NetApp DS4486 SAS Drive Issue - FreeNAS/TrueNAS

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
Running FreeNAS-11.3-U3. Hoping to upgrade to TrueNAS after these disk replacements.

I’ve got a DS4486 disk shelf with 12 x WD Red SATA drives. Dell R710 with an LSI SAS9200-16e HBA. When setting it up originally FreeNAS couldn’t get any info on the drives if they weren’t in pool yet. As soon as the drives were used the serial, model, and size would populate in the web UI (if values were present before being used they would be values from a previous device that was either connected to the disk shelf or FreeNAS. Not sure exactly where that info was coming from). I also can’t run SMART tests with this disk shelf, but that seems to be a common problem with SAS to SATA.

With all that said, I’m able to use it (although replacing a drive can be a little tricky because the serial isn’t readable until it’s already started the replacement process).

Where I’m running into an wall now is I just bought some SAS drives and I can’t do anything with them in FreeNAS. It can see the devices, but no info, I can’t wipe them or use them to replace another drive.

I get this error in the web UI:
Code:
[EFAULT] Command gpart create -s gpt /dev/da6 failed (code 1): gpart: provider: Operation not supported by device


I also tried using sg_format in the cli and got this:

Code:
root@freenas:~ # sg_format -v /dev/da6
ST316081 N/A DB 4321 peripheral_type: disk [0x0]
PROTECT=0
Unit serial number: 4LS60ZYW
LU name: 0▒b
mode sense(10) cdb: 5a 00 01 00 00 00 00 00 fc 00
Mode Sense (block descriptor) data, prior to changes:
Number of blocks=312500000 [0x12a05f20]
Block size=512 [0x200]
read capacity(10) cdb: 25 00 00 00 00 00 00 00 00 00
read capacity(10):
Descriptor format, current; Sense key: Aborted Command
Additional sense: Internal target failure
Descriptor type: Field replaceable unit code: 0x3d
READ CAPACITY (10): Aborted command, type: sense key, other than protection related (asc=0x10)
No changes made. To format use ‘–format’. To resize use ‘–resize’
root@freenas:~ # sg_format --format -v /dev/da6
ST316081 N/A DB 4321 peripheral_type: disk [0x0]
PROTECT=0
Unit serial number: 4LS60ZYW
LU name: 0▒b
mode sense(10) cdb: 5a 00 01 00 00 00 00 00 fc 00
Mode Sense (block descriptor) data, prior to changes:
Number of blocks=312500000 [0x12a05f20]
Block size=512 [0x200]



A FORMAT UNIT will commence in 15 seconds
ALL data on /dev/da6 will be DESTROYED
Press control-C to abort



A FORMAT UNIT will commence in 10 seconds
ALL data on /dev/da6 will be DESTROYED
Press control-C to abort



A FORMAT UNIT will commence in 5 seconds
ALL data on /dev/da6 will be DESTROYED
Press control-C to abort
Format unit cdb: 04 18 00 00 00 00
Format unit:
Descriptor format, current; Sense key: Illegal Request
Additional sense: Invalid field in cdb
Descriptor type: Sense key specific: Field pointer:
Error in Command: byte 1 bit 3
Format unit command: Illegal request, type: sense key, apart from Invalid opcode
FORMAT UNIT failed


Do you think it could be a firmware issue or misconfiguration in FreeNAS? From what I’ve read, it seems like most this stuff it plug-and-play in FreeNAS.
 

HoneyBadger

actually does care
Administrator
Moderator
iXsystems
Joined
Feb 6, 2014
Messages
5,112
Used SAS drives are often configured with 520-byte sectors or other T10 protection information, but yours seems to call them out as 512-byte. Maybe the protection info isn't being shown that way.

Dump the output of smartctl -a /dev/da6 and that may help determine.
 

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
Dump the output of smartctl -a /dev/da6 and that may help determine.
I have two of these drives. They were brand new stock, still in the sealed bags from Seagate. It's very possible that info is inaccurate in the output from my last post; even the serial numbers in the output below are from past drives that were in this system. Don't know why it does that but it's always done that with every drive that's not part of a pool. This is the output for both:
Code:
smartctl -a /dev/da6
smartctl 7.0 2018-12-30 r4883 [FreeBSD 11.3-RELEASE-p9 amd64] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Vendor:               ST316081
Product:              N/A           DB
Revision:             4321
Compliance:           SPC-3
Serial number:        4LS60ZYW
Device type:          disk
Transport protocol:   SAS (SPL-3)
Local Time is:        Tue Dec  8 13:37:12 2020 CST
device Test Unit Ready  [aborted command]
A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options.

smartctl -a /dev/da7
smartctl 7.0 2018-12-30 r4883 [FreeBSD 11.3-RELEASE-p9 amd64] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Vendor:               SEAGATE
Product:              ST6000NM0024  DB
Revision:             NA03
Compliance:           SPC-3
Logical Unit id:      0x5000c5007b58c913
Serial number:        Z4D2B6S8
Device type:          disk
Transport protocol:   SAS (SPL-3)
Local Time is:        Tue Dec  8 13:40:04 2020 CST
device Test Unit Ready  [aborted command]
A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options.


That's were this gets really confusing is because these drives are ST8000NT065 models. I don't have any clue why it shows the other model and serials.

When I add "-T short" I get the same output.

My NAS at home has direct connections to the HBA (no backplane) so I may try to get them working there. For some reason this disk shelf doesn't seem to allow the HBA to communicate fully with the drives. I don't understand why though.
 

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
Do you think the disk shelf would benefit from a firmware update? I wouldn't even know where to start on that. The Netapp stuff is kind of Greek to me but I wouldn't be afraid of digging into it if that would fix my issue.
 
