Re: WD Blue 510 SSD and strange write performance

From: mike tancsa <mike_at_sentex.net>
Date: Thu, 14 Mar 2024 20:26:15 UTC
On 3/14/2024 3:56 PM, mike tancsa wrote:
> On 3/14/2024 3:48 PM, Frank Leonhardt wrote:
>> "CAM status: SCSI Status Error" suggests to me that the drive was 
>> just too busy when asked. I'm not saying it's nothing to worry about, 
>> but neither am I saying it is.
>
> Given enough of them it does cause checksum errors on the test pool 
> unfortunately.  Could a buggy TRIM play a role here too ? I noticed a 
> commit the other day for a Segate SSD that had a broken NCQ TRIM. 
> Could these units suffer from that ?
> https://cgit.freebsd.org/src/commit/?h=stable/14&id=47fff7407c22c2c4b36b4f9f27ddfa70bb8f3fee 
>
>
> Is there a way to turn that off via camcontrol ? Or perhaps instrument 
> some other settings ?  I am not wedded to this hardware, but it would 
> be good to know if they can be made workable without too much effort.
>
On another test box with an MPR controller and same WD drives, a few 
more messages after the zfs send is about 50% done on a dataset thats 
about 1TB but compressed to about 260G. Some 29 million files. But with 
Samsungs, reliably no issue :(


(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 48 1e 9b 90 00 00 80 00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 897 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 1358 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 1742 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 1187 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 1006 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 16 SMID 758 loginfo 
31110f00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 48 1e 9c 10 00 00 b8 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 46 93 47 18 00 00 08 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): READ(10). CDB: 28 00 1c c7 dc 40 00 01 00 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): READ(10). CDB: 28 00 1c c7 d9 30 00 00 f8 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 46 93 47 10 00 00 08 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): READ(10). CDB: 28 00 1c c7 d8 30 00 01 00 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 49 55 29 20 00 00 08 00
(da6:mpr0:0:16:0): CAM status: CCB request completed with an error
(da6:mpr0:0:16:0): Retrying command, 3 more tries remain
(da6:mpr0:0:16:0): WRITE(10). CDB: 2a 00 48 1e 9b 90 00 00 80 00
(da6:mpr0:0:16:0): CAM status: SCSI Status Error
(da6:mpr0:0:16:0): SCSI status: Check Condition
(da6:mpr0:0:16:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, 
or bus device reset occurred)
(da6:mpr0:0:16:0): Retrying command (per sense data)
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 1023 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 297 loginfo 
31110f00
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c 49 50 18 00 00 a0 00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 1999 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 280 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 1970 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 859 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 1652 loginfo 
31110f00
mpr0: Controller reported scsi ioc terminated tgt 13 SMID 613 loginfo 
31110f00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c 49 4e a8 00 01 00 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c 49 4f a8 00 00 70 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c bc 65 30 00 01 08 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): READ(10). CDB: 28 00 2c 99 68 80 00 01 00 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 46 98 1e b8 00 00 18 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): READ(10). CDB: 28 00 1e cd a8 28 00 00 18 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c bc 64 30 00 01 00 00
(da3:mpr0:0:13:0): CAM status: CCB request completed with an error
(da3:mpr0:0:13:0): Retrying command, 3 more tries remain
(da3:mpr0:0:13:0): WRITE(10). CDB: 2a 00 4c 49 4e a8 00 01 00 00
(da3:mpr0:0:13:0): CAM status: SCSI Status Error
(da3:mpr0:0:13:0): SCSI status: Check Condition
(da3:mpr0:0:13:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, 
or bus device reset occurred)
(da3:mpr0:0:13:0): Retrying command (per sense data)