Do you recognize this repetitive HDD sound? Looking for help...

Currently reading
Do you recognize this repetitive HDD sound? Looking for help...

17
1
NAS
918+
Operating system
  1. Windows
Mobile operating system
  1. Android
For some weeks now, I have been hearing a repeating sound for a long time.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.


It lasts for a while and then it stops again. Then it starts again. Quite random.
As soon as I do something with the DSM the sound stops. But if I leave the system alone for a while, the problem starts again.

I have DSM 6.2.4-25556 Update 6

I have already looked at the task manager of the resource monitor, but I cannot pinpoint which process or service is causing the disk activity.

It's starting to drive me crazy. I bought 6TB of WD60EFAX disks 2 years ago. I would easily buy new disks to get rid of the problem. However, I do not know if the problem is in the HD or if it is a software/OS problem. Disk health says all disks are okay.

Hopefully someone can help me what the problem is or how to investigate further.

Thanks!
 
Hi Fredbert, thanks for welcoming.

If I observe correctly the sound is mostly audible when the NAS get in a idle state. Or put it different; as soon as I use anything in the DSM interface the sound disappears. So I wonder how the fan can be a suspect. I know the fan speed could increase when the temperature rises. But sudden use of DSM could not affect fan drastically, I assume.

Am I wrong? Maybe the audio file has too much irrelevant noise. To me I recognise disc activity. Without a doubt. But only a weird kind of disc activity that is repetitive and lasts for a while.
Sure the disc made noise for many years. But never repetitive


Welcome to the forum.

Are you sure it’s the HDs? To me it sounded like air/fan, so would check out varying the fan speeds or physically stopping them. It should be easy to discount the fan it it’s not them.
 
On earphones there was a low frequency, warbling type of sound. I wondered if that was related to the fans or air mvement, and it's easy to test. As would be noises from the chassis. The HDs are sealed units so more tricky to investigate. But when you start to use the HDs then the heat will increase and, depending on your settings, the fans may pick up speed.

My NAS always have a constant whoosh from the fans with random clacks and rattles from the HDs.
 
I had to amplify the recording and therefore noise increased.

It is definitely 'clacks and rattles' from the HDs. But as said before the repetitive sound is odd.
 
I might be wrong, but are these 6TB EFAX the SMR disks? SMR disks are re-writing all the tracks after the initial write action to make sure the data is properly aligned. And causing noise for that reason.

SMR is blacklisted by most NAS suppliers.
 
I missed that. And you are right that WD60EFAX are the SMR WD Red, not the older CMR WD Red which became the CMR WD Red Plus. Here's a description of how SMR requires rewriting a whole segment when reusing one of its locations.


Here the link to the DS918+ compatibility list, filtered for NAS HDD.
 
I might be wrong, but are these 6TB EFAX the SMR disks? SMR disks are re-writing all the tracks after the initial write action to make sure the data is properly aligned. And causing noise for that reason.

SMR is blacklisted by most NAS suppliers.
Hi true, I am aware of that. But the problem is just recently.
 
Pull the network cable, does it stop?
Hi, if I pull the cable the noise stops. This will stop the noise. And it will increase disc activity shortly (as a result of loosing the connection). After a short while the recurring sound pattern emerges again.
 
This sounds to me like a server in your network that may send out broadcast to your other devices. Do you have a linux computer running, or a tv mediabox and no igmp snooping?
 
If I leave the Network cable operational and if the repetitive sound comes up, when I login than the repetition is gone for a moment. The same happens wheln I unplug the network. both cause a increase in disc activity. Making the noise go away. It seems that the increase in disc activity avoids the problem shortly. A hdd still remains a sequentual device. The disc heads cannot do things parallel. So for me it sounds logical that the problem stops as a result of increased disc activity. But the question is, what continues the problem after the increase in disc activity comes to an end
 
I did a extended S.M.A.R.T. test on all 4 drives (using DSM storage manager) . No problems found.
Can I therefore exclude that the sound is the cause of a failing hard disk? Or can you not rely on the SMART test?
 
