Macbook Pro (early 2011) lagging, fan spinning, kernel_task taking over

Hello,


My daughter just inherited a Macbook Pro from her grandfather who said it started behaving oddly lately and got himself a new computer. His complaints were related to an unusual kernel-task activity that would reach 600-700% and would leave the computer unusable.

Most of the time the fan starts spinning really fast when the start-up bar has just halfway through. Other times, it starts just a couple of minutes after logging in. The cursor behaves slowly and erratically.

The computer isn't particularly hot.

— I reformated the disk and did a clean install. The problem persisted.

— I ran the Apple Hardware Test — no problems found.

— I replaced the old battery, expecting it to the culprit, but the problem persisted.

— I tried SMC reset, and the problem persisted.


I decided to try to reset the PRAM and the computer apparently started running normally. But as soon as I restart it the problem comes back.

Restarting and resetting the PRAM again, and it seems to go away again...


Any ideas? Thank you!


I'm adding an Etrecheck report which I run with the computer running in Safe Mode, which also seemed to prevent the problem.




MacBook Pro

Posted on Mar 25, 2021 12:58 PM

Reply
Question marked as Top-ranking reply

Posted on Mar 25, 2021 5:17 PM

The end of the report has two mentions of memory so perhaps you have loose or bad memory, or a bad memory slot. Try reseating the memory. When inserting the memory nearest the Bottom Case it can make the other module pop loose. Plus the memory slot nearest the Bottom Case can fail due to cracked solder joints from years of pressure exerted on the base of the memory slot/Logic Board. When you have the cracked solder joints just putting pressure on the black base of the memory slot will cause the laptop to freeze and/or get graphical glitches.


Another possibility is the SSD may need a firmware update or the SSD may be failing or needs to be reset to factory defaults. To check the health of the SSD run DriveDx and post the report here using the "Additional Text" icon which looks like a piece of paper. To reset the SSD to factory reset requires a bootable Linux drive and utility to access the SSD's built-in hardware secure erase feature. I can provide instructions if necessary.


Another possibility is the hard drive SATA Cable is bad. You can test this theory by removing the SSD and booting the SSD externally using a USB to SATA Adapter, drive dock, or enclosure. If you don't have the problem with the SSD booting externally, then the internal cable is likely at fault. You can try reseating the cable to the Logic Board in case it has become loose.


When you performed the clean install of macOS did you first erase the whole physical drive as opposed to just the Macintosh HD volume? Unfortunately Disk Utility hides the physical drive from view beginning with macOS 10.13. Within Disk Utility click "View" and select "Show All Devices" so that the physical drive appears on the left pane of Disk Utility.


Similar questions

39 replies
Question marked as Top-ranking reply

Mar 25, 2021 5:17 PM in response to pedrocm

The end of the report has two mentions of memory so perhaps you have loose or bad memory, or a bad memory slot. Try reseating the memory. When inserting the memory nearest the Bottom Case it can make the other module pop loose. Plus the memory slot nearest the Bottom Case can fail due to cracked solder joints from years of pressure exerted on the base of the memory slot/Logic Board. When you have the cracked solder joints just putting pressure on the black base of the memory slot will cause the laptop to freeze and/or get graphical glitches.


Another possibility is the SSD may need a firmware update or the SSD may be failing or needs to be reset to factory defaults. To check the health of the SSD run DriveDx and post the report here using the "Additional Text" icon which looks like a piece of paper. To reset the SSD to factory reset requires a bootable Linux drive and utility to access the SSD's built-in hardware secure erase feature. I can provide instructions if necessary.


Another possibility is the hard drive SATA Cable is bad. You can test this theory by removing the SSD and booting the SSD externally using a USB to SATA Adapter, drive dock, or enclosure. If you don't have the problem with the SSD booting externally, then the internal cable is likely at fault. You can try reseating the cable to the Logic Board in case it has become loose.


When you performed the clean install of macOS did you first erase the whole physical drive as opposed to just the Macintosh HD volume? Unfortunately Disk Utility hides the physical drive from view beginning with macOS 10.13. Within Disk Utility click "View" and select "Show All Devices" so that the physical drive appears on the left pane of Disk Utility.


