Mysterious errors with SSD
The past 2 weeks or so, I have been seeing this error early in the boot process:
Quote:
Code:
sudo smartctl -a -d ata /dev/sda Thanks for any input! Regards, Matt |
Hello,
I'm not an expert, but I see nothing wrong in your smartcl report. The drive seems to be in good health. You can try to run a short and/or a long selftest : Code:
$ smartctl --test=short /dev/sda Code:
$ smartctl -l selftest /dev/sda -- SeB |
i had a disk that died, showing such error. I remember the I/O error thing. Backup if you can.
|
As phenixia2003 said, your drive doesn't seem to be showing any smart errors. The main one to look out for specifically with SSDs is Reallocated_Event_Count. This "error" will start cropping up when you start going beyond the usable write amounts for a cell and it has to start moving data off worn out cells. Your value is still at 0, so nothing has worn out according to smart data.
I would also do as phenixia2003 recommended and check your cable that it's fully seated or replace it with another cable. But then, it is always possible that the error showing up in your dmesg is something that SMART doesn't log, so your drive could be a dud. If you have another computer, it might be worth checking in there to see if you get the same warning (which would remove the motherboard being the problem from the equation). As always, it wouldn't hurt to back up your important stuff, just in case things go sideways. |
Like phenixia2003, I also think that the smartctl output looks OK, except maybe
Code:
174 Unexpect_Power_Loss_Ct 0x0030 000 000 000 Old_age Offline - 76 |
1337_powerslacker --
Like Ilgar, it seems to me that the Unexpect_Power_Loss_Ct is suspicious. If the cable or the interface on the MoBo is bad, all the drive would know is 'power loss' OTOH, the OS might see some sort of request error. Check the Cable ( and Card, if one exists ) ? -- kjh( :) the only other oddity I see is that I've never had a drive in the smartctl database ... mine are all reported as not in the DataBase :) ) |
http://lkcl.net/reports/ssd_analysis.html
https://news.ycombinator.com/item?id=10552218 https://forums.anandtech.com/threads...ction.2452606/ 1. Check filesystem OR 2. Do a so called "secure erase procedure" https://www.thomas-krenn.com/en/wiki/SSD_Secure_Erase https://www.unixmen.com/secure-erase-your-ssd/ |
Quote:
I've not read the content of the 'Secure Erase' links, but unless they've discovered such-a-thing as a 'non-destructive Secure Erase' :) then #2 above sounds like a last resort ? Or am I missing something ? If I understand Secure Erase and the drive has actually gone bad then a Sledge Hammer is MUCH quicker than Secure Erase and it take MUCH LESS effort to erase an SSD with a Sledge Hammer than it does with HDDs :) :) Looking at 1337_powerslacker's `smartctl` Report, the drive itself looks OK. IMO, check the Interface Components ( Cable and SATA Connector and optionally any SATA Card ) before doing anything else. -- kjh |
SE is there to bring back disk to life if possible, not mobo,sata interface, filesystem etc.
That must be checked first. It costs lost data. If that fail after that your Sledge Hammer is very reasonable option but not first. "Report, the drive itself looks OK" Ya "itself" but telling nothing about possible power failure controller:-) Sometimes SMART means not so smart. |
Well, as has been suggested several times, I opened my case and re-seated the power and SATA cables, so that there's no question of incomplete electrical contact on either. I booted up, and no errors occurred. So I think this was a one-off, where there may have been some inadvertent jostling of cables when I was fiddling in my computer's internals.
Thanks for the suggestions, everyone! It was much appreciated! Happy Slacking! |
All times are GMT -5. The time now is 01:45 AM. |