My approach

Turn off the NAS.
Eject the disks from the NAS. Don't forget to mark which disk you had in which bay.

1. HW check
Now step by step, each disc separately:
Connect the disk via SATA/USB adapter to the computer.
Note: If you don't have it yet, buy it. It is a must-have for the NAS owner. Do not forget, you need to buy an adapter with an external power source.
Switch on the adapter.
Listen.
It is convenient to have the disk placed on a pad, which will allow reducing vibrations.
Feel free to change the pad to something hard. e.g. table and listen again.
Repeat with each disk.

2. NAS check
diskless NAS
turn on the NAS
Listen.
You are using 918+, you have two fans there.
If you didn't hear anything like that, turn off the NAS.
Check the dust around the fans. If necessary, dismantle, and clean. Similarly in the NAS case.

3. Insert the disks into the NAS.
Start the NAS and listen.
Open SSH and run:
Bash:
htop
Use F6 (SortBy) and select "IO_Write_Rate"
Listen and check the processes running with the noise + you will see Write IO Rate
Repeat it with "IO_Read_Rate"
Pull or plug the Ethernet cable when the noise is running then check the htop screen + listen.

So, this is my approach for such never observed noise. Without such tests, it's just shooting in the dark.

P.S.
Forget the Syno GUI SMART test and put here a result from:
Bash:
smartctl -x /dev/sd<x>
for each disk
when you don't know syntax use:
Bash:
smartctl --scan-open
post the result here
 
Thank you for your comprehensive help!

regarding 1 HW check. I do have external housing, used it for testing HD connected to W10 pc. Am I correct that you wan't to rule out any vibrations with your suggestion instead of do external surface test? There is no vibration issue.
So I want to make sure if you suggest this step only to rule out vibartion.
Removing each disc, mount them in external bracket for each disc would only useful to me if I use specific disc analysis tool to check disc condition. This will only be beneficial if I cannot rely on the extended test I did with SMART test of DSM storage manager. What's your viewpoint/advice?

regarding 2 NAS check
Has already been checked based on earlier suggestions

regarding 3
Cool! This will hopefully help me find the root cause. Will investigate tomorrow

regarding smartctl in P.S.:

smartctl 6.5 (build date Mar 2 2021) [x86_64-linux-4.4.59+] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Vendor: WDC
Product: WD60EFAX-68SHWN0
Revision: 0A82
User Capacity: 6,001,175,126,016 bytes [6.00 TB]
Logical block size: 512 bytes
Physical block size: 4096 bytes
LB provisioning type: not reported [LBPME=1, LBPRZ=0]
Rotation Rate: 5400 rpm
Form Factor: 3.5 inches
Logical Unit id: 0x50014ee26791de4d
Serial number: **serial removed
Device type: disk
Local Time is: Wed Aug 3 16:42:51 2022 CEST
SMART support is: Unavailable - device lacks SMART capability.
Read Cache is: Enabled
Writeback Cache is: Enabled

=== START OF READ SMART DATA SECTION ===
Current Drive Temperature: 0 C
Drive Trip Temperature: 0 C

Error Counter logging not supported


[GLTSD (Global Logging Target Save Disable) set. Enable Save with '-S on']
Device does not support Self Test logging
Device does not support Background scan results logging
 
Am I correct that you wan't to rule out any vibrations with your suggestion instead of do external surface test?
correct

Removing each disc, mount them in external bracket for each disc would only useful to me if I use specific disc analysis tool to check disc condition
negative
It will help you to discover vibrations, if they exist from another drive

SMART support is: Unavailable - device lacks SMART capability.
sucks
 
My approach

Turn off the NAS.
Eject the disks from the NAS. Don't forget to mark which disk you had in which bay.

1. HW check
Now step by step, each disc separately:
Connect the disk via SATA/USB adapter to the computer.
Note: If you don't have it yet, buy it. It is a must-have for the NAS owner. Do not forget, you need to buy an adapter with an external power source.
Switch on the adapter.
Listen.
It is convenient to have the disk placed on a pad, which will allow reducing vibrations.
Feel free to change the pad to something hard. e.g. table and listen again.
Repeat with each disk.

