Crucial M550 SSD Error

[Dopeyman]

Not sure if this technically belongs in the mini forum, but here goes.

I have a Crucial M550 512GB SSD that I installed in my 2012 Mac Mini Server. I recently purchased the SSD via a craigslist ad. I ran it through the SMART Utility app and it came out with a Reallocated Event Count Error.

What exactly is that error? Can it be cleared up? The drive is blank so I don't mind formatting it or whatever.

I don't know if it matters but it's in the upper bay.

 

[treekram]

How many reallocated blocks do you have? SSD manufacturers will allocate some amount of storage (typically 7%) for replacing bad blocks that occur so that the SSD maintains it's stated storage capacity. Once reallocated, you cannot un-reallocate (and why would you? - you want to start using a bad block again?).

If the M550 has 7% set aside for reallocated blocks, that would be about 36GB. The question is how many reallocated blocks do you have. Block size varies depending on the SSD. Maybe the software you're using has a percentage of reallocated blocks. If not, you can do some research or ask Crucial, to find out what percentage is being used. You want to pay attention to whether or not the reallocated count is growing on a regular basis.

There are two common scenarios that I see on these forums:
1) People run these SMART attributes utilities just because and they get freaked out when the software reports some anomaly. A lot of times, for various reasons that would take some time to explain, it's a false alarm. If you're worried, just make sure you're vigilant about backup and keep an eye as to whether or not the event numbers are increasing.
2) People run into some problem which they suspect which could be the SSD and then run some SMART attribute utility. In this case, you do see more instances where the SSD is a problem but it's not uncommon for the exact cause of the problem to not be captured by the SMART events.

A better diagnostic tool would be the Crucial diagnostic tool. Generic SMART utility tools may or may not be customized for each SSD. Manufacturer software would be customized to the specific SSD. The problem is that there is no MacOS-specific SSD diagnostic software from any of the major SSD manufacturers. (If you have a Windows computer, you can run the Crucial software - I don't know if it supports USB drives or not). And depending on the manufacturer, all you may get is a very non-specific health percentage rating.

 

[Dopeyman]

How many reallocated blocks do you have? SSD manufacturers will allocate some amount of storage (typically 7%) for replacing bad blocks that occur so that the SSD maintains it's stated storage capacity. Once reallocated, you cannot un-reallocate (and why would you? - you want to start using a bad block again?).

If the M550 has 7% set aside for reallocated blocks, that would be about 36GB. The question is how many reallocated blocks do you have. Block size varies depending on the SSD. Maybe the software you're using has a percentage of reallocated blocks. If not, you can do some research or ask Crucial, to find out what percentage is being used. You want to pay attention to whether or not the reallocated count is growing on a regular basis.

There are two common scenarios that I see on these forums:
1) People run these SMART attributes utilities just because and they get freaked out when the software reports some anomaly. A lot of times, for various reasons that would take some time to explain, it's a false alarm. If you're worried, just make sure you're vigilant about backup and keep an eye as to whether or not the event numbers are increasing.
2) People run into some problem which they suspect which could be the SSD and then run some SMART attribute utility. In this case, you do see more instances where the SSD is a problem but it's not uncommon for the exact cause of the problem to not be captured by the SMART events.

A better diagnostic tool would be the Crucial diagnostic tool. Generic SMART utility tools may or may not be customized for each SSD. Manufacturer software would be customized to the specific SSD. The problem is that there is no MacOS-specific SSD diagnostic software from any of the major SSD manufacturers. (If you have a Windows computer, you can run the Crucial software - I don't know if it supports USB drives or not). And depending on the manufacturer, all you may get is a very non-specific health percentage rating.

I've attached some screenshots of DriveDx and SMART Utility results. I also used the latest version of TechTool Pro and that came out clean with no errors.

