LinuxLists.cc - Faulty seagate drives, are going to be blacklisted?

2009-01-19 23:29:23

Subject: Faulty seagate drives, are going to be blacklisted?

Tech sites are reporting everywhere a massive flaw in seagate drives that
can lock up the drive and make it unusable (the bios doesn't detect it, you
can't read the data). Haven't read anything about it here on the lists.
Seagate has ack'ed the problem:
http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931

So, apparently there're a lot of drives on the market (including mine)
that can die any day. Are those drives going to be blacklisted? It's
still not clear if the firmware update is safe (some affected but
working drives are dying after the firmware update), so some people
like me is still waiting (and hoping that the drive doesn't die) for
more stable firmware updates...

Here is the list of drives+firmware affected, according to the support site
as of now. Some models are still being diagnosed.

Seagate Barracuda 7200.11 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207951)

Models Affected:
ST3500320AS
ST3640330AS
ST3750330AS
ST31000340AS
Firmware Affected
SD15, SD16, SD17, SD18, SD19, AD14
Recommended Firmware Update
SD1A

Seagate Barracuda 7200.11, page 2 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207957)
Models Affected:
ST31500341AS
ST31000333AS
ST3640323AS
ST3640623AS
ST3320613AS
ST3320813AS
ST3160813AS
Firmware Affected
Still Unknow
Recommended Firmware Update
Still Unknow

Seagate Barracuda ES.2 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207963)
Models Affected:
ST3250310NS
ST3500320NS
ST3750330NS
ST31000340NS
Firmware Affected
Still Unknow
Recommended Firmware Update
Still Unknow

DiamondMax 22 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207969)
Models Affected:
STM3500320AS
STM3750330AS
STM31000340AS
Firmware Affected
MX15 (or higher)
Recommended Firmware Update
MX1A

DiamondMax 22 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207975)
Models Affected:
STM31000334AS
STM3320614AS
STM3160813AS
Firmware Affected
Still Unknow
Recommended Firmware Update
Still Unknow

2009-01-20 00:25:24

by David Rees

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

On Mon, Jan 19, 2009 at 3:29 PM, Diego Calleja <[email protected]> wrote:
> So, apparently there're a lot of drives on the market (including mine)
> that can die any day. Are those drives going to be blacklisted? It's
> still not clear if the firmware update is safe (some affected but
> working drives are dying after the firmware update), so some people
> like me is still waiting (and hoping that the drive doesn't die) for
> more stable firmware updates...

What would blacklisting these buggy drives achieve? There isn't
anything that can be done except warn the user that they have known
buggy firmware and let them know they should contact the vendor for a
firmware update. But until that bug hits, it doesn't seem to
otherwise affect the performance or functionality of the drives.

-Dave

2009-01-20 02:55:28

by Robert Hancock

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

2009-01-20 03:32:45

by Valdis Klētnieks

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

On Tue, 20 Jan 2009 00:29:23 +0100, Diego Calleja said:
> Tech sites are reporting everywhere a massive flaw in seagate drives that
> can lock up the drive and make it unusable (the bios doesn't detect it, you
> can't read the data). Haven't read anything about it here on the lists.
> Seagate has ack'ed the problem:
> http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931
>
> So, apparently there're a lot of drives on the market (including mine)
> that can die any day. Are those drives going to be blacklisted?

The $64 question is, of course: What exactly should the operating system
*do* if it detects one of these drives? Prohibit it from bricking later
by essentially bricking it *now*? What if the drive already has a lot of
production data on it?

Attachments:

(No filename) (226.00 B)

2009-01-20 15:33:21

by Diego Calleja

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

El Mon, 19 Jan 2009 20:55:05 -0600, Robert Hancock <[email protected]> escribió:

> There are a few drives which are currently marked to disable NCQ and
> warn the user that the firmware that should be upgraded:
>
> ST31500341AS
> ST31000333AS
> ST3640623AS
> ST3640323AS
> ST3320813AS
> ST3320613AS
>
> all for firmware versions SD15 through SD19.

Yes, I saw them, but apparently the NCQ bug is unrelated to this one.

2009-01-20 15:39:20

by Diego Calleja

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

El Mon, 19 Jan 2009 22:32:25 -0500, [email protected] escribió:

> The $64 question is, of course: What exactly should the operating system
> *do* if it detects one of these drives? Prohibit it from bricking later
> by essentially bricking it *now*? What if the drive already has a lot of
> production data on it?

Yeah, that's why I asked. Now that I think about it, it should probably be
the HAL people who should add one of those desktop "bubbles" warning the
users about the possible failure (they already do it for faulty batteries)

2009-01-20 17:24:33

by Valdis Klētnieks

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

On Tue, 20 Jan 2009 16:30:28 +0100, Diego Calleja said:
> Yeah, that's why I asked. Now that I think about it, it should probably be
> the HAL people who should add one of those desktop "bubbles" warning the
> users about the possible failure (they already do it for faulty batteries)

Probably a better approach, as long as we leave enough info visible in various
/sys files for HAL to figure it out - but I'm pretty sure we already do that...

