Re: 520 BPS data integrity vendors

Re: 520 BPS data integrity vendors

[James Bottomley]

On Thu, 2010-07-01 at 17:11 -0400, Dan Porat wrote:
> Hi ,
> 
> In the documentation of 2.6.34 it says "We have been working with
> several FC/SAS HBA vendors to enable the protection information to be
> transferred to and from their controllers"
> I would like to test SAS 520.

So I'm having a little difficulty parsing this: SAS 520 is actually a
financial audit standard (which I really wish I didn't know about) ... I
assume you're actually talking about DIF/DIX?  The integrity standard
for SCSI?

> Which Vendor and what specific SAS HBA model will support it ?
> Thanks

I've added linux-scsi, since this is a SCSI question not an IDE one. 

The only SAS HBA supporting DIF I know of with a working Linux driver is
the mpt2sas.  It supports DIF types 1 and 3.

James

Re: 520 BPS data integrity vendors

[Dan Porat]

On Fri, Jul 2, 2010 at 2:06 PM, James Bottomley <James.Bottomley@suse.de> wrote:
> On Thu, 2010-07-01 at 17:11 -0400, Dan Porat wrote:
>> Hi ,
>>
>> In the documentation of 2.6.34 it says "We have been working with
>> several FC/SAS HBA vendors to enable the protection information to be
>> transferred to and from their controllers"
>> I would like to test SAS 520.
>
> So I'm having a little difficulty parsing this: SAS 520 is actually a
> financial audit standard (which I really wish I didn't know about) ... I
> assume you're actually talking about DIF/DIX?  The integrity standard
> for SCSI?
>
>>>You are right on the money.
>>>I mean DIF/DIX .
>>>The final purpose it to run a SATA 520 through a DIF supporting HBA , and a SAS/SATA dongle (interposer card).
>>>As initial stage we would like to check SAS DIF field.
>>>Afterwards just put it through a dongle to a SATA 520 sector size drive.

>> Which Vendor and what specific SAS HBA model will support it ?
>> Thanks
>
> I've added linux-scsi, since this is a SCSI question not an IDE one.
>
> The only SAS HBA supporting DIF I know of with a working Linux driver is
> the mpt2sas.  It supports DIF types 1 and 3.
>
>>Would you know of any specific LSI SAS/HBAmodel which will support SCSI DIF/DIX ?
> James
>
>
>

Thanks
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 520 BPS data integrity vendors

[Martin K. Petersen]

>>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:

>>> Would you know of any specific LSI SAS/HBAmodel which will support
>>> SCSI DIF/DIX ?

LSI's 92xx series HBAs have the LSISAS2008 chip that's required for DIF
operation. These boards do not support DIX.

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Dan Porat]

We checked with kernel 2.6.34 , Seagate Cheetah 15K and the following HBA:

LSI 9212-4i4e

Still , no success.

Any Ideas?

Thanks



On Sat, Jul 3, 2010 at 3:33 AM, Martin K. Petersen
<martin.petersen@oracle.com> wrote:
>
> >>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:
>
> >>> Would you know of any specific LSI SAS/HBAmodel which will support
> >>> SCSI DIF/DIX ?
>
> LSI's 92xx series HBAs have the LSISAS2008 chip that's required for DIF
> operation. These boards do not support DIX.
>
> --
> Martin K. Petersen      Oracle Linux Engineering
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 520 BPS data integrity vendors

[Martin K. Petersen]

>>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:

Dan> We checked with kernel 2.6.34 , Seagate Cheetah 15K and the
Dan> following HBA: LSI 9212-4i4e

Dan> Still , no success.

Send us the output of:

# lsscsi -p
# sg_inq -x /dev/sdN
# sg_readcap -l /dev/sdN


Dan> Any Ideas?

And you are sure that the drive is formatted with DIF?  Seagate's
standard firmware doesn't support it, fwiw.  It's a build-to-order
option.

Hitachi drives support DIF out of the box.

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Dan Porat]

Hi,

Thanks for the elaborate answer.
We want to buy the drive which supports (rather than currently
investigating , probably your assumption is right and our seagate does
not have the required FW).

Regarding the specific model ,
Will any of those do ?

