/dev/loop

[Patrick Dupre]

Hello,

When I pluck a USB Hard Drive, it mounts as /dev/loop0 and /dev/loop3

df does not show the disk.

What is wrong?

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================

[Patrick Dupre]

Sorry,

fdisk does not show the disk (df shows loopx)

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 12:35 AM
> From: "Patrick Dupre" 
> To: fedora 
> Subject: /dev/loop
>
> Hello,
>
> When I pluck a USB Hard Drive, it mounts as /dev/loop0 and /dev/loop3
>
> df does not show the disk.
>
> What is wrong?
>
> ===========================================================================
>  Patrick DUPRÉ                                 | | email: [e-mail]
>  Laboratoire de Physico-Chimie de l'Atmosphère | |
>  Université du Littoral-Côte d'Opale           | |
>  Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
>  189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
> ===========================================================================
> _______________________________________________
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Rick Stevens]

On 07/13/2017 03:38 PM, Patrick Dupre wrote:
> Sorry,
>
> fdisk does not show the disk (df shows loopx)

Huh? If "df" shows something on a loop device, then it's mounted
somewhere or df wouldn't know how to extract the size data. For example,
with an ISO image mounted via loop to /mnt/Misc using the command

mount -t loop /path/to/iso/image.iso /mnt/Misc

I see:

[root@prophead ~]# df -h
    ...
    /dev/loop0               1.2G  1.2G     0 100% /mnt/Misc

It shows up in fdisk using an "fdisk -l":

[root@prophead ~]# fdisk -l
    ...
    Disk /dev/loop0: 1.1 GiB, 1222639616 bytes, 2387968 sectors
    Units: sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disklabel type: dos
    Disk identifier: 0x6b8b4567

Device       Boot  Start     End Sectors  Size Id Type
    /dev/loop0p1 *         0 2387967 2387968  1.1G  0 Empty
    /dev/loop0p2      105336  118339   13004  6.4M ef EFI (FAT-12/16/32)
    /dev/loop0p3      118340  146419   28080 13.7M  0 Empty

If you mount external media via the GUI or if it's automounted, it
should mount in

/run/media/your-user-name/blah

with "blah" being either the media's filesystem label or some string
that identifies the USB device. If you don't see somewhat similar
things, then check dmesg and your logs for errors and/or hints.

And you know better than to top-post on this list, right? You've been
a member long enough.
> > Sent: Friday, July 14, 2017 at 12:35 AM
> > From: "Patrick Dupre" 
> > To: fedora 
> > Subject: /dev/loop
> >
> > Hello,
> >
> > When I pluck a USB Hard Drive, it mounts as /dev/loop0 and /dev/loop3
> >
> > df does not show the disk.
> >
> > What is wrong?

----------------------------------------------------------------------
- Rick Stevens, Systems Engineer, AllDigital    [e-mail] -
- AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
-                                                                    -
-    Overweight:  When you step on your dog's tail...and it dies.    -
----------------------------------------------------------------------

[Patrick Dupre]

Hello,

I guess that it is an issue with the partition table.
I have 3 disks in teh same machine.
Yesterday I turned on the machine and turn it on this morning.
I could not boot. Then I decided to mount them in an
external USB enclosure. They all behave the same.

testdisk
Partition sector doesn't have the endmark 0xAA55

Disk /dev/sdc: 232.9 GiB, 250059350016 bytes, 488397168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes

I never though that I could lost 3 disks at the same time.
I have a copy of the partition table on the disks.

