Question SAS drive IO errors. Am I using the wrong connectors with my HBA?

[Daylon Rankin]

Hi,

I recently bought:
an LSI SAS9201-16e HBA (4 external SFF-8088 connectors)
a cable (SFF-8088 to 4 sata (Link))
and a couple of 2TB SAS HDDs (Hitachi HUS72302 CLAR2000) Model: (HUS723020ALS640)

and I happened to have a sas-to-sata interface adapter like this one Link.

I plugged everything up, the LSI HBA is working great and reads my sata drives just fine, but when I try to initialize either drive using disk management (I am using Windows Server 2016 GUI) it locks up for a few minutes then gives me the message "the request could not be performed because of an I/O device error".
I also tried using the "convert" command with DISKPART and get the same error. The event log for this error says "VDS fails to write boot code on a disk during clean operation. Error code: 8007045D@02070008". This happens both with MBR and GPT.

I am completely new to SAS drives and had a heck of a hard time finding information about compatibility on the Internet, so I'm assuming I don't have the right combination of hardware to get these disks to work. I have tried using the interface adapter I mention in the first paragraph, as well as trimming the sata connector to make it fit directly to the drive.

Any help would be greatly appreciated.

 

[FireWire2]

Your issue most likely in the HBA BIOS, flash with the latest firmware and load the latest drivers, it should solve the issue

 

[Daylon Rankin]

It turned out that the drives had come from an EMC array and though that MODEL drive is typically compatible with the hardware I have, they had been preconfigured with a non-standard sector size that Windows will not initialize. I solved the problem using this guide: https://forums.servethehome.com/index.php?threads/how-to-reformat-hdd-ssd-to-512b-sector-size.4968/

The drives now work, however they perform at about 60MB/s write rather than the 150MB/s speed I get with the same model drive that did not come from an array and has Hitachi's stock firmware. Read speed seems to be unaffected. It is possible that there is a way to flash the other firmware and restore full write speed, but I'm not going to risk it because write speed isn't a major factor for my project. However, if someone reads this and finds a way to safely fix the speed issue, please let me know.

In case the link ends up dead, the guide says to use CentOS and " yum install sg3_utils ", run " sg_scan -i " to list the drives then run " sg_format --format --size=512 /dev/sg8 " where sg8 is the ID of the drive you want to format. This takes a few hours and will set the sector size to 512 so Windows can initialize it.