Attachments:

(No filename) (226.00 B)

2009-01-20 18:20:21

by Diego Calleja

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

El Tue, 20 Jan 2009 12:24:07 -0500, [email protected] escribió:

> Probably a better approach, as long as we leave enough info visible in various
> /sys files for HAL to figure it out - but I'm pretty sure we already do that...

Yeah, it's all there already, and HAL has support for it. It just needs
the neccesary .fdi files.

2009-01-21 00:30:44

by Robert Hancock

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

Diego Calleja wrote:
> El Mon, 19 Jan 2009 20:55:05 -0600, Robert Hancock <[email protected]> escribió:
>
>> There are a few drives which are currently marked to disable NCQ and
>> warn the user that the firmware that should be upgraded:
>>
>> ST31500341AS
>> ST31000333AS
>> ST3640623AS
>> ST3640323AS
>> ST3320813AS
>> ST3320613AS
>>
>> all for firmware versions SD15 through SD19.
>
>
> Yes, I saw them, but apparently the NCQ bug is unrelated to this one.

I suspect it might be related, given that the firmware versions seem to
partially overlap.

With this issue though, there isn't anything the kernel can do about the
problem, so blacklisting doesn't seem to really make much sense.

2009-01-21 10:28:01

by Patrick Horn

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

Diego Calleja wrote:
> Tech sites are reporting everywhere a massive flaw in seagate drives that
> can lock up the drive and make it unusable (the bios doesn't detect it, you
> can't read the data). Haven't read anything about it here on the lists.
> Seagate has ack'ed the problem:
> http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931
>
> So, apparently there're a lot of drives on the market (including mine)
> that can die any day. Are those drives going to be blacklisted? It's
> still not clear if the firmware update is safe (some affected but
> working drives are dying after the firmware update), so some people
> like me is still waiting (and hoping that the drive doesn't die) for
> more stable firmware updates...
>
> Here is the list of drives+firmware affected, according to the support site
> as of now. Some models are still being diagnosed.
>
>
> Seagate Barracuda 7200.11 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207951)
>
> Models Affected:
> ST3500320AS
> ST3640330AS
> ST3750330AS
> ST31000340AS
> Firmware Affected
> SD15, SD16, SD17, SD18, SD19, AD14
> Recommended Firmware Update
> SD1A
>
> Seagate Barracuda 7200.11, page 2 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207957)
> Models Affected:
> ST31500341AS
> ST31000333AS
> ST3640323AS
> ST3640623AS
> ST3320613AS
> ST3320813AS
> ST3160813AS
> Firmware Affected
> Still Unknow
> Recommended Firmware Update
> Still Unknow
>
>
> Seagate Barracuda ES.2 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207963)
> Models Affected:
> ST3250310NS
> ST3500320NS
> ST3750330NS
> ST31000340NS
> Firmware Affected
> Still Unknow
> Recommended Firmware Update
> Still Unknow
>
> DiamondMax 22 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207969)
> Models Affected:
> STM3500320AS
> STM3750330AS
> STM31000340AS
> Firmware Affected
> MX15 (or higher)
> Recommended Firmware Update
> MX1A
>
> DiamondMax 22 (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207975)
> Models Affected:
> STM31000334AS
> STM3320614AS
> STM3160813AS
> Firmware Affected
> Still Unknow
> Recommended Firmware Update
> Still Unknow
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/

Hi,

I have another drive which doesn't seem to be on any list, and a google search
comes up with very little information about this one.

I have two raided SATA 1TB "MAXTOR STM31000333AS" drives, firmware MX15, one of
which "failed" last weekend. I have since rebuilt the array and it has had no
further problems, but I know it's only a matter of time before it happens again.

I checked SMART, and both drives are essentially identical with nothing anywhere
near failure.
I am on Ubuntu kernel 2.6.28-4-generic #5-Ubuntu but I will be happy to build a
kernel if this becomes at all reproducible.

At first I thought that this NCQ problem might apply to me, but my drive is
(gasp) one letter different from two of those listed (both seagate and maxtor
variants):
http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931
And MX15 is listed as a faulty firmware for the STM31000340AS/334AS

I have been using these drives for just three weeks up to now, before having the
one drive fail (and later it gave a bunch of errors at bootup, which was solved
when it reset the SATA link). The other drive has luckily not had any issues.

Is this error just coincidence, or did Seagate forget to mention my drive?
(And what happened to the firmware updates--they seem to be "In Validation")
Is seagate the only site with information about this? Any public blacklist of
every affected drive? What can I see in dmesg that indicates that NCQ is the cause?