But how can an SSD have errors/bad blocks. Aren't they chips/processors? I can understand a regular platter hard drive, but not an SSD

 

[treekram]

But how can an SSD have errors/bad blocks. Aren't they chips/processors? I can understand a regular platter hard drive, but not an SSD

It's the nature of the technology. If you're interested, you can read the Wikipedia page - if you don't like crowd-sourced documents, then just look at the footnotes - there's 90 references to look at. Look at the "Limitations" section - particularly that of "Memory wear".
https://en.wikipedia.org/wiki/Flash_memory

As the screenshots you've provided provided show, the interpretation of the raw data differs. With the screenshots you've provided, this is what I see:
1) I don't why Reallocate_NAND_Blk_Cnt (raw value 0) is so different from the Reallocated Event Count (raw value 16). Even at 16, this by itself is not an issue. Crucial should have information on this (not that they have/would divulge that).
2) It says Power On Hours (raw value) is 32. Was this SSD represented as new or used when you bought it? Maybe Crucial has a multiplier to get the true value (I don't think that's the case)?
3) Yet, the Power Cycle Count (raw value) is 516 which seems out of whack with the hours.
4) There's also the Unexpected Power Loss (raw value) is 62. Are you doing that? There was a recent post from somebody who had a high unexpected power loss count and it turns out that the SSD was freezing the computer which is why they had to shut it down - causing this value to increment. The SSD in that case appears to have been going bad. But again, that seems out of whack with the Power On Hours.
5) The CRC Error count (raw value 8757) is something you see in the MBP's of this age where the cable has gone bad. To me, since you have a Mini and haven't mentioned that you're having issues with the SSD, maybe that was occurring with the previous owner with a MBP and perhaps they thought the SSD was going bad when it was really the cable - pure speculation on my part. But then perhaps it makes sense in that it could have also caused problems where the previous owner had to shutdown to get out of their problems. The only thing is that the M550 is a pretty old SSD (introduced March 2014, I think they've had 4 successor models since then) so I don't know if the previous owner kept it for a while before selling it or what - there maybe an interesting story there. In any case, this, along with both reallocated counts, should be monitored to see if it increases on a regular basis.

 

[Dopeyman]

It's the nature of the technology. If you're interested, you can read the Wikipedia page - if you don't like crowd-sourced documents, then just look at the footnotes - there's 90 references to look at. Look at the "Limitations" section - particularly that of "Memory wear".
https://en.wikipedia.org/wiki/Flash_memory

As the screenshots you've provided provided show, the interpretation of the raw data differs. With the screenshots you've provided, this is what I see:
1) I don't why Reallocate_NAND_Blk_Cnt (raw value 0) is so different from the Reallocated Event Count (raw value 16). Even at 16, this by itself is not an issue. Crucial should have information on this (not that they have/would divulge that).
2) It says Power On Hours (raw value) is 32. Was this SSD represented as new or used when you bought it? Maybe Crucial has a multiplier to get the true value (I don't think that's the case)?
3) Yet, the Power Cycle Count (raw value) is 516 which seems out of whack with the hours.
4) There's also the Unexpected Power Loss (raw value) is 62. Are you doing that? There was a recent post from somebody who had a high unexpected power loss count and it turns out that the SSD was freezing the computer which is why they had to shut it down - causing this value to increment. The SSD in that case appears to have been going bad. But again, that seems out of whack with the Power On Hours.
5) The CRC Error count (raw value 8757) is something you see in the MBP's of this age where the cable has gone bad. To me, since you have a Mini and haven't mentioned that you're having issues with the SSD, maybe that was occurring with the previous owner with a MBP and perhaps they thought the SSD was going bad when it was really the cable - pure speculation on my part. But then perhaps it makes sense in that it could have also caused problems where the previous owner had to shutdown to get out of their problems. The only thing is that the M550 is a pretty old SSD (introduced March 2014, I think they've had 4 successor models since then) so I don't know if the previous owner kept it for a while before selling it or what - there maybe an interesting story there. In any case, this, along with both reallocated counts, should be monitored to see if it increases on a regular basis.

1. I don't understand French.
2. I was listed as slightly used
3. Mmmmmkay
4. I attached another pic where the app DriveDx mentions something about a cable. Maybe I should try swapping drives and see if the Samsung gives any errors?