2. NAS check
diskless NAS
turn on the NAS
Listen.
You are using 918+, you have two fans there.
If you didn't hear anything like that, turn off the NAS.
Check the dust around the fans. If necessary, dismantle, and clean. Similarly in the NAS case.

3. Insert the disks into the NAS.
Start the NAS and listen.
Open SSH and run:
Bash:
htop
Use F6 (SortBy) and select "IO_Write_Rate"
Listen and check the processes running with the noise + you will see Write IO Rate
Repeat it with "IO_Read_Rate"
Pull or plug the Ethernet cable when the noise is running then check the htop screen + listen.

So, this is my approach for such never observed noise. Without such tests, it's just shooting in the dark.

P.S.
Forget the Syno GUI SMART test and put here a result from:
Bash:
smartctl -x /dev/sd<x>
for each disk
when you don't know syntax use:
Bash:
smartctl --scan-open
post the result here
Okay. I found the problem, thanks to your tips.

Your tip with HTOP brought the solution. There was sound while I could not observe any read and write activity in htop. I checked the htop manual to make sure I was interpreting the data correctly.

Then I removed the casing from the nas and put the disks back, to see if I could detect vibration with my fingertips. That turned out to be quite difficult, but I managed to feel the vibration.

I put that disk in an external enclosure. and after a while you hear the problem sound.

Thanks for your help!
 
you welcome
-------------------------------

Back to the smartctl result at #17:
Product: WD60EFAX-68SHWN0
Revision: 0A82
SMART support is: Unavailable - device lacks SMART capability.
it is very interesting that your output really contains only basic information. Which would indicate what is stated:
SMART support is: Unavailable - device lacks SMART capability.
However, if I looked in the smartmontools tickets (3y old record):
Device Model: WDC WD60EFAX-68SHWN0
Firmware Version: 82.00A82
contains the normal (extended) output I would expect.

However, if you check a detail, the recording method for the Device model:
WD60EFAX-68SHWN0
vs
WDC WD60EFAX-68SHWN0

and

Firmware:
0A82
vs
82.00A82

both have been changed. It is a peanut that only an attentive eye will notice. But it suggests that the same firmware can contain two different SMART records.
I'm thinking that the Product Managers at WD do not have clearly set guidelines for SMART standardization. Which is something similar to the WD Blue SSD range case mentioned here 3 months ago.
... too many coincidences from this vendor.

My advice, if you buy a new drive, don't be lazy and test the drive before using it in the NAS (or in any other device, computer, laptop, ...). Make a copy of the result.
For such purpose use the above-named command:
first
Bash:
smartctl -x /dev/sd<x>
Note: if the output displays a message - SMART support is: "Unavailable - device lacks SMART capability." So return the drive back to the shop. A drive without SMART support is like a TV without HDMI or RJ45 input. If any SMART ID counter (RAW VALUE or VALUE) reports a number greater than zero, return it. As if you bought a new car and the "engine" light was on.
Don't be afraid to ask about something you don't understand.

then use a Long SMART test:
Bash:
smartctl -t=long /dev/sd<x>

Finally when you can't read the test output as pro-grade, then use:
Bash:
smartctl -H /dev/sd<x>


Some useful notes:
WD for three years now, for an unknown reason, has been calculating (interpreting) the completely opposite value of the Media Wearout Indicator. So such a drive starts with a value of ZERO, which should mean that it is suitable for throwing in the trash.

Last:
Bash:
smartctl --scan-open
will help you to find appropriate naming for your drives
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Similar threads

Replies
9
Views
1,434
thanks @Telos @Coop777 and @fredbert . Now I got the device and doing the migration (from my older NAS)...
Replies
10
Views
1,271
  • Question
If you " really don't want to lose all that data": is there no backup? Any disk will fail one day.
Replies
2
Views
764
  • Locked
There is an unofficial way to activate SHR support on XS+ models. Search the forum for a solution...
Replies
6
Views
1,903

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Back
Top