Mar 27, 2021 1:08 PM in response to pedrocm

I cannot be sure what the UDMA CRC Error Count is from the DriveDx report due to the formatting issue in the report. If there are UDMA CRC Errors, then it most likely indicates a bad hard drive SATA Cable (assuming this SSD was not installed into another computer with those issues). This health attribute only tells us this SSD was at one time connected in a manner which produced these errors. If the "RAW value" of this attribute is increasing, then it means the current cable used for the SSD is most likely causing the problem.


You can use the Terminal app to run the following command to provide another health report in a slightly different layout of the report. Begin by typing "sudo " making sure to leave at least one space at the end. Then drag & drop the DriveDx app from the Finder onto the open Terminal window to auto-fill the correct path to the DriveDx app. Then press the "Delete" key or "Backspace" key to remove the trailing space on the line and add the following immediately after the last character:

/Contents/Resources/smartctl  -a  /dev/disk0  >  ~/Desktop/SSD-report.txt

so the final command looks something like this (it will differ slightly due to the different location of the DriveDx app and your user account name):

sudo  /Users/hwtech/Downloads/DriveDx.app/Contents/Resources/smartctl  -a  /dev/disk0  >  ~/Desktop/SSD-report.txt

Press the "Return" key at the end of the line to execute the command. You will be prompted for your admin password. Nothing will appear on the screen as you type the password. Press the "Return" key to submit the password. The command will save another report to your Desktop in a file called "SSD-report.txt". Paste the contents of the report here using the "Additional Text" icon which looks like a piece of paper.


It appears your SSD's firmware is a bit out of date so it would be a good idea to update the SSD's firmware, but I think you will need to burn the firmware update .iso file to a CD since it won't work as a bootable USB drive on a Mac (or UEFI only booting computer like a Mac). Before updating the SSD's firmware we want to address the UDMA CRC Errors (if any) so they don't affect the firmware upgrade. If the SSD firmware update is interrupted it could make the SSD completely unusable and inaccessible.

https://www.crucial.com/support/ssd-support/mx200-support


FYI, you can also modify the previous command to run the SSD's long/extended self-diagnostic as well using this command as a template:

sudo  /Users/hwtech/Downloads/DriveDx.app/Contents/Resources/smartctl  -t  long  /dev/disk0


You can review the results of the long/extended selftest within the DriveDx app. The output of the last command will provide an estimate of how long it will normally take the selftest to complete. Of course it may take much longer for the selftest to complete if you are using the SSD or if the SSD is actually having performance issues.

Mar 29, 2021 8:28 PM in response to pedrocm

pedrocm wrote:

Hello again.
Do you think it is safe to try and do the SSD firmware update?

Probably, since you don't have issues while booted into Safe Mode and to update the SSD firmware you will be booting from a CD. You will need to burn the SSD firmware updater .iso image to a CD since the firmware updater doesn't have support for UEFI booting from a USB drive which is required on a Mac.


I don't know what this might mean but the fact is the computer works fine whenever I start it in Safe Mode or when I reset the PRAM.

Usually if Safe Mode works, but normal booting does not it means you have some sort of software issue. Out of all the items shown in the EtreCheck report the Huion Tablet driver is the most likely culprit as it is the only item that is listed as "loaded".

User Launch Agents:
[Loaded] com.huion.HuionTablet.plist (Shenzhen Huion Animation Technology Co.,LTD - installed 2021-03-25)


In both "Normal" and "Runaway Kernel Task" versions of the EtreCheck report something is causing extreme "System Load". So if you don't have a problem in Safe Mode, then try uninstalling the Huion Tablet driver by following the developer's instructions. Are you using the latest version of this driver?


As far as the second SSD health report I think the 24,780 UDMA CRC Errors is an incorrect interpretation of the RAW value since the "Value" and "Worst" columns are still at their original values and I don't think that would be true if there were that many UDMA CRC errors. Plus the RAW value for this attribute did not change between the two SSD health reports.


Apr 11, 2021 5:13 AM in response to pedrocm

In the meantime, I decided to again erase the disk and reinstall everything from scratch. I add to install Lion, and then Sierra, and then upgrade to High Sierra. Everything seem to run ok in Lion and Sierra (although I didn't run it for very long), but again as I was running Etrecheck after intalling High Sierra, the problem came back...

