[smartmontools-support] OS hangs and COMRESET events: failing drive?

[Gruff Hacker]

On Tuesday, May 26, 2020, 01:43:56 PM EDT, Christian Franke wrote: 

>> I have a LITEONIT LGT-256M6G M.2 SATA SSD on a Windows 10 system.
>> For a few days the OS has been hanging periodically and the only thing I can see in Windows events logs is this:
>>
>> Source:        iaStorA
>> Event ID:      129
>> Reset to device, \Device\RaidPort0, was issued.
>>
>> There are apparently a large number of issues that can cause this, but the only other thing I noticed is this in the smartmontools output:
>> ID      Size     Value  Description
>> 0x000a  2          218  Device-to-host register FISes sent due to a COMRESET
>>
>> Every time one of these hangs happens, the counter of that COMRESET event in the smartmontools output will increment.


>COMRESET is an OOB signal on the SATA interface used to reset the 
>device. The "... due to a COMRESET" counter is cleared on power on and 
>usually increases by 2 during each (warm-)boot due to COMRESETs from 
>BIOS and device driver.
>
>In the above case, the counter increased further because the device 
>driver decided to reset the device several times. Unfortunately this 
>provides no hint why the resets were issued.

Thanks Christian.  It's helpful to know this.  Would be nice if Intel could update their driver to be more verbose.  I assume the driver has some additional knowledge about why it was resetting the device but that knowledge is hidden.

>> 199 UDMA_CRC_Error_Count    PO----   100   100   000    -    9

>SMART values show no signs of trouble. If the raw value of ID 199 
>increases further, check for M.2 connector problems.

It did increase further.  I also recalled that this same issue had happened to me with this same drive several years earlier.
I didn't see any M.2 connector issues, but I did decide to replace the drive to be cautious.