ST373207LC uncorrected read error, cannot find

Hi,

One of my servers crashed today, logging nothing on the way down, and 
when it came back up the system disk, a Seagate ST373207LC looked like 
this (sorry about the formatting):

Device: SEAGATE  ST373207LC       Version: 0005
Serial number: 3KT4Z42L00007651Q8TR
Device type: disk
Transport protocol: Parallel SCSI (SPI-4)
Local Time is: Fri Aug 28 10:34:59 2015 PDT
Device supports SMART and is Enabled
Temperature Warning Enabled
SMART Health Status: OK

Current Drive Temperature:     20 C
Drive Trip Temperature:        68 C
Elements in grown defect list: 35
Vendor (Seagate) cache information
   Blocks sent to initiator = 2471614632
   Blocks received from initiator = 540061326
   Blocks read from cache and sent to initiator = 300088228
   Number of read and write commands whose size <= segment size = 
1553209039
   Number of read and write commands whose size > segment size = 1968677
Vendor (Seagate/Hitachi) factory information
   number of hours powered up = 77316.85
   number of minutes until next internal SMART test = 112

Error counter log:
   Errors Corrected by      Total   Correction     Gigabytes    Total
            ECC   rereads/    errors   algorithm      processed    
uncorrected
       fast | delayed   rewrites  corrected  invocations   [10^9 bytes]  
err                                          ors
read:   2753408433 888 793 2753410114   2753443556 96109.405   3
write:           0   0   0          0            0  1961.867   0
Non-medium error count:        3

So a long test was run, and it passed with no errors.  Then this was 
done:

  dd if=/dev/sda bs=8192 of=/dev/null

and that completed without error too.  Nevertheless the errors (3 
uncorrected in read, 3 in non-medium error count) were unchanged.

What gives?  Some electronic fault during a read?  How can I clear the 
read error
if reads do not throw more errors???  Maybe this was a stray gamma ray 
passing through some bit of electronics elsewhere brought the system 
down and the disk reads showed up, innocent bystander caught in the 
cross fire?  There doesn't seem to be any problem with the blocks the 
drive is currently using.

Hmm, I just noticed that the grown defect list has a count of 35.  Not 
sure if that is high or low for a disk of this age.  The other disks in 
the system all read 0 there, but none of them see as much activity as 
this one does.

Thanks,

David Mathog
[email protected]
Manager, Sequence Analysis Facility, Biology Division, Caltech

------------------------------------------------------------------------------