HUS151414VL3800
HUS151473VL3800
HUS151436VL3800
HUS151414VL3600
HUS151473VL3600
HUS151436VL3600

Thanks

Dan Porat


On Tue, Jul 6, 2010 at 2:53 PM, Martin K. Petersen
<martin.petersen@oracle.com> wrote:
>
> >>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:
>
> Dan> We checked with kernel 2.6.34 , Seagate Cheetah 15K and the
> Dan> following HBA: LSI 9212-4i4e
>
> Dan> Still , no success.
>
> Send us the output of:
>
> # lsscsi -p
> # sg_inq -x /dev/sdN
> # sg_readcap -l /dev/sdN
>
>
> Dan> Any Ideas?
>
> And you are sure that the drive is formatted with DIF?  Seagate's
> standard firmware doesn't support it, fwiw.  It's a build-to-order
> option.
>
> Hitachi drives support DIF out of the box.
>
> --
> Martin K. Petersen      Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Martin K. Petersen]

>>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:

Dan> Regarding the specific model , Will any of those do ?

Dan> HUS151414VL3800 HUS151473VL3800 HUS151436VL3800 HUS151414VL3600
Dan> HUS151473VL3600 HUS151436VL3600

I'm not familiar with those specific models.  But as far as I know any
SAS/FC drive shipped by Hitachi within the last 3 years is DIF-capable.
Check the datasheet for "end-2-end protection".

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Dan Porat]

Following is the output with the Hitachi drive:

[root@fedora13 ~]# lsscsi -p

[0:0:0:0]    cd/dvd  hp       CDDVDW TS-H653R  0E00  /dev/sr0

[1:0:0:0]    disk    ATA      ST31000528AS     CC38  /dev/sda

[6:0:0:0]    disk    HITACHI  HUS151473VLS300  A420  /dev/sdb

[root@fedora13 ~]# sg_inq /dev/sdb

standard INQUIRY:

  PQual=0  Device_type=0  RMB=0  version=0x05  [SPC-3]

  [AERC=0]  [TrmTsk=0]  NormACA=0  HiSUP=0  Resp_data_format=2

  SCCS=0  ACC=0  TPGS=0  3PC=0  Protect=0  BQue=0

  EncServ=0  MultiP=1 (VS=0)  [MChngr=0]  [ACKREQQ=0]  Addr16=0

  [RelAdr=0]  WBus16=0  Sync=0  Linked=0  [TranDis=0]  CmdQue=1

  [SPI: Clocking=0x0  QAS=0  IUS=0]

    length=164 (0xa4)   Peripheral device type: disk

 Vendor identification: HITACHI

 Product identification: HUS151473VLS300  Product revision level: A420

[root@fedora13 ~]# sg_readcap /dev/sdb

Read Capacity results:

   Last logical block address=140174231 (0x85ae397), Number of blocks=140174232

   Logical block length=520 bytes

Hence:

   Device size: 72890600640 bytes, 69513.9 MiB, 72.89 GB

[root@fedora13 ~]#


Thanks


On Tue, Jul 6, 2010 at 2:53 PM, Martin K. Petersen
<martin.petersen@oracle.com> wrote:
>
> >>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:
>
> Dan> We checked with kernel 2.6.34 , Seagate Cheetah 15K and the
> Dan> following HBA: LSI 9212-4i4e
>
> Dan> Still , no success.
>
> Send us the output of:
>
> # lsscsi -p
> # sg_inq -x /dev/sdN
> # sg_readcap -l /dev/sdN
>
>
> Dan> Any Ideas?
>
> And you are sure that the drive is formatted with DIF?  Seagate's
> standard firmware doesn't support it, fwiw.  It's a build-to-order
> option.
>
> Hitachi drives support DIF out of the box.
>
> --
> Martin K. Petersen      Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Martin K. Petersen]

>>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:

Dan> [root@fedora13 ~]# sg_inq /dev/sdb
[...]
Dan> [root@fedora13 ~]# sg_readcap /dev/sdb
[...]

I asked for EXTENDED INQUIRY and READ CAPACITY(16).

# sg_inq -x /dev/sdb
# sg_readcap -l /dev/sdb

It does not look like you have reformatted the drive, though...

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: 520 BPS data integrity vendors

[Dan Porat]

Thank you all for your great advice.
Actually we did succeed with the Hitachi drives.

Thank you very very much.

Dan Porat

On Thu, Jul 15, 2010 at 9:47 AM, Martin K. Petersen
<martin.petersen@oracle.com> wrote:
>
> >>>>> "Dan" == Dan Porat <dan.porat@gmail.com> writes:
>
> Dan> [root@fedora13 ~]# sg_inq /dev/sdb
> [...]
> Dan> [root@fedora13 ~]# sg_readcap /dev/sdb
> [...]
>
> I asked for EXTENDED INQUIRY and READ CAPACITY(16).
>
> # sg_inq -x /dev/sdb
> # sg_readcap -l /dev/sdb
>
> It does not look like you have reformatted the drive, though...
>
> --
> Martin K. Petersen      Oracle Linux Engineering
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html