[smartmontools-support] long test reliability

[Gandalf Corvotempesta]

Hi to all
I'm running long test every weeks, but they seems to be totally unreliable:

SMART Self-test log
Num  Test              Status                 segment  LifeTime
LBA_first_err [SK ASC ASQ]
     Description                              number   (hours)
# 1  Background short  Completed                   -   16715
      - [-   -    -]
# 2  Background short  Completed                   -   16691
      - [-   -    -]
# 3  Background short  Completed                   -   16667
      - [-   -    -]
# 4  Background short  Completed                   -   16643
      - [-   -    -]
# 5  Background short  Completed                   -   16619
      - [-   -    -]
# 6  Background short  Completed                   -   16595
      - [-   -    -]
# 7  Background long   Completed                   -   16573
      - [-   -    -]
# 8  Background short  Completed                   -   16571
      - [-   -    -]
# 9  Background short  Completed                   -   16547
      - [-   -    -]
#10  Background short  Completed                   -   16523
      - [-   -    -]
#11  Background short  Completed                   -   16499
      - [-   -    -]
#12  Background short  Completed                   -   16475
      - [-   -    -]
#13  Background short  Completed                   -   16451
      - [-   -    -]
#14  Background short  Completed                   -   16427
      - [-   -    -]
#15  Background long   Completed                   -   16405
      - [-   -    -]
#16  Background short  Completed                   -   16403
      - [-   -    -]
#17  Background short  Completed                   -   16379
      - [-   -    -]
#18  Background short  Completed                   -   16355
      - [-   -    -]
#19  Background short  Completed                   -   16331
      - [-   -    -]
#20  Background short  Completed                   -   16307
      - [-   -    -]


everything is ok, but, from the RAID controller:

02/06/18 21:38:52: EVT#24763-02/06/18 21:38:52: 113=Unexpected sense:
PD 03(e0x20/s3) Path 5000c5008f72555d, CDB: 28 00 03 fb 6a 80 00 00 70
00, Sense: 3/11/00

03/02/18  3:25:40: EVT#25373-03/02/18  3:25:40: 113=Unexpected sense:
PD 03(e0x20/s3) Path 5000c5008f72555d, CDB: 28 00 03 fb 6a 80 00 00 70
00, Sense: 3/11/00

03/16/18  4:13:09: EVT#26010-03/16/18  4:13:09: 113=Unexpected sense:
PD 03(e0x20/s3) Path 5000c5008f72555d, CDB: 28 00 00 34 c3 00 00 00 58
00, Sense: 3/11/00

03/16/18  4:13:10: EVT#26011-03/16/18  4:13:10: 113=Unexpected sense:
PD 03(e0x20/s3) Path 5000c5008f72555d, CDB: 28 00 00 34 c3 05 00 00 53
00, Sense: 3/11/00

03/16/18  4:13:12: EVT#26012-03/16/18  4:13:12: 113=Unexpected sense:
PD 03(e0x20/s3) Path 5000c5008f72555d, CDB: 28 00 00 34 c3 06 00 00 52
00, Sense: 3/11/00



I have multiple read errors (sectors properly reallocated by the
controller probably by issuing a write)

Even the grown element list counter doesn't increase accordingly:

Elements in grown defect list: 2

it should be at least 5

any idea ?