You can attempt recovery using photorec and testdisk or if your data is really important to you use a professional.
I agree that this is a hardware failure of the M.2 disc, this might be a long shot but you can try this:
Remove the M.2 device from the motherboard
Clean the contacts with a clean piece of printer paper , tear of a piece, fold it around the contacts (avoid contact with your fingers and discharge yourself before doing this)
Gently slide the paper over the contacts, you will notice that it will become black, that is normal.
Re seat the M.2 device and check if it will work again.
If this doesn’t help, but the M.2 device is still recognized by UEFI/bios you can try to wipe the disk and re partition it but you will lose all your data.
If this doesn’t help/work the device is definitively broken…
The same trick can be used on memory modules but be careful not to touch the contacts with your fingers and discharge yourself up front
@linuxislife
How do you do it? I mean i literally pound my drives into the ground with installs and re-installs and partitioning and re-partitioning.
I know, right
Pffffffffffffffffffffffffffffffffffffff, there is so many worst part to that story… main one being I’m not the only person depending on those data…
And YES, I know, I know, I know… (well “depending” is a big word…euh… I hope )
no no, but I tried on different boards, so…
and well, if I ever recover a single bit of data, then this stuff goes nowhere near a computer anymore
but yep, sound advice otherwise, cheers
Sounds like you’ve had a bad experience with a drive. But the 5000h uptime (over 200 days straight?), and 7TB written, doesn’t make a whole lot of sense in the context of your saying it’s brand new, - unless it was resold as “new” to you, in which case it’s a claim against the seller.
Going forwards, - automate your backups, and at the very least, set them up so they’re little more than a bash alias you can manually run at will. 1 Backup = No backup. Online, Offline, Air-gapped. Good luck on your journey, these things happen.
Sure doesn’t seem brand new to me.
lol, yeah, “brand new” to my standards
and still… manufacturer announces like 200 TBW and MTBF at 2 million hours for this drive… well he announces whatever he wants anyway, but we could admit it wasn’t quite EOL yet… well at least on paper
well, you missed those parts:
but you’re so right, i know…
at least now we can agree that it is NOT brand new anymore
One of my SSDs at the moment is on 15548 powered on hours currently, and another at 7376 hours, and a third one at 12435 hours, and another at 6244 hours still going. So even for 5000 hours and 7tb wrote that seems like it failed early, although 7TB could be a lot of wear depending on the full capacity of this SSD due to sector writes or flash wear, and other factors could break an SSD such as how many times it’s turned on and off.
But then SSDs have a habit of potentially failing with no notice at any time if you are unlucky unlike HDDs which at least usually give you a hint beforehand indicating an eventual complete failure. And is why I have a multiple backups due to this risk. And for cold data storage (On HDDs) I do a full disk hash on the drive so I can verify the checksums of all files stored on them and know if any are different, indicating possible corruption.
@linuxislife If you ever get a new SSD I would check the S.M.A.R.T status of it as soon as you receive it, and for me I also use kdiskmark to verify it’s speeds are correct (But in settings you have to change to NVME setting to test). This is what I always do so if I find an issue it’s easier to report to seller, I also verify the serial number matches the box it come in as well. Depends on the seller but I sometimes even record opening the box with my phone depending on the cost to prove the serial numbers don’t match out of the box (If this is the case), many places you can just say the device is broken to return but if it come with a mismatched SSD they can sometimes deny return thinking you swapped a good SSD with a bad one to get a free SSD.
Age has nothing to do with hardware going bad. Sometimes hardware goes bad from the testing done at the factory before its sent out but since it past the test no one knows it is broke until a customer gets it and there is an issue and a return is made.
I’ve had drives last from a month to over ten years. Sometimes things happen and every now and then you get a lemon.
I didn’t see where the name brand of the NVME was mentioned. Just out of curiosity, what is the name brand and model of the failed NVME?
Pudge
I have had roughly 20% of my SSDs fail in the first 1-2 years of life. When I worked in IT, our failure rate across a much larger set of drives was ~10% in the first year. Yes, my luck with drives is…bad.
Sometimes hardware just fails. No brand is immune to failure. I only buy drives from reputable brands. Samsung, Western Digital, Sandisk, Intel, etc.
Samsung is the only brand where I never had any ssd failure.
All of my SSDs are currently Samsung in my system (Around 10 of them including two external ones). But I have had Intel and Sandisk in my company too that have been fine too with no issues. The disclaimer is I haven’t used every SSD brand out there so I can’t say for those.
The only brand I could confidently advise against, in my experience, is Kingston. My company insisted on buying them to replace all laptops and PCs that had traditional HDDs and we had at least 10-14 of them fail out of around 100 bought (We had a pile of them) within a year or two of use, and around 4-5 were broken on arrival out of the sealed package, not detected by BIOS or anything (They all has the same symptom or issue) but I guess there is a possibility of being unlucky with a bad SSD batch. While these were cheaper SSDs within £29-30 (For 500Gb), it has put me off risking it with the rest.
I wouldn’t have selected Kingston personally but it wasn’t my choice to make and a lot of these PCs were from 2013-2015 so it was a cheap solution to speed up PCs with failing and old hardware and the users were not exactly power users, just complaints of slowness due to old HDDs or they are used to SSDs by 2021.
brand new in paper !
Thanks a lot @_Six for these terrific advice in the rest of your post!!
so, it appears to be a PNY XLR8 CS3030 500 GB
Well I had sata 870 evo epically failed on me after very very short time of use (even to normal people standards I mean)
but I might have the same luck than @dalto so…
F$*K I just mounted 2 fury renegade in my friend’s gaming computer… well… it’s not mine so it’s probably the good batch
I had a 970 EVO fail on me.
Again, there is a luck factor involved here.
I think that the environment that the SSD / NVME operate in is a factor.
Since the days of HDD, I have never had a SSD / NVME fail. But my computers are in our basement (my man cave) where the ammbient temperature is 67 or 68 degrees F in the summer, and today it is 62.5 degrees F. Add to that, all of my NVME are encased in a NVME heat sink.
Pudge
I will clarify the ones we had are cheap Sata Kingston ones. With SSDs not every model is bad under a brand. A brand can have good SSDs and bad SSDs but some brands have a higher failure than others, I think the ones I have used in companies etc Intels seem to be the most robust, I don’t know about now though. These SSDs we had are not Fury Renegade ones. I should have clarified that in my Message rather than indicate the whole brand is faulty, but I haven’t had the more expensive Kingstons to know.
While I have all Samsung at the moment, as they are old and built them up over time, I have had one Samsung fail, an EVO 840 which had a known flaw where it would become as slow as a HDD after a while unless the sectors were refreshed each month. Intel and Sandisk, no issues so far.
regarding the 870 evo I’ve read there were batches during the first year or so that encounter problems, apparently Samsung replaced those without even asking question… of course I didn’t ask for mine…yet…? I dunno maybe I still can… I think it’s 5 years warranty with samsung ssd, idk