I am trying to run testdisk

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 1:06 AM
> From: "Rick Stevens" 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> On 07/13/2017 03:38 PM, Patrick Dupre wrote:
> > Sorry,
> >
> > fdisk does not show the disk (df shows loopx)
>
> Huh? If "df" shows something on a loop device, then it's mounted
> somewhere or df wouldn't know how to extract the size data. For example,
> with an ISO image mounted via loop to /mnt/Misc using the command
>
> mount -t loop /path/to/iso/image.iso /mnt/Misc
>
> I see:
>
> [root@prophead ~]# df -h
>    ...
>    /dev/loop0               1.2G  1.2G     0 100% /mnt/Misc
>
> It shows up in fdisk using an "fdisk -l":
>
> [root@prophead ~]# fdisk -l
>    ...
>    Disk /dev/loop0: 1.1 GiB, 1222639616 bytes, 2387968 sectors
>    Units: sectors of 1 * 512 = 512 bytes
>    Sector size (logical/physical): 512 bytes / 512 bytes
>    I/O size (minimum/optimal): 512 bytes / 512 bytes
>    Disklabel type: dos
>    Disk identifier: 0x6b8b4567
>
> Device       Boot  Start     End Sectors  Size Id Type
>    /dev/loop0p1 *         0 2387967 2387968  1.1G  0 Empty
>    /dev/loop0p2      105336  118339   13004  6.4M ef EFI (FAT-12/16/32)
>    /dev/loop0p3      118340  146419   28080 13.7M  0 Empty
>
> If you mount external media via the GUI or if it's automounted, it
> should mount in
>
> /run/media/your-user-name/blah
>
> with "blah" being either the media's filesystem label or some string
> that identifies the USB device. If you don't see somewhat similar
> things, then check dmesg and your logs for errors and/or hints.
>
> And you know better than to top-post on this list, right? You've been
> a member long enough.
> > > Sent: Friday, July 14, 2017 at 12:35 AM
> > > From: "Patrick Dupre" 
> > > To: fedora 
> > > Subject: /dev/loop
> > >
> > > Hello,
> > >
> > > When I pluck a USB Hard Drive, it mounts as /dev/loop0 and /dev/loop3
> > >
> > > df does not show the disk.
> > >
> > > What is wrong?
> Rick Stevens, Systems Engineer, AllDigital    [e-mail] -
> AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
>                                                                -
> Overweight:  When you step on your dog's tail...and it dies.    -
>
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Rick Stevens]

On 07/13/2017 04:29 PM, Patrick Dupre wrote:
> Hello,
>
> I guess that it is an issue with the partition table.
> I have 3 disks in teh same machine.
> Yesterday I turned on the machine and turn it on this morning.
> I could not boot. Then I decided to mount them in an
> external USB enclosure. They all behave the same

Ok, I'm a bit confused here. Do you mean that you install one drive at a
time into a USB enclosure to test?
> testdisk
> Partition sector doesn't have the endmark 0xAA55
>
> Disk /dev/sdc: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
>
> I never though that I could lost 3 disks at the same time.
> I have a copy of the partition table on the disks.
>
> I am trying to run testdisk

How? Via a rescue CD or something? I'm not familiar with testdisk but I
see it's available on a bunch of rescue CD images.
----------------------------------------------------------------------
- Rick Stevens, Systems Engineer, AllDigital    [e-mail] -
- AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
-                                                                    -
-         Okay, who put a "stop payment" on my reality check?        -
----------------------------------------------------------------------

[Patrick Dupre]

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> > I guess that it is an issue with the partition table.
> > I have 3 disks in teh same machine.
> > Yesterday I turned on the machine and turn it on this morning.
> > I could not boot. Then I decided to mount them in an
> > external USB enclosure. They all behave the same
>
> Ok, I'm a bit confused here. Do you mean that you install one drive at a
> time into a USB enclosure to test?

Yes, on another machine

I tried to recover the partition table for 1 disk.
I worked, but there are plenty of issues.

Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x00000000

Device     Boot     Start       End   Sectors   Size Id Type
/dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
/dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
/dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
/dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
/dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
/dev/sdd6       317001728 405065727  88064000    42G 83 Linux

As you can see this table is wrong

But cannot mount /dev/sdd6 for example.
Disk /dev/sdd6 - 45 GB / 41 GiB - CHS 5481 255 63

The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
Check the harddisk size: HD jumpers settings, BIOS detection...