 

[Fishrrman]

You purchased the SSD (standalone) on Craigslist?
Hmmm... now we know why the seller was sellin' it !

What I'd try:
Download the "Parted Magic" iso and burn a CD.
Boot from the CD, run Parted Magic's "secure ATA reset" command (I -think- that's what it's called).
This should reset the drive back to it's "factory fresh state".

Then, give it another go.

If the errors still keep happening, and you suspect the cable, I'd take the drive OUT OF the Mini, attach it via USB3, and run it that way for a week or two.
If the errors disappear, it's the cable.
If they keep occurring, either toss the drive or relegate it to "non-critical storage"...

 

[treekram]

1. I don't understand French.

For at least one vendor, Kingston, for a specific SSD, Reallocate_NAND_Blk_Cnt and Reallocated Event Count are synonymous. Obviously, for Crucial, for the M550, it isn't. It's a mystery to me why the event count is 16 but it actually, apparently didn't reallocate a block (Reallocate_NAND_Blk_Cnt = 0). As you can see, one SMART attributes tool chose to place emphasis on one of these values another chose to emphasize the other value. Who knows why they did this. Again, this shows the inherent issues with these tools. And the reallocated block count is really one of the crucial (no pun intended) attributes for a SSD.

2. I was listed as slightly used

This was not inaccurate.

3. Mmmmmkay
4. I attached another pic where the app DriveDx mentions something about a cable. Maybe I should try swapping drives and see if the Samsung gives any errors?

If all of the CRC errors occurred because the previous owner had a bad cable, this attribute has no relevance to you except you should keep a record of what the current value is as it will serve as a baseline. When a cable goes bad and CRC errors start occurring, they usually will increment at a pretty high rate (hundreds, thousands).

I noticed in another post that a particular SMART attribute software tool showed timestamps of the last event for some of the attributes. You should check with the software tools you're using to see if they can report this. There's can be some issues in this. First, maybe the manufacturer is not writing the timestamp to the log. Second, the amount of storage allocated to the timestamp is not large so the clock used for the timestamp may be reset every so often (in other words, the timestamp may not be accurate). If you can actually get a timestamp, you can see when the various attributes of interest last occurred (maybe).

What happened in the past, before you bought the SSD, has happened. It's not necessarily an indicator that the SSD is bad. However, it's also not an indicator that the SSD doesn't have some issue. I'll repeat it again. If there is a problem that the SSD has that isn't being captured by the SMART attributes (or there's a fault and it's failing to capture the event), then one won't know that the SSD has problems just by looking at the SMART attributes. Since the SSD has a questionable provenance (that is a French word, late 18th century), you should periodically see if these values are increasing - the critical ones being the reallocated block, CRC and the unexpected power loss attributes. If these values are not increasing, then my guess is that the previous owner(s) had some issue which was not the fault of the SSD and they sold it because they couldn't handle it. If the values increase on a regular basis, then the SSD may be suspect.

If the fact that the SMART event log for your SSD isn't perfect bothers you, just sell the SSD.

Doing a secure ATA reset won't reset the SMART attributes.

 

[treekram]

It turns out that Micron (Crucial) has a document detailing the SMART attributes for a number of their SSD's, including the M550.
http://www.micron.com/~/media/docum...torage/tnfd22_client_ssd_smart_attributes.pdf
[This link may automatically download the document in your browser.]

In an earlier search some months ago, this didn't turn up. I would have noticed it if it did since I own a M500, one of the SSD's also covered by this document.

My reading of the document indicates that there should be no difference in the reallocated block count and reallocation event count for the M550. There is some ambiguity where perhaps the reallocated block count could be more than the reallocation event count. So one can interpret this as one would like. Perhaps the document is inaccurate. Perhaps something is not quite right with the SSD. Perhaps there's an addendum to this document or the firmware was revised and Crucial changed how these parameters were used. In any case, if one believes that 16 is the correct number, 0.3 percent of the allocated bad blocks have been used on this SSD. According to the Micron document, the 16 count would not include any bad blocks at the time the SSD shipped from the factory.

