SSD failing?
Hi,
So apparently I can't boot my PC on linux anymore. I get some kind of recovery mode. I can't copy/paste in this mode (or at least I don't know how to) but amongst other things, it says "failed command: read FPDMA QUEUED" and "Buffer I/O error on dev sdc1, logical block 7340064 async page read" I checked the sata cable to make sure they were plugged in correctly. but I don't know what else to do. I'm using my windows dual boot atm. Is it possible that my SSD is dead? Is there anything I can do? Could I get some help please? :'( |
Do you have a spare sata cable you could swap in?
Are you able, either through Windows or a Linux boot media, to get at and paste here the SSD's SMART data? Linux: smartctl -a /dev/sda (replace sda with the relevant device) |
You almost need to take a known working system and then install just the ssd to it and run s.m.a.r.t linux tests or the OEM's diag suite. Nothing in your current system can be trusted just yet.
|
Well I have no spare sata cable.
I tried to make a ubuntu usb to access the SSD but it fails trying to boot ubuntu. I can't access the SSD through windows because I don't think it has a format windows could read. And I have no other system to try the SSD on. I think I'm gonna have to go see some profesionnals so they can take a look at it. The only thing I did recently is use an air canister to remove some dust inside. Is it possible that that could have damaged something? |
The SMART data is maintained separately on the disk from any partitions/filesystems - you should be able to access this from a Windows system.
I used to use HD Tune on Windows for this purpose, but you might even be able to get the SSD manufacturer's specific software. |
columini,
You may find these links of use: https://superuser.com/questions/2914...sd-drive-dying https://communities.intel.com/thread/77801 In some cases it is simply a matter of changing the SATA 3 cable to the SSD and making sure that the ends are firmly engaged. If your motherboard has SATA 3 ports, make sure you connect to one of those. Also ensure that the cable is not kinked. A replacent quality SATA 3 cable costs pennies. I would start there first and then run SMART or follow Crucial's advice. https://www.amazon.co.uk/kenable-LOC.../dp/B004HTFDDE Note that these cables are round and do not kink. They also have lockable end fittings. In other cases a firmware update is needed. This is what Crucial recommend using their Power Cycle: http://forums.crucial.com/t5/Crucial...tem/ta-p/65215 |
http://hddscan.com/
HDDScan is a free utility for Windows. You can test your drive and read SMART data from Windows. |
1 Attachment(s)
Quote:
I'll buy a new SATA cable tomorrow to try it. |
Ok, I ran "e2fsck -c /dev/sda1" and it found a read error in one block and fixed it.
Now I can boot on linux again. Is it a concern for long term? Should I try to get a new one with the warranty? |
Now that you're back on Linux, can you please paste the output from sudo smartctl -a /dev/sda
That will allow a better look at the overall health of the SSD. Then, if you want, you can also run a couple of tests on it. |
Sure, here are the results:
Code:
|
Thanks columini.
It all looks good to my untrained eye, but the attribute that *may* be a bit concerning is #170 (Reserved Block Count), described as "On an SSD, this attribute describes the state of the reserve block pool. The value of the attribute shows the percentage of the pool remaining. The Raw value sometimes contains the actual number of used reserve blocks." So, that would indicate that you have 68% of your reserve blocks left and 298 used reserve blocks. I don't know what the value should be for an SSD of the model and age of the one you have. If you want, you can run a short and extended test (both non-destructive). The output that you provided above indicates that the former will take you approx. 4 minutes, the latter 32 minutes. You can see the results afterwards by re-issuing sudo smartctl -a /dev/sda and checking the data at the foot of the report. smartctl -t short /dev/sda smartctl -t long /dev/sda |
Oh well I'm glad I read the smartctl report a bit. my sda1 disk is my Galax 120GB disk which means its the one that has windows on. I assumed the failing disk was the other ssd on which I have linux because I could boot on windows but not linux. That's kinda odd.
That SSD is pretty rarely used too and is only 9 month old so I find it weird that it would wear off that soon :/ I ran those 2 other tests and here is the result afterwards: Code:
Just in case, here is my linux SSD (Samsung 750 evo, 9 month old): Code:
|
Wait, what, you have two SSDs? :)
When I gave the command to show the SMART data and run the tests on /dev/sda, I assumed that was the SSD on which you had been experiencing problems and, if not, that you would change the drive designation accordingly. Right, as regards /dev/sda, which we call the "Windows SSD", that figure of 68% may or may not be a problem, and I am going to recommend that you contact Galax customer support to ask them. As regards the "Linux SSD", please paste the SMART data here after having run the tests. From your opening post, I now see that it appears to be /dev/sdc. |
Ok I'm going to clarify the situation then. I have 3 drives:
sda is my windows SSD and is also the boot drive (with grub on it) sdb is my linux SSD, with a swap partition on it as well sdc is a regular HDD used for storage The recovery mod was telling me that sda had a problem and (I forgot which drive was which) I assumed sda was the linux SSD because I could still boot on windows but not linux. Then I ran "e2fsck -c /dev/sda1" and it fixed the boot problem. Although now the boot time is getting very long compared to before (the PC stays a few tens of seconds on the asrock screen) and windows is getting really slow and get stuck whenever I try to open the file explorer. Linux is fine though. Here is the smart report for the linux SSD: Code:
|
All times are GMT -5. The time now is 11:22 AM. |