Synology Hyper Backup
Hyper Backup on DSM backs up your NAS to DanubeData Object Storage over the S3 API. This page covers first-time setup, and the change you need to make if your backups started failing on or after 23 August 2026.
Action needed before 30 August 2026
On 23 August 2026 we applied a security update to DanubeData Object Storage. It requires that a Content-Type header, when present, is covered by the request signature — see S3 Signature Requirements for the full rule and the reasoning.
Hyper Backup bundles the AWS SDK for PHP, which does not cover that header when signing with Signature Version 4. Backup tasks configured that way are rejected.
You are not blocked right now. A temporary compatibility setting keeps existing tasks working until 30 August 2026. After that date, a Hyper Backup task still set to Signature Version 4 will fail.
The fix is to switch the task to Signature Version 2, which covers Content-Type as part of the signature by design. Your existing backups do not need to be re-uploaded.
Am I affected?
You are affected if all of these are true:
- you back up to DanubeData Object Storage using Hyper Backup, and
- the task's Signature Version is set to V4, and
- the task is on DSM 7.x
To check, open Hyper Backup, select the task, open its destination settings, and look at the Signature Version field.
Symptoms once the compatibility window closes:
- the task fails partway through with a permission or connection error
- the task log shows
403orAccess Deniedon upload - restores and browsing existing backup versions still work, because those are read operations
If your task already uses Signature Version 2, no action is needed.
Why the setting cannot simply be edited
Hyper Backup does not allow the Signature Version or the server URL of an existing task to be changed after the task is created. The destination panel only lets you re-enter the access key and secret key.
The supported way to change a destination setting is to create a new task and use Relink to existing task, which reuses the backup data already in your bucket instead of uploading everything again.
Fix: new task with Signature Version 2, relinked
Do this before 30 August 2026. It takes a few minutes and does not re-upload your data.
Before you start, note your current task's settings — you will re-enter them. Open the existing task's settings and record: the bucket name, the directory, which folders and applications are backed up, the schedule, the retention/rotation policy, and whether encryption or compression is enabled. A screenshot of each tab is the quickest way.
Steps
- Open Hyper Backup and create a new backup task with the + button.
- Choose S3 Storage as the backup destination.
- On the Backup Destination Settings page, set:
- S3 Server:
Custom Server URL - Server address:
s3.danubedata.ro(ornew-s3.danubedata.roif that is what your current task uses). Do not includehttps://— Hyper Backup adds it. - Signature Version: V2
- Access Key and Secret Key: the same keys your existing task uses
- Bucket Name: select your existing bucket from the dropdown
- Directory: select the existing backup directory, the one ending in
.hbk
- S3 Server:
- Choose Relink to existing task rather than creating a fresh backup. This reuses the backup data already in the bucket.
- Work through the remaining wizard pages, re-applying the source folders, applications, schedule and retention settings you noted earlier.
- Run the task once, manually, and confirm it completes successfully.
- Only after that succeeds, remove the old task.
Keep the old task until the new one has completed a successful backup. Your backup data stays in the bucket either way, so a failed relink costs you time, not data. Deleting the old task first removes your ability to fall back.
If the bucket dropdown does not populate
The bucket list is fetched using the keys you entered. If it stays empty, the keys are wrong or lack access to that bucket — re-enter them, and confirm in your dashboard that the key has access to the bucket in question.
Path-style addressing
If your current task addresses the bucket in path style — the bucket appears after the server address, as in s3.danubedata.ro/your-bucket — note that Synology documents path-style buckets as supported only with Signature Version 4.
Selecting Signature Version 2 therefore also moves the task to virtual-hosted-style addressing, where the bucket name becomes part of the hostname (your-bucket.s3.danubedata.ro). Both of our endpoints support both styles, so no change is needed on our side, and the relink process is otherwise identical.
If you are unsure which style your task uses, open a ticket and we will tell you from your bucket's request logs.
Older DSM versions
DSM 6.x ships a much older Hyper Backup and a different S3 client, and may not expose a Signature Version control at all. If your NAS runs DSM 6.x, open a ticket before changing anything — we will check your bucket's request logs, confirm whether your backups are actually affected, and advise from there. Do not delete or recreate a working task on the basis of this page.
First-time setup
For a new backup to DanubeData Object Storage:
- Open Hyper Backup and create a task with +.
- Choose S3 Storage.
- Configure:
- S3 Server:
Custom Server URL - Server address:
s3.danubedata.ro(withouthttps://) - Signature Version: V2
- Access Key / Secret Key: from your dashboard, under your bucket's Access Keys tab
- Bucket Name: select your bucket
- Directory: the folder to store the backup in
- S3 Server:
- Select the folders and applications to back up.
- Set the schedule, retention policy, and optionally encryption and compression.
- Apply, then run the task once to confirm it works.
Your access key needs write access to the bucket. See S3 Access Keys for how to create one.
Troubleshooting
| Symptom | Cause and fix |
|---|---|
403 / Access Denied on upload, restores still work | Signature Version is V4. Follow the fix above. |
Failed to connect to the server | Check the server address has no https:// prefix and no trailing slash. Confirm the NAS can resolve s3.danubedata.ro. |
Failed to read the bucket list | Authentication failed. Re-enter the access key and secret key, and confirm the key has access to that bucket. |
| Bucket dropdown is empty | Same as above — the keys are wrong or lack access to the bucket. |
| Certificate or SSL errors | Update DSM so its CA certificate store is current. |
PartNotFound or InvalidPart | A multipart upload was interrupted. Cancel the run and start a new one; the next run re-uploads the affected files cleanly. |
| Task reports the destination is offline | The backup data cannot be found at the configured path. Confirm the bucket and directory, then relink as described above. |
Alternatives
If Hyper Backup does not suit your needs, a Synology NAS can also reach DanubeData Object Storage through Cloud Sync (file-level sync rather than versioned backup) or through rclone, which can be run from the NAS. Note that Cloud Sync mirrors files rather than keeping the versioned, deduplicated backup history Hyper Backup produces, so it is not a like-for-like replacement for a backup task.
Need help?
Open a ticket from your dashboard. We can see from your bucket's request logs whether your backups are currently succeeding, which signature version and addressing style your task uses, and whether anything you run will break on 30 August 2026. If you would like us to walk through the relink with you before making changes, say so in the ticket.