Question What up with my Crucial SSD's?

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
I've got 2 Crucial 500GB MX500 SSD's in 2 different PC's... one is the OS drive, and the other, after serving as the OS drive in another rig, has been relegated to scratch drive duty in my main desktop. What concerns me is the Health Status... both of these drives have fairly low hours and writes, but are already showing reduced life. I don't get it.

I've got a crappy old OCZ 60GB drive that has been thrashed... it's showing 99% life even after 8 years of use (20TB of writes over 10K hours, on about 80% less nand?)

I have a Crucial M550 525GB drive (formerly the OS drive in my desktop before the update to M.2) and it shows nothing for wear, with far more hours and writes.

Is this the result of needless TRIM? (These are/were on W7 systems.) Were the MX500's so 'budget driven' that they fail sooner?

What am I missing?
 

Attachments

  • HTPC MX500.jpg
    HTPC MX500.jpg
    191 KB · Views: 67
  • IPS5 MX500.jpg
    IPS5 MX500.jpg
    151.4 KB · Views: 57

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
But I wouldn't personally be worried about the drive health indication in the pictures.

I do use Storage Executive on both drives, along with CrystalDisk to monitor all of them... but there must be a reason it's triggering, even CrystalDisk alerts on it when I open it.

Back in the Old Days, my Samsung 840Pro OS drive up and died on me with no warning... it's got me a little skeert. Granted, the one MX500 drive that is serving as OS drive is not on a critical PC... but I wouldn't want to be forced into an emergency reimage if at all possible.
 

UsandThem

Elite Member
May 4, 2000
16,068
7,382
146
I do use Storage Executive on both drives, along with CrystalDisk to monitor all of them... but there must be a reason it's triggering, even CrystalDisk alerts on it when I open it.

Back in the Old Days, my Samsung 840Pro OS drive up and died on me with no warning... it's got me a little skeert. Granted, the one MX500 drive that is serving as OS drive is not on a critical PC... but I wouldn't want to be forced into an emergency reimage if at all possible.
Just keep an occasional eye on their reported health, and replace if they get significantly worse. They have a 5 year warranty on them, so you should be covered. Other than that, there's not much to do (or worry about) at this point with one showing 99% and the other 93%.

The part that is triggering the lowered life is 'average block erase account', and I've seen various reasons on what impacts that. Some users have discovered it was due to their antivirus program: https://hardforum.com/threads/what-could-cause-this-extremely-high-wear-leveling-crucial-m4.1892200/
 

Furious_Styles

Senior member
Jan 17, 2019
492
228
116
I do use Storage Executive on both drives, along with CrystalDisk to monitor all of them... but there must be a reason it's triggering, even CrystalDisk alerts on it when I open it.

Back in the Old Days, my Samsung 840Pro OS drive up and died on me with no warning... it's got me a little skeert. Granted, the one MX500 drive that is serving as OS drive is not on a critical PC... but I wouldn't want to be forced into an emergency reimage if at all possible.

Sorry to hear about the ol' 840 pro. Still using mine as an OS drive and no signs of wear. That is always the one drawback of SSDs though, you rarely get warnings that they are failing, you just drive straight off the cliff.
 
  • Like
Reactions: Charlie98

NewMaxx

Senior member
Aug 11, 2007
250
31
91
Health is determined by the P/E rating of the drive. For example in the 93% drive: AD value is 69, or 105 in decimal, so (105)/(.07) = 1500. This means the drive is at 105 drive writes (P/E) and it's rated for 1500, although most likely the flash on the MX500 will survive at least 3-5K in my experience. Your write amplification is relatively high, though.
 

Lucretia19

Member
Feb 9, 2020
25
5
41
I too have a Crucial MX500 500GB ssd (SATA) and I'm alarmed by its accelerating decrease of Remaining Life. For example, it dropped from 94% to 93% after the pc wrote only 138 GB. That's terrible. Previously it dropped from 95% to 94% after writing 390 GB, almost as terrible. I've used the F7 and F8 SMART attributes to calculate its Write Amplification Factor (WAF = "1 + ΔF8/ΔF7" where the Δ refers to the increase of the attribute over a period of time) over recent days, and WAF has averaged approximately 50 over the last three days. WAF should be closer to 1.5 or 2.

