Synology checksum mismatch errors? What to do first

Your Synology NAS has started reporting “checksum mismatch” errors, or a data scrubbing run has finished with errors. Perhaps a few photos now open half grey, a spreadsheet will not load, or Log Center lists file paths you did not expect to see there. These warnings are your NAS telling you that some of the data it read back is not the same as the data it originally stored. That is worth taking seriously, but it is not a reason to panic or to start clicking repair options. This guide explains what the warnings mean, what usually causes them and the safe order of steps to protect your files.

What a checksum mismatch actually means

When a shared folder on a Btrfs volume has data checksums enabled, DSM stores a small fingerprint, called a checksum, for each block of data it writes. Every time that block is read back, DSM recalculates the fingerprint and compares the two. If they differ, the data has changed somewhere between being written and being read, and DSM logs a checksum mismatch.

What happens next depends on how your storage is set up:

• With redundancy (SHR with two or more disks, RAID 1, 5, 6 or 10) and data checksums enabled, DSM can often read a good copy of the block from the redundant data and repair the damaged one automatically. Synology calls this file self-healing.
• Without redundancy (a single disk, Basic, JBOD or RAID 0), there is no second copy to repair from. DSM can tell you the file is damaged, but it cannot fix it.
• On ext4 volumes, or shared folders without data checksums, this detection does not happen at file level at all, so silent damage may go unnoticed until someone opens the file.

Data scrubbing, which you can run or schedule from Storage Manager, reads through the whole volume to find these problems before you stumble on them.

Common causes

• A failing hard drive. Drives developing bad or unstable sectors are the most frequent cause. Look for rising reallocated or pending sector counts in the drive’s health information.
• Memory problems. Faulty or incompatible RAM, including third-party memory added to upgrade the NAS, can corrupt data before it reaches the disks.
• Cables, backplanes and drive bays. On units with removable trays, a poorly seated drive or a fault in a bay can cause read errors.
• Power cuts and improper shutdowns. Interrupted writes can leave blocks that no longer match their checksums.
• A degraded array. If one disk has already failed, the redundancy that self-healing relies on is no longer there, so any further errors cannot be corrected.

What not to do

• Do not run file system repair commands over SSH, such as Btrfs check with the repair option. These can make a damaged volume unreadable and are not meant as a first response.
• Do not replace or pull several disks at once to “find the bad one”. If the array is already degraded, removing another disk can take the whole volume offline.
• Do not delete or overwrite the files listed in the log until you know whether good copies exist elsewhere. Those paths tell you exactly what to restore.
• Do not start a RAID repair or expansion while the NAS is reporting errors and you have no up-to-date copy of the important data.
• Do not rely on snapshots alone. Snapshots on the same volume share the same underlying blocks, so a damaged block can affect the snapshot copy too.

Safe steps, in order

1. Write down the details. Export or screenshot the Log Center entries, including the file paths and dates, and note your RAID type and DSM version.
2. Check disk health. In Storage Manager, look at each drive’s status and health information. A drive showing warnings, or a storage pool marked as degraded, changes the priorities.
3. Copy your most important data off to a separate external drive or another storage device, starting with anything that is not backed up. Do this before running further scrubs or making hardware changes, because heavy reading can push a weak drive over the edge.
4. Restore damaged files from a backup held outside the NAS, such as Hyper Backup to a USB drive or a cloud destination, if you have one.
5. Consider the memory. If you have added non-Synology RAM, refer to Synology’s own guidance on testing memory and consider removing the extra module once your data is safe.
6. Replace a failing disk only once your data is protected, one disk at a time, and let the repair finish before doing anything else.

When to stop and get help

Stop and switch the NAS off if the volume becomes read-only or crashed, more than one disk shows warnings, the errors keep growing with each scrub, or the files being reported are ones you have no other copy of. At that point continued use risks turning a handful of damaged files into a failed volume. Professional NAS recovery normally starts by imaging each disk, so the array can be reassembled and the file system read from copies without writing anything to the original drives. See our Synology NAS data recovery page, or our wider NAS data recovery service for other brands.

Talk to us before you repair anything

If your Synology is reporting checksum mismatches and the affected files matter to you, call Data Retrieval on 020 7064 4996. We offer a free assessment with no obligation, so you will know where you stand before any work is carried out. You can also request a quote online.