Again, I erased the disk, but this time, before installing the system, I decided to re-run the Apple Hardware Test (which previously had reported no errors): this time, though, it reported error 4HDD/11/40000000: SATA(0, 0).

So, maybe the problem resides with the SDD disk or the SATA cable.


Apr 12, 2021 8:25 AM in response to pedrocm

pedrocm wrote:

I think for now I'm going to try something you suggested in your first reply, and which I haven't been able to try yet:
"removing the SSD and booting the SSD externally using a USB to SATA Adapter, drive dock, or enclosure."

Good idea for seeing if the internal cable is at fault.


I will try both that and replacing the SSD with an old HDD I have here. I think that will be enough to conclude on either a faulty cable ou faulty disk (or both... or none...)

Not necessarily. A bad internal hard drive SATA Cable tends to fail more often with an SSD due to the faster speeds of the SSD. Even a different model SSD could possibly behave differently. If a hard drive SATA Cable fails it can fail in several different ways. I've personally seen multiple types of symptoms from these SATA Cable failures. Testing the SSD externally is the best way to narrow down the issue.


Mar 25, 2021 6:21 PM in response to pedrocm

I forgot to mention you can perform a more thorough memory test by using Memtest86. Use the downloaded Memtest86 .iso or .img file as a source for Etcher which will create a bootable Memtest86 USB stick for you. Option Boot the Memtest86 USB drive and select the orange icon labeled "EFI". The older v7.5 version of Memtest86 allows you to run more than the default four loops.


Apr 10, 2021 9:41 PM in response to pedrocm

If the SSD firmware update seemed to help, then you should maybe try performing a hardware secure erase on the SSD which will reset the SSD to factory defaults which can resolve any odd issues with the SSD. Updating the SSD firmware may have reset a few internal settings of the SSD, but the hardware reset will reset all of the SSD's hardware. I know that with some SSD firmware updates that a hardware secure erase is necessary to fix some pre-existing issues. A hardware secure erase will destroy all data on the SSD so you must make sure to have a good backup (or two or three) plus it requires using a Linux boot disk and utility to activate the SSD's hardware secure erase feature. I can provide instructions if you are interested.


I also suggest running the v7.5 Memtest86 over several days to see if any memory issues are detected. You may also want to try running the service diagnostics that can be downloaded here. Instructions for creating a bootable USB stick is included in a .pdf file once you mount the first .dmg file included in the .zip archive.



Apr 11, 2021 9:23 PM in response to pedrocm

If the AHT previously passed and it failed this time, then it is possible the hard drive SATA Cable is bad. I've heard this cable does go bad on these older models especially when using an SSD, but I've never personally seen it happen on any of our organization's 2011 laptops (although it is very common specifically on the MBPro 13" mid-2012 model).


Normally would suggest creating a bootable Parted Magic USB drive using the last free version since it has a custom easy to use app, but I was unable to get the free version of Parted Magic to boot a few months ago for some reason (perhaps a later laptop system firmware update broke it?). Unfortunately I haven't had time to investigate.


It is possible to create a bootable Knoppix Linux USB drive to perform the hardware reset of the SSD, but it does require using the command line. You can use the downloaded Knoppix .iso file as a source for Etcher which will "burn" the Knoppix .iso file to a USB stick. Make sure to grab the v8.6.1 Knoppix with the "EN" in the file name for ENglish. Option Boot the Knoppix USB drive and select the orange icon labeled. "EFI". The Mac may appear to be frozen on the Apple boot picker menu while Knoppix is booting so make sure to give Knoppix lots of time to finish booting.


Once Knoppix boots to the desktop you need to acquire the drive identifier for the Crucial SSD. You can get this information by clicking on the "Start" menu icon on the lower left corner of the Taskbar and navigate the menus to "System Tools --> GSmartControl". Within the GSmartControl app it should provide you the drive identifier which will likely be "sda" or something similar.


You can also get the drive identifier using the command line using one of the multiple Terminal apps and using the following command:

sudo  lsblk  -f  |  grep  -i  'sd'


In each of the following commands make sure to replace "sdX" with the correct drive identifier.


