Sample Header Ad - 728x90

sas drive i/o error 5 via raid card - sed locked partition table

1 vote
1 answer
383 views
I am trying to find a better way to access my newly purchased SAS drives (deepdiscountserver 3tb IBM platters(used or refurb i think yes)), using a 2x4 LSI PCIe3 card. I think the card is only half working based on a few things mentioned below. For now, I'm simply trying to put a file system on it, running into error 5, and i/o errors(need to remove SED encryption?). I worked out a way to do it with dd, see below. I don't have options in my SAS/RAID controller card or I would reset it there. When I press the hotkey during boot, it responds says it will load, but never loads. If i understand correctly, this should be where one woud normally turn off the encryption, right? sedutil-cli, and hdparm report this, not helpful.
sedutil-cli --scan
...
/dev/sdf   No                                                                                             sdc      8:32   0   2.7T  0 disk
The Kernel flag libata.allow_tpm is not set correctly                                                     ├─sdc1   8:33   0   2.7T  0 part
Please see the readme note about setting the libata.allow_tpm  
...
hdparm -I /dev/sdd
/dev/sdd:
SG_IO: bad/missing sense data, sb[]:  72 05 20 00 00 00 00 1c 02 06 00 00 cf 00 00 00 03 02 00 01 80 0e 0
0 00 05 20 00 01 ff ff ff ff
SG_IO: bad/missing sense data, sb[]:  72 05 20 00 00 00 00 1c 02 06 00 00 cf 00 00 00 03 02 00 01 80 0e 0
0 00 05 20 00 01 ff ff ff ff
 HDIO_DRIVE_CMD(identify) failed: Input/output error
I will do what worked for me last time. These drives are going to be a pain in the ass. See below, and also let me know If I do this a better way; dd /dev/sdd if=/dev/zero of=/dev/sdd bs=1M
dd: error writing ‘/dev/disk/by-id/scsi-35000c50084889cf3’: No space left on device
2861589+0 records in
2861588+0 records out
3000592982016 bytes (3.0 TB) copied, 21272.1 s, 141 MB/s
SED is gone, should no longer receive I/O errors. Now use gdisk to lay in a new partition table. If you run gdisk before you dd the entire drive, it reports the I/O error, usually that means i didnt DD enough of the drive. usualy I think thats some stuff at the very end of the drive.
(? for help): o
This option deletes all partitions and creates a new protective MBR.
Proceed? (Y/N): y

Command (? for help): w

Final checks complete. About to write GPT data. THIS WILL OVERWRITE EXISTING
PARTITIONS!!

Do you want to proceed? (Y/N): y
OK; writing new GUID partition table (GPT) to /dev/sdd.
The operation has completed successfully.
I've tried parted, fdisk, etc.. If I don't run dd at all, or run it just on the beginning, SED takes over I think, and I/O errors, error 5, and everything points to bad controller card, cabling, or drive. However, trying every other cable, and running through smartctl says thats a false alarm, that things are fine. I even have some other IBM drives to compare to on the system, and can tell you, this is normal behaviour, for these drives in their received condition. If I could get into the SAS settings, I would just clear the drive there. So in the end, dd is all that's working for me, thankfully at least :-) is there is a better (quicker) way? Is it normal to not access to sedutil, and hdparm for SAS drives? I better be careful when I choose my next RAID controller card based on this.
Asked by Brian Thomas (734 rep)
Oct 10, 2021, 03:27 AM
Last activity: Feb 13, 2022, 11:11 AM