05-25-2017 04:39 AM
I'm quite happy with my new Crucial MX300 SSD but today it started reallocating sectors.
I'm a bit worried now that this drive will fail early on me.
It replaced an old Crucial M4 SSD which never had any issue only it was too small (only 64GB).
TRIM is enabled, there is free space, disk activity is mainly Steam downloading games and updates...
Is it normal to encounter read errors this early with this SSD model?
05-25-2017 10:53 AM
It's had one block failure - but according to B4 it also has 1258 reserve blocks left.
If it's not giving you any trouble it's probably nothing to worry about. All those reserve blocks and error correction and RAIN are there for a reason.
You should of course ALWAYS backup any important data no matter what it is stored on.
05-26-2017 01:28 PM - edited 05-26-2017 01:29 PM
Thanks for reply.
Only I'm unsure, if the SSD replaced a block, how much data does one block entail, and does that mean one of my files is now corrupt somewhere?
05-27-2017 02:45 AM - edited 05-27-2017 02:46 AM
A block is 24MB.
Hard to say on the corruption front. You'd typically receive an I/O error, CRC error, or hang from Windows if you attempted to read something that it was unable to.
You also have successful RAIN recovery (RAIN is sort of a built in RAID like redundancy) so it could be it needed to use RAIN to read that block of data then marked it as not to use since RAIN was required.
If you want to check Windows is OK, bring up an administrative command prompt and type:
This will check and restore any corrupt or missing files. Note: this will go as far counting stuff you've deleted from your start menu as a fault and bring it right back again. So certain customisations you've made are just as likely to be counted as errors as actual errors.
If you have any doubts, feel free to contact Crucial Support via the link in my signature.
05-27-2017 06:23 AM
Thanks for reply.
I will cross fingers and hope for the best for now. If n it happens again I would contact Crucial and see what they say.
Maybe they would replace it maybe they won't, I don't know if this is supposed to happen at all, but if it happens again either way I will replace it...
Backups are one thing but if you don't even know what the problem was and which file might be affected, and when it might happen again in the future. It's difficult to check every single file for possible corruption and if I can't trust the device I'd rather get another.
03-17-2018 08:25 AM - edited 03-17-2018 08:26 AM
Some time has passed since my last post. Nothing changed... until now.
About a month ago, the Reallocated NAND Blocks rose from 1 to 2.
Two days ago, it rose from 2 to 3.
Today, it went from 3 to 4.
I am still on M0CR040 firmware - would updating change anything about the reallocations?
03-17-2018 01:51 PM
I would definitely update the firmware to M0CR60 as I've encountered other issues with the MX300 which required me to Secure Erase the SSD. I'm not sure the firmware update will stop what is happening since these look like normal block failures, but you never know and hopefully it keeps you from encountering the issues I had. You should definitely keep a close eye on the situation especially as the SSD nears the end of its warranty period. As long as you are not seeing Uncorrectable Errors you shouldn't have to perform a Secure Erase.
targetbsp's information and advice in previous posts is correct. Plus you run a similar risk of data corruption as you would with a traditional hard drive. There are some new filesystems that provide built-in checksumming, but unfortunately there is none for a Windows boot drive. If you are concerned about file integretity, I believe there are tools available which will provide checksums for your files.
It is perfectly normal for SSDs to report bad blocks like this as long as you don't get a huge number of them in a short period of time or run out of reserved blocks. The MX300 is using TLC NAND which is the new current style of flash memory being used by the current generation of SSD so there may still be growing pains.
Here is the beginning of a very interesting article series from 2013 detailing how SSDs tend to fail during a multi-year 24/7 endurance test on some earlier SSDs. Here is the final article in the series which provides a final summary.
03-18-2018 06:53 AM
Thank you for quick reply.
I updated the firmware to MR0C60 successfully.
However, immediately after the update, Raw Read Error went 4->5, Reallocate NAND stayed 4, Current Pending went 0 -> 1.
I run a SMART selftest at this point. During the selftest, Current Pending went 1 -> 0 and Reallocate NAND 4 -> 5. RAIN Recovery is 20.
This makes it the third reallocation within a few days. I can't help but be a little worried about it after all.