. For a home or small business backup, Standard retrieval is the usual choice. You cannot open a file directly from Glacier like you would from a normal folder; you must first initiate a restore job, wait for it to complete, then download the data.

How Synology Cloud Sync Works with Glacier

Cloud Sync is a package you install from the Synology Package Center. It supports many cloud providers, including Amazon S3 and S3 Glacier. When you set up a connection to Glacier, Cloud Sync treats it like a remote folder. It can upload new files, download files, and sync deletions, depending on the mode you pick.

The key detail is that Cloud Sync does not use the Glacier API directly. Instead, it uploads files to an S3 bucket that has a lifecycle rule to move objects to Glacier after a set number of days. You create the bucket in the AWS console, add a lifecycle rule (for example, move to Glacier after 30 days), then point Cloud Sync at that bucket. Files are stored in S3 Standard temporarily, then transitioned to Glacier automatically.

This matters because it affects how you restore. If you try to download a file that has already moved to Glacier, the request fails or takes hours. To restore, you must either temporarily change the lifecycle rule, use the AWS console to initiate a restore, or use the S3 API. Cloud Sync itself does not handle Glacier restores well; it often shows an error or a hanging task.

Setting Up Synology Glacier Backup

The setup process has several steps, and getting them right avoids problems later. Here is the general workflow:

• Create an AWS account and note your access key and secret key.
• Create an S3 bucket in a region near you.
• Add a lifecycle rule to that bucket: transition to Glacier after X days (30 is common).
• On the Synology NAS, install Cloud Sync from Package Center.
• Open Cloud Sync, choose Amazon S3, and enter your AWS credentials.
• Select the bucket, choose a local folder to sync, and pick a sync mode (one-way upload is typical for backup).
• Set a schedule if you want uploads to happen at night.

One-way upload means files go from NAS to cloud only; deletions on the NAS do not delete cloud files. This is the safest mode for backup. Two-way sync mirrors changes both ways, which can accidentally delete cloud data if you delete a local file by mistake. For a true backup, use one-way upload.

Another option is to use Hyper Backup instead of Cloud Sync. Hyper Backup has a dedicated S3 Glacier destination and handles versioning and restore more gracefully. It also compresses and deduplicates data, which can lower costs. The tradeoff is that Hyper Backup stores data in its own format, so you cannot browse files directly; you must use Hyper Backup to restore.

Costs and Retrieval Fees

Glacier’s appeal is low storage cost, but the total bill depends on more than the per-gigabyte price. You pay for storage, upload requests, retrieval requests, and data transfer out. Uploading to S3 Standard first means you pay S3 Standard rates for the first 30 days (or whatever your lifecycle rule says), then Glacier rates after that. If you delete a file before the minimum 90-day storage period, you pay an early deletion fee.

Retrieval is where costs can surprise you. Glacier charges per request and per gigabyte retrieved. A large restore can cost more than the original upload. AWS also charges for data transfer out of the region, which applies when you download from Glacier to your NAS. If you restore frequently, Glacier is not the right choice; a normal S3 bucket or a different backup service would be cheaper.

There is also a minimum object size of 128 KB for Glacier. Smaller files are charged as if they were 128 KB, so many tiny files inflate the bill. If your NAS holds thousands of small config files, Glacier may not be cost-effective.

Common Problems and Misconceptions

The most common issue is trying to download a file that has already transitioned to Glacier. The download fails with an error like “operation failed” or hangs indefinitely. The fix is to restore the object in AWS first, then retry. This is not obvious from the Synology interface, and many users think the backup is broken when it is just the nature of Glacier.

Another misconception is that Glacier is a backup service. It is not. It is a storage class. If you accidentally delete files on your NAS and the sync is set to two-way, those deletions propagate to the cloud, and your Glacier backup is gone. Glacier does not protect against ransomware either; if your NAS is infected and the sync runs, encrypted files overwrite the good ones in the cloud. For real protection, you need versioning or immutable backups, which Glacier alone does not provide.

Also, Glacier is not designed for compliance with specific regulations on its own. You must configure the bucket policy, enable versioning, and set appropriate retention rules yourself. AWS offers S3 Object Lock for this, but it is not automatically enabled.

Glacier vs. Other Synology Backup Destinations

Synology offers several cloud backup targets, and Glacier is only one of them. Here is how it compares:

• S3 Standard: Fast retrieval, higher storage cost. Good for data you might need soon.
• S3 Glacier: Cheap storage, slow and costly retrieval. Good for long-term archives.
• S3 Glacier Deep Archive: Even cheaper, but retrieval takes up to 12 hours. Good for data you almost never need.
• Backblaze B2: Flat pricing, no retrieval fees. Often simpler for home users.
• Wasabi: No retrieval fees, but a minimum storage duration.
• Synology C2: Synology’s own cloud, integrates tightly with Hyper Backup, but more expensive per GB.

For most home users, Glacier’s complexity and retrieval fees outweigh the storage savings. If you have a few hundred gigabytes of family photos you will never touch, Glacier is fine. If you have a few terabytes and might need to restore after a drive failure, a service with free retrieval is easier to live with.

Security and Encryption

By default, Cloud Sync encrypts data in transit using TLS. For encryption at rest, you have two options: use AWS’s server-side encryption (SSE) on the bucket, or enable client-side encryption in Cloud Sync. Client-side encryption means files are encrypted on the NAS before upload, and only you hold the key. This is stronger because even AWS cannot read your files. However, if you lose the encryption key, you lose the data. Store the key somewhere safe, not just on the NAS.

If you use server-side encryption, AWS manages the keys, and you can still read files through the AWS console. This is simpler but less private. For sensitive data, client-side encryption is the right call. Cloud Sync also supports encryption for S3 Glacier connections, but the interface is not always clear about which encryption is active. Check the settings before enabling sync.

Frequently Asked Questions

Can I use Cloud Sync to back up to Glacier directly? Yes, but it works through an S3 bucket with a lifecycle rule, not the Glacier API directly. Files land in S3 Standard first, then move to Glacier after the rule triggers.

Why can’t I download a file from my Glacier backup? Because the file has already transitioned to Glacier. You must restore it in AWS first, which takes hours, then download it. Cloud Sync does not do this automatically.

Is Glacier backup free? No. You pay for storage, requests, and data transfer. The storage cost is low, but retrieval and early deletion can add up. There is no free tier for Glacier.

What is the difference between Cloud Sync and Hyper Backup for Glacier? Cloud Sync gives you browsable files but poor restore handling. Hyper Backup uses a proprietary format but handles restore and versioning better. For a true backup, Hyper Backup is usually the safer choice.

How long does it take to restore from Glacier? Standard retrieval takes 3-5 hours, Bulk up to 12 hours, and Expedited a few minutes but costs more. The restore job must complete before you can download the data.

Can I use Glacier for a ransomware-proof backup? Not by itself. If your NAS is compromised and sync runs, encrypted files can overwrite the cloud copies. You need versioning or Object Lock, which you must configure separately.

Related

• Baldur’s Gate 3 Crashing: Fixes That Work
• Disk Inserted Not Readable? Fixes That Work
• Verify x402 Payment Requests Before API Calls