Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.hardware > #515 > unrolled thread

new WD passport filesystem?

Back to article view | Back to comp.os.linux.hardware

Contents

  new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-22 11:31 -0400
    Re: new WD passport filesystem? Michael Black <et472@ncf.ca> - 2011-07-22 11:37 -0400
    Re: new WD passport filesystem? Richard Kettlewell <rjk@greenend.org.uk> - 2011-07-22 16:46 +0100
      Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-22 12:28 -0400
        Re: new WD passport filesystem? "Trevor Hemsley" <Trevor.Hemsley@mytrousers.ntlworld.com> - 2011-07-22 17:43 -0500
          Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-22 21:27 -0400
    Re: new WD passport filesystem? J G Miller <miller@yoyo.ORG> - 2011-07-22 15:50 +0000
      Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-22 12:37 -0400
        Re: new WD passport filesystem? J G Miller <miller@yoyo.ORG> - 2011-07-22 17:36 +0000
        Re: new WD passport filesystem? Bill Marcum <bill@lat.localnet> - 2011-07-22 14:03 -0400
        Re: new WD passport filesystem? Richard Kettlewell <rjk@greenend.org.uk> - 2011-07-22 19:42 +0100
          Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-22 21:56 -0400
            Re: new WD passport filesystem? felmon <nemo@nowhere.INVALID> - 2011-07-22 22:50 -0500
              Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-23 05:01 -0400
                Re: new WD passport filesystem? felmon <nemo@nowhere.INVALID> - 2011-07-23 13:32 -0500
                  Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-24 08:28 -0400
                  Re: new WD passport filesystem? scott@alfter.diespammersdie.us (Scott Alfter) - 2011-07-26 16:08 +0000
                    Re: new WD passport filesystem? Haines Brown <haines@HistoricalMaterialism.info> - 2011-07-27 06:26 -0400
                      Re: new WD passport filesystem? felmon <nemo@nowhere.INVALID> - 2011-07-28 18:46 -0500

#515 — new WD passport filesystem?

I've been using various WD Passport external USB drives over the years
with luck and so recently purchased a 1 Tb USB 3.0 drive (I understand
that the debian Squeeze kernel supports USB 3). I'm having trouble
mounting it. I had presumed it came formatted as VFAT16, but:

 $ sudo fdisk -l /dev/sdb
 Disk /dev/sdb: 1000.2 GB, 1000170586112 bytes
 255 heads, 63 sectors/track, 121597 cylinders
 Units = cylinders of 16065 * 512 = 8225280 bytes
 Sector size (logical/physical): 512 bytes / 512 bytes
 I/O size (minimum/optimal): 512 bytes / 512 bytes
 Disk identifier: 0x00042ada

  Device Boot      Start         End      Blocks   Id  System
 /dev/sdb1               1    121598   976728064    7  HPFS/NTFS

Formatted HPFS (OS/2?)/NTFS? So I try:

  # mount -t hpfs /dev/sdb1 /media/backup
  mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
               ...
 
  $ ls /dev | grep sdb1
  sdb1

  $ dmesg
  ...
  ... sd 50:0:0:0: [sdb] 1953458176 512-byte logical 
                   blocks: (1.00 TB/931 GiB)
  ... sd 50:0:0:0: [sdb] Write Protect is off
  ... sd 50:0:0:0: [sdb] Mode Sense: 47 00 10 08
  ...
  ... sdb: sdb1
  ... sd 50:0:0:0: [sdb] Assuming drive cache: write through
  ... sd 50:0:0:0: [sdb] Attached SCSI disk
  ... FAT: utf8 is not a recommended IO charset for FAT filesystems, 
           filesystem will be case sensitive!
  ... FAT: bogus number of reserved sectors
  ... VFS: Can't find a valid FAT filesystem on dev sdb1.
  ... FAT: utf8 is not a recommended IO charset...
  ... FAT: bogus number of reserved sectors
  ... VFS: Can't find a valid FAT filesystem on dev sdb1.
  ... HPFS: Bad magic ... probably not HPFS
  ... HPFS: Bad magic ... probably not HPFS

So is it HPFS or not? If it is, why the FAT error messages? If it is
not, why did fdisk tell a lie? Has hpfs/ntfs become the default 
file system on devices these days?

