Mailing List Archive

On 05/06/2012 12:29 PM, Frank Peters wrote:
>
> Internet searches only provide comments from Ubuntu or Fedora users
> stating that "I plugged it in and it worked," which are next to useless.
> I need to know what modules (drivers) and basic commands can get a Western
> Digital USB external HDD to successfully mount.
>
> Frank Peters
>
I don't have one of those drives at the moment, but on my system I
typically have usb-common, usb-core, and usb-storage. Only thing I use
sg for is burning CD's/DVD's. You didn't mention this specifically, so
for what it's worth I access such devices as /dev/sd* just like I would
any other internal drive.

--
Stan & HD Tashi Grad 10/08 Edgewood, NM SWR
PR - Cindy and Jenny - Sammamish, WA NWR
http://www.cci.org

On Sun, 06 May 2012 13:18:23 -0600
Stan Sander <stsander@sblan.net> wrote:

> I don't have one of those drives at the moment, but on my system I
> typically have usb-common, usb-core, and usb-storage. Only thing I use
> sg for is burning CD's/DVD's. You didn't mention this specifically, so
> for what it's worth I access such devices as /dev/sd* just like I would
> any other internal drive.
>

Since my keyboard and mouse are both USB devices, I have the basic USB modules
for built into the kernel, and those would include, I believe, usb-core
and usb-common.

For other USB devices, such as printers and mass storage, I need to load
some more modules before using them. I would assume that a Western Digital
USB HDD is just another mass storage device, like a thumb drive, that would
call for sg and usb-storage, but before I spend more money on acquiring one
I need to know exactly how to set things up.

Yes, a thumb USB drive, and presumable all USB mass storage devices, are
recognized as SCSI drives and accessed via /dev/sdX.

Some might ask why not just utilize udev or similar to automount the device.
For me, Linux means choice and control and I would rather understand my own
system even if it means a little less convenience. For that reason I avoid udev.

Frank Peters

On Sun, May 6, 2012 at 1:29 PM, Frank Peters <frank.peters@comcast.net> wrote:
>
> My USB thumb drives can all be successfully recognized with the sg and
> usb-storage modules (as shown in the commands above).  Is this all I would
> need to mount a Western Digital USB HDD or is there some other module that
> needs to be loaded?

Yes, that should be all you need. Virtually every USB hard drive or
USB IDE/SATA adapter work as regular mass storage devices. Nothing
special required. Same as a flash drive or generic memory card reader.

/Some/ USB hard drives are "intelligent", having software
pre-installed on them, a "one touch" backup button, weird immutable
partitions that show up as CDROM drives, etc. but the overwhelming
majority are just a plain USB mass storage device. If you can find one
that lacks special features like buttons or anything, it's probably
more likely to work without hassle.

Hi,

Am 06.05.2012 20:29, schrieb Frank Peters:
> Hello,
>
> I recently acquired a Seagate USB external HDD but could not get it
> to function on my Gentoo Linux.

Back in the olden days to most common mistake with custom build kernels
was not having SCSI disk support compiled in, but now that basically
everyone uses SATA that should not be what you're missing. Note that the
SCSI disk driver is not the sg driver. sg works for everything attached
to SCSI (a scanner, for example) and doesn't have the required
functionality. What you also need is sd (which creates the accordingly
named /dev/sd? device).

If it still doesn't work if you'd post your dmesg output so we can check
what is missing/unusual.

With kind regards,
Thomas

On Sun, 6 May 2012 14:29:20 -0400
Frank Peters wrote:

... <snip> ...

> Using a Gentoo Live DVD to boot Linux, the drive was recognized
> without any read errors. In this case fdisk could recognize the
> drive. I then attempted to reformat the drive with an ext2 file
> system but the format failed.
... <snip> ...

With the LiveCD running, try the following commands

"lsusb -v" to list information on the usb devices found by
Gentoo

"lsmod" to list the modules loaded.

HTH,

David

On Mon, 07 May 2012 12:51:58 +0200
Thomas Rösner <thomas.roesner@digital-trauma.de> wrote:

>
> What you also need is sd (which creates the accordingly
> named /dev/sd? device).
>

Since I use SATA exclusively, sd is built in to the kernel.

As Paul Hartman has indicated, all that should be necessary
are the sg and usb-storage modules.

The Seagate drive has already been returned to the seller.
I am expecting a Western Digital Elements 2TB USB HDD to
arrive in a few days. If this new drive also fails then I
will post more information.

