Apple Event: May 7th at 7 am PT

Looks like no one’s replied in a while. To start the conversation again, simply ask a new question.

Old iMac doing {click-click-beep} every 60 sec during sleep

I have a {click-click-beep} cycling every 60.0 sec on my sleeping iMac27” (mid 2010, maxOS 10.13.6)


When in sleep mode, the Hitachi 2TB drive fails to sleep, keeps spinning but cycles every 60.0 seconds (precisely!) with an audible {click, 1 sec pause, click, then single faint beep}. The sounds are coming from inside the iMac case, not the speakers. 


The 1st click causes 'power-off-like' spin-down; the 2nd click invokes ‘power-surging-on-like' spin-up, all within 2-3 sec. 


The iMac sleeps much of the day and all night, and this {click-click-beep} happens every minute of every hours of sleep.


Hitachi model = HDS722020ALA330, and S.M.A.R.T. Status = verified, so the drive is not indicating failing nor generating any errors.


I have reset the NVRAM, and enabled “Put HD to sleep when possible” in Power settings. I understand the hidden-default wait-to-sleep is set to 10 min.


Where is this very-precise snooze-wake cycle, every 60.0 sec, coming from? How to keep-asleep this disk? Help!


Which is least wear: never sleep mode so disk spins, or this {click-click-beep} spin-down-up once a minute 1400 x per day, everyday forever.

iMac 27″, macOS 10.13

Posted on Jan 29, 2021 6:11 PM

Reply
Question marked as Best reply

Posted on Jan 31, 2021 11:29 AM

I know that magnetic 'write' is slower than 'read', and my new 2021 Seagate 2TB USB-external drive has {read / write} throughput of {6.2 / 4.1} MB/s (likely throttled by its USB port). My old iMac internal Hitachi HD is 12 y.o. technology, and currently has {read / write} throughput of {26.2 / 1.2} MB/s -- and I agree that is a very slow write-rate, but still a very brisk read-rate (for old bones). Is the slow write-rate due to aging, and a sign of looming failure? Or is that typical of 2010 tech? The 2010 read-rate is so fast compared to the UBS-throttled 2021 Seagate drive. So, I don't know. Your comments?


I do agree with you about the Seek-Time Performance having a status of 50% and raw value 40 (ms?), but DriveDx judges that to be 'ok'. Furthermore, the Problems Summary says all the life-indicators are ok and there are zero (!) I/O errors, and the 4 Important Health Indicators all have 100% ok status. So, DriveDx says no errors and no fail indicators...


I would think that my {click, click, beep} 1400x per day for months, if due to sticking (getting stuck) drive heads would generate some I/O error accumulation over time, but perhaps not. As you say, DriveDX can't give a full diagnosis, so that suspicion (sticking heads) can't be ruled in nor out.


I now wonder if the {click, click, beep} was due to a macOS bug in the iMac sleep cycle. The fact that this noise ONLY occurred when the iMac was in supposed-sleep mode (display black, Hitachi stayed spinning, but no I/O demands) AND it occurred exactly every 60.0 sec (zero randomness) during sleep, and ceased occurring when awakened, gives credence to some bug in the macOS that was periodically (60.0 sec intervals) disturbing the sleep. Adding to this suspicion, a manual AppleMenu-Sleep selection would immediately cause display-sleep (screen blacken) but sometimes it would self-awake a second later, and I would have to select Sleep again, and sometimes even do this a 3rd time before the iMac would stay asleep, and finally 30 sec later, stop the disks. So, buggy macOS? Also, during the many months of experiencing the noisy (click, click, beep} sleep, that rhythmically-cycled while in long-term Mac sleep, the disk kept spinning anyway, and the click-pause-click was due to a spin-down followed by a re-spin-up, then faint beep. This really did not sound like random sticking drive-heads (to me).


I use the past-tense above, because now the {click, click, beep} is gone. And I don't know why. And I can't reproduce the conditions that ever started it. Now when the iMac sleeps (display dims to black), 32 sec later is a {click} and the drives power-off and spin down (Hitachi internal & Seagate external). So, my problem refuses to manifest. Good or blah? Over the last few days, I've done several re-boots, possibly I cleared the bug-condition. I just don't know. I can only say that now sleep is sleep as it should be, with disks powered down and silent.


HWTech, thank you for recommending DriveDx. I will buy the full version & run it in background to be alerted to looming drive failure. I am also grateful for your experienced consultation, which definitely addressed and resolved my own wondering if it might be a drive-related problem. At this point, with drive silence, I'll monitor the drive and also the macOS behaviour. I also run TimeMachine as a BU policy.

