HWINFO: HDD clicking, when SMART monitoring is enabled.

HDDs have heads and clicking noises coming from it is either the drive moving these heads, parking or unparking them. Apparently there are models which do this when they're being queried for their SMART attribute values or ATA Statistics. I've never seen this in my personal experience, but I only buy WD drives, so...

Is it normal behavior for drives to be doing that? I don't know. In my opinion, all a drive should do when the drive is queried for those values, is maybe unpark the heads and then park them again after some time - but more or less quietly, and definitely without thrashing the drive. Usually, the drive should be able to answer back to the software without (un)parking the heads. Because remember that SMART has been existing for decades, and you'd think that manufacturers had enough time to figure this out.

What you could do is this: Get smartmontools (smartctl specifically) and query the drive with that tool. If it behaves the same way, then it's "normal" for the drive to be doing such things. As normal as its firmware allows, anyway. Many Linux Live systems have this tool available, but there's also a Windows port available. Querying the drive might be unusual for users who are not familiar with Linux systems. For the first drive use
Code:
smartctl -Ai /dev/sda
, the second drive is named /dev/sdb, the third is /dev/sdc and so on. Remember to run this command on an elevated CMD.

Regards
Dalai
 
Is it normal behavior for drives to be doing that? ...............................................

Regards
Dalai
SMART this is information stored within a chip, at HDD controller, there is no need for Heads activity.
Seagate has a pack of old repair tools, HDD Check along drive mechanics calibration, along HDD physical space repair.
I did had recently an old Maxtor ( Maxtor branding now belong to Seagate), I did use Seagate tools, HDD this was recovered at full functional state, but the miracle lasted only for few months.
Heads parking when again off calibration, and I did remove this HDD permanently.

 
Drive health information (SMART or Statistics) are periodically evaluated by the firmware and kept in persistent memory.
When a client requests this set of information there's no reason for the drive to perform any operations over the media/disk and to spin up/down the heads or motor. If the firmware is properly designed, it should only return the actual set of information without affecting the media or motor. It seems that most drives act so, but some others also perform some undesired operations during such query.
So it would be really useful to determine which drives (vendor and model) cause such clicking sounds.
 
Well, I remember that older Seagate drives (7200.11 series) stored parts of their firmware on the platters. Not sure if that's still the case with recent models. But it's possible that some drives store the results of (short and extended) SMART self tests on the platters, making it necessary to move the heads. However, just querying the SMART attribute values shouldn't initiate a SMART self test, unless explicitely specified (can also be done with smartctl mentioned above).

Regards
Dalai
 
I could provide 2 points:
  1. There are at least 3x ways to read temp on a SATA drive (SAS, NVMe are different):
    • SMART attribute, which is never a standard way, but it's the tranditional way, and most HDD/SDD vendor will follow same format, some vendors specially did optimization so that it does not require wake up from idle/standby mode. But some software will read not only SMART attributes but also thresholds, which may still result in waking up from idle/standby.
    • SCT status command, which is introduced in ATA8-ACS in ~2008, the SCT (SMART Command Transport) status provides some overview of the drive including temp. Vendor could also optimize firmware to avoid unnecessary wake up.
    • Device Statistics log, which is introduced in ATA8-ACS also. It provides dedicated details including history about temperature. Vendor may not optimize for it, since it's `Statistics` and it's not intended to be polled.
  2. @Dalai is partially right about smartctl. smartlctl is the front end part of smartmontools suite, while smartmontools provide a daemon called smartd. smartd could be configured to use `CHECK POWER MODE` to check drive mode before query the drive to avoid waking up an idle/standby drive.
So overall, SMART attributes are preferred, and `CHECK POWER MODE` first would be even better.
 
  • @Dalai is partially right about smartctl. smartlctl is the front end part of smartmontools suite, while smartmontools provide a daemon called smartd. smartd could be configured to use `CHECK POWER MODE` to check drive mode before query the drive to avoid waking up an idle/standby drive.
This can also be done via smartctl:
Code:
smartctl -Ai --nocheck=standby /dev/sda
It's the approach taken by the hddtemp_smartl Munin plugin, for example.

It's still not clear to me if this issue is about the drive spinning up or just moving its heads. If it's about the former (which I doubt), the idle timeout in the Windows power profile is shorter than the SMART cycle in HWiNFO, and one of them should be changed accordingly. If it's about the noise from the heads, well, I already gave an option to check if this is normal for the drive in question.

Regards
Dalai
 
My last message in this topic.
Its more than clear to me that complaining OP's has not enough technical background, to troubleshot the issue and to offer any usable feedback.
Seagate alone, a known troublesome brand, millions of HDD failures, short life, too much noise on operation, too much heat.

Practically any one whom has the preference at setting HDD to sleep, he does not have any high performance workstation PC, he does not need HWINFO to evaluate anything.
 
I found also another bug - triggering SMART/emb with low values causes cpu utilization spiking up, games stuttering every smart/emb check. Cpu i5-10400.
But increasing to above 1000 solves this problem. I tested it with only nvme ssd.
 
I found also another bug - triggering SMART/emb with low values causes cpu utilization spiking up, games stuttering every smart/emb check. Cpu i5-10400.
But increasing to above 1000 solves this problem. I tested it with only nvme ssd.
That's not a bug, but by design.
 
I have this problem as well, with default settings (IDE/(S)ATA Drive Scan set as ‘Safe Mode’ HWinfo wakes up my hard drive every time the condition for Disk S.M.A.R.T every xxxx cycles is met.

The drive obviously wakes up, starts spinning, I can hear the sounds of heads moving, etc. (these are all very loud because it is an Enterprise grade drive, specifically an HC550 Ultrastar 18GB by WD).

I had to circumvent this by setting the value to 99999 circles, but I’m not sure whether this value is honored because when I am using the PC, my hard drives still wakes up from time to time, but not as frequent so I can’t tell the exact time span anymore to pinpoint the cause. I have tried procmon and several other monitoring tools but was not able to tell as of why.

I will try running HWinfo with ‘Disable Drive Scan’ forca while and provide more information.
 
Back
Top