The following partition can't be recovered:
     Partition               Start        End    Size in sectors
> Linux                 1889  40  8  7370 226 24   88064000 [Backup]
> > testdisk
> > Partition sector doesn't have the endmark 0xAA55
> >
> > Disk /dev/sdc: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> > Units: sectors of 1 * 512 = 512 bytes
> > Sector size (logical/physical): 512 bytes / 512 bytes
> > I/O size (minimum/optimal): 512 bytes / 512 bytes
> >
> > I never though that I could lost 3 disks at the same time.
> > I have a copy of the partition table on the disks.
> >
> > I am trying to run testdisk
>
> How? Via a rescue CD or something? I'm not familiar with testdisk but I
> see it's available on a bunch of rescue CD images.
> Rick Stevens, Systems Engineer, AllDigital    [e-mail] -
> AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
>                                                                -
>     Okay, who put a "stop payment" on my reality check?        -
>
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Rick Stevens]

On 07/13/2017 05:11 PM, Patrick Dupre wrote:
> ===========================================================================
>  Patrick DUPRÉ                                 | | email: [e-mail]
>  Laboratoire de Physico-Chimie de l'Atmosphère | |
>  Université du Littoral-Côte d'Opale           | |
>  Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
>  189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
> ===========================================================================
> > > I guess that it is an issue with the partition table.
> > > I have 3 disks in teh same machine.
> > > Yesterday I turned on the machine and turn it on this morning.
> > > I could not boot. Then I decided to mount them in an
> > > external USB enclosure. They all behave the same
> >
> > Ok, I'm a bit confused here. Do you mean that you install one drive at a
> > time into a USB enclosure to test?
>
> Yes, on another machine
>
> I tried to recover the partition table for 1 disk.
> I worked, but there are plenty of issues.
>
> Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> Disklabel type: dos
> Disk identifier: 0x00000000
>
> Device     Boot     Start       End   Sectors   Size Id Type
> /dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
> /dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
> /dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
> /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> /dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
> /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
>
> As you can see this table is wrong
>
> But cannot mount /dev/sdd6 for example.
> Disk /dev/sdd6 - 45 GB / 41 GiB - CHS 5481 255 63

Well, yes. A DOS disk (not-GPT) can only have four primary partitions.
If you're going to have more than four, then at least one must be
an extended partition (typically the fourth partition or /dev/sdd4) and
any partitions above 4 live INSIDE partition 4. In the above example,
/dev/sdd4 should be type 5 (EXTENDED), then /dev/sdd5 and /dev/sdd6
would live inside /dev/sdd4.

In some ways, that makes sense in that /dev/sdd4 is the biggest
partition and could hold sdd5 and sdd6, but I have no idea what this
stuff looked like before. Anything I say would be a wild guess. Also the
mixture of straight Linux (type 83) and LVM (type 8e) partitions is
worrying. Again, without knowing how things were laid out before makes
any advice I offer dangerous.
> The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
> Check the harddisk size: HD jumpers settings, BIOS detection...
>
> The following partition can't be recovered:
>      Partition               Start        End    Size in sectors
> > Linux                 1889  40  8  7370 226 24   88064000 [Backup]
> > > testdisk
> > > Partition sector doesn't have the endmark 0xAA55
> > >
> > > Disk /dev/sdc: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> > > Units: sectors of 1 * 512 = 512 bytes
> > > Sector size (logical/physical): 512 bytes / 512 bytes
> > > I/O size (minimum/optimal): 512 bytes / 512 bytes
> > >
> > > I never though that I could lost 3 disks at the same time.
> > > I have a copy of the partition table on the disks.
> > >
> > > I am trying to run testdisk
> >
> > How? Via a rescue CD or something? I'm not familiar with testdisk but I
> > see it's available on a bunch of rescue CD images.

----------------------------------------------------------------------
- Rick Stevens, Systems Engineer, AllDigital    [e-mail] -
- AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
-                                                                    -
-  You know the old saying--any technology sufficiently advanced is  -
-               indistinguishable from a Perl script                 -
-                                 --Programming Perl, 2nd Edition    -
----------------------------------------------------------------------

[Samuel Sieb]

On 07/13/2017 05:29 PM, Rick Stevens wrote:
> > /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> Well, yes. A DOS disk (not-GPT) can only have four primary partitions.
> If you're going to have more than four, then at least one must be
> an extended partition (typically the fourth partition or /dev/sdd4) and
> any partitions above 4 live INSIDE partition 4. In the above example,
> /dev/sdd4 should be type 5 (EXTENDED), then /dev/sdd5 and /dev/sdd6
> would live inside /dev/sdd4.