Haines Brown

[toc] | [next] | [standalone]

#516

On Fri, 22 Jul 2011, Haines Brown wrote:

> I've been using various WD Passport external USB drives over the years
> with luck and so recently purchased a 1 Tb USB 3.0 drive (I understand
> that the debian Squeeze kernel supports USB 3). I'm having trouble
> mounting it. I had presumed it came formatted as VFAT16, but:
>
Try mount -t auto ....

That may not work if it's a real odd format, but if it can be mounted, 
then simply running mount by itself will reveal what the drive has been 
mounted as.

   Michael

  > $ sudo fdisk -l /dev/sdb
> Disk /dev/sdb: 1000.2 GB, 1000170586112 bytes
> 255 heads, 63 sectors/track, 121597 cylinders
> Units = cylinders of 16065 * 512 = 8225280 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> Disk identifier: 0x00042ada
>
>  Device Boot      Start         End      Blocks   Id  System
> /dev/sdb1               1    121598   976728064    7  HPFS/NTFS
>
> Formatted HPFS (OS/2?)/NTFS? So I try:
>
>  # mount -t hpfs /dev/sdb1 /media/backup
>  mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
>               ...
>
>  $ ls /dev | grep sdb1
>  sdb1
>
>  $ dmesg
>  ...
>  ... sd 50:0:0:0: [sdb] 1953458176 512-byte logical
>                   blocks: (1.00 TB/931 GiB)
>  ... sd 50:0:0:0: [sdb] Write Protect is off
>  ... sd 50:0:0:0: [sdb] Mode Sense: 47 00 10 08
>  ...
>  ... sdb: sdb1
>  ... sd 50:0:0:0: [sdb] Assuming drive cache: write through
>  ... sd 50:0:0:0: [sdb] Attached SCSI disk
>  ... FAT: utf8 is not a recommended IO charset for FAT filesystems,
>           filesystem will be case sensitive!
>  ... FAT: bogus number of reserved sectors
>  ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>  ... FAT: utf8 is not a recommended IO charset...
>  ... FAT: bogus number of reserved sectors
>  ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>  ... HPFS: Bad magic ... probably not HPFS
>  ... HPFS: Bad magic ... probably not HPFS
>
> So is it HPFS or not? If it is, why the FAT error messages? If it is
> not, why did fdisk tell a lie? Has hpfs/ntfs become the default
> file system on devices these days?
>
> Haines Brown
>

[toc] | [prev] | [next] | [standalone]

#517

Haines Brown <haines@HistoricalMaterialism.info> writes:

> I've been using various WD Passport external USB drives over the years
> with luck and so recently purchased a 1 Tb USB 3.0 drive (I understand
> that the debian Squeeze kernel supports USB 3). I'm having trouble
> mounting it. I had presumed it came formatted as VFAT16, but:
>
>  $ sudo fdisk -l /dev/sdb
>  Disk /dev/sdb: 1000.2 GB, 1000170586112 bytes
>  255 heads, 63 sectors/track, 121597 cylinders
>  Units = cylinders of 16065 * 512 = 8225280 bytes
>  Sector size (logical/physical): 512 bytes / 512 bytes
>  I/O size (minimum/optimal): 512 bytes / 512 bytes
>  Disk identifier: 0x00042ada
>
>   Device Boot      Start         End      Blocks   Id  System
>  /dev/sdb1               1    121598   976728064    7  HPFS/NTFS
>
> Formatted HPFS (OS/2?)/NTFS? So I try:
>
>   # mount -t hpfs /dev/sdb1 /media/backup
>   mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
>                ...
>  
>   $ ls /dev | grep sdb1
>   sdb1
>
>   $ dmesg
>   ...
>   ... sd 50:0:0:0: [sdb] 1953458176 512-byte logical 
>                    blocks: (1.00 TB/931 GiB)
>   ... sd 50:0:0:0: [sdb] Write Protect is off
>   ... sd 50:0:0:0: [sdb] Mode Sense: 47 00 10 08
>   ...
>   ... sdb: sdb1
>   ... sd 50:0:0:0: [sdb] Assuming drive cache: write through
>   ... sd 50:0:0:0: [sdb] Attached SCSI disk
>   ... FAT: utf8 is not a recommended IO charset for FAT filesystems, 
>            filesystem will be case sensitive!
>   ... FAT: bogus number of reserved sectors
>   ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>   ... FAT: utf8 is not a recommended IO charset...
>   ... FAT: bogus number of reserved sectors
>   ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>   ... HPFS: Bad magic ... probably not HPFS
>   ... HPFS: Bad magic ... probably not HPFS
>
> So is it HPFS or not? If it is, why the FAT error messages? If it is
> not, why did fdisk tell a lie? Has hpfs/ntfs become the default 
> file system on devices these days?

