Is this critical??
I have this error's over my kern.log file and when i type dmesg..
Whats this all about? HD problems or some kernel conflict?
--
dmesg
db: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=14912550,
sector=10719504
end_request: I/O error, dev 03:42 (hdb), sector 10719504
--
--
kern.log
May 29 18:03:58 euclid ls[55] exited with preempt_count 9
May 29 18:03:58 euclid depscan.sh[54] exited with preempt_count 2
May 29 18:03:58 euclid ls[219] exited with preempt_count 1
May 29 18:03:58 euclid ls[409] exited with preempt_count 1
May 29 18:03:58 euclid depscan.sh[31] exited with preempt_count 76
May 29 18:03:58 euclid Adding Swap: 2096472k swap-space (priority -1)
May 29 18:03:58 euclid Adding Swap: 2096440k swap-space (priority -2)
May 29 18:03:58 euclid swapon[942] exited with preempt_count 3
May 29 18:03:58 euclid dmesg[943] exited with preempt_count 2
May 29 18:03:58 euclid ls[947] exited with preempt_count 1
--
Thanks,
Louie...
On Monday 16 September 2002 15:31, louie miranda wrote:
> Is this critical??
> I have this error's over my kern.log file and when i type dmesg..
> Whats this all about? HD problems or some kernel conflict?
This is the drive reporting a hardware error. It is probably dying, backups
are to be made ASAP.
DK
--
James Joyce -- an essentially private man who wished his total
indifference to public notice to be universally recognized.
-- Tom Stoppard
yes, this is critical.
It means that your HD is going to break soon.
Luigi
On Mon, 16 Sep 2002, louie miranda wrote:
> Date: Mon, 16 Sep 2002 21:31:18 +0800
> From: louie miranda <[email protected]>
> To: [email protected]
> Subject: Hi is this critical??
>
> Is this critical??
> I have this error's over my kern.log file and when i type dmesg..
> Whats this all about? HD problems or some kernel conflict?
>
>
> --
> dmesg
> db: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=14912550,
> sector=10719504
> end_request: I/O error, dev 03:42 (hdb), sector 10719504
> --
>
>
> --
> kern.log
> May 29 18:03:58 euclid ls[55] exited with preempt_count 9
> May 29 18:03:58 euclid depscan.sh[54] exited with preempt_count 2
> May 29 18:03:58 euclid ls[219] exited with preempt_count 1
> May 29 18:03:58 euclid ls[409] exited with preempt_count 1
> May 29 18:03:58 euclid depscan.sh[31] exited with preempt_count 76
> May 29 18:03:58 euclid Adding Swap: 2096472k swap-space (priority -1)
> May 29 18:03:58 euclid Adding Swap: 2096440k swap-space (priority -2)
> May 29 18:03:58 euclid swapon[942] exited with preempt_count 3
> May 29 18:03:58 euclid dmesg[943] exited with preempt_count 2
> May 29 18:03:58 euclid ls[947] exited with preempt_count 1
> --
>
>
>
> Thanks,
> Louie...
>
> -
> 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/
>
Le lun 16/09/2002 ? 15:37, [email protected] a ?crit :
>
> yes, this is critical.
> It means that your HD is going to break soon.
>
Maybe these error messages should be a bit less cryptic for the
uninitiated. Or is there a userspace utility to convert theses to
luser-understandable messages ?
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Monday 16 September 2002 16:47, Xavier Bestel wrote:
> Le lun 16/09/2002 ? 15:37, [email protected] a ?crit :
> > yes, this is critical.
> > It means that your HD is going to break soon.
>
> Maybe these error messages should be a bit less cryptic for the
> uninitiated. Or is there a userspace utility to convert theses to
> luser-understandable messages ?
1. No user-space utility exists/will exist for that (it will only make
maintaince of the kernel a bigger problem than it is today as there will be a
need to sync the two systems and since messages change all the time and there
is no standard format for error message which is strictly adhered too then
this is a much bigger problem than you talk about).
2. The user who posted the question is under no circumstances a "looser"
(mind the oo instead of the u...). His question is very valid and the fact
that he read dmesg puts him way past any standard computer user.
3. I don't think it is appropriate to call him with such a name and it even
hurts kernel development as it makes him (and other people that are watching
the list) abstain from submitting questions. While some of the questions are
not the interest of this list they could be politely told so but we do NOT
want to lose the really interesting bits (some user with a special IDE setup
that does not work and will be afraid to report the messages he's getting on
the ground of being called a looser).