In the thread at hardforum.com linked above, some people several years ago blamed BitDefender for a WAF that was approximately 25 (assuming I calculated correctly) in a Crucial M4 256 GB ssd, but in my opinion a huge WAF should always be blamed on the ssd, not on software running on the pc.

My pc has Comodo antivirus suite, and perhaps I'll see what happens if I disable it.

Another problem I've had is that Momentum Cache wouldn't activate.

Crucial/Micron tech support have been unable to figure out either problem.

Below is a screencapture that I made after the Remaining Life reached 93%. The spreadsheet at the left of the image shows the recent rapid acceleration of the decrease of Remaining Life.
Crucial MX500 500GB SSD 93% Remaining Life, 6310 GB Written (2020-02-04 screencapture 16 color...png
 

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
Sorry to hear about the ol' 840 pro. Still using mine as an OS drive and no signs of wear. That is always the one drawback of SSDs though, you rarely get warnings that they are failing, you just drive straight off the cliff.

Oddly enough, I got a refurb 840Pro in return when I RMA'ed it... but I couldn't trust it. On a fluke, I installed it on my mom's Dell, to replace the smaller Intel SSD... and it's running wonderfully 4 years later.

Some users have discovered it was due to their antivirus program

I'll have to think about that. I've installed some newer utilities lately. Dumb question... is Garbage Collection / TRIM still a 'thing' with SSD's on updated OS like W10? I know it was quite an issue back in the day with W7, when SSD's became more mainstream.

My previous OS drive, a M550... I just pulled it out to wipe it to maybe reuse it somewhere... has 31K hours on it and only shows 97%... but that drive was on W7. I'm probably overthinking this... but when a drive with less than 10K hours on it starts to show it's age... I get a little queasy.
 

Furious_Styles

Senior member
Jan 17, 2019
492
228
116
Oddly enough, I got a refurb 840Pro in return when I RMA'ed it... but I couldn't trust it. On a fluke, I installed it on my mom's Dell, to replace the smaller Intel SSD... and it's running wonderfully 4 years later.



I'll have to think about that. I've installed some newer utilities lately. Dumb question... is Garbage Collection / TRIM still a 'thing' with SSD's on updated OS like W10? I know it was quite an issue back in the day with W7, when SSD's became more mainstream.

My previous OS drive, a M550... I just pulled it out to wipe it to maybe reuse it somewhere... has 31K hours on it and only shows 97%... but that drive was on W7. I'm probably overthinking this... but when a drive with less than 10K hours on it starts to show it's age... I get a little queasy.

Garbage collection/trim is certainly still used. Win 7 (at least post SP1) has no problems I know of with SSDs, I'm still on 7 with that 840 pro!
 
  • Haha
Reactions: Charlie98

UsandThem

Elite Member
May 4, 2000
16,068
7,382
146
I'll have to think about that. I've installed some newer utilities lately. Dumb question... is Garbage Collection / TRIM still a 'thing' with SSD's on updated OS like W10? I know it was quite an issue back in the day with W7, when SSD's became more mainstream.
Like Furious_Styles stated, yes it is. It's a standard function of SSDs, and I've never had an issue with it on any of my SSDs (Intel, Samsung, Crucial, WD, Kingston, Crucial, etc.) on Windows 7 or Windows 10.

Most of the time, it will likely be a program that is the problem. For example, Norton Internet Security would automatically "optimize" my disk during idle time. The problem being instead of running a TRIM command it would run a full defrag, which you don't do on a SSD as it is not needed at all, and really can lower the drive's life if run on a regular basis.

So I learned with that program, I needed to go into the program settings and disable that function.
 

Lucretia19

Member
Feb 9, 2020
25
5
41
I have some "good" news to report: My Crucial MX500's Write Amplification Factor is down to about 11 during the last 24 hours. That's still poor, but much better than its 50-ish average during the previous 3 days (when I began recording the ssd's "daily WAF," calculated using the daily increases of the F7 & F8 S.M.A.R.T. attributes).