Similar questions

9 replies
Question marked as Best reply

Jan 31, 2021 11:29 AM in response to HWTech

I know that magnetic 'write' is slower than 'read', and my new 2021 Seagate 2TB USB-external drive has {read / write} throughput of {6.2 / 4.1} MB/s (likely throttled by its USB port). My old iMac internal Hitachi HD is 12 y.o. technology, and currently has {read / write} throughput of {26.2 / 1.2} MB/s -- and I agree that is a very slow write-rate, but still a very brisk read-rate (for old bones). Is the slow write-rate due to aging, and a sign of looming failure? Or is that typical of 2010 tech? The 2010 read-rate is so fast compared to the UBS-throttled 2021 Seagate drive. So, I don't know. Your comments?


I do agree with you about the Seek-Time Performance having a status of 50% and raw value 40 (ms?), but DriveDx judges that to be 'ok'. Furthermore, the Problems Summary says all the life-indicators are ok and there are zero (!) I/O errors, and the 4 Important Health Indicators all have 100% ok status. So, DriveDx says no errors and no fail indicators...


I would think that my {click, click, beep} 1400x per day for months, if due to sticking (getting stuck) drive heads would generate some I/O error accumulation over time, but perhaps not. As you say, DriveDX can't give a full diagnosis, so that suspicion (sticking heads) can't be ruled in nor out.


I now wonder if the {click, click, beep} was due to a macOS bug in the iMac sleep cycle. The fact that this noise ONLY occurred when the iMac was in supposed-sleep mode (display black, Hitachi stayed spinning, but no I/O demands) AND it occurred exactly every 60.0 sec (zero randomness) during sleep, and ceased occurring when awakened, gives credence to some bug in the macOS that was periodically (60.0 sec intervals) disturbing the sleep. Adding to this suspicion, a manual AppleMenu-Sleep selection would immediately cause display-sleep (screen blacken) but sometimes it would self-awake a second later, and I would have to select Sleep again, and sometimes even do this a 3rd time before the iMac would stay asleep, and finally 30 sec later, stop the disks. So, buggy macOS? Also, during the many months of experiencing the noisy (click, click, beep} sleep, that rhythmically-cycled while in long-term Mac sleep, the disk kept spinning anyway, and the click-pause-click was due to a spin-down followed by a re-spin-up, then faint beep. This really did not sound like random sticking drive-heads (to me).


I use the past-tense above, because now the {click, click, beep} is gone. And I don't know why. And I can't reproduce the conditions that ever started it. Now when the iMac sleeps (display dims to black), 32 sec later is a {click} and the drives power-off and spin down (Hitachi internal & Seagate external). So, my problem refuses to manifest. Good or blah? Over the last few days, I've done several re-boots, possibly I cleared the bug-condition. I just don't know. I can only say that now sleep is sleep as it should be, with disks powered down and silent.


HWTech, thank you for recommending DriveDx. I will buy the full version & run it in background to be alerted to looming drive failure. I am also grateful for your experienced consultation, which definitely addressed and resolved my own wondering if it might be a drive-related problem. At this point, with drive silence, I'll monitor the drive and also the macOS behaviour. I also run TimeMachine as a BU policy.

Jan 29, 2021 6:30 PM in response to JimPek

To check the health of the hard drive run DriveDx and post the report here using the "Additional Text" icon which looks like a piece of paper. The macOS & Disk Utility "SMART Status = verified" only fails when the drive is near death. I've personally only ever seen this status as failed two times in 20 years. The actual health information (aka SMART Attributes) on the drive itself will give a better indication of the health of the drive.


Jan 30, 2021 6:07 PM in response to JimPek

My guess is the drive is failing. The health report is not as definitive as I would like to see, but there are a couple of clues such as the very slow write speed compared to the read speed. Assuming the DriveDx percentages are correct, then the "Throughput Performance" and "Seek Time Performance" are not good and indicate a problem. The "click-click-beep" sounds like the heads on the drive are getting stuck and these two attributes would make sense in that context. You do have 39K hours on the hard drive which is a decent amount and typically around the time I begin to see 3.5" hard drives begin to fail these days. Chances are this noise will occur more & more.


With this type of drive failure booting the iMac with an external drive probably won't work because the internal drive failure will probably begin to interfere.

Jan 31, 2021 4:42 PM in response to JimPek

JimPek wrote:

I know that magnetic 'write' is slower than 'read', and my new 2021 Seagate 2TB USB-external drive has {read / write} throughput of {6.2 / 4.1} MB/s (likely throttled by its USB port). My old iMac internal Hitachi HD is 12 y.o. technology, and currently has {read / write} throughput of {26.2 / 1.2} MB/s -- and I agree that is a very slow write-rate, but still a very brisk read-rate (for old bones). Is the slow write-rate due to aging, and a sign of looming failure? Or is that typical of 2010 tech? The 2010 read-rate is so fast compared to the UBS-throttled 2021 Seagate drive. So, I don't know. Your comments?