Mark.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9heeqxlxDIcceXTgRAkYcAJ9OlVd2xUvH+ZAVM5Nwdv7umKHGiACggBYK
nmnRud3bf8NGq9OcDhuGUWY=
=oliB
-----END PGP SIGNATURE-----
Le lun 16/09/2002 ? 16:16, Mark Veltzer a ?crit :
> 2. The user who posted the question is under no circumstances a "looser"
> (mind the oo instead of the u...). His question is very valid and the fact
> that he read dmesg puts him way past any standard computer user.
Well, actually I didn't want to depict *him* as a looser. I was talking
about me :) I've already been confronted with message from the IDE
drivers (and that's when I see them. I'm not always at the console or
reading syslog) and I never remember if they are critical or harmless. I
have to either dig through lkml archives to find what they mean, or ask
lkml what do they mean (to the luser I am).
IDE error/status message aren't visible enough. I'd like to know when my
drive is near failing, without looking at syslog.
Xav - luser
On Mon, 2002-09-16 at 14:31, louie miranda wrote:
> Is this critical??
> I have this error's over my kern.log file and when i type dmesg..
> Whats this all about? HD problems or some kernel conflict?
>
>
> --
> dmesg
> db: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=14912550,
> sector=10719504
> end_request: I/O error, dev 03:42 (hdb), sector 10719504
Your drive thinks it has a bad block. While your kernel is full of logs
showing pre-empt usage and breakage its very very improbable that such
an error would occur any other way but a genuine error report by the
drive
On Mon, Sep 16, 2002 at 03:47:20PM +0200, Xavier Bestel wrote:
> Le lun 16/09/2002 ? 15:37, [email protected] a ?crit :
> >
> > yes, this is critical.
> > It means that your HD is going to break soon.
> >
>
> Maybe these error messages should be a bit less cryptic for the
> uninitiated. Or is there a userspace utility to convert theses to
> luser-understandable messages ?
I had a patch in the Daleki 2.5 IDE which would easily allow for
alternative descriptions to be added with very little pain.
The patch was really trivial, and converted the code that gives
these messages into a data structure instead. (A good balance
between data structures and code is an important point IMHO.)
However, due to the method by which Martins stuff was ripped out
of the kernel, its going to be pretty hard for me to recover the
patch. I was hoping BK might help here, but it doesn't appear
to.
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
If so, why not to use S.M.A.R.T with smartd and smartctl?
I think you will like them (loock on freshmeat for the link).
On 16 Sep 2002, Xavier Bestel wrote:
> Date: 16 Sep 2002 16:14:54 +0200
> From: Xavier Bestel <[email protected]>
> To: Mark Veltzer <[email protected]>
> Cc: Linux Kernel Mailing List <[email protected]>
> Subject: Re: Hi is this critical??
>
> Le lun 16/09/2002 ? 16:16, Mark Veltzer a ?crit :
>
> > 2. The user who posted the question is under no circumstances a "looser"
> > (mind the oo instead of the u...). His question is very valid and the fact
> > that he read dmesg puts him way past any standard computer user.
>
> Well, actually I didn't want to depict *him* as a looser. I was talking
> about me :) I've already been confronted with message from the IDE
> drivers (and that's when I see them. I'm not always at the console or
> reading syslog) and I never remember if they are critical or harmless. I
> have to either dig through lkml archives to find what they mean, or ask
> lkml what do they mean (to the luser I am).
>
> IDE error/status message aren't visible enough. I'd like to know when my
> drive is near failing, without looking at syslog.
>
> Xav - luser
>
> -
> 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/
>
On Mon, 16 Sep 2002 [email protected] wrote:
> If so, why not to use S.M.A.R.T with smartd and smartctl?
> I think you will like them (loock on freshmeat for the link).
>
I don't think S.M.A.R.T should be the all mighty god of dying hard drive
detections. Many times I had dying IDE drives that wouldn't show much
difference on S.M.A.R.T tests and would stay in operationnal parameters
until it refuses to spin up or was just covered in bad sectors.
> > If so, why not to use S.M.A.R.T with smartd and smartctl?
> > I think you will like them (loock on freshmeat for the link).
