How to replace a failing drive inside a NAS without losing data

No drive will last forever. Whether it’s a hard disk drive (HDD) or solid-state drive (SSD), inside network-attached storage (NAS), or a desktop PC, it will inevitably fail. It’s something we know will happen but planning for a drive failure inside a NAS can take us by surprise. This guide will run you through what to do should you encounter one of your HDDs or SSDs giving up.

Firstly; don’t panic!

So long as you have backups, you’re good

The humane response to a self-monitoring, analysis, and reporting technology (S.M.A.R.T.) alert on a drive is panic but I recommend staying calm. Just because a S.M.A.R.T. check has deemed one of your drives to be on its way out, it doesn’t mean you’re set to see a failure right now. The first task is to ensure your backups are all updated. Ensure everything stored on the device has multiple copies and is stored across different systems to avoid data loss. Running drives in a RAID can even allow you to replace and repair the storage pool without losing a single file.

Related

What is RAID (and how to set it up in a NAS)

RAID is worth checking out if you desire fast transfer speeds or want to protect your data against drive corruption (or even both, at the same time)

After you’ve made the necessary backups, run a S.M.A.R.T. health check on all drives, as well as a manufacturer-supplied tool if available — see the IronWolf health check in our above screenshot. This will provide insight into how your drives are performing, whether they have bad sectors, are starting to fail, and need replacing. Each drive should have a status, Synology’s DiskStation Manager (DSM) uses “Healthy” representing a drive in working order, “Warning” for any that require attention, and “Failing” for drives that should be replaced as soon as possible.

Source a replacement drive

Remember to buy drives designed for NAS

Source: XDA

It’s important to replace the failing drive with one that’s equal to or larger in capacity. Depending on the RAID you’re using on the NAS, this will impact the available storage space for storing data. Upgrading the capacity is a good idea when replacing a failed drive, hitting two birds with a single stone. Like any NAS or server, it’s vital to pick a drive designed for continuous operation. Standard consumer-grade drives for a laptop or desktop PC won’t cut it and you could experience a fault sooner than you would with a NAS drive.

Related

Best hard drives for NAS

If you’re looking for new hard drives to increase the storage capacity of your NAS, you’ve come to the right place

Note that it’s not a good idea to mix HDDs and SSDs in a RAID, so if you have hard drives, grab new mechanical drives. The same goes for SSDs. Combining both into a single RAID would cause issues such as performance imbalance, excessive wear through rebuilds and maintenance, and having to manage mismatched drives in terms of speed and capacity. If a drive or two fail and you’ve had an SSD upgrade in mind, you will need to replace all drives with SSD counterparts.

Replace and rebuild

One out, another in

Depending on the NAS or server in question, it may be possible to how-swap the failing drive with its replacement. This can be as straightforward as removing the failed drive from the array. You can safely do this with the system powered on and although the NAS won’t like the drive being removed, it will hold out until you install a new one to rebuild the RAID. If your NAS does not support drive hot-swapping, you must power off the system before disconnecting the drive from the SATA port.

Some NAS operating systems, including Synology’s DSM, will even allow you to configure spare drives in free bays. These can then be spun up and used should another drive within the RAID fails. Once a fresh drive has been detected by the OS, it should be an easy process to add it to the existing RAID and have it rebuilt itself. The duration of this depends on how much data is stored on the RAID and which type of RAID you are using. When finished, the NAS will be fully functional again without losing a single byte.

Don’t rely on RAID for data redundancy

RAID is fantastic and works well with multiple drives to create storage pools with certain tolerances, but you shouldn’t rely on a RAID as a means of backing up your data. Things can go wrong and you could lose more drives than the RAID can recover from. Always follow the 3-2-1 backup rule (or better) when working with data you cannot afford to lose.