It is probably NTFS.  fdisk only reports the partition identifier, which
is the same for HPFS and NTFS; it does not tell you what the actual
formatting is.  What you find on a newly bought disk is up the vendor.
The FAT error messages indicate something tried to mount it with that
and failed.

-- 
http://www.greenend.org.uk/rjk/

[toc] | [prev] | [next] | [standalone]

#520

Richard Kettlewell <rjk@greenend.org.uk> writes:

> Haines Brown <haines@HistoricalMaterialism.info> writes:
>
>> I've been using various WD Passport external USB drives over the years
>> with luck and so recently purchased a 1 Tb USB 3.0 drive (I understand
>> that the debian Squeeze kernel supports USB 3). I'm having trouble
>> mounting it. I had presumed it came formatted as VFAT16, but:

>>   $ ls /dev | grep sdb1
>>   sdb1
>>
>>   $ dmesg
>>   ...
>>   ... sd 50:0:0:0: [sdb] 1953458176 512-byte logical 
>>                    blocks: (1.00 TB/931 GiB)
>>   ... sd 50:0:0:0: [sdb] Write Protect is off
>>   ... sd 50:0:0:0: [sdb] Mode Sense: 47 00 10 08
>>   ...
>>   ... sdb: sdb1
>>   ... sd 50:0:0:0: [sdb] Assuming drive cache: write through
>>   ... sd 50:0:0:0: [sdb] Attached SCSI disk
>>   ... FAT: utf8 is not a recommended IO charset for FAT filesystems, 
>>            filesystem will be case sensitive!
>>   ... FAT: bogus number of reserved sectors
>>   ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>>   ... FAT: utf8 is not a recommended IO charset...
>>   ... FAT: bogus number of reserved sectors
>>   ... VFS: Can't find a valid FAT filesystem on dev sdb1.
>>   ... HPFS: Bad magic ... probably not HPFS
>>   ... HPFS: Bad magic ... probably not HPFS
>>
>> So is it HPFS or not? If it is, why the FAT error messages? If it is
>> not, why did fdisk tell a lie? Has hpfs/ntfs become the default 
>> file system on devices these days?
>
> It is probably NTFS.  fdisk only reports the partition identifier, which
> is the same for HPFS and NTFS; it does not tell you what the actual
> formatting is.  What you find on a newly bought disk is up the vendor.
> The FAT error messages indicate something tried to mount it with that
> and failed.

Thanks for the clarification. But meanwhile I did manage to mount the
disk successfully with mount -t hpfs. For this new drive, WD decided to
make its file system HPFS/NTFS instead of VFAT as in the past, which
strikes me as a smart move, although perhaps it looses the
cross-platform advantage of FAT. I wonder if this is a trend.

Haines Brown

[toc] | [prev] | [next] | [standalone]

#525

On Fri, 22 Jul 2011 16:28:58 UTC in comp.os.linux.hardware, Haines Brown 
<haines@HistoricalMaterialism.info> wrote:

> meanwhile I did manage to mount the
> disk successfully with mount -t hpfs.

I would be astonished if this were true as HPFS is understood only by OS/2, 
Windows NT 3.1 and Linux. I cannot imagine WD restricting their market to that 
vanishingly small segment! I suspect you actually meant NTFS which is a 
filesystem used by more recent versions of Windows installed a few billion 
machines. While HPFS and NTFS share the same partition table id, there is not 
much else that they share.