>
> I don't think S.M.A.R.T should be the all mighty god of dying hard drive
> detections. Many times I had dying IDE drives that wouldn't show much
> difference on S.M.A.R.T tests and would stay in operationnal parameters
> until it refuses to spin up or was just covered in bad sectors.
S.M.A.R.T. is useful to prove that a drive is dying, but it is not useful to prove that it is not.
I.E. if you get a S.M.A.R.T. message saying that the drive is dying, then believe it, and back up your data, but don't rely solely on S.M.A.R.T. to detect a dying drive.
In any case, though, check out:
http://csl.cse.ucsc.edu/smart.shtml
John.
On Mon, 16 Sep 2002 [email protected] wrote:
>
> S.M.A.R.T. is useful to prove that a drive is dying, but it is not useful to prove that it is not.
Yes, of course, and this was exaclty what was asked here in the mail
from xavier that started this thread. The point is if S.M.A.R.T will
advice before you see seek errors messages from the kernel or not.
Luigi
On Tue, 17 Sep 2002 [email protected] wrote:
> On Mon, 16 Sep 2002 [email protected] wrote:
>
> >
> > S.M.A.R.T. is useful to prove that a drive is dying, but it is not useful to prove that it is not.
>
> Yes, of course, and this was exaclty what was asked here in the mail
> from xavier that started this thread. The point is if S.M.A.R.T will
> advice before you see seek errors messages from the kernel or not.
It never advised me before seeing problems in the kernel (I had about 5
drives dying running Linux machines).
It should be trivial to just grep the kernel log for the error and mail it
to some address.
Another way to prove a dead drive is (after a backup) to drop it some
until it made a nice broken drive sound (a very high pitch shriek) and
bring it to the store that sold it to you (ideally a small one as they are
less technically challenged then big chains).
> On Tue, 17 Sep 2002 [email protected] wrote:
>
> > On Mon, 16 Sep 2002 [email protected] wrote:
> >
> > >
> > > S.M.A.R.T. is useful to prove that a drive is dying, but it is not useful to prove that it is not.
> >
> > Yes, of course, and this was exaclty what was asked here in the mail
> > from xavier that started this thread. The point is if S.M.A.R.T will
> > advice before you see seek errors messages from the kernel or not.
>
> It never advised me before seeing problems in the kernel (I had about 5
> drives dying running Linux machines).
Were you actually monitoring the S.M.A.R.T. information before you suspected that the drive was failing, though?
S.M.A.R.T. data is only meaningful when it's monitored over a period of time. The actual numbers are fairly meaningless, and are bound to vary from drive to drive, but if you suddenly find that, for example, spinning up the disk is taking longer than normal, then you might suspect that there is something wrong with the motor/spindle assembly.
S.M.A.R.T. is far from being perfect, but it is another tool that we have at our disposal. All modern discs support S.M.A.R.T. so you might as well enable it. If nothing else, you could replace the disk after, say, 10,000 power on hours, or 2500 spin up, (wash, rinse, spin), and spin down, cycles. I.E. Preventative maintainance.
> It should be trivial to just grep the kernel log for the error and mail it
> to some address.
Exactly, you can do this in user space, using tail -f and a perl script.
> Another way to prove a dead drive is (after a backup) to drop it some
> until it made a nice broken drive sound (a very high pitch shriek) and
> bring it to the store that sold it to you (ideally a small one as they are
> less technically challenged then big chains).
I think they would probably say that you voided the warranty.
John.
> > Another way to prove a dead drive is (after a backup) to drop it some
> > until it made a nice broken drive sound (a very high pitch shriek) and
> > bring it to the store that sold it to you (ideally a small one as they are
> > less technically challenged then big chains).
>
> I think they would probably say that you voided the warranty.
Correct!
MORONS that think the drive vendors are not clued into the issue.
I have to read and vote on NASTY proposals, whose intent is to check for
G-Force damage. If you think that record is not findable, even if you
rake the heads over the media you are wrong.
They are loosing money because of returns and are about to have the means
to tell you nice try, does warrenty fraud mean anything?
Cheers,
Andre Hedrick
LAD Storage Consulting Group
On Mon, Sep 16, 2002 at 03:47:20PM +0200, Xavier Bestel wrote:
> Le lun 16/09/2002 ? 15:37, [email protected] a ?crit :
> >
> > yes, this is critical.
> > It means that your HD is going to break soon.
