My Mac mini 2011 model, has been freezing off lately. I am using ADATA SSD on my Mac mini. The freezing happens everyday, mostly when playing videos or compiling code. The apple disk utility command diskutil info disk0 | grep SMART. I get the SMART status verified response. Using an app called DriveDX I came to know that my Hard Drive is failing. I am uploading the screenshot of the result. I think that the drive needs to be replaced, your advice is appreciated. Thanks in advance.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Mac mini freezes abruptly
- Thread starter markiv810
- Start date
- Sort by reaction score
It started happening a few months back, the freezing was not that frequent. I am using High Sierra (APFS), and the freezing now happens daily, I attached the screenshot for reference. The hard drive shows no errors when using Apple's Disk Utility/ first aid but DriveDX report states that the advanced SMART status is failing. I have a feeling that the hard drive is failing and has bad sectors. Just wanted to confirm if my analysis about the SSD drive is ok.
I would surely give that a try, the freezing happens every 10 minutes. I could be surfing the web or just playing music via iTunes. When booting in verbose mode I did see errors related to block count. Hence I ran the DriveDX app which indicated that the SSD's Advanced SMART status is failing. I did attach screenshot of the report in the original message. Btw how reliable are the ADATA SSDs. I have attached the error report I got through DriveDX app
Attachments
OP:
Something you need to do if you haven't done it already:
Create a BOOTABLE CLONED BACKUP in case the SSD suddenly "goes dark" on you.
A Time Machine backup WILL NOT DO.
It MUST be a bootable cloned backup using either CarbonCopyCloner or SuperDuper.
Both are free to download and use for 30 days, and it will "cost you nothing" software-wise to do this.
That way, if the SSD fails, you just connect the bootable backup, boot from it, and you'll be "right back where you left off" (an exact working copy of the SSD at the last time you cloned it).
Is the SSD mounted internally?
If so, a suggestion.
- create the cloned backup
- download the "Parted Magic" iso and burn a CD from it (or create a bootable USB drive from it, if that can be done)
- DISCONNECT the cloned backup for the moment
- Boot from Parted Magic
- Use the "ATA reset" command to set the SSD back to it's "factory-fresh" state (this will DESTROY everything on the drive, that's why you need a cloned backup)
- Now, quit Parted Magic.
- Connect the cloned backup and boot from it.
- RE-CLONE the backup BACK TO the SSD.
Note: the "ATA reset" command (I -think- that's what it's called) works ONLY on internally-mounted drives. It can't be used [to my knowledge] over USB...
Something you need to do if you haven't done it already:
Create a BOOTABLE CLONED BACKUP in case the SSD suddenly "goes dark" on you.
A Time Machine backup WILL NOT DO.
It MUST be a bootable cloned backup using either CarbonCopyCloner or SuperDuper.
Both are free to download and use for 30 days, and it will "cost you nothing" software-wise to do this.
That way, if the SSD fails, you just connect the bootable backup, boot from it, and you'll be "right back where you left off" (an exact working copy of the SSD at the last time you cloned it).
Is the SSD mounted internally?
If so, a suggestion.
- create the cloned backup
- download the "Parted Magic" iso and burn a CD from it (or create a bootable USB drive from it, if that can be done)
- DISCONNECT the cloned backup for the moment
- Boot from Parted Magic
- Use the "ATA reset" command to set the SSD back to it's "factory-fresh" state (this will DESTROY everything on the drive, that's why you need a cloned backup)
- Now, quit Parted Magic.
- Connect the cloned backup and boot from it.
- RE-CLONE the backup BACK TO the SSD.
Note: the "ATA reset" command (I -think- that's what it's called) works ONLY on internally-mounted drives. It can't be used [to my knowledge] over USB...
Thanks a lot for your advice, I really appreciate it. I shall be doing exactly that, btw I do have the SSD installed internally. I shall inform how this process goes. Going to create a clone backup today, rest of the stuff on Saturday. Thanks once again.
Take the interpretation that these programs give with a grain of salt. What I'm seeing is that on April 7 (probably), something happened that caused the SSD to retire some data blocks. SSD's will typically reserve some space (typically in gigabytes) for the eventuality that some data blocks will fail and that's what happened to your SSD, probably on April 7. I say probably because it takes more bytes to store a large calendar so it's typical that the timestamp values for the individual attributes will recycle at some point. But April 7 sounds plausible. The amount of bytes that went bad would depend on the block size of the SSD, but I would guess it's in the tens of megabytes (under 50). So if April 7 is the last time a block got retired and you have 324 blocks total retired, unless the manufacturer has over-provisioned a very small amount, I don't see how this can be characterized as a failing SSD. I very well could be wrong - you really have to ask the vendor for a proper interpretation and an overall characterization of your SSD's health. But if this is an econo-SSD from a company that caters to the econo segment of the market, you might not get support to get a proper interpretation.