Frank Peters

On Mon, May 7, 2012 at 6:34 AM, Frank Peters <frank.peters@comcast.net> wrote:
> On Mon, 07 May 2012 12:51:58 +0200
> Thomas Rösner <thomas.roesner@digital-trauma.de> wrote:
>
>>
>> What you also need is sd (which creates the accordingly
>> named /dev/sd? device).
>>
>
> Since I use SATA exclusively, sd is built in to the kernel.
>
> As Paul Hartman has indicated, all that should be necessary
> are the sg and usb-storage modules.
>
> The Seagate drive has already been returned to the seller.
> I am expecting a Western Digital Elements 2TB USB HDD to
> arrive in a few days.  If this new drive also fails then I
> will post more information.
>
> Frank Peters
>
>

Hi Frank,
Sorry I'm sort of late to the conversation. I don't have anything
to add in terms of what to build in the kernel source tree. I will
offer however that in my experience not all USB ports are created
equal. I've got about 8 USB drives here that I've collected over the
years and a couple of them do not work in all USB ports on all my
machines so there is history of that sort of problem here. Nominally I
can see some problems in dmesg, such as a new USB device is added but
it's not recognized as a disk, etc., so look there.

Additionally, and if you know already this please excuse my
teaching side coming out, but not all USB ports in most modern
machines come from the same semiconductor chip. There are (Intel/AMD)
chipset-based USB ports and then most machine use 1 or 2 additional
USB controllers to get you up to 6-8 USB ports. There are 3 or now I
think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
3.0 is using. This means multiple USB drivers might need to be loaded
to get all chips working.

Best thing to do in my experience is try the drive in every USB
port you have available to see whether one of them does the job.

Good luck,
Mark

On Mon, May 7, 2012 at 10:00 AM, Mark Knecht <markknecht@gmail.com> wrote:
> here are 3 or now I
> think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
> 3.0 is using

XHCI

Furthermore, USB 3.0 has 9-pin ports and cables (for type-A) versus
the 4-pin of USB 1/2. The USB 3.0 sockets are backward-compatible with
older USB devices and cables, but older USB cables are not
forward-compatible with USB 3.0.

On Mon, May 7, 2012 at 8:21 AM, Paul Hartman
<paul.hartman+gentoo@gmail.com> wrote:
> On Mon, May 7, 2012 at 10:00 AM, Mark Knecht <markknecht@gmail.com> wrote:
>> here are 3 or now I
>> think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
>> 3.0 is using
>
> XHCI
>
> Furthermore, USB 3.0 has 9-pin ports and cables (for type-A) versus
> the 4-pin of USB 1/2. The USB 3.0 sockets are backward-compatible with
> older USB devices and cables, but older USB cables are not
> forward-compatible with USB 3.0.
>

Good point. None of my USB 2.0 drives work in my USB 3.0 port. I do
have a USB 3.0 drivers, supposedly, and one drive that claims to be
eSata/USB 3.0 compatible, but I've not tried to get it working using
USB 3.0.

- Mark

Seems this either a typo, or a contradiction:
"The USB 3.0 sockets are backward-compatible with older ... cables"
would mean I can use the old cables for 3.0, but then
"but older USB cables are not forward-compatible with USB 3.0"
says I can NOT use the old cables.
I have no 3.0, but some day I probably will, so I'd like to know what
exactly you mean.

kind regards

michael

--
Michael Scherer
Univ.klinik f. Psychiatrie
email: michael.scherer@meduniwien.ac.at
phone: +43 6991 941 22 54

----- Original Message -----
From: "Mark Knecht" <markknecht@gmail.com>
To: <gentoo-amd64@lists.gentoo.org>
Sent: Monday, 07 May, 2012 17:43
Subject: Re: [gentoo-amd64] Drivers For USB HDD


> On Mon, May 7, 2012 at 8:21 AM, Paul Hartman
> <paul.hartman+gentoo@gmail.com> wrote:
>> On Mon, May 7, 2012 at 10:00 AM, Mark Knecht <markknecht@gmail.com> wrote:
>>> here are 3 or now I
>>> think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
>>> 3.0 is using
>>
>> XHCI
>>
>> Furthermore, USB 3.0 has 9-pin ports and cables (for type-A) versus
>> the 4-pin of USB 1/2. The USB 3.0 sockets are backward-compatible with
>> older USB devices and cables, but older USB cables are not
>> forward-compatible with USB 3.0.
>>
>
> Good point. None of my USB 2.0 drives work in my USB 3.0 port. I do
> have a USB 3.0 drivers, supposedly, and one drive that claims to be
> eSata/USB 3.0 compatible, but I've not tried to get it working using
> USB 3.0.
>
> - Mark
>
>

