Input/Output error hard drive failure?
 

This isn't really a Slackware issue, or at least I don't think it is. I was using my laptop the other day and some programs just stopped working. I attempted to open a terminal and use dmesg to figure out what was going on and all I got was "Input output error" regardless of the command. I even tried halt and reboot with the same result. I assume this is related to inability to read the hard drive.

I was forced to manually reboot and attempted to use the Slackware DVD to perform fsck /dev/sda fsck returned "bad superblock." I've only had to deal with the bad superblock error once before when my file system got corrupt (due to a power outage) but I should note that everything has been running fine since. Is it possible that the hard drive just needed to shut down temporarily and get restarted? This laptop is almost brand new. I bought it about two months ago. A hard drive issue would be unfortunate.

Like I said, everything is working fine now, but why would fsck return bad superblock?

First check if it's a hardware or software problem with "smartctl -a /dev/sdX" where sdX is your hard disk device (probably sdX = sda).

Smartctl did return some errors, but I don't know how to read it.

They all have something to do with set transfer mode it seems. Can anyone shed some light?

This disk has already had 23 errors after that short time. I would recommend to backup your data and return it to the vendor for a hardware check.

Is it possible that all of these errors are related to the same issue? For instance maybe each command ended up bringing an error and 23 commands were run? Also what would the vendor do for a hardware check that I can't? I'll give it back to the vendor if required, but I would like to figure this out on my own if possible. Mostly because this is my only computer and I would prefer not to go without.

This rather looks like a hardware error, it might be the controller, but more likely the disk. You can download the disk manufacturer's diagnosis software and test the disk. Before doing anything else I would recommend to backup your data, hardware tests usually stress the hardware, so that an possible error might get worse.

Well I'm going to try and find out if Asus offers a warranty. Like I said it's only two months old, but IDK if they offer a limited warranty or not. Looking into it...

Edit: Talked to an Asus rep and they have opened up an RMA case for me. By the way it sounds I will have to re-install Slackware though. They seem dead set on putting Windows 8 on here. Hopefully the fact that I removed the recovery partition will stop that, but I doubt it.

There are some errors, but nothing conclusive. Try running a SMART long test:
Wait for it to finish and post 'smartctl -a /dev/sda' again.

I ran the long test and smartctl -a again. The results look the same to me.

Please be patient ...

Whoops. Let me try that again. :redface:

Looks like it completed without error, and the attributes are still good. I don't think it is failing.

There are those errors, but one of my drives also has some of those and continues to work fine. They tend to appear after sudden loss of power.

I wouldn't consider a disk that after working only 1600 hours has already 23 errors in the log and already 47 reallocated sectors to be in a good state. All my disks have exactly zero errors and zero reallocated sectors after running 20000+ hours and if they would have I would backup the data on them and look for a replacement. This is especially true as long as I have guarantee on the disk, a 2 months old disk shouldn't show any of those signs.

No, that is really not good.

Just out of curiosity -- why is Reallocated_Event_Count 47 while both Current_Pending_Sector and Reallocated_Sector_Ct are zero?

If the drive is under warranty and you believe it to be faulty, send it in for replacement. If not, I don't think it is failing.

Member Response
 

Hi,

First, I would make the backup as suggested by other members. Then get the manufactures diagnostic set if you wish to test further. If the drive is in question and you did state that a RMA was given then ship it back to the manufacture. I agree that you should not be getting errors on a new drive with low hours.

You should use the diagnostic set to test then make your decision(s) based on the test results.

Asus said they do not offer a diagnostic tool to end users. I don't know of any others, except maybe fsck? But I am likely going to send it in for the RMA. There was a sudden loss of power issue about a week or two before this happened.

I notice that the 5 errors listed all occurred at about the same time (1615 hours) and refer to the same LBA. I also see that the G-Sense_Error_Rate (parameter 191) had a raw value of 1 in the first report (at 1637 hours) but returned to 0 in the second report (at 1657 hours). That in combination with the oddity of a Reallocated_Event_Count of 47 while both Current_Pending_Sector and Reallocated_Sector_Ct are zero suggests that some external event happened at around 1615 hours to cause the errors. Since the long test currently passes without error, I believe that the drive is now fine.

You might have difficulty getting a warranty replacement for a drive that now passes all tests and has zero reallocated or pending sectors.

It's a Hitachi drive.

http://www.hgst.com/hdd/support/down...2_v416_b00.iso

http://www.hgst.com/support/troubles...eshooting-tips
A windows version here if that's an option.
http://www.hgst.com/support/downloads/#WINDFT