How much free space do you have on the SSD? Some SSD's do not handle low free space conditions very well and that might be what's happening here.
How much free space do you have on the SSD? Some SSD's do not handle low free space conditions very well and that might be what's happening here.
Last edited:
One thing I noticed earlier but forgot to write in my previous post. Your screenshot shows a "Unsafe shutdown count" of 188. Any explanation for that? If not, is your SSD encrypted?
The computer would freeze abruptly after a few minutes, thats probably the most consistent thing with my Mac Mini. Every ten minutes or so Mac Mini freezes, hangs and I would hear the fans speed increasing. So I have power off the Mac Mini (force shutdown by pressing the power button). I had an issue prior to installing High Sierra once, I was using Sierra and the computer would not boot (about two years back). After reinstalling Sierra everything was ok. I have been having the issue with the Mac Mini for past 2 months, I could be doing nothing on the computer and it would freeze and I would get no response. The only way to shutdown the computer would be the unsafe shutdown. Btw the Mac Mini is barely utilising half of the disk space. I have a 256 GB SSD of which on 110 GB approx. is being utilised. MY SSD is pretty cheap/economical, the SSD is almost 3.5 years old, thought SSD's were more durable than the conventional HDDs. My SSD is unencrypted.
Your explanation helps understand what's going on.
I asked about the encryption because there was a post some time ago from somebody who noticed that external encrypted HDD's would not be "ejected" when the eject button was pressed in the Finder and when the USB cable was pulled, the head would make an audible noise. Indeed, I looked into the issue and while encrypted external USB drives are unmounted after pressing the eject button, they aren't actually given the eject command (which does happen if the HDD is unencrypted) so I was wondering if the same thing was happening with the SSD (even though it's internal) and that was causing the issue with the unsafe shutdown. But I don't know if the unsafe shutdown has anything to do with your issues. (I looked into this issue because I have an encrypted disk which makes a "ping" when the cord is pulled after it's supposedly ejected through the Finder. It doesn't make the "ping" if the eject command is given - either through Disk Utility or through an OS command.)
Your SSD had it's issue about 3 months ago and you started noticing issues 2 months although it doesn't appear (again, remember what I mentioned about timestamps with SMART attributes) that it has recorded the same anomaly since then. At this point, I would complete the backup and you can try re-installing the OS. It doesn't have to be a clean install - just re-install the OS over your existing OS - this can clear some issues. It may be that some file(s) got corrupted during the April incident and re-installing will help. I recently had an issue where I cloned a HDD of an old install to a SSD. I noticed that one of the drives had a discconnect but I thought it may have happened before the copy and I just ignored it since the program I have to clone the disk will detect read errors and retry. It was probably a write error (due to the disconnect). The clone actually booted but some programs would take forever to start and other strange things happened. I sync'd the files between the two disks and that solved the problem.
I could write a lot about SSD vs. HDD durability and SSD price and manufacturer reputation vs. quality. In this case, I'll just say that it's a tough business to be in because the big companies in this space manufacturer their own NAND chips and some of them make their own controller chips (Samsung) or if they don't, they have very close relationships with the controller makers because they rely on each other for the success of their products. You compare that with the smaller companies who have to buy NAND chips from these same companies that are selling competing products and who don't have the same type of relationship with the controller makers. These smaller companies also don't have the same economies of scale and yet they have to price it below the big name companies' prices.
Thanks a bunch for replying, I have reinstalled the OS (just the way you mentioned over the existing OS) about 3-4 times in past 2 months thinking it would solve my problem. But that clearly has not helped. Booting into the verbose mode I did notice errors relating to sectors, but I could not read the entire thing (the verbose message scrolls really fast). The unsafe shutdown was the only way I could shutdown the Mac Mini (I have the power supply via APC 1KVA UPS) as it had stopped responding. I already have cloned my SSD, I will try and format the it today. Lets see how it goes, there goes my weekend.
I am going to make a bootable USB for Parted Magic and try it today. Thanks Again.
Wow, a lot of people are missing the obvious. The disk has quite a few bad sectors and uncorrectable errors. It is in the process of failing. Start saving for a new drive, and start backing up now.
