SSD Problems

[ZVH]

DriveDX provides an "SSD Lifetime Left" indication.

DriveDX is an interface for smartmontools smarctl, which is a SMART monitor. My question isn't whether it can be detected or not but whether the Mac itself would give out a warning.

If any one has been following the SSD endurance experiment, it took a long time but finally all the SSDs in the experiment died. Here's what I think is the last link in the experiment:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

Pretty interesting stuff, IMHO.

 

[OldGuyTom]

Thanks for the link. Now we know what to expect.

 

[simonsi]

DriveDX is an interface for smartmontools smarctl, which is a SMART monitor. My question isn't whether it can be detected or not but whether the Mac itself would give out a warning.

If any one has been following the SSD endurance experiment, it took a long time but finally all the SSDs in the experiment died. Here's what I think is the last link in the experiment:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

Pretty interesting stuff, IMHO.

Yes I know, but OS X doesnt provide a lifetime remaining estimate on HDDs based on allocated spare blocks either.

 

[ZVH]

Yes I know, but OS X doesnt provide a lifetime remaining estimate on HDDs based on allocated spare blocks either.

HDDs usually lose blocks because of a pre-existing defect during manufacturing or a head crash. These are both, sort of, one-off events. SSDs on the other hand will begin to progressively lose blocks, so I think predicting their demise might be easier.

If you haven't already, see the linked article I posted above because I think you might find it interesting.

 

[simonsi]

HDDs usually lose blocks because of a pre-existing defect during manufacturing or a head crash. These are both, sort of, one-off events. SSDs on the other hand will begin to progressively lose blocks, so I think predicting their demise might be easier.

If you haven't already, see the linked article I posted above because I think you might find it interesting.

Yes I read the article, TBH it still supports treating data the same irrespective of media, SSD or HDD it <may> disappear without warning at any time.

Manufacturers are probably wary of being seen to predict the life of such a component for various reasons...

 

[ZVH]

I personally suspect the reason they don't really bother checking stuff is because both SSDs and HDDs really last a fairly long time, at least for most people. I was really quite impressed and pleased to see that the SSDs held up that well…or at least I thought they held up well. That's a lot of abuse, IMHO.

For me an HDD usually gets swapped out because I've run out of space, not because the drive failed. A few years ago a fair number of SSDs seemed to suffer from a complete loss of data problem, which seems to be unique to OCZ SSDs from years ago, but I no longer see that sort of complaint, at least not that often.

If they can fix the SSD wipeout problem when the power is cut, the SSD will truly be a reliable entity.

Just my opinions, of course.

 

[simonsi]

If they can fix the SSD wipeout problem when the power is cut, the SSD will truly be a reliable entity.

Yep - hopefully thats an iMac/MP/Mini only problem...I was a bit disappointed no Crucials in that test, just swapped my 120GB out for a 512GB - because of space not issues.

 

[B-Eugen]

Yep - hopefully thats an iMac/MP/Mini only problem...I was a bit disappointed no Crucials in that test, just swapped my 120GB out for a 512GB - because of space not issues.

That problem exists with SSDs regardless of the system they're running on. It's not specific to specific computers.

 

[simonsi]

That problem exists with SSDs regardless of the system they're running on. It's not specific to specific computers.

My suggestion that only those machines might be affected is because....they are NOT battery powered.

Significantly harder to experience a sudden loss of power on a battery powered device. Not impossible but several orders of magnitude less likely.

 

[B-Eugen]

My suggestion that only those machines might be affected is because....they are NOT battery powered.

Significantly harder to experience a sudden loss of power on a battery powered device. Not impossible but several orders of magnitude less likely.

You're right. I wasn't thinking. I thought that you were implying that the failure report was done using only Mac's which it wasn't.

 

[ZVH]

So far on an SSD we basically have two problems, total wipeout that may or may not be recoverable, and blocks dropping.

How about something like data being corrupted or stored improperly? I'd think an SSD would have problems similar to RAM that's bad.

After a few years, I'm really beginning to hold on to my thought that an SSD is still sort of no-man's land. They're still a bit too new for a lot of people to know what's wrong with them.

 

[simonsi]

So far on an SSD we basically have two problems, total wipeout that may or may not be recoverable, and blocks dropping.

How about something like data being corrupted or stored improperly? I'd think an SSD would have problems similar to RAM that's bad.