Perhaps the WAF improvement was caused by one the following actions yesterday:
1. I stopped running CrystalDiskInfo.
2. I stopped running HWiNFO.
3. I briefly shut down the pc.
(The pc had been powered for many weeks, with occasional restarts but no shutdowns, so I theorized that the ssd might need to be reset by a power off/on, similar to how Windows must be restarted every few weeks to prevent Windows from misbehaving or stop it from misbehaving.)

I'll continue to log the MX500's daily WAF. In a few days I'll resume running CrystalDiskInfo or HWiNFO to learn whether they're related to the 50-ish WAF.

My interpretation of the thread linked in a post above, about the antivirus program, is that one user of Bitdefender reported about 4 years ago that it appeared to be a major contributor to his ssd's high amount of writing. Several people in that thread concurred that Bitdefender was the culprit, because the ssd started writing much less to NAND after the user uninstalled Bitdefender. However, the thread didn't indicate the experiment was properly completed; in other words the user didn't verify Bitdefender actually was the problem by reinstalling it to see if the problem would recur. It's possible the problem actually went away due to the pc restart that was presumably one of the uninstall steps, in which case Bitdefender may have been completely innocent with regard to the ssd's large WAF. (The WAF was approximately 25.)

On the other hand, Bitdefender apparently wrote a lot of temporary files to the user's ssd. That wouldn't explain the ssd's large WAF, but it's possible that Bitdefender could have been reconfigured not to write the temporary files to the ssd. On my pc, Comodo's weekly full virus scan wrote a lot of temporary files (once a week) to the ssd until I decided to disable scanning of compressed archives. I assume Comodo needs to uncompress archives in order to scan their contents, but if Comodo were smarter it would give the user a choice of where to uncompress... to a hard disk, for example, or perhaps to ram. (It should be straight-forward to create a Windows symlink so that Comodo will uncompress to a hard drive, if I learn which ssd folder is the normal destination for uncompression.)
 
Last edited:

R81Z3N1

Member
Jul 15, 2017
77
24
81
I would be intrested is seeing a Linux Luks comparision, right now I have a M500 sata 500G model and get this error in my logs.

"WARNING this firmware returns bogus raw values in attribute 197"

When looking up this error, it says it's mostly harmless, but I don't expect an update anytime soon. Also this is on a Luks container, with discards disabled. With a basic /etc/fstab file set as defaults.

It would be nice if Micron or Crucial could update the firmware, I am not that worried about it but. I am more ticked off at Motherboard beta bios as I lost stability. I do get sudden lockup's since beta bios, but part of me wonders if it's the drive. Bad bios, and bad firmware in ssd, probably doesn't give the warm fuzzy's. But if problem continues, I might just RMA the board and make MSI eat the cost for such bad support.

R81Z3N1
 

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
I'll continue to log the MX500's daily WAF. In a few days I'll resume running CrystalDiskInfo or HWiNFO to learn whether they're related to the 50-ish WAF.

Please post your results here, I'll be curious to see them. I'm going to start my own diagnostic process and see if I can find the culprit. In my case, it's a little different... the high mileage is on the W7 OS drive, the younger of the drives, on a machine that has largely remained unchanged in about 5 years.
 

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
Most of the time, it will likely be a program that is the problem. For example, Norton Internet Security would automatically "optimize" my disk during idle time. The problem being instead of running a TRIM command it would run a full defrag, which you don't do on a SSD as it is not needed at all, and really can lower the drive's life if run on a regular basis.

So I learned with that program, I needed to go into the program settings and disable that function.

I see how much I've forgotten... on a system that has been rock solid for 8 years. I'd forgotten about defrag and such... I'll have to dig out my old notes and start going through the system.

EDIT: FWIW, the W10 system recognizes both SSD's and it looks like it does cleanup once a week... maybe longer. The W7 system... I still have defrag turned off completely, for all the drives.

The process continues... :)
 
Last edited:

Lucretia19

Member
Feb 9, 2020
25
5
41
@Charlie98: Consider logging the SMART F7 and F8 attributes on a daily basis, so you can use the changes in F7 and F8 to calculate the "daily Write Amplification Factor" (or WAF during some other short time period) to help you see the effects of experiments you try.
 