> >
>
> Maybe these error messages should be a bit less cryptic for the
> uninitiated. Or is there a userspace utility to convert theses to
> luser-understandable messages ?
The problem is that you're going to leave out details to do this.
This means that even the experts don't know what happened.
>From the reported message I was able to pinpoint the exact spot on
his hdb2 partition where his drive has a bad spot, and I can calculate
the size of his hdb1 partition (In this case probably unneccesary
information, but still...).
If the error message says something like:
"your harddrive hdb seems to be failing. Replace it NOW!"
then I can't make a valid decision about this. I have had a WD 31600
drive which had the trouble that the last 400M would go "bad": Lots of
bad blocks. In that situation I had to monitor the drive to just have
bad blocks in the last 400M or so, while I used only the first 1000M.
(Yes, only for "tmp" stuff.)
I also have a drive that has exactly ONE bad block. Run it through
"badblocks", and you can use the disk just fine. A read-ahead
might hit on the bad block, leading to funny error messages. However
I DO need to know which block.
MAC and Windows both try to give "simple" error messages. As a
knowledgeable person I then am confronted with:
"Something went wrong. Contact your sysadmin if you can't
fix it yourself".
Well, I -=AM=- the sysadmin, and would like to know what went wrong.
Otherwise I can't fix it. This is NOT the way to go.
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* The Worlds Ecosystem is a stable system. Stable systems may experience *
* excursions from the stable situation. We are currenly in such an *
* excursion: The stable situation does not include humans. ***************
> MAC and Windows both try to give "simple" error messages. As a
> knowledgeable person I then am confronted with:
>
> "Something went wrong. Contact your sysadmin if you can't
> fix it yourself".
>
> Well, I -=AM=- the sysadmin, and would like to know what went wrong.
> Otherwise I can't fix it. This is NOT the way to go.
Exactly, because the sysadmin can't learn any more from that error than the user can, and instead he or she just defaults to:
"Well, I don't know, why are you asking me? Reboot it and see what happens. Good luck."
John.
On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
> MORONS that think the drive vendors are not clued into the issue.
> I have to read and vote on NASTY proposals, whose intent is to check for
> G-Force damage. If you think that record is not findable, even if you
Sounds good news for honest users. What it does want though is the
ability of users to check that data when the disk arrives, because we
have delivery people, and they think that if looks like a box its
probably a football.
I've noticed putting one way temperature strips into things is becoming
popular to which is nice, but again its done in a way the user can't get
the data, while for high end enterprise/business use knowing you
accidentally cooked a disk is really important for making decisions both
about new fans/aircon and also about whether to replace
On Tue, Sep 17, 2002 at 11:41:44AM +0100, Alan Cox wrote:
> On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
> > MORONS that think the drive vendors are not clued into the issue.
> > I have to read and vote on NASTY proposals, whose intent is to check for
> > G-Force damage. If you think that record is not findable, even if you
>
> Sounds good news for honest users. What it does want though is the
> ability of users to check that data when the disk arrives, because we
> have delivery people, and they think that if looks like a box its
> probably a football.
No. Sounds _bad_ news for honest users. Lets be realistic here.
Parcel company uses package as a football, then delivers it.
How many parcel companies will wait while you unpack the hard drive,
dismantle your machine, connect the drive, power it up and check to
see if the drive has suffered too much shock?
Yep, that's right, none. If you try to, the parcel delivery person
will get really ratty. So you sign for it after checking that the
outside box is undamaged.
Then you find out that its been used in the world cup. You try to
return it to the vendor, but the vendor says its your fault for
dropping the drive. You protest, but the vendor refuses to listen
because they've got their technology that says so in their product.
So, its NOT great for honest users. Its another form of "Digital
Rights Management."
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
On Tue, 2002-09-17 at 12:09, Russell King wrote:
> Then you find out that its been used in the world cup. You try to
> return it to the vendor, but the vendor says its your fault for
> dropping the drive. You protest, but the vendor refuses to listen
> because they've got their technology that says so in their product.
Then you take them to the small claims court. Its their burden of proof.
Or in the US I imagine you file a class action lawsuit, but before you
can file it the features go away because someone else sues them for
patent infringement instead ;)
Alan
> On Tue, 2002-09-17 at 12:09, Russell King wrote:
> > Then you find out that its been used in the world cup. You try to
> > return it to the vendor, but the vendor says its your fault for
> > dropping the drive. You protest, but the vendor refuses to listen
> > because they've got their technology that says so in their product.