On Mon, May 7, 2012 at 2:18 PM, Michael Scherer
<a6702894@unet.univie.ac.at> wrote:
> Seems this either a typo, or a contradiction:
> "The USB 3.0 sockets are backward-compatible with older ... cables"
> would mean I can use the old cables for 3.0, but then
>
> "but older USB cables are not forward-compatible with USB 3.0"
> says I can NOT use the old cables.
> I have no 3.0, but some day I probably will, so I'd like to know what
> exactly you mean.
>
> kind regards
>
> michael
>
> --
> Michael Scherer
> Univ.klinik f. Psychiatrie
> email: michael.scherer@meduniwien.ac.at
> phone: +43 6991 941 22 54
>

I believe what Paul was saying was that _IF_ a USB 3.0 Controller is
designed to support both 3.0 AND 2.0 devices then the socket is
compatible with both 2.0 & 3.0 cables. However if you use a 2.0 cable
you have to use a 2.0 device. You cannot, TTBOMK, use a 2.0 cable with
a 3.0 device.

At least that's how I read his response.

HTH,
Mark

On Mon, May 7, 2012 at 4:18 PM, Michael Scherer
<a6702894@unet.univie.ac.at> wrote:
>> On Mon, May 7, 2012 at 8:21 AM, Paul Hartman
>> <paul.hartman+gentoo@gmail.com> wrote:
>>>
>>> On Mon, May 7, 2012 at 10:00 AM, Mark Knecht <markknecht@gmail.com>
>>> wrote:
>>>>
>>>> here are 3 or now I
>>>> think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
>>>> 3.0 is using
>>>
>>>
>>> XHCI
>>>
>>> Furthermore, USB 3.0 has 9-pin ports and cables (for type-A) versus
>>> the 4-pin of USB 1/2. The USB 3.0 sockets are backward-compatible with
>>> older USB devices and cables, but older USB cables are not
>>> forward-compatible with USB 3.0.
>>>
>>
>> Good point. None of my USB 2.0 drives work in my USB 3.0 port. I do
>> have a USB 3.0 drivers, supposedly, and one drive that claims to be
>> eSata/USB 3.0 compatible, but I've not tried to get it working using
>> USB 3.0.
>>
>> - Mark
>>
>>
>

> Seems this either a typo, or a contradiction:
> "The USB 3.0 sockets are backward-compatible with older ... cables"
> would mean I can use the old cables for 3.0, but then
>
> "but older USB cables are not forward-compatible with USB 3.0"
> says I can NOT use the old cables.
> I have no 3.0, but some day I probably will, so I'd like to know what
> exactly you mean.
>
> kind regards
>
> michael

Hi,

Here is what I mean:

USB 3.0 socket is backwards compatible = USB 2.0 cable and devices
still work as USB 2.0 when plugged into USB 3.0 port.

USB 2.0 cable is not forward-compatible = you can not use USB 2.0
cable with USB 3.0 device as USB 3.0. It will only behave as USB 2.0
in that case.

If you want to truly use USB 3.0 you need to have all three: USB 3.0
port, USB 3.0 cable, USB 3.0 device.

meaning the controller then needs different outlets for 3.0 and 2.0 cables,
so I need to use 3.0 cable on the 3.0 one, but 2.0 cable on the other.
sounds reasonable.

thanks

michael

--
Michael Scherer
Univ.klinik f. Psychiatrie
email: michael.scherer@meduniwien.ac.at
phone: +43 6991 941 22 54

----- Original Message -----
From: "Paul Hartman" <paul.hartman+gentoo@gmail.com>
To: <gentoo-amd64@lists.gentoo.org>
Sent: Monday, 07 May, 2012 23:45
Subject: Re: [gentoo-amd64] Drivers For USB HDD


