Hard Drives for use in my Synology NAS

 Why I purchased Seagate IronWolf drives

I recently purchased two 6TB Seagate IronWolf drives. They are similarly priced to the WD Reds but have a faster RPM of 7200 compared to 5400 on the Reds. The IronWolf also have 128MB of cache vs 64MB on the Red. To get a drive similar in spec to the IronWolf would require purchasing Red Pro drives, which are a lot more expensive.

Using IronWolf or Red for NAS units

My Synology DS916+ can hold up to 4 hard drives. I currently occupy 3 slots with:

• Two WD Green 5TB drives
• One Toshiba 5TB drive

I want to replace one of the WD Green drives with the new IronWolf and add the second IronWolf as the 4th drive. 

Additional storage space is just one of the reasons for the swap, but the other is to get NAS grade hard drives. But what does that mean? Seagate marketing states IronWolfs come with:

• Rotation Vibration (RV) mitigation, which helps with 'performance and reliability' when you have multiple drives in a unit.
• Advanced power management
• IronWolf Health Management (IHM), which is compatible with different NAS units- including Synology. It is similar to SMART, but better in the ability to track and analyze SMART (plus additional drive parameters) through time for a more reliable health and life-span report.
• RAID performance Error Recovery Control

Can I make my existing drives 'NAS Ready' ?

Advanced power management (APM) includes how long the drive stays awake before the hard drive parks it's head and spins down. This can affect lifespan and performance latency of the drive. This can be controlled on WD drives on Linux/Synology with the package idle3-tools. 

ash-4.3# ./idle3ctl -d /dev/sda
Idle3 timer disabled

The same can be done for my Toshiba drive with hdparm:

ash-4.3# hdparm -B 255 /dev/sdc

/dev/sdc:
 setting Advanced Power Management level to disabled
 APM_level = off

Error Recovery Control (ERC), for RAID arrays, essentially tells the hard drive how long to try and recover from an error. This is important in a RAID setup where other drives can pick up the slack if the current disk is having read/write issues with a file- so it should not make the RAID controller wait long. In a single disk setup, you want the hard drive to try and recover from the error for as long as it needs since it is your only source for the data. The smartctl utility allows us to configure this if allowed.

This is what my new IronWolf drive shows:

ash-4.3# smartctl -l scterc /dev/sdd
smartctl 6.5 (build date Mar  7 2018) [x86_64-linux-3.10.102] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

SCT Error Recovery Control:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)

Let's see what my two WD Green drives report:

ash-4.3# smartctl -d sat -l scterc /dev/sda
smartctl 6.5 (build date Mar  7 2018) [x86_64-linux-3.10.102] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

SCT Error Recovery Control command not supported

Unfortunately, WD disables ERC on their Green drives. Apparently this was not the case almost a decade ago. But it makes sense WD would prefer to sell Red drives instead of modifiable Green drives.

Maybe I will have better luck with the Toshiba...

ash-4.3# smartctl -d sat -l scterc /dev/sdc
smartctl 6.5 (build date Mar  7 2018) [x86_64-linux-3.10.102] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

SCT Error Recovery Control:
           Read: Disabled
          Write: Disabled

Great! Since ERC is disabled, what happens if I try to enable it?

ash-4.3# smartctl -d sat -l scterc,70,70 /dev/sdc
smartctl 6.5 (build date Mar  7 2018) [x86_64-linux-3.10.102] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

SCT Error Recovery Control set to:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)

It looks like it worked! But it will get reset after a reboot. A boot-up script that runs that command will need to be created in Task Scheduler. 

Since I had better luck trying to convert the Toshiba drive into a "NAS/RAID" optimized drive, I decided to replace one of the WD Green drives instead.

Are these things critical and needed in a home environment? Probably not, but it's fun to tinker and get all the drives to try and behave the same.