>
> Then you take them to the small claims court. Its their burden of proof.
> Or in the US I imagine you file a class action lawsuit, but before you
> can file it the features go away because someone else sues them for
> patent infringement instead ;)
I want to file a patent on using hard disks as footballs. I claim prior art, because I kicked a, (non-functional), disk across a room in front of the person who owned it, (with their permission), about five years ago, and they would verify that.
So, how do I go about filing a patent request in the EU, Japan, and the U.S.A.?
Incidently unlimited permission is granted to use disks containing only GNU/Linux software as footballs without royalties. All other usage of disks as footballs will be considered on a per case basis.
Note that this only applies to hard disks, (IDE and SCSI), *NOT* floppy disks, or optical media.
John.
On Tue, Sep 17, 2002 at 12:17:53PM +0100, Alan Cox wrote:
> On Tue, 2002-09-17 at 12:09, Russell King wrote:
> > Then you find out that its been used in the world cup. You try to
> > return it to the vendor, but the vendor says its your fault for
> > dropping the drive. You protest, but the vendor refuses to listen
> > because they've got their technology that says so in their product.
>
> Then you take them to the small claims court. Its their burden of proof.
Ok. So the vendor gets out their technology that says "this drive
has been exposed to excessive G force." So they have proof that it
has been dropped by someone. However, the vendor can't prove that:
1. the vendor didn't drop it
2. the parcel company didn't drop it
3. you didn't drop it
Conversely, you can't prove:
1. that the vendor dropped it
2. the parcel company dropped it
3. you dropped it (not that you'd want to)
So its now their word against yours, and as you say, its up to the
vendor to prove that _you_ caused the damage. Fortunately, this
technology doesn't do that, so honest users might be safe for the
time being.
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
> On Tue, Sep 17, 2002 at 12:17:53PM +0100, Alan Cox wrote:
> > On Tue, 2002-09-17 at 12:09, Russell King wrote:
> > > Then you find out that its been used in the world cup. You try to
> > > return it to the vendor, but the vendor says its your fault for
> > > dropping the drive. You protest, but the vendor refuses to listen
> > > because they've got their technology that says so in their product.
> >
> > Then you take them to the small claims court. Its their burden of proof.
>
> Ok. So the vendor gets out their technology that says "this drive
> has been exposed to excessive G force." So they have proof that it
> has been dropped by someone. However, the vendor can't prove that:
>
> 1. the vendor didn't drop it
> 2. the parcel company didn't drop it
> 3. you didn't drop it
>
> Conversely, you can't prove:
>
> 1. that the vendor dropped it
> 2. the parcel company dropped it
> 3. you dropped it (not that you'd want to)
>
> So its now their word against yours, and as you say, its up to the
> vendor to prove that _you_ caused the damage. Fortunately, this
> technology doesn't do that, so honest users might be safe for the
> time being.
Am I missing something, though, or is it not the case that if it has been dropped by the parcel company or the vendor, errors are likely to show up pretty soon if they are going to show up at all. You can't phone up 35 months in to a three year warranty, and expect them to believe, "Oh, the parcel company broke it, and it hasn't shown up because it damaged the last sector on the disk", (especially since most disks park the heads on the outside of the disk, so if that is the problem, you're going to notice it straight away).
I suppose that you could have a problem like a cracked spindle that didn't show up straight away, but realistically, if it's not your fault it got broken, then it's not likely to show up after a couple of years, is it?
John.
On Tue, 17 Sep 2002, Russell King wrote:
> On Tue, Sep 17, 2002 at 11:41:44AM +0100, Alan Cox wrote:
> > On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
> > > MORONS that think the drive vendors are not clued into the issue.
> > > I have to read and vote on NASTY proposals, whose intent is to check for
> > > G-Force damage. If you think that record is not findable, even if you
> >
> > Sounds good news for honest users. What it does want though is the
> > ability of users to check that data when the disk arrives, because we
> > have delivery people, and they think that if looks like a box its
> > probably a football.
>
> No. Sounds _bad_ news for honest users. Lets be realistic here.
> Parcel company uses package as a football, then delivers it.
>
> How many parcel companies will wait while you unpack the hard drive,
> dismantle your machine, connect the drive, power it up and check to
> see if the drive has suffered too much shock?