Lucretia19

Member
Feb 9, 2020
25
5
41
@Charlie98: What did you mean, earlier in this thread, when you said it's been "triggering" and CrystalDiskInfo "alerts" about it? Your image of CrystalDiskInfo doesn't show a yellow alert; it's all blue. (The limited portion that we can see, that is... in the future you might want to resize the Crystal window when you screencapture, so we will be able to see all of the SMART data.)

Regarding the "good" news I reported yesterday about the daily WAF shrinking to 11... unforunately it's gone up considerably during the approximately 21 hours since then, to about 30. (In a few hours I'll check again, when I log my "official" daily values.) Possibly the 11 was a fluke, or perhaps a short term benefit of briefly turning off the pc (resetting the ssd). Hopefully the next few days of logging will shed more light.

Regarding the possibility of a defragger running, which you and someone else discussed earlier, couldn't that possibility have been ruled out by the small 3,961 GB Total Host Writes? The problem clearly seems to be the ssd's large Write Amplification Factor, not the host pc writing more than necessary. Given the small Total Host Writes, I think the only thing the pc might be doing wrong is failing to convert frequent small random writes into occasional large sequential writes by failing to cache writes cleverly.

I'm wondering if the high WAF could be caused by an ssd firmware "bug" that reveals itself only when the pc writes very little to the ssd. In other words, perhaps the ssd has a background "optimization" process that's scheduled to run frequently, much more frequently than necessary in the case when little is written to the ssd. Micron's software engineers might have the process scheduled to run periodically, with a period that's far too short when the pc writes at a low rate. Both you and I have very low rates of host pc writing to the ssd. After Remaining Life reached 95% in December, I noticed (using HWiNFO) that the pc was averaging more than 1 MByte/second writing to the ssd, and I reduced the average to less than 0.1 MByte/second by moving some frequently written files from ssd to hard drive and disabling the scanning of archive files during Comodo's weekly full virus scan. Since then, the reduction in Remaining Life has accelerated, paradoxically. But I don't know if the reduced writing was the cause of the acceleration; it could be coincidence. This theory could be tested by moving the frequently written files back to the ssd from hard drive, and reenabling Comodo's scanning of archive files, to see if it reduces WAF and paradoxically slows down the decrease of Remaining Life.

Does your ssd's 1550 Power On Hours look correct to you? Storage Executive reports my ssd has had 991 POH. But my pc has been set for the last 5 months to Never Sleep (and Never Turn Off Hard Disk) so I don't understand how the ssd could have spent much time in a low power state during those months. When I observe HWiNFO, it appears the pc never goes more than a few seconds without writing something to ssd. Perhaps the strangely low POH implies the ssd goes into one of the low power states as soon as it completes an operation, rather than waiting for an idle timer to eventually trigger the switch to a low power state. Perhaps most computers write so much more to their ssd that their ssd rarely enters a low power state, and perhaps the transition to or from low power involves writing extra to the ssd.

Another untested theory is that VMware Player has a virtual disk driver that's buggy and interferes with Windows' write caching, so that many small random writes to ssd occur instead of a few large sequential writes, as if the Windows write cache were turned off. I almost never run a vm anymore, but assuming VMware has a disk driver installed, it could in principle affect disk writing even when Player isn't running.

Did you try to enable Crucial/Micron's Momentum Cache? I tried to install it in January, after Remaining Life reached 95%. After I installed Storage Executive (which includes Momentum Cache) it reported that Momentum Cache was "inactive," and tech support has been unable to find the reason why it won't activate. On the theory that the "inactive" Momentum Cache driver might be interfering with Windows write caching, a few days ago I "disabled" Momentum Cache, which I think caused Storage Executive to delete the Momentum Cache driver before it restarted the pc. That apparently didn't solve the problem. In a few days, I plan to uninstall Storage Executive entirely to see if that helps. (After I uninstall Storage Executive, I'll need to reenable CrystalDiskInfo or install some other SMART monitor to be able to continue measuring the daily WAF.)
 

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
A few things...

I don't run CrystalDisk actively, I only open it when I want to check on the drives. When I open it, it sounds the little melody that usually means there's an issue (such as my first portable HDD, which has bad sectors, for example.) Nothing shows caution... but what I'm thinking is the health status being at 93% is below some sort of triggering threshold, but not enough to highlight the health status box to yellow.