The speeds reported by DriveDx are not the real speeds and are always much slower than the real speed if you checked the speed using a regular drive speed check app. When reading the DriveDx report I only use the speed listed as a basic reference relative to read & write on the same drive. A hard drive's speed can be anywhere from 20MB/s to 150MB/s with most somewhere in the 40MB/s to 80MB/s. The speed can vary even on a single drive depending if you are doing the test on the inner tracks or outer tracks.


I do agree with you about the Seek-Time Performance having a status of 50% and raw value 40 (ms?), but DriveDx judges that to be 'ok'. Furthermore, the Problems Summary says all the life-indicators are ok and there are zero (!) I/O errors, and the 4 Important Health Indicators all have 100% ok status. So, DriveDx says no errors and no fail indicators...

Those percentages reported by DriveDx may not mean exactly what you think. In theory the percentage listed is a gauge of how close the drive is to the failing or warning conditions the manufacturer has set for the drive. I don't normally look at these particular attributes since the different drive manufacturers implement the reporting of these particular attributes differently so it is hard to interpret their meaning. If a computer is having issues with no other reasons for the issues and I see these particular attributes look like this, then I tend to equate them with the issue (some of it is a bit of experience with similar model of drives).


The best & most important attributes for accurately judging hard drive health (SSDs are different) are the Reallocated Sectors & Events and Pending Sector Counts with a couple other important attributes for some brands of drives. As soon as the RAW value goes above zero on these particular attributes it indicates the hard drive is failing (the % of "Value/Worst" & "Threshold" may still be near 100%, but the drive is considered bad & failing). SSDs are interpreted differently.


If these two attributes keep declining, then it would definitely be an indication that the drive is failing or wearing out whether it is caused by the hard drive itself or whether a bug in macOS is causing stress on the drive leading to premature drive failure.


I would think that my {click, click, beep} 1400x per day for months, if due to sticking (getting stuck) drive heads would generate some I/O error accumulation over time, but perhaps not. As you say, DriveDX can't give a full diagnosis, so that suspicion (sticking heads) can't be ruled in nor out.

The clicking is something mechanical and sticking heads is common. Maybe something is activating the heads to move and park repeatedly (that is not good for the health of the drive). A beeping or chirping sound from a hard drive is usually fatal.


I now wonder if the {click, click, beep} was due to a macOS bug in the iMac sleep cycle.

I use the past-tense above, because now the {click, click, beep} is gone. And I don't know why. And I can't reproduce the conditions that ever started it. Now when the iMac sleeps (display dims to black), 32 sec later is a {click} and the drives power-off and spin down (Hitachi internal & Seagate external). So, my problem refuses to manifest. Good or blah?

Sounds like there may indeed be a bug in macOS or some corrupt system preference. Hitachi hard drives had the best reputation for being reliable hard drives. It is a shame they don't really exist anymore.


HWTech, thank you for recommending DriveDx. I will buy the full version & run it in background to be alerted to looming drive failure. .... At this point, with drive silence, I'll monitor the drive and also the macOS behaviour. I also run TimeMachine as a BU policy.

DriveDx is very good at assessing the health of a hard drive so if you see any "Warnings" or "Failures" listed, then the drive should be replaced. However, DriveDx (and other drive monitoring apps) don't really assess the health of an SSD properly so a manual assessment of an SSD's SMART Attributes is necessary since a bad block on an SSD is not fatal and is expected as normal as long as only a couple blocks go bad in a short period of time.


Jan 31, 2021 7:22 PM in response to HWTech

HWTech, thank you again for sharing your insights: I've learned a lot in these exchanges. I will purchase & run DriveDx as a backgrounder, and keep an eye out for any non-zero raw values for the Reallocated Sectors & Events and Pending Sector Counts for this drive (so far, all zero). Especially if the {click, click, beep} returns and a re-boot doesn't clear it. I can live with an OS bug (like, by doing reboots) as long as it does not harm the HW or my files.


I am also heartened by your favourable rating of these old Hitachi drives -- this one still seems robust after 10+ years of service. And the nice thing for now, is that the iMac sleeps like it should sleep -- in silence with the HDs powered off. No {click, click, beep}. Problem abated, if not solved.

Old iMac doing {click-click-beep} every 60 sec during sleep

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple ID.