Quietly it is the SHIPPER's Test.
There should be an appilcation you run when you get the device.
If you fail to run it when you get first powerup the device, you are at
fault. To many devices have been smoked in shipping.
> Yep, that's right, none. If you try to, the parcel delivery person
> will get really ratty. So you sign for it after checking that the
> outside box is undamaged.
>
> Then you find out that its been used in the world cup. You try to
> return it to the vendor, but the vendor says its your fault for
> dropping the drive. You protest, but the vendor refuses to listen
> because they've got their technology that says so in their product.
>
> So, its NOT great for honest users. Its another form of "Digital
> Rights Management."
>
> --
> Russell King ([email protected]) The developer of ARM Linux
> http://www.arm.linux.org.uk/personal/aboutme.html
>
Andre Hedrick
LAD Storage Consulting Group
On Tue, Sep 17, 2002 at 10:34:10AM -0700, Andre Hedrick wrote:
> Quietly it is the SHIPPER's Test.
>
> There should be an appilcation you run when you get the device.
> If you fail to run it when you get first powerup the device, you are at
> fault. To many devices have been smoked in shipping.
Umm, that's crap. If the program is an x86 binary, I don't have
the facilities to run it here.
Its their problem. Its a crap system. Period.
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
On 17 Sep 2002, Alan Cox wrote:
> On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
> > MORONS that think the drive vendors are not clued into the issue.
> > I have to read and vote on NASTY proposals, whose intent is to check for
> > G-Force damage. If you think that record is not findable, even if you
>
> Sounds good news for honest users. What it does want though is the
> ability of users to check that data when the disk arrives, because we
> have delivery people, and they think that if looks like a box its
> probably a football.
Yes, some delivery men have misunderstood the term "drop ship."
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
On Tue, 17 Sep 2002, Russell King wrote:
> On Tue, Sep 17, 2002 at 11:41:44AM +0100, Alan Cox wrote:
> > On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
> > > MORONS that think the drive vendors are not clued into the issue.
> > > I have to read and vote on NASTY proposals, whose intent is to check for
> > > G-Force damage. If you think that record is not findable, even if you
> >
> > Sounds good news for honest users. What it does want though is the
> > ability of users to check that data when the disk arrives, because we
> > have delivery people, and they think that if looks like a box its
> > probably a football.
>
> No. Sounds _bad_ news for honest users. Lets be realistic here.
> Parcel company uses package as a football, then delivers it.
>
> How many parcel companies will wait while you unpack the hard drive,
> dismantle your machine, connect the drive, power it up and check to
> see if the drive has suffered too much shock?
First, there are cheap "tell-tale" indicators which can be attached to the
contents of the container where you can see them as soon as you open the
package. That the vendors ship without them tells you they have found that
it's cheaper to eat a few DOA units than spend even a few cents on proof.
Second, having attached such a device to an old hard drive and repacked
it, I dropped it from eye level (about two meters) on a hardwoord floor.
The 5G triggered, the 20G didn't. Many drives will take 20G power off, so
you really need to beat on the package to do harm. Tossing it into a truck
or dropping it a foot onto a conveyor belt is unlikely to cause damage.
Yes, anything is possible.
Finally, I have returned a number of drives, and at least WD has just
asked for the serial number so they could check the build date. I'm told
Seagate is also good on that, asking only about power surges, etc. Someone
else always seems to deal with that, so that's lack of complaint, not
experience.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
On Tue, 17 Sep 2002 [email protected] wrote:
> I want to file a patent on using hard disks as footballs. I claim prior
> art, because I kicked a, (non-functional), disk across a room in front
> of the person who owned it, (with their permission), about five years
> ago, and they would verify that.
Sorry, prior art exists. When iomega (the ZIP people) first shipped, they
brought out a unit to demo. They pulled the media out, tossed it against
the far wall, kicked it back, wiped it and ran from it. Then lifted the
front of the case about two inches and dropped it.
In those days the "hard drives" were the same size, 10MB (sic), and the
heads would crash if a mouse farted in the next room. Bought a bunch, and
had very few problems.
> Incidently unlimited permission is granted to use disks containing only
> GNU/Linux software as footballs without royalties. All other usage of
> disks as footballs will be considered on a per case basis.
