Building a new PC

Last edited:
That looks fine but you might want to check the SMART and Information tabs for any anomalies. Report > Send Test Report to Developer will enable the dev to check for you if there's anything that looks out of place, Questions • Hard Disk Monitoring is probably the best area for a query. In my experience, the dev is very responsive.
 
satrow said:
That looks fine but you might want to check the SMART and Information tabs for any anomalies. Report > Send Test Report to Developer will enable the dev to check for you if there's anything that looks out of place, Questions • Hard Disk Monitoring is probably the best area for a query. In my experience, the dev is very responsive.
Click to expand...

SMART and information all checks out through the tool. I sent the report and will let you know if/when I get a response.
 
Digerati said:
Is the SSD more than 3 years old? That's the warranty period.
Click to expand...

It's currently two and a half years since purchasing. I have considered that, but I'll wait and see how things go. I'm sure the Crucial support rep will let me know if/when a replacement is in order.
 
Not sure I am buying that. If it was a first generation SSD on a old operating system, maybe but I think you would be getting read and/or write errors, not missing drive errors.
 
softwaremaniac said:
I had the exact same issue when friend's SSD was dying. Switched to a Samsung and has been running problemfree ever since.
Click to expand...

Well that is not a comforting post to read. :frown6:

The drive has been more stable since I did the power cycle, but it's only been 36 hours or so. Since the problem was typically every 20-30 hours, I'm not confident that the drive is working properly yet. I imagine I'll have to go through the steps with the garbage collection to get to an RMA process either way, so I'll try to perform that sometime this week. I'll probably re-image to an hdd so I can keep the system up and running during the garbage collection process.
 
I have only today had the time to check the topic since my last reply and everything matches up (including the 97% Health in HD Sentinel). Feel free to try it, I'd be happy to learn something if that in fact does do the trick for you :)
 
Digerati said:
Not sure I am buying that. If it was a first generation SSD on a old operating system, maybe but I think you would be getting read and/or write errors, not missing drive errors.
Click to expand...

I had similar doubts. The support rep could be right, though. This system is rarely idle for 6-8 hour periods, so it may not have had a full garbage collection in a while. The system has been more stable the past couple days after the power cycle and garbage collection process. I cannot say for sure which one helped if either as not enough time has passed to really know and both happened within 56 hours of each other. 56 hours was a long time without it happening, so I'd lean more toward the power cycle. The system does seem a little faster since the garbage collection, too.

The last crash was around 10:00 PM on August 4th.
 
This system is rarely idle for 6-8 hour periods
Click to expand...
Yeah, but a busy CPU and RAM does not imply constant drive access. And garbage collection is important for writes. Reads are no big deal, other than they keep the drive active. I also note that typically, garbage collection is basically a fail safe for when TRIM fails or is unable to run.

I found this Crucial SSD article an interesting read - especially the part about garbage collection with the data cable disconnected.
 
Digerati said:
Yeah, but a busy CPU and RAM does not imply constant drive access.
Click to expand...

True, however this is a system drive with Windows installed on it. That means there is a lot more read/write activity than if it were an additional drive.

I use Macrium Reflect to backup my drive daily through differential backups with a full backup monthly. My monthly backup ran last weekend at 3 AM, and my daily backup ran subsequently at 4 AM. There were 0.5 GB worth of changes in that hour alone, which means that 0.5 GB of the sectors changed. I believe that shows that the drive is not idle very often, and probably not for a full duration of 6-8 hours while Windows is up and running.


Digerati said:
I also note that typically, garbage collection is basically a fail safe for when TRIM fails or is unable to run.
Click to expand...

There seems to be varying opinions on that point. See the below quote for reference (From page 2 of Garbage Collection and TRIM in SSDs Explained - An SSD Primer | The SSD Review). Essentially, it's possible that TRIM merely sets sectors aside that are ready to be removed/overwritten through the garbage collection process. If garbage collection does not run, it would seem that data can accumulate and still take up sectors on the drive. While TRIM allows the drive to run faster for subsequent writes, it does not ensure that garbage collection keeps the drive free to do more writes later (see the physical drive representation in the image that follows the quote). If that is the case, forcing garbage collection as outlined in the Crucial article may still be necessary even with a newer OS, at least from time to time.
TRIM command[:] In this case, the process is the same when the user writes the four original files. When the user erases file C in column two, however, the old file C space is immediately marked as invalid in preparation for GC because the OS uses the TRIM command. The space originally holding file C is now free space to the SSD and becomes available as dynamic over provisioning meaning the SSD has more free space during GC which helps improve the overall SSD performance.
In column three when the user finally writes file E, the total free space on the SSD is now the same as the drive without TRIM. The difference is that the SSD with TRIM support knows what data is invalid and it can be considered free space during GC and prevents having to move the invalid data to another block.
This produces three key benefits:
Lower write amplification. Less data is re-written and more free space is available during GC (more space to write equals fewer writes needed);
Higher throughput. With the TRIM command, there is less data to move during GC and the drive runs faster. Throughput is bottlenecked at the flash an SSD is only as fast as it can write to the flash memory. During the time it is doing GC, the drive has to stop some of the data transfer from the host while it moves data around. This is why it’s beneficial for the SSD to know which data is invalid so it doesn’t have to be moved during GC.
Improved endurance, because the drive is writing less to the flash by not rewriting invalid data.
Click to expand...

Figure-3.png​
 
Last edited:
There seems to be varying opinions on that point.
Click to expand...
Well, I certainly am no expert in that area but I would like to think that Crucial knows how their SSDs work.
 
Digerati said:
Well, I certainly am no expert in that area but I would like to think that Crucial knows how their SSDs work.
Click to expand...

Me too, which is why I went ahead with the forced garbage collection as recommended by Crucial's support rep. If he thought it was a possible reason for the crashes/lost drive, it could be depending on how their garbage collection works on that drive.
 
Digerati said:
Well, it seems clear it helped so that's a good thing!
Click to expand...

It helped for ~5 days. I walked out to the kitchen to get a drink of water, and I came back to a black screen saying the system could not find my boot device. Power cycling the PSU didn't help, nor did loading optimized defaults in the BIOS. I tried power cycling and then loading, and I tried loading and then power cycling. Neither worked. The drive is not found by the system at all.

I am currently running off the hard drive I imaged onto this past Monday. The SSD is sitting in an enclosure and working just fine. For some reason, my SSD does not like this motherboard or vice versa. The other Crucial SSD I have is always detected without issues.


I may run for a week or two with this hard drive connected and see how the system behaves to determine whether it might be a motherboard issue. Of course, this hard drive has its own issues since it doesn't pass a short drive test through SeaTools. :thud:
 
Last edited:
softwaremaniac said:
Could you please try usingh this and telling me what the findings are? Crucial Storage Executive Tool | Firmware Download | Crucial.com
That is the official Crucial tool for monitoring the drive, so it may help.
Click to expand...
Items of interest and warnings:
  • Performance and Optimization

    Disable 8.3 Filename Creation

    Supporting legacy MS-DOS file-naming conventions will reduce performance of an NTFS volume. Unless absolutely necessary, 8.3 filename creation should be disabled.
    Click to expand...
  • It is recommended that you enable Momentum Cache on your boot drive.
    Click to expand...

Outside of those, the drive is shown to be in good health and has the same data presented previously through Hard Disk Sentinel
 

Has Sysnative Forums helped you? Please consider donating to help us support the site!

Back
Top