Once you have a drive identifier you can verify you have correct drive identifier for the Crucial SSD by using this command:

sudo  hdparm  -I  /dev/sdX  |  grep  -i  'model'


Now that you have the correct drive identifier for the Crucial SSD you need to sleep and wake the laptop to "unfreeze" the SSD. It is best to do this by clicking the "Start" menu icon and selecting "Logout" followed by "Suspend". Then wake the laptop up and run the following command to make sure the SSD is not "frozen":

sudo  hdparm  -I  /dev/sdX  |  grep  -iE  '^security'  -A5


The output should look something like:

Security: 
	Master password revision code = 65534
		supported
	not	enabled
	not	locked
	not	frozen


Now you must enable security mode on the SSD:

sudo  hdparm  --user-master  u  --security-set-pass  Eins  /dev/sdX


Followed by the command to initiate the SSD's built-in hardware secure erase feature:

sudo  hdparm  --user-master  u  --security-erase  Eins  /dev/sdX


This last command can take anywhere from 15 seconds to several minutes to complete (sometimes longer -- up to 20 minutes). If the secure erase completes successfully, then it automatically disables the SSD's security mode set earlier in these instructions. If the secure erase command fails or is interrupted for any reason, then you will need to issue this command to disable the security mode previously set so you can use the SSD normally:

sudo  hdparm  --user-master  u  --security-disable  Eins  /dev/sdX


You can confirm the SSD's security mode is disabled by running this command again and making sure "not locked" is listed:

sudo  hdparm  -I  /dev/sdX  |  grep  -iE  '^security'  -A5


Make sure the laptop is using the charger so the SSD is not damaged if the battery fails.


I know this seems like a lot, but as long as you remember to replace "sdX" with the proper drive identifier in each command line, then there should be no problems.



Mar 27, 2021 11:29 AM in response to pedrocm

Performance:

System Load: 58.79 (1 min ago) 48.11 (5 min ago) 21.99 (15 min ago)

Nominal I/O speed: 0.50 MB/s

File system: 81.55 seconds

Write speed: 95 MB/s

Read speed: 489 MB/s


that is because of this:

Drives:

disk0 - Crucial_CT1024MX200SSD1 1.00 TB (Solid State - TRIM: No)


SSD drives need TRIM. you can enble TRIM in Terminal with this command:


sudo trimforce enable  


reference this article, that simply explains the advantages of TRIM without getting lost in the gory details:

https://www.lifewire.com/enable-trim-for-ssd-in-os-x-yosemite-2260789


.


Mar 27, 2021 12:02 PM in response to pedrocm

Great, now you are ready for the second part of the process.


TRIMforce sets a variable in the running MacOS that allows NEW deletes to be TRIM-ed, but that does not get rid of the old deleted data.


Restart in Safe Mode (hold shift at startup). This first loads the kernel, then performs a disk check --which will remove all old deleted data blocks. Then it adds a minimal set of Apple-only extensions, not including graphics acceleration extensions. Screen updates will be wonky and slow, but ultimately correct. You will need you userid and password to proceed, even if you 'normally' auto-login.


later, when you restart in regular mode, your drive write speed will be substantially faster.



Apr 14, 2021 2:01 AM in response to HWTech

Hello, again

So, I did all that — I even replaced the cable, out of despair — and the problem persists. The only difference is that now the AHT reports problem (4sns/1/C0000008: TC0D- -124) which seems to be related to a thermal sensor malfunction. (Etrecheck still only reports the runaway kernel task and the slow operation.)

I'm afraid there won't be an easy cure for this.

Mar 30, 2021 4:59 AM in response to HWTech

Well, I decided to try the firmware update and, apparently — fingers crossed! — that seemed to have solved the problem!...

I already restarted the computer twice in Normal Mode and after a 15 minutes or so (more than usual) the kernel task still hasn't started! I'm going to leave it on for a longer period, just to be sure and I'll come back to let you know — and, hopefully, properly thank you for your help.

I include the new DriveDX report.


This thread has been closed by the system or the community team. You may vote for any posts you find helpful, or search the Community for additional answers.

Macbook Pro (early 2011) lagging, fan spinning, kernel_task taking over

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