> On Mon, May 7, 2012 at 4:18 PM, Michael Scherer
> <a6702894@unet.univie.ac.at> wrote:
>>> On Mon, May 7, 2012 at 8:21 AM, Paul Hartman
>>> <paul.hartman+gentoo@gmail.com> wrote:
>>>>
>>>> On Mon, May 7, 2012 at 10:00 AM, Mark Knecht <markknecht@gmail.com>
>>>> wrote:
>>>>>
>>>>> here are 3 or now I
>>>>> think 4 different USB specs - EHCI, OHCI, UHCI, and then whatever USB
>>>>> 3.0 is using
>>>>
>>>>
>>>> XHCI
>>>>
>>>> Furthermore, USB 3.0 has 9-pin ports and cables (for type-A) versus
>>>> the 4-pin of USB 1/2. The USB 3.0 sockets are backward-compatible with
>>>> older USB devices and cables, but older USB cables are not
>>>> forward-compatible with USB 3.0.
>>>>
>>>
>>> Good point. None of my USB 2.0 drives work in my USB 3.0 port. I do
>>> have a USB 3.0 drivers, supposedly, and one drive that claims to be
>>> eSata/USB 3.0 compatible, but I've not tried to get it working using
>>> USB 3.0.
>>>
>>> - Mark
>>>
>>>
>>
>
>> Seems this either a typo, or a contradiction:
>> "The USB 3.0 sockets are backward-compatible with older ... cables"
>> would mean I can use the old cables for 3.0, but then
>>
>> "but older USB cables are not forward-compatible with USB 3.0"
>> says I can NOT use the old cables.
>> I have no 3.0, but some day I probably will, so I'd like to know what
>> exactly you mean.
>>
>> kind regards
>>
>> michael
>
> Hi,
>
> Here is what I mean:
>
> USB 3.0 socket is backwards compatible = USB 2.0 cable and devices
> still work as USB 2.0 when plugged into USB 3.0 port.
>
> USB 2.0 cable is not forward-compatible = you can not use USB 2.0
> cable with USB 3.0 device as USB 3.0. It will only behave as USB 2.0
> in that case.
>
> If you want to truly use USB 3.0 you need to have all three: USB 3.0
> port, USB 3.0 cable, USB 3.0 device.
>
>

Am 07.05.2012 23:58, schrieb Michael Scherer:
> meaning the controller then needs different outlets for 3.0 and 2.0
> cables,
> so I need to use 3.0 cable on the 3.0 one, but 2.0 cable on the other.

Not quite :). A 3.0 USB client (like a 3.0 HDD) will usually come with a
new 3.0 USB cable with additional contacts. The client head is different
than with a USB 2.0 cable. The master head (that plugs into the
controller, your computer or hub) is backwards compatible - if you plug
it into an USB 2.0 port, the additional contacts won't connect and the
HDD will speak USB 2.0. The same goes for a 2.0 client in a 3.0 port.

So, controller side, you can plug everything into everything. Only
client side you have a new outlet (which is good because it prevents you
from using an old 2.0 cable and falling back to slow speed if both your
PC and client speak USB 3.0)

Regards,
Thomas

On Mon, 7 May 2012 09:34:58 -0400
Frank Peters <frank.peters@comcast.net> wrote:

>
> The Seagate drive has already been returned to the seller.
> I am expecting a Western Digital Elements 2TB USB HDD to
> arrive in a few days. If this new drive also fails then I
> will post more information.
>

I just want to follow up on my previous posts for the sake of
anyone who may be interested.

The Western Digital Elements 2TB USB HDD arrived and it
works with Linux right out of the box. The extra modules
needed are sg and usb-storage, assuming that all the usb
core stuff is built into the kernel.

The drive is pre-formatted with a ntfs filesystem, but using
cfdisk I re-partitioned the drive and formatted the new partitions
with ext3. Everything works nicely.

This WD drive is USB 2.0, whereas the non-functional Seagate was
USB 3.0. Whether or not this can explain the problem I cannot
say.

Thanks again to all who responded.

Frank Peters

Den 2012-05-10 04:49, Frank Peters skrev:

> The drive is pre-formatted with a ntfs filesystem, but using cfdisk I
> re-partitioned the drive and formatted the new partitions with ext3.
> Everything works nicely.

why not ext4 ?, imho ext4 is more caple of so big drives, specially on
fsck, my own qnap ts 419 p+ supports both so it must be good :=)

Indeed, EXT4 is the new standard IMOO.

2012/5/10 Benny Pedersen <me@junc.org>

> Den 2012-05-10 04:49, Frank Peters skrev:
>
> The drive is pre-formatted with a ntfs filesystem, but using cfdisk I
>> re-partitioned the drive and formatted the new partitions with ext3.
>> Everything works nicely.
>>
>
> why not ext4 ?, imho ext4 is more caple of so big drives, specially on
> fsck, my own qnap ts 419 p+ supports both so it must be good :=)
>
>
>
>
>


--
Salut
alp
Sylvain

On Thu, 10 May 2012 15:07:09 +0200
Benny Pedersen <me@junc.org> wrote:

>
> why not ext4 ?, imho ext4 is more caple of so big drives, specially on
> fsck, my own qnap ts 419 p+ supports both so it must be good :=)
>

My plan for the drive is long-term and portable storage of digital files.
The idea is basically write once, read many. Also, the drive has been
formatted into several smaller partitions. For this kind of use I don't
need the capabilities of ext4. In fact I don't even need a journal and
could probably use ext2 just as well.

