2

I am getting "UNC" errors on a hard disk on Ubuntu 9.10. It slows down my system, applications have not been responding for a long time. But when I mount the filesystem on another computer, it works properly.

disk: SAMSUNG HD161HJ (SATA)

syslog:

Apr 25 00:28:25 vare6gin kernel: [  885.773839] ata3.00: exception Emask 0x1 SAct 0x1e 
SErr 0x0 action 0x6 frozen
Apr 25 00:28:25 vare6gin kernel: [  885.773845] ata3.00: Ata error. fis:0x21
Apr 25 00:28:25 vare6gin kernel: [  885.773861] ata3.00: cmd 60/08:08:3f:00:ad/00:00:10:00:00/40 tag 1 ncq 4096 in
Apr 25 00:28:25 vare6gin kernel: [  885.773864]          res 51/40:24:67:c8:91/40:00:05:00:00/40 Emask 0x9 (media error)
Apr 25 00:28:25 vare6gin kernel: [  885.773871] ata3.00: status: { DRDY ERR }
Apr 25 00:28:25 vare6gin kernel: [  885.773877] ata3.00: error: { UNC }

[...snip 3 similar repeats of last 4 lines; see revision history for full log...]

Apr 25 00:28:25 vare6gin kernel: [  885.773970] ata3: hard resetting link
Apr 25 00:28:25 vare6gin kernel: [  885.773974] ata3: nv: skipping hardreset on occupied port
Apr 25 00:28:25 vare6gin kernel: [  886.240073] ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Apr 25 00:28:25 vare6gin kernel: [  886.256277] ata3.00: configured for UDMA/133
Apr 25 00:28:25 vare6gin kernel: [  886.256305] ata3: EH complete
Apr 25 00:28:27 vare6gin kernel: [  888.176088] ata3: EH in SWNCQ mode,QC:qc_active 0xF sactive 0xF
Apr 25 00:28:27 vare6gin kernel: [  888.176099] ata3: SWNCQ:qc_active 0xF defer_bits 0x0 last_issue_tag 0x3
Apr 25 00:28:27 vare6gin kernel: [  888.176102]   dhfis 0xF dmafis 0x1 sdbfis 0x0
Apr 25 00:28:27 vare6gin kernel: [  888.176109] ata3: ATA_REG 0x51 ERR_REG 0x40
Apr 25 00:28:27 vare6gin kernel: [  888.176113] ata3: tag : dhfis dmafis sdbfis sacitve
Apr 25 00:28:27 vare6gin kernel: [  888.176120] ata3: tag 0x0: 1 1 0 1  
Apr 25 00:28:27 vare6gin kernel: [  888.176126] ata3: tag 0x1: 1 0 0 1  
Apr 25 00:28:27 vare6gin kernel: [  888.176131] ata3: tag 0x2: 1 0 0 1  
Apr 25 00:28:27 vare6gin kernel: [  888.176136] ata3: tag 0x3: 1 0 0 1
Improve this question
1
  • "nv: skipping hardreset on occupied port" looks like this is from an Nvidia chipset of some kind. controller chipset details please?
    – quack quixote
    Apr 25, 2010 at 20:17

2 Answers 2

Reset to default
3

disk might be faulty or controller/cables.

run:

sudo apt-get install smartmontools

and then

sudo smartctl -d ata -a /dev/sdX|more

[eg sda - if it's first sata disk ]. read the output - possibly you will find something about surface errors or errors in communication.

then run self test:

sudo smartctl -d ata -t long /dev/sdX

and after 2h or so check result with the previous command. if you don't see traces of problems - replace sata cable, check if drive works on another motherboard.

Improve this answer
2
  • s/sudu/sudo/g ?
    – Javier
    Apr 25, 2010 at 16:11
  • 1
    @javier right. damn spelling - i was never good at it. thanks!
    – pQd
    Apr 25, 2010 at 16:45
0

I had a similar problem just today. It was just after a hard reset because I unplugged the power from my laptop, wrongly thinking it has the battery on.
It avoided booting, but not in safe mode.
Mounting it was neither a problem.
So from recovery mode or a live CD (you cannot do it safely with the target filesystem mounted) I did

sudo e2fsck -cvf /dev/sdaX

where sdaX is your ubuntu partition. In my case, it showed problems with some files in the /var/log directory. So the problem was there. The default action of e2fsck for some of them was ignore them so it did not solved the issue directly. But knowing the problem pointed to .log files in /var/log...

mkdir /var/log/old
mv /var/log/*.log /var/log/old

And the problem was gone. Hope it helps other with these DRDY ERR UNC media error exception.

Improve this answer

You must log in to answer this question.