After a few years, I'm really beginning to hold on to my thought that an SSD is still sort of no-man's land. They're still a bit too new for a lot of people to know what's wrong with them.

But I'm not sure either of these issues are occurring at any higher rate than with HDDs, and certainly the trade-off for speed appears worth it. I certainly had zero issues with my first SSD - quite unlike my very first HDD for example...

 

[B-Eugen]

So far on an SSD we basically have two problems, total wipeout that may or may not be recoverable, and blocks dropping.

How about something like data being corrupted or stored improperly? I'd think an SSD would have problems similar to RAM that's bad.

After a few years, I'm really beginning to hold on to my thought that an SSD is still sort of no-man's land. They're still a bit too new for a lot of people to know what's wrong with them.

Just keep good backups….on a hard drive.

 

[alex0002]

So far on an SSD we basically have two problems, total wipeout that may or may not be recoverable, and blocks dropping.

How about something like data being corrupted or stored improperly? I'd think an SSD would have problems similar to RAM that's bad.

Unless we are talking about ECC-RAM then RAM has no protection against bad blocks. They can't be mapped out of the usable area like on an SSD. Technologies like RAIN help too.

I've been using SSD's since 2012 with no data loss due to the SSD. I'm also working with servers where I see a lot of 15,000 rpm SAS hard drives failing. Just make backups like you should already do for hard disks.

 

[ZVH]

Bad blocks can refer to a block of data just going bad, similar to a bad sector on an HDD, or it can refer to an entire chip in the process of failing. This is what happened to the OP because he kept getting more and more and more bad blocks on a new SSD and he says it was replaced by the manufacturer. I'm assuming one of the SSD RAM chips was failing.

On the other hand, theoretically if it's a single occurrence due to some oddity and it's in an existing file, I'm uncertain SSD management software would correct it. For example suppose a file exists on a system and it consists of blocks A, B, C, and D, and block C fails. The file is now corrupt. When the user accesses the file and the SSD controller detects an error, does it leave in in place, or remove it? If the file is left in place blocks A, B, and D could be recoverable, but if the SSD just decides the file is corrupt it could kill the whole thing (all blocks).

The reason I mention this is because a year or two ago someone posting here was having that very problem: an SSD would drop a block in the middle of a file and just leave it there. It acted just like a bad sector on a hard drive. If I recall correctly when it happened he had to re-initialize the drive then restore the whole thing from backup.

 

[TheBSDGuy]

Controllers typically don't delete bad blocks or sectors in existing files until the user initiates changes to the file. They'll just sit there until they're either deleted by the user or they're overwritten/re-used. When the user initiated change occurs, then the controller will, theoretically, do any sector or block remapping as needed.

 

[ZVH]

Well, here's another SSD problem, not really an SSD problem but a Yosemite problem, and that has to do with the inability of 3rd party TRIM to be enabled in Yosemite. I saw posts about this early on and just assumed it was because drivers/kext files were outdated, but apparently Apple has put part of its kext signing in the NVRAM of all places and without manually adjusting NVRAM settings to disable the security feature, third party TRIM won't work.

I haven't tried this yet because, like many, I put Yosemite on an external drive all by itself (and that's where it's going to stay.)

 

[simonsi]

Well, here's another SSD problem, not really an SSD problem but a Yosemite problem, and that has to do with the inability of 3rd party TRIM to be enabled in Yosemite. I saw posts about this early on and just assumed it was because drivers/kext files were outdated, but apparently Apple has put part of its kext signing in the NVRAM of all places and without manually adjusting NVRAM settings to disable the security feature, third party TRIM won't work.

I haven't tried this yet because, like many, I put Yosemite on an external drive all by itself (and that's where it's going to stay.)

Thats the point, its done that way so an app cant defeat the security. Ive run without Trim for 6mths with no ill effects on two different SSDs, IMHO Trim is over-rated. There are various work-arounds in any case so not really a problem whichever way you want to go, I stayed with security YMMV.

 

[B-Eugen]

Thats the point, its done that way so an app cant defeat the security. Ive run without Trim for 6mths with no ill effects on two different SSDs, IMHO Trim is over-rated. There are various work-arounds in any case so not really a problem whichever way you want to go, I stayed with security YMMV.