-- 
Trevor Hemsley, Brighton, UK
Trevor dot Hemsley at ntlworld dot com

[toc] | [prev] | [next] | [standalone]

#527

"Trevor Hemsley" <Trevor.Hemsley@mytrousers.ntlworld.com> writes:

> On Fri, 22 Jul 2011 16:28:58 UTC in comp.os.linux.hardware, Haines Brown 
> <haines@HistoricalMaterialism.info> wrote:
>
>> meanwhile I did manage to mount the
>> disk successfully with mount -t hpfs.
>
> I would be astonished if this were true as HPFS is understood only by
> OS/2, Windows NT 3.1 and Linux. I cannot imagine WD restricting their
> market to that vanishingly small segment! I suspect you actually meant
> NTFS which is a filesystem used by more recent versions of Windows
> installed a few billion machines. While HPFS and NTFS share the same
> partition table id, there is not much else that they share.

No, I mounted the disk as hpfs and am currently doing a backup to
it. Apparently the WD external drive is NTFS rather than VFAT, but my
mount command that specified HPFS nevertheless somehow mounted it.

         # fsck /dev/sdb1
         fsck from util-linux-ng 2.17.2
         fsck: fsck.ntfs: not found
         fsck: Error 2 while executing fsck.ntfs for /dev/sdb1

This suggests the drive is definitely NTFS.

         # fdisk -l /dev/sdb
         Disk /dev/sdb: 1000.2 GB, 1000170586112 bytes
         255 heads, 63 sectors/track, 121597 cylinders
         Units = cylinders of 16065 * 512 = 8225280 bytes
         Sector size (logical/physical): 512 bytes / 512 bytes
         I/O size (minimum/optimal): 512 bytes / 512 bytes
         Disk identifier: 0x00042ada

         Device Boot  Start End      Blocks     Id  System
         /dev/sdb1    1     121598   976728064  7   HPFS/NTFS

My vague recollection (which has little to recommend it) is that
Microsoft "borrowed" HPFS from IBM.

I see this:  

         $ mount
         ...
         /dev/sda8 on /var type ext4 (rw)
         fusectl on /sys/fs/fuse/connections type fusectl (rw)
         /dev/sdb1 on /media/backup type fuseblk 
                   (rw,allow_other,blksize=4096)

I don't understand FUSE, but could it be salvaging my mount? In
otherwords, is it allowing type "fuseblk" to substitute for an unknown
HPFS type?

Haines Brown

[toc] | [prev] | [next] | [standalone]

#518

On Friday, July 22nd, 2011 at 11:31:31h -0400, Haines Brown explained:

>   Device Boot      Start         End      Blocks   Id  System
>  /dev/sdb1               1    121598   976728064    7  HPFS/NTFS

What does it stay for TYPE when you do

         blkid /dev/sdb1


Examples --

blkid /dev/sda1
/dev/sda1: LABEL="WINDOWS XP" UUID="54F07CDDF07CC730" TYPE="ntfs"

blkid /dev/sda5
/dev/sda5: LABEL="/" UUID="6083c507-9764-4164-8b39-962abb04bba4" TYPE="ext4"

blkid /dev/sde1
/dev/sde1: LABEL="WD_2TB-Disk0" UUID="0c64585d-a891-41f9-9e38-4ca935466a00" TYPE="xfs" 

[toc] | [prev] | [next] | [standalone]

#521

J G Miller <miller@yoyo.ORG> writes:

> On Friday, July 22nd, 2011 at 11:31:31h -0400, Haines Brown explained:
>
>>   Device Boot      Start         End      Blocks   Id  System
>>  /dev/sdb1               1    121598   976728064    7  HPFS/NTFS
>
> What does it stay for TYPE when you do
>
>          blkid /dev/sdb1

I had tried to use blkid, but without sucess. For my Debian, Squeeze,
there is no blkid package, but only libblkid1, which is installed. However,
I have no blkid or blkid1 command available to me. I guess mount and
fsck use that library, but I don't know of any user program that does
so.

Haines Brown

[toc] | [prev] | [next] | [standalone]

#522

On Friday, July 22nd, 2011 at 12:37:00h -0400, Haines Brown complained:

> For my Debian, Squeeze, there is no blkid package

You are correct that there is no blkid package.

The program /sbin/blkid is in the package util-linux.