I think those partition types are equivalent, but that's definitely the 
least of the problems there.

[Samuel Sieb]

On 07/13/2017 05:11 PM, Patrick Dupre wrote:
> I tried to recover the partition table for 1 disk.
> I worked, but there are plenty of issues.
>
> Device     Boot     Start       End   Sectors   Size Id Type
> /dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
> /dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
> /dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
> /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> /dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
> /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
>
> As you can see this table is wrong

You don't say why you think it's wrong.  I do see that there are some 
strange gaps.   Particularly the first partition is nowhere near the 
start of the disk.
> The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
> Check the harddisk size: HD jumpers settings, BIOS detection...
> Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors

Where are you getting that number?  Fdisk says that the harddisk size is 
233 GB, or 250GB in marketing numbers.

[Patrick Dupre]

Yes, I did not partition the disk with gaps.

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 2:59 AM
> From: "Samuel Sieb" 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> On 07/13/2017 05:11 PM, Patrick Dupre wrote:
> > I tried to recover the partition table for 1 disk.
> > I worked, but there are plenty of issues.
> >
> > Device     Boot     Start       End   Sectors   Size Id Type
> > /dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
> > /dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
> > /dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
> > /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> > /dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
> > /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
> >
> > As you can see this table is wrong
>
> You don't say why you think it's wrong.  I do see that there are some 
> strange gaps.   Particularly the first partition is nowhere near the 
> start of the disk.
> > The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
> > Check the harddisk size: HD jumpers settings, BIOS detection...
> > Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors
>
> Where are you getting that number?  Fdisk says that the harddisk size is 
> 233 GB, or 250GB in marketing numbers.

testdisk says
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Patrick Dupre]

With another disk, I have,

Disk /dev/sdc: 465.8 GiB, 500107862016 bytes, 976773168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x00000000

Device     Boot     Start       End   Sectors   Size Id Type
/dev/sdc1  *     27265024  39847935  12582912     6G  7 HPFS/NTFS/exFAT
/dev/sdc2        39847936  50333695  10485760     5G  7 HPFS/NTFS/exFAT
/dev/sdc3        50335744  62918655  12582912     6G 82 Linux swap / Solaris
/dev/sdc4        62918656 976773119 913854464 435.8G  f W95 Ext'd (LBA)
/dev/sdc5        62920704  63944703   1024000   500M 83 Linux
/dev/sdc6       310751232 399998975  89247744  42.6G 83 Linux

or

Disk /dev/sdc - 500 GB / 465 GiB - CHS 60801 255 63
Current partition structure:
     Partition                  Start        End    Size in sectors

1 * HPFS - NTFS           1697  43 11  2480 106 58   12582912
 2 P HPFS - NTFS           2480 106 59  3133  32 35   10485760
 3 P Linux Swap            3133  65  5  3916 128 52   12582912
 4 E extended LBA          3916 128 53 60801  80 15  913854464
 5 L Linux                 3916 161 22  3980  95 19    1024000
   X extended             19343  93  1 24898 200  6   89247822
 6 L Linux                19343  94 16 24898 200  6   89247744 [backup]

Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63
Current partition structure:
     Partition                  Start        End    Size in sectors

Partition sector doesn't have the endmark 0xAA55

The most important partition is the sdc6 where I have my backups
Actually, only a few numbers of files (but big).
There are gaps again, lvm partitions.

sdc5 is a /boot, I can mount it, but there are few information.

Any tools that you could suggest?

testdisk on /dev/sd6 gives:
Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63

The harddisk (45 GB / 42 GiB) seems too small! (< 65 GB / 60 GiB)
Check the harddisk size: HD jumpers settings, BIOS detection...

The following partitions can't be recovered:
     Partition               Start        End    Size in sectors
> Linux                 1409 145 17  6964 251  7   89247744 [backup]

Linux                 2404 112 13  7959 218  3   89247744 [backup]

It looks like that the LVM partitions are the most of issues.