This has NOTHING to do with the file system. An ATA reset will not fix the underlying fact that the drive is dying, plain and simple. And when SSDs die, it is damn difficult to get data off of it aside from sending it into a recovery specialist. Healthy SSDs should have near-zero errors of any kind.
Not sure on ADATA drive reliability, but personally I don't use anything but Samsung SSDs. They are the only manufacturer that makes the entire drive (chips, board, etc) in-house, and their QA is pretty damn good (although like anything there are some baddies that get missed).
To put it into perspective, my 5+ year old samsung SSD with over 100 TB of data written to it over the years has zero reallocated sectors, zero uncorrectable errors, and is still running as snappy as day 1. One of my servers has an 840 pro that has over 300 TB written, with zero errors.
Whether or not the drive is related to the freezing is anyone's guess. It could be, or it could be some other hardware failing.
This has NOTHING to do with the file system. An ATA reset will not fix the underlying fact that the drive is dying, plain and simple. And when SSDs die, it is damn difficult to get data off of it aside from sending it into a recovery specialist. Healthy SSDs should have near-zero errors of any kind.
Not sure on ADATA drive reliability, but personally I don't use anything but Samsung SSDs. They are the only manufacturer that makes the entire drive (chips, board, etc) in-house, and their QA is pretty damn good (although like anything there are some baddies that get missed).
To put it into perspective, my 5+ year old samsung SSD with over 100 TB of data written to it over the years has zero reallocated sectors, zero uncorrectable errors, and is still running as snappy as day 1. One of my servers has an 840 pro that has over 300 TB written, with zero errors.
Whether or not the drive is related to the freezing is anyone's guess. It could be, or it could be some other hardware failing.
Last edited:
A problem with the SSD could very well be causing the OP problems and given the cost of high-quality SSD's now, I would probably just buy a new one and set the old one aside to figure out if it could be salvaged. But I would disagree with the idea that only perfect or near-perfect SSD's should be used. Manufacturers over-provision SSD storage - the typical percentage seems to be 7%. If manufacturers believed that the only good SSD is a perfect/near-perfect SSD, that percentage would be lower. The OP has 324 retired blocks, the typical block size is between 16K-512K, so if the SSD in question has a 512K block size and the typical over-provisioning percentage, this would be under 1% of the over-provision size. The OP also has a screen shot which shows that the last time a block was retired was on April 7. A failing SSD should have more recent error events. Again, maybe there's something else wrong with the SSD not captured by the SMART events or maybe the SMART event capture mechanism is borked, but I think other possible causes should be looked at.
If you use the Console app (in the Applications/Utilities folder), click on system.log and you can search for the word "sector".
At this point, I would also try the typical things - SMC reset, NVRAM reset. You should also try Safe Mode. Maybe there's some software causing the issue. Safe mode will prevent 3rd-party background programs from starting so that may hinder the normal use of the computer.
https://support.apple.com/en-us/ht201262
If you use the Console app (in the Applications/Utilities folder), click on system.log and you can search for the word "sector".
At this point, I would also try the typical things - SMC reset, NVRAM reset. You should also try Safe Mode. Maybe there's some software causing the issue. Safe mode will prevent 3rd-party background programs from starting so that may hinder the normal use of the computer.
https://support.apple.com/en-us/ht201262[/QUOTE]
I already did SMC reset and NVRAM reset many a times. I have borrowed SSD from a friend. I have cloned to the new SSD. I have mounted the new SSD via USB and am using that system. It seems to be ok so far. I will test it some more. The new SSD I got from my friend is Crucial MX300, so far it has not had any freezing issues.
At this point, I would also try the typical things - SMC reset, NVRAM reset. You should also try Safe Mode. Maybe there's some software causing the issue. Safe mode will prevent 3rd-party background programs from starting so that may hinder the normal use of the computer.
https://support.apple.com/en-us/ht201262[/QUOTE]
I already did SMC reset and NVRAM reset many a times. I have borrowed SSD from a friend. I have cloned to the new SSD. I have mounted the new SSD via USB and am using that system. It seems to be ok so far. I will test it some more. The new SSD I got from my friend is Crucial MX300, so far it has not had any freezing issues.
If the borrowed SSD works fine from the USB enclosure, I would just buy a new SSD and then if you wish, you can do additional testing by swapping SSD's - the new one internal and the old one from USB and see what happens. It's generous of your friend to loan the SSD to you for this type of testing.
That's exactly what I shall be doing, I am testing the new SSD via USB enclosure. Would swap the hard drives in a couple of days. If the things go ok, I shall buy this SSD from my friend.