  Description: Miscellaneous system utilities

  This package contains a number of important utilities, most of which
  are oriented towards maintenance of your system.  Some of the more
  important utilities included in this package allow you to partition
  your hard disk, view kernel messages, and create new filesystems.

> However, I have no blkid or blkid1 command available to me.

You will after doing 

       apt-get update && apt-get install util-linux

May I suggest you install the apt-file package so that you can
easily determine which package contains a particular file even
if the package is not installed.

It is particularly invaluable when compiling programs from
source and one needs to know which development package to
install because a header file is not present on the system.

[toc] | [prev] | [next] | [standalone]

#523

On 2011-07-22, Haines Brown <haines@HistoricalMaterialism.info> wrote:
> J G Miller <miller@yoyo.ORG> writes:
>
>> On Friday, July 22nd, 2011 at 11:31:31h -0400, Haines Brown explained:
>>
>>>   Device Boot      Start         End      Blocks   Id  System
>>>  /dev/sdb1               1    121598   976728064    7  HPFS/NTFS
>>
>> What does it stay for TYPE when you do
>>
>>          blkid /dev/sdb1
>
> I had tried to use blkid, but without sucess. For my Debian, Squeeze,
> there is no blkid package, but only libblkid1, which is installed. However,
> I have no blkid or blkid1 command available to me. I guess mount and
> fsck use that library, but I don't know of any user program that does
> so.
>
> Haines Brown

"dpkg -S blkid" finds /sbin/blkid in the util-linux package. I think dpkg only
searches installed packages. apt-file might search for a filename in all packages.



-- 
Inside every older person is a younger person wondering what the hell
happened.

[toc] | [prev] | [next] | [standalone]

#524

Haines Brown <haines@HistoricalMaterialism.info> writes:
> J G Miller <miller@yoyo.ORG> writes:

>> On Friday, July 22nd, 2011 at 11:31:31h -0400, Haines Brown explained:
>>
>>>   Device Boot      Start         End      Blocks   Id  System
>>>  /dev/sdb1               1    121598   976728064    7  HPFS/NTFS
>>
>> What does it stay for TYPE when you do
>>
>>          blkid /dev/sdb1
>
> I had tried to use blkid, but without sucess. For my Debian, Squeeze,
> there is no blkid package, but only libblkid1, which is installed. However,
> I have no blkid or blkid1 command available to me. I guess mount and
> fsck use that library, but I don't know of any user program that does
> so.

You need to type /sbin/blkid; evidentally JGM has /sbin on their path
but you do not.

-- 
http://www.greenend.org.uk/rjk/

[toc] | [prev] | [next] | [standalone]

#528

Richard Kettlewell <rjk@greenend.org.uk> writes:

> You need to type /sbin/blkid; evidentally JGM has /sbin on their path
> but you do not.

No, I foolishly tried to run it as user. When run as root, I was able
to extract the UUID for the drive to which I'm backing up, and it
verified that it is formatted as NTFS.

Haines Brown

[toc] | [prev] | [next] | [standalone]

#530

On Fri, 22 Jul 2011 21:56:51 -0400, Haines Brown wrote:

> Richard Kettlewell <rjk@greenend.org.uk> writes:
> 
>> You need to type /sbin/blkid; evidentally JGM has /sbin on their path
>> but you do not.
> 
> No, I foolishly tried to run it as user. When run as root, I was able to
> extract the UUID for the drive to which I'm backing up, and it verified
> that it is formatted as NTFS.
> 
> Haines Brown

I know you said you mounted it with a 'hpfs' flag but there is a driver 
for using ntfs; I mount my ntfs partition thus:

sudo mount -t ntfs-3g /dev/sda2 /Win7

I wonder if this wouldn't be safer.

used to use hpfs in my os/2 days. I believe it was IBM and then Microsoft 
worked on it and perhaps used it as a basis for ntfs (but my memory is 
not necessarily accurate).

Felmon

[toc] | [prev] | [next] | [standalone]

#532

felmon <nemo@nowhere.INVALID> writes:

> On Fri, 22 Jul 2011 21:56:51 -0400, Haines Brown wrote:
>
>> Richard Kettlewell <rjk@greenend.org.uk> writes:

> used to use hpfs in my os/2 days. I believe it was IBM and then Microsoft 
> worked on it and perhaps used it as a basis for ntfs (but my memory is 
> not necessarily accurate).