It's interesting, I had a Plextor SSD for a while, it showed hardly any power on hours... on a PC that ran 24/7 (this one.) I've seen other SSD's do the same thing, but I've had so many SSD's I don't recall which ones are equivalent to what I would expect. The SSD that I pulled out of this before the upgrade to the AMD chip, a Crucial M550, shows 31K hours... which is more or less correct, and I believe the ~1500 hours on the 93% drive is correct, also. That drive is in my HTPC, which was used very little in the past year since I upgraded to that drive, but that PC is now turning into a 24/7 PC as my wife is starting to use it more... and the hours are starting to roll up, like they should... but it definitely has some sort of power save mode... it only shows 2 power on hours since yesterday, and it's been on since Saturday.

I use Storage Exec, but like CrystalDisk, it's not always open... I only open it to check the FW, or check the status of the drive. I've found it to be a rather testy program... I can't even get it to open on my game rig... with an MX300.

The AV thing might be the culprit. I use both SAS and Malwarebytes, along with Defender. I had SAS set to scan ALL files, including ZIP files, and files larger than 4MB because our home network was hacked last year, so I have been going overboard on virus scans. I've disabled the large and ZIP file scan... we will see if that makes a difference.
 

Lucretia19

Member
Feb 9, 2020
25
5
41
@Charlie98: You can check whether CrystalDiskInfo is set to caution you at 93% by using its menu (Function | Advanced Feature | Health Status Setting). The default setting appears to be 10%, since I don't recall ever tinkering with that setting. CDI also appears able to sound an alert for excessive sector errors and excessive temperature.

Do you have software that can display the realtime rate of writing to the ssd? I normally use HWiNFO (one of the two monitoring programs I've stopped running since Saturday). HWiNFO helped me discover that Comodo was writing heavily during its weekly full virus scan... by chance I looked at the HWiNFO ssd write rate while Comodo was running its weekly virus scan in background, and when I saw the high write rate I looked around some more and found Comodo was scanning. I didn't understand why a virus scan might write so much, and after exploring Comodo's scan options the only one that seemed plausibly related to writing is the scanning of compressed archives. Disabling that option did indeed significantly reduce the ssd write rate during Comodo full scans, but due to the WAF I don't know if it's helping or hurting the ssd.
 

Lucretia19

Member
Feb 9, 2020
25
5
41
After collecting WAF data for a couple more days, it appears the 11 daily WAF two days ago was just a fluke. Yesterday's daily WAF was about 37 and today's about 53. See the column in red:
Date​
Total Host Writes (GB)​
S.M.A.R.T. F7​
S.M.A.R.T. F8​
WAF = 1 + F8/F7​
Total Amplified Writes (GB)​
ΔF7 (1 row)​
ΔF8 (1 row)​
Recent WAF = 1 + ΔF8/ΔF7​
02/06/2020
6,323
219,805,860
1,229,734,020
6.59​
41,698​
02/07/2020
6,329
220,037,004
1,242,628,588
6.65​
42,071​
231,144​
12,894,568​
56.79
02/08/2020
6,334
220,297,938
1,252,694,764
6.69​
42,351​
260,934​
10,066,176​
39.58
02/09/2020
6,342
220,575,966
1,269,273,190
6.75​
42,836​
278,028​
16,578,426​
60.63
02/10/2020
6,351
220,857,490
1,272,080,434
6.76​
42,931​
281,524​
2,807,244​
10.97
02/11/2020
6,357
221,087,760
1,280,283,705
6.79​
43,169​
230,270​
8,203,271​
36.62
02/12/2020
6,365
221,357,482
1,294,326,214
6.85​
43,583​
269,722​
14,042,509​
53.06

Cumulative WAF (over the life of the drive) has been increasing monotonically, now up to 6.85. ABEC incremented again, to 110.

Each day since I stopped running CrystalDiskInfo and HWiNFO, I've launched Crystal for about a minute to obtain Total Host Writes. (That's for consistency. Storage Executive displays it in TB and for an unknown reason SE's value is about 400 GB higher than the Crystal value of THW.) Although it seems theoretically possible that briefly running Crystal each day hurts WAF for much longer than the brief periods, that seems very unlikely, so my tentative conclusion is that neither CrystalDiskInfo nor HWiNFO have been relevant to the large WAF, and I'll resume running them.

