always go for double redundancy

Posted on by Daniel Kirstenpfad

As the replacement drive for yesterdays hard drive crash was put into place the storage array started to re-silver the newly added empty drive. This process takes a while – about 8 hours for this particular type of array.

Interestingly just 2 minutes into the process another drive dropped a bombshell:

Apparently disk 8 holds together it’s business so far but dropped a couple of parity errors into the equation.

This is bad news. But so far science still is on my side of things and no data has been lost.

But now redundancy is down completely. There’s no redundancy for now – until the replaced hard disk is fully integrated. My policy for these sized drives demands a minimum of 2-disk redundancy and for today this policy saved the day (data).

Actually let’s dive a bit into what it’s doing there to achieve 2-disk redundancy:

Synology Hybrid RAID (SHR) is an automated RAID management system from Synology, designed to make storage volume deployment quick and easy. If you don’t know much about RAID, SHR is recommended to set up the storage volume on your Synology NAS.

You will learn different types of SHR and their advantages/disadvantages over classic single disk/RAID setups. In the end, you will be able to choose a type of RAID or SHR for the best interest of your storage volume. This article assumes that as the admin of your Synology NAS, you are also an experienced network administrator with a firm grasp of RAID management.

Synology Hybrid Raid

So you trade data redundancy and safety for useable disk space. Here this is compared to traditional RAID 5:

disk down, rinse and repeat.

Posted on by Daniel Kirstenpfad

After the sudden death of a hard drive in one of the house’ storage arrays (after 55997 hours of service) beginning of this month it has happened again:

With less than half the runtime of the previously dead disk this one is an early failure. Well within the warranty. Therefore the disk is already en route to be replaced by an RMA (Western Digital RMA process so far is spotless!)

Anyhow: This was a 4 TB drive. It’s in an array with 2-disk redundancy and 8 other drives. So the array still is operating with redundancy right now. Additionally a full backup exists as well as a hot-standby (but slower) offsite mirror.

I am quite confident to not loose data. But this raid-sync is going to take a bit longer. As drives get bigger, syncs get longer.