Joined
Dec 29, 2014
Messages
1,135

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
It looks like that solution may work but I'm running into this issue:
Code:
root@freenas:~ # diskinfo -v da6
diskinfo: da6: ioctl(DIOCGMEDIASIZE) failed, probably not a disk.

root@freenas:~ # camcontrol cmd da6 -v -c "15 10 0 0 v:i1 0" 12 -o 12 "0 0 0 8 0 0:i3 0 v:i3" 512

root@freenas:~ # camcontrol format da6 -q -y
camcontrol: Unexpected SCSI error during format
(pass12:mps1:0:28:1): TEST UNIT READY. CDB: 00 00 00 00 00 00
(pass12:mps1:0:28:1): CAM status: SCSI Status Error
(pass12:mps1:0:28:1): SCSI status: Check Condition
(pass12:mps1:0:28:1): SCSI sense: ABORTED COMMAND asc:44,0 (Internal target failure)
(pass12:mps1:0:28:1): Field Replaceable Unit: 61


Any thoughts? I'm going to try bypassing the disk shelf and connecting the drive directly to the HBA.
 
Joined
Dec 29, 2014
Messages
1,135
It looks like that solution may work but I'm running into this issue:
Code:
root@freenas:~ # diskinfo -v da6
diskinfo: da6: ioctl(DIOCGMEDIASIZE) failed, probably not a disk.

root@freenas:~ # camcontrol cmd da6 -v -c "15 10 0 0 v:i1 0" 12 -o 12 "0 0 0 8 0 0:i3 0 v:i3" 512

root@freenas:~ # camcontrol format da6 -q -y
camcontrol: Unexpected SCSI error during format
(pass12:mps1:0:28:1): TEST UNIT READY. CDB: 00 00 00 00 00 00
(pass12:mps1:0:28:1): CAM status: SCSI Status Error
(pass12:mps1:0:28:1): SCSI status: Check Condition
(pass12:mps1:0:28:1): SCSI sense: ABORTED COMMAND asc:44,0 (Internal target failure)
(pass12:mps1:0:28:1): Field Replaceable Unit: 61


Any thoughts? I'm going to try bypassing the disk shelf and connecting the drive directly to the HBA.
It was a couple of years ago when I did this. My recollection was using the format command from that article verbatim, but I can't say 100%. I know the format command took a REALLY LONG TIME to complete. I was using Netapp drives in an HP shelf, so it isn't exactly the same thing.
 

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
So this is good and a little disappointing at the same time; I plugged the drive directly into the HBA on my home NAS the drive works perfectly as expected! Agh! That means the issue has something to do with my disk shelf; bummer.

Any suggestions on where to start looking on that side? Is it possible the disk shelf just isn't compatible with FreeBSD? I got what seemed to be a killer deal at the time... Maybe that's why. lol
 

HoneyBadger

actually does care
Administrator
Moderator
iXsystems
Joined
Feb 6, 2014
Messages
5,112
I'd like to think that the shelf itself wouldn't have any specific compatibility issues; NetApp was normally pretty good about that. A firmware update for the backplane though might be the ticket. Unfortunately my knowledge with wiring NetApp stuff up to work outside their box is rather limited.
 

photomatix

Dabbler
Joined
May 30, 2018
Messages
24
Sorry for the late reply. Was out sick for a week.

Any suggestions on where to start with the shelf? Or the IO cards rather. I haven't been able to find anything on config or updates for the NetApp gear.
 

WSkinny

Cadet
Joined
Dec 12, 2022
Messages
2
Well, I have only recently moved to TrueNAS as my base OS for my NAS from OpenMediaVault, mostly because I do add a lot of stuff on it that does not come or is usually possible to install much less supported on TrueNAS but I can tell you what is wrong.

So to the point the issue is the same I have 4x4Tb HGST SAS disks branded NETAPP the basterds at NETAPP change the drives firmware so it will only work with their OS basically, the only solution here being somehow restoring the disks to their original HGST firmware, question is how.

Personally I am lucky as I do have a few other HGST 4Tb SAS disks that are the exact same mode, that said I am still researching the practicality of changing the firmware manually on this NETAPP disks without any funny business, which I will likely won't have any luck with and end having to attempt to directly upgrade the firmware on the devices EEPROMS/FLASH assuming that is not blocked.

If anyone has any way to address this in a confirmed working way please do advise!
 

WSkinny

Cadet
Joined
Dec 12, 2022
Messages
2
Well, I have only recently moved to TrueNAS as my base OS for my NAS from OpenMediaVault, mostly because I do add a lot of stuff on it that does not come or is usually possible to install much less supported on TrueNAS but I can tell you what is wrong.

So to the point the issue is the same I have 4x4Tb HGST SAS disks branded NETAPP the basterds at NETAPP change the drives firmware so it will only work with their OS basically, the only solution here being somehow restoring the disks to their original HGST firmware, question is how.

Personally I am lucky as I do have a few other HGST 4Tb SAS disks that are the exact same mode, that said I am still researching the practicality of changing the firmware manually on this NETAPP disks without any funny business, which I will likely won't have any luck with and end having to attempt to directly upgrade the firmware on the devices EEPROMS/FLASH assuming that is not blocked.

If anyone has any way to address this in a confirmed working way please do advise!

By the way I seem to have found a place where you can get the HDDs original or netapp firmware apparently!
Link for the HGST HUS724020ALS640 SAS HDD Original FW if anyone needs it, looking for how to install it now!

DO THIS AT YOUR OWN RESPONSIBILITY! THIS IS A DANGEROUS PROCEDURE THAT CAN DEFINITELY KILL YOUR DISK
 
Top