I would be inclined to allow this for Windows drives instead. They contain
no useful data.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
> On Tue, Sep 17, 2002 at 10:34:10AM -0700, Andre Hedrick wrote:
> > Quietly it is the SHIPPER's Test.
> >
> > There should be an appilcation you run when you get the device.
> > If you fail to run it when you get first powerup the device, you are at
> > fault. To many devices have been smoked in shipping.
>
> Umm, that's crap. If the program is an x86 binary, I don't have
> the facilities to run it here.
No, it's not an x86 binary, it's in the firmware of the drive:
Execute Extended Self Test
smartctl -x /dev/hda?
so if you can send a S.M.A.R.T. command, you can run the test.
> Its their problem. Its a crap system. Period.
No, I don't think that you can expect drive manufacturers to replace a disc that you haven't taken care of, especially if it has been used as a football, (isn't that against FIFA regulations, though? Especially in the world cup example given earlier in this thread. I thought the specs of the football were quite strict).
John.
On Tuesday, September 17, 2002, at 07:09 AM, Russell King wrote:
> On Tue, Sep 17, 2002 at 11:41:44AM +0100, Alan Cox wrote:
>> On Tue, 2002-09-17 at 09:20, Andre Hedrick wrote:
>>> MORONS that think the drive vendors are not clued into the issue.
>>> I have to read and vote on NASTY proposals, whose intent is to check
>>> for
>>> G-Force damage. If you think that record is not findable, even if
>>> you
>>
>> Sounds good news for honest users. What it does want though is the
>> ability of users to check that data when the disk arrives, because we
>> have delivery people, and they think that if looks like a box its
>> probably a football.
>
> No. Sounds _bad_ news for honest users. Lets be realistic here.
> Parcel company uses package as a football, then delivers it.
>
> How many parcel companies will wait while you unpack the hard drive,
> dismantle your machine, connect the drive, power it up and check to
> see if the drive has suffered too much shock?
>
> Yep, that's right, none. If you try to, the parcel delivery person
> will get really ratty. So you sign for it after checking that the
> outside box is undamaged.
>
> Then you find out that its been used in the world cup. You try to
> return it to the vendor, but the vendor says its your fault for
> dropping the drive. You protest, but the vendor refuses to listen
> because they've got their technology that says so in their product.
>
Then you notice that nowhere did you agree that your warranty validity
is dependent on some little piece of technology.
You ask them to point out where you agreed to this.
They can't.
You threaten them, they ignore you.
You start a class action suit (cost to you: Zero in straight money, but
a *lot* of time)
You win 7 years later (average time to resolve a lawsuit these days).
They get a big black eye.
If you don't want a class action, you take them to small claims court.
You'll win much quicker.
In short, they need to be able to prove that *you* damaged the drive.
They can't.
--Dan
On Tuesday, September 17, 2002, at 08:02 AM, [email protected]
wrote:
>> On Tue, 2002-09-17 at 12:09, Russell King wrote:
>>> Then you find out that its been used in the world cup. You try to
>>> return it to the vendor, but the vendor says its your fault for
>>> dropping the drive. You protest, but the vendor refuses to listen
>>> because they've got their technology that says so in their product.
>>
>> Then you take them to the small claims court. Its their burden of
>> proof.
>> Or in the US I imagine you file a class action lawsuit, but before you
>> can file it the features go away because someone else sues them for
>> patent infringement instead ;)
>
> I want to file a patent on using hard disks as footballs. I claim
> prior art, because I kicked a, (non-functional), disk across a room in
> front of the person who owned it, (with their permission), about five
> years ago, and they would verify that.
>
You made no attempt to ensure their confidentiality, and thus, have
just admitted you used it publicly more than one year before you've
filed.
Thus, it is unpatentable.
> So, how do I go about filing a patent request in the EU, Japan, and
> the U.S.A.?
>
This is tricky, actually, because if the patent issues in the EU or
Japan before you file in the USA, yer screwed (your EU and Japan
patents will be prior art, and thus, your US patent will be rejected).
> Incidently unlimited permission is granted to use disks containing
> only GNU/Linux software as footballs without royalties. All other
> usage of disks as footballs will be considered on a per case basis.
>
> Note that this only applies to hard disks, (IDE and SCSI), *NOT*
> floppy disks, or optical media.
>
> John.
I'm the weird combo of "GCC/GDB hacker and 2nd year law student",
Dan