Frank Peters

It's not a good idea to not do journalling for an external drive. No matter
how careful you are, it is just a matter of when you will trip up your USB
cable in the middle of a write; and then you should just pray that fsck can
save your drive without the journal.

Also, it probably won't hurt using ext4, given that ext4 had several
performance improvements over ext3. However, as Theodore T'so - the primary
developer of ext4 - said, ext4 is a stopgap until btrfs is ready. IMO, from
my experience of using btrfs some time ago, I'd say that btrfs is pretty
much ready nowadays.

Lie Ryan, mused, then expounded:
> It's not a good idea to not do journalling for an external drive. No matter
> how careful you are, it is just a matter of when you will trip up your USB
> cable in the middle of a write; and then you should just pray that fsck can
> save your drive without the journal.
>

If there is concern about that, then the two filesystems to use would be
XFS and EXT3.

But, don't take my work for it, simply set up a bootable image with each
filesystem, create a bootable USB stick, boot it, then pull it without
unmounting it. Even pull power. With current >= 2.6.32 kernels those
will survive better than the others.

Lie Ryan posted on Fri, 11 May 2012 02:31:27 +1000 as excerpted:

> Also, it probably won't hurt using ext4, given that ext4 had several
> performance improvements over ext3. However, as Theodore T'so - the
> primary developer of ext4 - said, ext4 is a stopgap until btrfs is
> ready. IMO, from my experience of using btrfs some time ago, I'd say
> that btrfs is pretty much ready nowadays.

As long as you don't have problems or want to do anything fancy like
multi-disk, btrfs can be fine. But it's still in active development and
officially experimental, it only recently (Feb/Mar) got an error
correcting btrfsck at all, and that still comes with "it may make the
problem worse instead of fixing it" warnings. I'm running a few
partitions of it now, but as I tell people on the btrfs list, while a
good admin will always have backups no matter the stability of the
filesystem, with something as experimental and under development as
btrfs, it's best to consider your btrfs copy an extra "testing" copy,
that may or may not be there the next time you access it. Your primary
copy, along with all backups you'd ordinarily have, should still exist
and be located on something other than btrfs.

We routinely see people on the list asking how to recover data, because
they didn't heed that advice. Sometimes it's recoverable, sometimes part
of it is, sometimes not.

Just watching the commits and related discussion on either the btrfs
lists or as they hit the mainline kernel, they're still actively fixing
code broken in one way or another, as well as continuing to add
features. raid5/6 mode is roadmapped for 3.5 (some preliminary prep
commits went into 3.4), and full n-way mirroring raid1, the current so-
called raid1 mode is only two-way-mirroring) is roadmapped after that as
it builds on it.

But with a few more kernels, say by the end of the year or early next,
btrfs should really begin to stabilize.

Meanwhile, anyone who does choose to run it should be keeping up with the
latest kernels. If you're not running the rc kernels at least by rc5 or
so, you're running old code with known problems patched in newer kernels.

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

On May 10, 2012 7:34 PM, "Lie Ryan" <lie.1296@gmail.com> wrote:
>
> It's not a good idea to not do journalling for an external drive. No
matter how careful you are, it is just a matter of when you will trip up
your USB cable in the middle of a write; and then you should just pray that
fsck can save your drive without the journal.
>
> Also, it probably won't hurt using ext4, given that ext4 had several
performance improvements over ext3. However, as Theodore T'so - the primary
developer of ext4 - said, ext4 is a stopgap until btrfs is ready. IMO, from
my experience of using btrfs some time ago, I'd say that btrfs is pretty
much ready nowadays.

for a system drive maybe, but I wouldn't trust my data on it yet. let it
mature a bit first :)

Alex | wired

On Fri, May 11, 2012 at 4:33 AM, Alex Alexander
<alex.alexander@gmail.com> wrote:
>
> for a system drive maybe, but I wouldn't trust my data on it yet. let it
> mature a bit first :)
>

For something like Gentoo that uses rolling releases, a system drive
might be one of btrfs's better use cases anyway. All those instant
snapshots can be very handy when managing upgrades/etc. I'm not sure
I'd even go so far as to use it for that purpose yet, however, unless
I didn't mind occasional downtime and had the ability to rapidly
rebuild the server. If you view all those snapsnots as a great way to
manage all the extra backups you'll be doing, then maybe it might work
out.

Rich