Also an OS/2 refugee.

> I know you said you mounted it with a 'hpfs' flag but there is a driver 
> for using ntfs; I mount my ntfs partition thus:
>
> sudo mount -t ntfs-3g /dev/sda2 /Win7

The mount command reports for my NTFS 1 Gb drive when using either your
suggested ntfs-3g driver as well as the ntfs driver as follows:

  /dev/sdb1 on /media/backup type fuseblk (rw,allow_other,blksize=4096)

Why is the type "fuseblk" rather than ntfs? 

I still don't understand if I still have a problem. Firsdt, when I plug
in my new 1 Tb Passport or the older drive, they automount and are
accessible. However, they don't dismount automatically. Shouldn't they?

For example, if I unplug a drive assigned to /dev/sdb1 and then plug in
another drive, mount says it is assigned to /dev/sdc1 and the old drive
still seen even though unplugged:

  $ mount
  ...
  fusectl on /sys/fs/fuse/connections type fusectl (rw)
  /dev/sdb1 on /media/mirror type ext3 (rw)
  /dev/sdc1 on /media/backup type fuseblk (rw,allow_other,blksize=4096)

However, blkip tells a different story:

  /dev/sda15: UUID="87e151ff-28b8-4c90-812c-13e3706262be" TYPE="ext4" 
  /dev/sdc1: LABEL="My Passport" UUID="8A2CF4F62CF4DE5F" TYPE="ntfs" 

Another thing I don't understand is that installation with udev created
these lines line in fstab:

  /dev/scd0       /media/cdrom0   udf,iso9660 user,noauto 0       0
  /dev/sda1       /media/usb0     auto    rw,user,noauto  0       0

However, sda1 is my HD:

  $ mount
  /dev/sda1 on /boot type ext4 (rw)

Shouldn't fstab look to the sr0 instead of the sda1 device file for
usb CDROM drive?

Haines

[toc] | [prev] | [next] | [standalone]

#533

On Sat, 23 Jul 2011 05:01:33 -0400, Haines Brown wrote:

> [...]
>>
>> sudo mount -t ntfs-3g /dev/sda2 /Win7
> 
> The mount command reports for my NTFS 1 Gb drive when using either your
> suggested ntfs-3g driver as well as the ntfs driver as follows:
> 
>   /dev/sdb1 on /media/backup type fuseblk (rw,allow_other,blksize=4096)
> 
> Why is the type "fuseblk" rather than ntfs?

I don't know the behind-the-scenes technical reason. fuseblk is a work-
around that allows mounting ntfs so I guess it is invoked with your -t 
ntfs switch also. 

I just used the '-t ntfs' and confirm you are right, it works now too 
(debian with 2.6.32-5-686). not sure when this happened but it's good to 
know!

> I still don't understand if I still have a problem. Firsdt, when I plug
> in my new 1 Tb Passport or the older drive, they automount and are
> accessible. However, they don't dismount automatically. Shouldn't they?
> 
> For example, if I unplug a drive assigned to /dev/sdb1 and then plug in
> another drive, mount says it is assigned to /dev/sdc1 and the old drive
> still seen even though unplugged:

I can't help since I don't auto-mount or auto-dismount. 

hmm, I do have Mint running on a laptop which automounts; I believe one 
is expected to manually do a 'safe' dismount though. 

I defer to someone who can help you with safely auto-dismounting and the 
other questions. I wasn't sure about the concerns about how the devices 
are identified, whether sr0, scd1 or sdb1. but do you really have /dev/
sda15?

about dismounting, I assume writes have to be sync-ed as they occur. I'd 
be scared to rely on automatic dismounting.