Thanks,
-Patrick

(I'll paste my dmesg as I don't know enough to tell if this is the same issue as
the other seagate drives--I trimmed the repetitive parts)

[ 7520.699730] ata2.00: exception Emask 0x10 SAct 0x7ff4f SErr 0x400100 action
0x6 frozen
[ 7520.699734] ata2.00: irq_stat 0x08000000, interface fatal error
[ 7520.699738] ata2: SError: { UnrecovData Handshk }
[ 7520.699743] ata2.00: cmd 61/50:00:89:4b:c0/00:00:01:00:00/40 tag 0 ncq 40960 out
[ 7520.699745] res 40/00:30:91:60:c0/00:00:01:00:00/40 Emask 0x10 (ATA
bus error)
[ 7520.699748] ata2.00: status: { DRDY }
[ 7520.699752] ata2.00: cmd 61/40:08:b1:4f:c0/00:00:01:00:00/40 tag 1 ncq 32768 out
[ 7520.699753] res 40/00:30:91:60:c0/00:00:01:00:00/40 Emask 0x10 (ATA
bus error)
[ 7520.699756] ata2.00: status: { DRDY }
[ 7520.699875] ata2: hard resetting link
[ 7521.180020] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[ 7521.250673] ata2.00: configured for UDMA/133
[ 7521.250724] ata2: EH complete
[ 7521.250812] sd 1:0:0:0: [sdb] 1953525168 512-byte hardware sectors: (1.00
TB/931 GiB)
[ 7521.250832] sd 1:0:0:0: [sdb] Write Protect is off
[ 7521.250835] sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
[ 7521.250865] sd 1:0:0:0: [sdb] Write cache: enabled, read cache: enabled,
doesn't support DPO or FUA
[ 7521.258968] ata2.00: exception Emask 0x10 SAct 0x7ffff SErr 0x400100 action
0x6 frozen
[ 7521.258972] ata2.00: irq_stat 0x08000000, interface fatal error
[ 7521.258975] ata2: SError: { UnrecovData Handshk }
... it then goes down to 1.5 Gbps but continues to give errors until it is
kicked from the raid array an hour later

[10477.764175] ata2.00: status: { DRDY }
[10477.764179] ata2: hard resetting link
[10478.248019] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[10478.318670] ata2.00: configured for UDMA/33
[10478.318679] end_request: I/O error, dev sdb, sector 989067690
[10478.318685] raid1: Disk failure on sdb3, disabling device.
[10478.318686] raid1: Operation continuing on 1 devices.

This drive also encountered a similar error on bootup the next day:
[ 9.389771] ata2.00: exception Emask 0x10 SAct 0xf SErr 0xc00000 action 0x6
frozen
[ 9.389774] ata2.00: irq_stat 0x0c000000, interface fatal error
[ 9.389776] ata2: SError: { Handshk LinkSeq }
[ 9.389780] ata2.00: cmd 60/02:00:3f:af:4e/00:00:00:00:00/40 tag 0 ncq 1024 in
[ 9.389781] res 40/00:10:41:af:4e/00:00:00:00:00/40 Emask 0x10 (ATA
bus error)
[ 9.389783] ata2.00: status: { DRDY }

From lspci -vvv:

0:1f.2 SATA controller: Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port
SATA AHCI Controller (rev 02) (prog-if 01)
Subsystem: ASUSTeK Computer Inc. Device 8277

Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx+
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0

Interrupt: pin B routed to IRQ 2299

Region 0: I/O ports at 9c00 [size=8]

Region 1: I/O ports at 9880 [size=4]

Region 2: I/O ports at 9800 [size=8]

Region 3: I/O ports at 9480 [size=4]

Region 4: I/O ports at 9400 [size=32]

Region 5: Memory at f9ffe800 (32-bit, non-prefetchable) [size=2K]

Capabilities: [80] Message Signalled Interrupts: Mask- 64bit- Queue=0/4
Enable+
Address: fee0f00c Data: 4181

Capabilities: [70] Power Management version 3

Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA
PME(D0-,D1-,D2-,D3hot+,D3cold-)
Status: D0 PME-Enable- DSel=0 DScale=0 PME-

Capabilities: [a8] SATA HBA <?>

Capabilities: [b0] Vendor Specific Information <?>

Kernel driver in use: ahci

Kernel modules: ahci

2009-01-25 01:13:05

by Tejun Heo

[permalink] [raw]

Subject: Re: Faulty seagate drives, are going to be blacklisted?

Hello, Patrick.

Patrick Horn wrote:
...
> Is this error just coincidence, or did Seagate forget to mention my drive?
> (And what happened to the firmware updates--they seem to be "In
> Validation")
> Is seagate the only site with information about this? Any public
> blacklist of every affected drive? What can I see in dmesg that
> indicates that NCQ is the cause?

I think it's coincidental. AFAIK, there was no report of increased
transmission failures. Two known problems with these firmwares are

1. timeout on FLUSH if NCQ is in use on certain drives

2. bricking after power off (so, the failure is almost always during
BIOS probing during boot)

> (I'll paste my dmesg as I don't know enough to tell if this is the same
> issue as the other seagate drives--I trimmed the repetitive parts)
>
> [ 7520.699730] ata2.00: exception Emask 0x10 SAct 0x7ff4f SErr 0x400100
> action 0x6 frozen
> [ 7520.699734] ata2.00: irq_stat 0x08000000, interface fatal error
> [ 7520.699738] ata2: SError: { UnrecovData Handshk }

This is transmission error. Most common causes are power related or
unreliable connection especially if backplanes are involved. Is the
problem still reproducible? If so, can you please try to move it to
different power connector and SATA port and see what changes?

Thanks.

--
tejun