Snapshot Creation Failed on Synology: Fix in 3 Steps

30-Second Fix: Check Free Space

Snapshots are space-hungry. The most common reason they fail? Your volume is nearly full. Synology needs at least 5% free space on the volume where the snapshot lives. If you're under that, the system won't even start the snapshot process.

Open Storage Manager → Storage Pool → check the volume's free space. If it's below 5%, you need to free up room. Delete old snapshots, move files off, or delete trash files (check the Recycle Bin).

One specific trigger: if you've been running a backup job that uses snapshots, and that volume is also your Time Machine backup target, those .sparsebundle files can eat space silently. Check hidden files in the home directory.

If you’ve got space and the error persists, move to the next step.

5-Minute Fix: Check Btrfs-Specific Limits

What's actually happening here is that Synology's Btrfs filesystem has a hard limit on the number of snapshots per shared folder – max 256. You might hit this without knowing it, especially if you've got automatic snapshot schedules running for months. The system logs the error as generic “snapshot creation failed” but it's really a quota issue.

Go to Snapshot Replication → Shared Folder Snapshot. Select the folder that's failing. Check the snapshot count in the bottom panel. If it's close to 256, you need to prune. Delete old snapshots – keep only the ones from the last week. I suggest keeping at most 32 hourly snapshots and 7 daily ones. Anything more is noise.

Also, Btrfs has a bug on DSM 6.2.4 and earlier: if the snapshot schedule uses “repeat every hour” without a max age, it can create a snapshot every hour forever until you hit 256. Upgrade to DSM 7.x or later to avoid this. On DSM 7, the system automatically prunes old snapshots when you set a retention policy. But even then, I've seen it fail if you set retention to “keep all” – don't do that.

If pruning fixes it, you're done. If not, something's wrong with the filesystem itself.

15+ Minute Fix: Repair the Btrfs Filesystem

If space and snapshot count are fine, the problem is almost certainly a corrupt Btrfs metadata tree. This happens after unexpected shutdowns, disk failures, or just bad luck. The snapshot creation process writes new metadata, and if the tree has an inconsistency, the write fails.

Start by checking the health of your disks. Run a Storage Manager → HDD/SSD → SMART test on all drives. Even one failing disk can cause metadata corruption. Replace any that show reallocated sectors or pending errors. Don't bother with long tests if you're in a hurry – quick test will catch the obvious issues.

If disks are fine, you need to run a Btrfs scrub and repair. SSH into your Synology:

sudo btrfs scrub start /volume1

Replace /volume1 with your actual volume path. This scans all data and metadata, and fixes any checksum errors. It can take hours on large volumes. Run it in a screen session or tmux if you're SSHing from a remote terminal – don't want it killed by a dropped connection.

After the scrub completes (check with btrfs scrub status /volume1), try creating a snapshot manually via Snapshot Replication. If it still fails, you've got deeper corruption. Next step is to run a read-only check:

sudo btrfs check --readonly /dev/mapper/cachedev_0

You need the right device mapper name. Run lsblk first to find it – it'll be something like cachedev_0 if you use SSD cache, or vgX-lvY for RAID volumes. This won't fix anything, just report errors. If you see “errors found” but no repair path, you have two options:

• Backup and restore – Copy all data off, reformat the volume, copy back. This is the nuclear option but the most reliable.
• Contact Synology support – They have internal tools that can repair some Btrfs issues their GUI can't. It's free for the first year. After that, you're on your own. I recommend the backup route anyway – trust the filesystem less after corruption.

One last thing: if you're running DSM 7.2 and using the new “folder-level encryption” on a snapshot-encrypted folder, there's a known bug where snapshot creation fails if the encryption key is rotated. The workaround is to disable snapshot encryption, take the snapshot, then re-enable. Not ideal, but it's what works.

Bottom line: 90% of snapshot failures are space or quota. 9% are Btrfs metadata. 1% are real hardware. Fix in that order.