Some people might argue that SSDs are over rated as well. I see an obvious difference in boot speed with an SSD, but when using applications, I really see no earth shattering change. If I did a lot of disk intensive stuff I know the difference would be noticeable but I don't. If my system was being bounced around a lot I would expect an SSD to be more reliable, but it doesn't happen to me.

Regarding TRIM, I think the real problem is with the way Cindori implemented TRIM Enabler. There are documents on the web from Apple indicating that they were going to implement this stuff a fairly long time ago. I also find it sort of hard to believe that TRIM absolutely must be implemented the way they did it. Why not do it with a background process that loads after the system is up and running and then runs in the background?

I also have to agree about the benefits of TRIM. When I put my first SSD in I got TRIM enabler, and after an update to the OS it became disabled. I never noticed the difference. I just happened one day to look at the system profile and then realized it became disabled during the OS update and I had been running it like that for months.

 

[simonsi]

Some people might argue that SSDs are over rated as well.

Yep, setting expectations that they fix any and all OS X performance problems is a fallacy.

Work well for me in Capture One though (or any photo editing I expect), as they are continually loading and writing files as part of workflow...

 

[Cyberpower678]

I thought SSDs were supposed to be problem free!!!

I've been battling periodic lock ups using my system. I finally pulled out a copy of Scannerz and did a test on it and it found a bad block. The only reason I ever got Scannerz was about a year ago I was having problems with my then 5 year old drive. It was toast. It had almost 10GB of bad data on it probably from a head crash. I pulled it and put in an OWC SSD because I thought these things weren't supposed to have problems until they're like 5 years old. Seemed reasonable.

I called the Scannerz people and they told me I could either use Phoenix to copy the data off the drive onto an HD, wipe the SSD and then clone the data back onto the SSD. They said Phoenix would flag any files that failed to be copied in the log files and that way I would know what to replace from back up. The other option they gave me was to completely re-initialize the SSD and restore from a back up. They said the SSD should correct itself using either of these techniques and if it doesn't I should contact OWC for a replacement.

The thing that's ticking me off is that it seems I'm back to being reliant on HDs YET AGAIN!!! I'm using HDs to clone to and restore from and it's time consuming. I used the old OEM drive in the system for 5 years without a problem. Here I am, just about a year later, and an SSD is basically giving me the same types of problems.

Why did the SSD leave a bad block on it and why didn't it correct it. I thought these things could automatically fix these things. If given an option to get a replacement SSD or an HD instead, should I opt for an HD instead? Am I just unlucky?

You are definitely unlucky, or I, and all of my friends, are extremely lucky. SSDs are incredibly reliable. I've been using my for 3 years now, and still no performance degradation. You may have gotten a defective unit, or you have it attached to defective hardware, or incompatible hardware.

P.S. sorry if I'm repeating. I only read the OP.

 

[BradHatter]

You are definitely unlucky, or I, and all of my friends, are extremely lucky. SSDs are incredibly reliable. I've been using my for 3 years now, and still no performance degradation. You may have gotten a defective unit, or you have it attached to defective hardware, or incompatible hardware.

P.S. sorry if I'm repeating. I only read the OP.

The SSD had a bad memory chip in it, so it was an oddity. I Received a new one months ago and I haven't had any problems with it.

The thread has been interesting especially some of the test reports about failures and longevity. Most of the hard drives I've had have been reliable, with many replaced because I wanted more space.

It should be interesting to see how SSDs hold up in the long run.

 

[Cyberpower678]

The SSD had a bad memory chip in it, so it was an oddity. I Received a new one months ago and I haven't had any problems with it.

The thread has been interesting especially some of the test reports about failures and longevity. Most of the hard drives I've had have been reliable, with many replaced because I wanted more space.

It should be interesting to see how SSDs hold up in the long run.

Glad to hear it.

 

[ZVH]

Thats the point, its done that way so an app cant defeat the security. Ive run without Trim for 6mths with no ill effects on two different SSDs, IMHO Trim is over-rated. There are various work-arounds in any case so not really a problem whichever way you want to go, I stayed with security YMMV.

I would think this might effect other things like some of WD's drive management software. They don't use TRIM because they're external hard drives, but never the less they still need to make low level kernel access.

 

[BradHatter]

I believe the external drive SAT drivers that are open source won't work properly under Yosemite unless that security feature is disabled as well.