Felmon

> 
>   $ mount
>   ...
>   fusectl on /sys/fs/fuse/connections type fusectl (rw) /dev/sdb1 on
>   /media/mirror type ext3 (rw) /dev/sdc1 on /media/backup type fuseblk
>   (rw,allow_other,blksize=4096)
> 
> However, blkip tells a different story:
> 
>   /dev/sda15: UUID="87e151ff-28b8-4c90-812c-13e3706262be" TYPE="ext4"
>   /dev/sdc1: LABEL="My Passport" UUID="8A2CF4F62CF4DE5F" TYPE="ntfs"
> 
> Another thing I don't understand is that installation with udev created
> these lines line in fstab:
> 
>   /dev/scd0       /media/cdrom0   udf,iso9660 user,noauto 0       0
>   /dev/sda1       /media/usb0     auto    rw,user,noauto  0       0
> 
> However, sda1 is my HD:
> 
>   $ mount
>   /dev/sda1 on /boot type ext4 (rw)
> 
> Shouldn't fstab look to the sr0 instead of the sda1 device file for usb
> CDROM drive?
> 
> Haines

[toc] | [prev] | [next] | [standalone]

#534

felmon <nemo@nowhere.INVALID> writes:

> I just used the '-t ntfs' and confirm you are right, it works now too 
> (debian with 2.6.32-5-686). not sure when this happened but it's good to 
> know!

Thanks. This is very reassuring. 

And yes, I do have a /dev/sda15. Partly due to a policy started many
years ago of worrying about partition location on the disk and the
advantages of breaking out certain partitions. In part also because of
my work, for which it makes sense to have huge separate partititions for
/info and /storage.

> hmm, I do have Mint running on a laptop which automounts; I believe one 
> is expected to manually do a 'safe' dismount though. 
>
> I defer to someone who can help you with safely auto-dismounting and the 
> other questions. I wasn't sure about the concerns about how the devices 
> are identified, whether sr0, scd1 or sdb1. but do you really have /dev/
> sda15?
>
> about dismounting, I assume writes have to be sync-ed as they occur. I'd 
> be scared to rely on automatic dismounting.

I hope someone can chime in here about whether auto-dismount is supposed
to work and if it is too dangerous to use. But I have an impression that
auto-dismount is not supported.

Haines

[toc] | [prev] | [next] | [standalone]

#543

In article <pan.2011.07.23.18.32.32@nowhere.INVALID>,
felmon  <nemo@nowhere.INVALID> wrote:
>On Sat, 23 Jul 2011 05:01:33 -0400, Haines Brown wrote:
>>> sudo mount -t ntfs-3g /dev/sda2 /Win7
>> 
>> The mount command reports for my NTFS 1 Gb drive when using either your
>> suggested ntfs-3g driver as well as the ntfs driver as follows:
>> 
>>   /dev/sdb1 on /media/backup type fuseblk (rw,allow_other,blksize=4096)
>> 
>> Why is the type "fuseblk" rather than ntfs?
>
>I don't know the behind-the-scenes technical reason. fuseblk is a work-
>around that allows mounting ntfs so I guess it is invoked with your -t 
>ntfs switch also. 

NTFS on Linux is implemented in userspace.  (There's a kernelspace
implementation as well, but it's read-only for all practical purposes and
hasn't seen much development in a long time.)  FUSE is the API that allows
userspace implementation of filesystems; fuseblk is the kernel module that
sits between ntfs-3g (the userspace NTFS implementation) and the kernel.

  _/_
 / v \ Scott Alfter (remove the obvious to send mail)
(IIGS( http://alfter.us/            Top-posting!
 \_^_/                              >What's the most annoying thing on Usenet?


--- Posted via news://freenews.netfront.net/ - Complaints to news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#544

scott@alfter.diespammersdie.us (Scott Alfter) writes:

> In article <pan.2011.07.23.18.32.32@nowhere.INVALID>,
> felmon  <nemo@nowhere.INVALID> wrote:

> NTFS on Linux is implemented in userspace.  (There's a kernelspace
> implementation as well, but it's read-only for all practical purposes
> and hasn't seen much development in a long time.)  FUSE is the API
> that allows userspace implementation of filesystems; fuseblk is the
> kernel module that sits between ntfs-3g (the userspace NTFS
> implementation) and the kernel.

Thank you for the explanation.

Haines Brown

[toc] | [prev] | [next] | [standalone]

#545

On Wed, 27 Jul 2011 06:26:28 -0400, Haines Brown wrote:


> Thank you for the explanation.

my thanks also. very clear and very useful. F.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.os.linux.hardware

csiph-web