It should be noted that the reallocated block count definition underwent a significant change on the M500 between firmware versions. And this points out, once again, that if the SMART attribute software doesn't take into account these SSD brand-model-firmware differences, the interpretation of the raw values could be wrong.

 

[Dopeyman]

Treekram and Fishrrman, thank you very much for all your input and information. Honestly I'm not really understanding all this. I guess basically my question is, is it good enough to use constantly as a startup drive?

Also, I switched positions of the drive from the Samsung to see if it was a cable issue and it turned out that it's not the cable.

 

[treekram]

Treekram and Fishrrman, thank you very much for all your input and information. Honestly I'm not really understanding all this. I guess basically my question is, is it good enough to use constantly as a startup drive?.

I probably should have made this clear at the outset. The SMART data that all these programs use capture events that have occurred since the SSD left the factory - so it covers what the previous owner(s) did and what you have done.

It's not possible to know if the SSD will be reliable based how on the information you have presented. There are inconsistencies in the SMART data of your SSD that likely date back to the previous owner. (I mentioned that you should see if timestamp data is available in these programs on your SSD - that would help.) Without knowing what the previous owner(s) did with the SSD, it's impossible to explain the inconsistencies. And as I mentioned before, the SMART data won't have all the data necessary to determine SSD health.

I'll be frank with you. If you wanted a SSD with the lowest uncertainty, you should have bought a premium new SSD (they're not that expensive). People who buy used from eBay or Craigslist (or wherever) should be prepared for some uncertainty in exchange for the lower price.

Also, I switched positions of the drive from the Samsung to see if it was a cable issue and it turned out that it's not the cable.

This one is easier to answer and I've answered it before. Use the software you have to check the "UDMA CRC Error Count". If the number goes up from what you have in your screenshots (8757), then there could be a problem with your cable. If this number does not go up, the previous owner probably had cable issues and this is why this number is as high as it is - it isn't reset to zero just because it got a new owner or because you erased the SSD. It's already been two days since the screenshot with the 8757 number. If you have used the Mini and written data to the SSD over the past couple of days and this number has not changed, there's a decent chance your cable is OK. I took apart my 2012 Mini to put in a SSD and my opinion would be that the chances you having a cable issue is lower than the chances of causing some damage by moving the SSD around.

 

[Cordorb]

I got a NEW SSD from a local store yesterday -- a MX500 one of the newer SSD
and there was already a NEW firmware update available

I keep a SATA cable and a power cable hanging out of an old PC just to do SSD firmware and formatting.

----
Interesting that even with same vendor the SMART data for different versions some times has :
Life remaining or says LIFE used. so zero not always GOOD

 

[Dopeyman]

A new but unrelated error has occurred.

I installed a few apps to check the status of the drive. And now I tried using App Uninstaller to uninstall them all. But when I get the pictured prompt, the computer freezes completely. Is there a way to check which app is causing the freezes? And can I manually delete the file or do it through terminal?

Regardless of whatever I attempt to do, whenever that prompt pops up, the computer freezes.

 

[treekram]

A new but unrelated error has occurred.

I installed a few apps to check the status of the drive. And now I tried using App Uninstaller to uninstall them all. But when I get the pictured prompt, the computer freezes completely. Is there a way to check which app is causing the freezes? And can I manually delete the file or do it through terminal?

Regardless of whatever I attempt to do, whenever that prompt pops up, the computer freezes.

This is likely a bug in the uninstall program. You should see if the software maker of whatever program you're trying to remove has instructions or a program to uninstall it. If the software maker doesn't have this, you can search on the web for instructions on how to uninstall a specific program but that usually returns results from companies that sell uninstall programs with generic instructions. If you don't find specific instructions to uninstall the program you have, you can follow the generic instructions. However, if you do that, there's always the risk that whatever action you take will do more harm than good.