For my next experiment, I'll uninstall Storage Executive (briefly shutting down the pc after uninstallation to ensure the ssd is reset), to see if SE had any effect on WAF. (Future experiments planned: uninstall of VMware Player; increasing host writes by moving frequently written files back to ssd from hard drive.)
 
  • Like
Reactions: Charlie98

Lucretia19

Member
Feb 9, 2020
25
5
41
To avoid redundancy, I'm going to post my ssd logs and experiment descriptions only in the following forum thread:

I posted my most recent results (a 3-day experiment) there a few minutes ago.

But I'll continue to respond here to posts here.
 
  • Like
Reactions: Charlie98

Charlie98

Diamond Member
Nov 6, 2011
6,294
64
91
Actually, I'm looking at my primary desktop data and nothing is going on.

I've got to log on to the HTPC (the one with the 93% drive) and see what's there...
 

Lucretia19

Member
Feb 9, 2020
25
5
41
Here's some SMART data from 14 Crucial MX500 500GB ssd drives:
SSD ID#​
Power On Hours​
Power Cycle Count​
F7 Host Page Writes​
F8 FTL Background Page Writes​
Average Block Erase Count​
Remaining Life %​
POH/PCC
WAF = 1 + F8/F7​
B9A1A3E78127​
889​
1067​
263343469​
210499276​
56​
97​
0.83​
1.80​
58A732AD4741​
1565​
1738​
100555421​
141793361​
18​
99​
0.90​
2.41​
D472FF24BE77​
571​
200​
113059179​
105747605​
24​
99​
2.86​
1.94​
78DA41D1D402​
5912​
545​
214838823​
758396846​
76​
95​
10.85​
4.53​
7E902D148339​
1813​
14​
132850575​
765980653​
65​
96​
129.50​
6.77​
M.2
0D62A18613DA​
427​
525​
174019463​
149379660​
35​
98​
0.81​
1.86​
186E7EB108B1​
1479​
955​
167561718​
199215317​
35​
98​
1.55​
2.19​
81317ED21602​
64​
15​
1079385​
226493​
0​
100​
4.27​
1.21​
E65B05D7DBCC​
920​
368​
49884486​
45245571​
10​
100​
2.50​
1.91​
1828E148B537​
13​
14​
1127008​
46217​
0​
100​
0.93​
1.04​
1813E134D584​
3468​
2058​
735469391​
3295350788​
288​
81​
1.69​
5.48​
1826E1466CB5​
85​
115​
12191194​
10203030​
2​
100​
0.74​
1.84​
Charlie98​
1550​
131​
105​
93​
11.83​
10
Mine 2020-01-15​
883​
94​
214422794​
959278784​
90​
94​
9.39​
5.47​
Mine 2020-02-17​
1029​
99​
223137063​
1344405734​
113​
93​
10.39​
7.03​
Mine 2020-02-19​
1035​
99​
223545998​
1365068972​
114​
93​
10.45​
7.11​

Four of the ssds are M.2 and the rest are 2.5". My hunch is that the form factor isn't relevant.

Three rows are for my ssd, recorded on different dates, because the changes over time can be informative. For example, the changes are needed to calculate that its WAF has been extremely high during recent weeks.

Only partial data is included for Charlie98's ssd (due to his cropped screencapture). My rough estimate of 10 for its WAF is based on its Total Host Writes (3961 GB) and Average Block Erase Count (105). Hopefully Charlie98 will start logging SMART data periodically, so the changes over time can be analyzed. In particular, we'll be able to determine whether its WAF over some recent period of time is much worse than it used to be, like mine is.

Much of the data is suggestive that WAF is helped by frequent power cycles (see the ratio of Power On Hours to Power Cycle Count) and hurt by ssd age (increases over time).

I probably should have copied each drive's firmware version, since a bug in the latest version might not have been present in previous versions. On the other hand, the version could be misleading because a drive might have had its firmware updated not long before its SMART data was posted. The drives' SMART dates (which I also neglected to copy) and Power On Hours might provide helpful hints about the validity of a version interpretation, since the firmware release dates can be looked up.
 
  • Like
Reactions: Charlie98