M1015 LSI IT mode / print_req_error: I/O error / spindown

[French Chamallow]

I just build my first server with X9 supermicro and 2 HBA M1015 in IT mod ( P20 driver ) SC836 ( direct attach backplane )
I'm a newbie in your world.
MergerFS + Snapraid / OMV 4 ( Debian stretch )
All my server work fine.
BUT I see some error in my syslog :

[ven. mars 15 13:48:54 2019] sd 0:0:5:0: [sdf] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:48:54 2019] sd 0:0:5:0: [sdf] tag#1771 CDB: Read(16) 88 00 00 00 00 04 8c 3f ff 00 00 00 00 08 00 00
[ven. mars 15 13:48:54 2019] print_req_error: I/O error, dev sdf, sector 19532873472
[ven. mars 15 13:49:04 2019] sd 0:0:4:0: [sde] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:49:04 2019] sd 0:0:4:0: [sde] tag#1771 CDB: Read(16) 88 00 00 00 00 04 8c 3f ff 00 00 00 00 08 00 00
[ven. mars 15 13:49:04 2019] print_req_error: I/O error, dev sde, sector 19532873472
[ven. mars 15 13:49:12 2019] sd 0:0:2:0: [sdc] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:49:12 2019] sd 0:0:2:0: [sdc] tag#1771 CDB: Read(16) 88 00 00 00 00 01 d1 c0 be 00 00 00 00 08 00 00
[ven. mars 15 13:49:12 2019] print_req_error: I/O error, dev sdc, sector 7814036992
[ven. mars 15 13:49:19 2019] sd 0:0:6:0: [sdg] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:49:19 2019] sd 0:0:6:0: [sdg] tag#1771 CDB: Read(16) 88 00 00 00 00 01 d1 c0 be 00 00 00 00 08 00 00
[ven. mars 15 13:49:19 2019] print_req_error: I/O error, dev sdg, sector 7814036992
[ven. mars 15 13:49:27 2019] sd 0:0:3:0: [sdd] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:49:27 2019] sd 0:0:3:0: [sdd] tag#1771 CDB: Read(16) 88 00 00 00 00 01 d1 c0 be 00 00 00 00 08 00 00
[ven. mars 15 13:49:27 2019] print_req_error: I/O error, dev sdd, sector 7814036992
[ven. mars 15 13:49:41 2019] sd 0:0:1:0: [sdb] tag#1771 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK
[ven. mars 15 13:49:41 2019] sd 0:0:1:0: [sdb] tag#1771 CDB: Read(16) 88 00 00 00 00 01 d1 c0 be 00 00 00 00 08 00 00
[ven. mars 15 13:49:41 2019] print_req_error: I/O error, dev sdb, sector 7814036992

I think it's related to the awakening of my hdd when they are in spindown.

When I remove spindown I haven't this error.
All smart are good long test.

synch and scrub good.

I see a long thread abot the same problem from users of ZFS ( ZFS io error when disks are in idle/standby/spindown mode · Issue #4713 · zfsonlinux/zfs)

But I don't find the solution...

Please can you help me ?

 

[French Chamallow]

I see similar problem for an unraid user : "Array has 9 disks with read errors"

with all these HBA 1015 on the forum nobody does spindown/spinup?

 

[rmweiss]

No solution (sorry), but I have the same problem: Best non-SLI SAS controller (for IT mode) : DataHoarder