Opened 3 years ago

Closed 3 years ago

Last modified 3 years ago

#1396 closed enhancement (fixed)

Seagate IronWolf Pro 125 SSD - False Positive SMART. 246 Unknown_Attribute

Reported by: Al Owned by: Christian Franke
Priority: major Milestone: Release 7.2
Component: drivedb Version:
Keywords: ssd Cc:

Description

As above.
Brand new SSD's populated in a Synology NAS all report failure on attribute 246 "Unknown_Attribute".

Synology NAS can't use the drives as they report total fail.

Attached -x report.

Attachments (1)

smartctl-SEAGATE-IRONWOLF-ZA3840NX1000.txt (7.1 KB ) - added by Al 3 years ago.

Download all attachments as: .zip

Change History (14)

comment:1 by Al, 3 years ago

Component: allsmartctl

comment:2 by Christian Franke, 3 years ago

Component: smartctldrivedb
Keywords: ssd added
Milestone: Release 7.2
Priority: blockermajor
Type: defectenhancement

Could be handled by a drive database entry. For a more general solution see ticket #1397.

Workaround for now: Add for example -v 246,hex64 or -v 246,raw64 to smartctl command line or create a local database entry which provides this option (see -B option on smartctl man page for details).

comment:3 by Al, 3 years ago

Thanks for the update. I am completely confused on how to implement the workarounds you mentioned.
I tried smartctl -v 246,hex64 /dev/hda, can't seem to make it ignore the dodgy attribute
"smartctl 6.5 (build date Aug 21 2020) [x86_64-linux-4.4.59+] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

ATA device successfully opened

Use 'smartctl -a' (or '-x') to print SMART (and more) information
"

As for -B, I looked at the man page and have no idea how to get that working.

comment:4 by Christian Franke, 3 years ago

I tried smartctl -v 246,hex64 /dev/hda, ...

Of course you need request the attribute table (at least -A) to see the effect of -v ... :-)

As for -B, I looked at the man page and have no idea how to get that working.

Copy this to a new local drive database file:

{
  "",
  "Seagate IronWolfPro ZA3840NX10001-2ZH104",
  "", "", "-v 246,hex64"
}

See -B at man page or smartctl -h output for the configured path for this file (usually /etc/smart_drivedb.h or /etc/smartmontools/smart_drivedb.h).

comment:5 by Al, 3 years ago

Ok, the only thing I've got anywhere is the following:

Created a new dir and file, /etc/smart_drivedb.h with the contents
{

"",
"Seagate IronWolfPro ZA3840NX10001-2ZH104",
"", "", "-v 246,hex64"

}

Output now is
smartctl 6.5 (build date Aug 21 2020) [x86_64-linux-4.4.59+] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

/B: Unable to detect device type
Please specify device type with the -d option.

Tried to force detect the device but no good.

I give up. How long does it take generally for these fixes to get updated typically?

in reply to:  5 comment:6 by Christian Franke, 3 years ago

Output now is ...

To get support for unexpected output, please also provide the input. In this case this means: The actual smartctl command line used.

/B: Unable to detect device type

/B is not a valid device name. Prefix / is not a valid option prefix (even in the Windows version of smartctl).

I give up. How long does it take generally for these fixes to get updated typically?

In general: When it's done :-) We are a very small team of volunteers maintaining smartmontools in unpaid spare time. Nobody pays for a SLA, so there is none.
In this case: See the Milestone.

comment:7 by Al, 3 years ago

My apologies, I was using the incorrect operator switch. I didn't get it to work but ill keep at it.
More apologies, I was directed here by Synology support. I thought this was a Synology product that they were working on, not palming me off to a open source dev.
Cheers for your help, much appreciated.

comment:8 by Christian Franke, 3 years ago

You're welcome.

If possible, please tell Synology that it is a bad idea to unconditionally block devices solely on failures of a single ATA SMART attribute.

Unlike the global SMART STATUS (SMART overall-health self-assessment test result: PASSED), attributes are not (and never were and never will be) part of any ATA standard. ATA attributes are "existing practice" which works in most cases. Long term experience (since 2004) shows that, in rare cases, drives may return anything as attributes, including random garbage.

If a related discussion with Synology is publicly available to non-customers, feel free to report a reference here.

comment:9 by Al, 3 years ago

Yes I am annoyed they used the sledgehammer approach too. The drives were still "Passed" but a 255/255 bogus attribute more or less killed the entire drive.

I've fed back the information and an admonishment for misleading people to think you are "synology".

Cheers.

comment:10 by Christian Franke, 3 years ago

Owner: set to Christian Franke
Status: newaccepted

comment:11 by Christian Franke, 3 years ago

IronWolf_Pro_125_SSD_Product_Manual_100866982_A.pdf says that the threshold of attribute 246 (F6h, "Write Protect Detail") is 253 (FDh) which is obviously not the case. This firmware violates early ATA SMART Attribute specs, see #1397.

comment:12 by Christian Franke, 3 years ago

Resolution: fixed
Status: acceptedclosed

comment:13 by Christian Franke, 3 years ago

Ticket #1383 has been marked as a duplicate of this ticket.

Note: See TracTickets for help on using tickets.