Re: Errors on SSD

[Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx>]

:Model Family:     SandForce Driven SSDs
:Device Model:     OCZ-VERTEX2
:Serial Number:    OCZ-CU25VMZ6117F3NFM

    SandForce controller... those have historically had many bugs over
    the years, and OCZ's firmware has also historically had many bugs
    over the years.  There might be a firmware update, which you should
    look into via OCZ's web site (google for it maybe).

    (Of course, updating firmware on a SSD has its own problems, usually
    requiring one to throw together a DRDOS bootable usb stick to run the
    firmware updating software).

:Firmware Version: 1.29
:User Capacity:    60,022,480,896 bytes

    60GB SDD.

: ERR=...., SC=...., LL=...., LM=...., LH=...., DEV=...., STS=....
:SMART overall-health self-assessment test result: PASSED
:Warning: This result is based on an Attribute check.

    Test result is mostly meaningless.

:ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
:  1 Raw_Read_Error_Rate     0x000f   110   102   050    Pre-fail  Always       -       0/32677182

    No read errors according to the device.

:  5 Retired_Block_Count     0x0033   100   100   003    Pre-fail  Always       -       0
:  9 Power_On_Hours_and_Msec 0x0032   100   100   000    Old_age   Always       -       681h+41m+10.690s

    681 hours of power-on time.  Neither here nor there.

: 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       387

    The device was power-cycled 387 times so far.  Neither here nor
    there.

:171 Program_Fail_Count      0x0032   000   000   000    Old_age   Always       -       0
:172 Erase_Fail_Count        0x0032   000   000   000    Old_age   Always       -       0

    No program or erase failures.  That's good.

:174 Unexpect_Power_Loss_Ct  0x0030   000   000   000    Old_age   Offline      -       156

    This just means the plug was pulled on the device without
    giving it the shutdown command, 156 times.  Usually not an issue.

:177 Wear_Range_Delta        0x0000   000   000   000    Old_age   Offline      -       0
:181 Program_Fail_Count      0x0032   000   000   000    Old_age   Always       -       0
:182 Erase_Fail_Count        0x0032   000   000   000    Old_age   Always       -       0
:187 Reported_Uncorrect      0x0032   100   100   000    Old_age   Always       -       0
:194 Temperature_Celsius     0x0022   030   129   000    Old_age   Always       -       30 (Min/Max 30/30)

    All good.

:195 ECC_Uncorr_Error_Count  0x001c   110   102   000    Old_age   Offline      -       0/32677182

    Not sure what this means but it looks ok.

:196 Reallocated_Event_Count 0x0033   100   100   000    Pre-fail  Always       -       0

    No reallocations needed.  Neither here nor there but 0 is good.

:231 SSD_Life_Left           0x0013   100   100   010    Pre-fail  Always       -       0

    The SSD has basically no wear on it.  100% life left,
    10% worth of reserve sectors (above and beyond the stated
    capacity) left.  Very good.

:233 SandForce_Internal      0x0000   000   000   000    Old_age   Offline      -       192
:234 SandForce_Internal      0x0032   000   000   000    Old_age   Always       -       384
:241 Lifetime_Writes_GiB     0x0032   000   000   000    Old_age   Always       -       384
:242 Lifetime_Reads_GiB      0x0032   000   000   000    Old_age   Always       -       576

    All good.

    Insofar as the SSD is concerned everything is working properly.
    At least that means there's a good chance that the errors you
    are seeing are NOT media errors, but firmware issues.

    On the plus side we know it's an OCZ SSD which means that you
    need to research possible firmware upgrades for the device.
    OCZ has had many, MANY bad firmwares over the years.  Far
    more than other SSD vendors.

    I'm not sure regarding the AHCI errors you are getting.
    Could it be related to the kernel?  Well, not likely.
    We've actually fixed a few bugs in the AHCI driver in
    later kernels, but I doubt any of those could cause new
    errors to occur.

    AHCI reports your SSD as:

    ahci0.2: Found DISK "OCZ-VERTEX2 1.29" serial="OCZ-CU25VMZ6117F3NFM"

    So, version 1.29 firmware. I went to OCZ's site and the current
    firmware release is 1.37.

    http://www.ocztechnology.com/ssd_tools/OCZ_Vertex_2,_Vertex_LE,_Agility_2/

    So I think it is definitely worth going through the hell of
    updating the firmware.

					-Matt
					Matthew Dillon 
					<dillon@backplane.com>