Momentary and frequent system freeze in Linux MInt
This is my dmesg error:
[10678.069113] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[10678.069119] ata1.00: irq_stat 0x40000001
[10678.069124] ata1.00: failed command: FLUSH CACHE EXT
[10678.069134] ata1.00: cmd ea/00:00:00:00:00/00:00:00:00:00/a0 tag 0
[10678.069136] res 51/04:00:34:cf:f3/00:00:00:00:00/a3 Emask 0x1 (device error)
[10678.069141] ata1.00: status: { DRDY ERR }
[10678.069145] ata1.00: error: { ABRT }
[10678.076036] ata1.00: configured for UDMA/100
[10678.076046] ata1: EH complete
And this is my lspci, http://pastebin.com/XbMPSV26 . How can I fix this?
This happens randomly, and lasts a few seconds, and is extremely annoying.
I recently went through another scenario causing these problems:
After a server crash I had to replace some pieces of my hardware. After setting up the new HW, some of my disks, which worked perfectly well before, threw those messages. Since I did not want to believe that all my disks failed at once, I digged a bit deeper into this and discovered the reason for my problem:
My previous raid controller had activated a low-level security feature and locked the disks (although I was using mdadm software raid) rendering them unusable without this controller.
It was a 3ware controller, which locked the disk by setting the password 3wareUserPassword (including 15 spaces).
After realizing this, I was able to recover by running
hdparm --security-unlock "3wareUserPassword " /dev/sdX
hdparm --security-disable "3wareUserPassword " /dev/sdX
hdparm --security-set-pass NULL /dev/sdX
with sdX being the device file of the drive.
Here is the source of this helpful wisdom: http://blog.chr.istoph.de/tag/hdpam/
Your system is momentarily hanging because a (most likely) disk fault is causing all kinds of interrupt errors and ATA errors.
You could investigate further but if i were you, i'd just double check all the connections of that storage unit (which seems to be using an IDE 40pin or IDE 44pin flat cable) and if they're not visibly damaged, just replace the hard drive.
This error is most likely related to spread centrum clocking (SSC) out of boundery on the sata bus. there are far too many dwords accumulated during the slow period to practically buffer for release in the fast periods. However sata supports downspreading, there should be an option in your BIOS to turn SSC off and drives like those from WD have a SSC disable jumper (normaly open), try tweaking see if it helps.
It helped me out to get rid of those { DRDY ERR } and { ABRT } errors once.