I really do not understand how 3 HD can fail simultaneously.

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 3:19 AM
> From: "Patrick Dupre" 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> Yes, I did not partition the disk with gaps.
>
> ===========================================================================
>  Patrick DUPRÉ                                 | | email: [e-mail]
>  Laboratoire de Physico-Chimie de l'Atmosphère | |
>  Université du Littoral-Côte d'Opale           | |
>  Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
>  189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
> ===========================================================================
> > Sent: Friday, July 14, 2017 at 2:59 AM
> > From: "Samuel Sieb" 
> > To: [e-mail]
> > Subject: Re: /dev/loop
> >
> > On 07/13/2017 05:11 PM, Patrick Dupre wrote:
> > > I tried to recover the partition table for 1 disk.
> > > I worked, but there are plenty of issues.
> > >
> > > Device     Boot     Start       End   Sectors   Size Id Type
> > > /dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
> > > /dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
> > > /dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
> > > /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> > > /dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
> > > /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
> > >
> > > As you can see this table is wrong
> >
> > You don't say why you think it's wrong.  I do see that there are some 
> > strange gaps.   Particularly the first partition is nowhere near the 
> > start of the disk.
> > > The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
> > > Check the harddisk size: HD jumpers settings, BIOS detection...
> > > Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> >
> > Where are you getting that number?  Fdisk says that the harddisk size is 
> > 233 GB, or 250GB in marketing numbers.
>
> testdisk says
> > users mailing list -- [e-mail]
> > To unsubscribe send an email to [e-mail]
>
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Samuel Sieb]

On 07/13/2017 06:47 PM, Patrick Dupre wrote:
> testdisk on /dev/sd6 gives:
> Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63
>
> The harddisk (45 GB / 42 GiB) seems too small! (< 65 GB / 60 GiB)
> Check the harddisk size: HD jumpers settings, BIOS detection...

Oh, now I see what you're doing.  You need to run testdisk on the whole 
disk, not a partition.  So "testdisk /dev/sdc".

[Patrick Dupre]

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> On 07/13/2017 06:47 PM, Patrick Dupre wrote:
> > testdisk on /dev/sd6 gives:
> > Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63
> >
> > The harddisk (45 GB / 42 GiB) seems too small! (< 65 GB / 60 GiB)
> > Check the harddisk size: HD jumpers settings, BIOS detection...
>
> Oh, now I see what you're doing.  You need to run testdisk on the whole 
> disk, not a partition.  So "testdisk /dev/sdc".

This is what I did to try to recover the partition table.
Then I tried on one specific partition.

In my opinion, the disks are probably OK, I "just"  need to recover 
properly the tables.

The bad thing is that there is a backup of the tables on the
HD them because. I anticipate such an issue, but not on the 3 disks
at the same time.

Do you know any other tools to do it?

Thank

[Samuel Sieb]

On 07/13/2017 06:47 PM, Patrick Dupre wrote:
> It looks like that the LVM partitions are the most of issues.

Do you have any idea how it was partitioned originally?
> I really do not understand how 3 HD can fail simultaneously.

That seems very strange.  Do you have any more background on what happened?

[Patrick Dupre]

Hello,

Here is my current thinking.
2 days ago, I received a brand new UBS key (Lexar S75).
I inserted it inside the UBS port of my Linux machine.
I open the key (it mounted automatically) and read the instructions
about the encryption coming with the key (by opening a file).
I did not pay too much attention because I was not interested,
I just changed the size of the 
first partition and I added a ext4 partition.
Later I turned off the PC.
The following day, I tried to restart te computer, but I was
unable to boot.
Thus, I try to read the key from another computer and gparted did not
like (fdisk was giving wrong partitions). Thus, I had to entirely
repartition the stick.

Anyway, I started to investigate the issues with the PC which could
not boot. And I arrived at the point that the partition tables of the
3 HD were mess up.

Now, I am making the link.
Can this stick have a virus which may have mess up the partition tables.

Please, note that one of the HD probably still had a XP OS bootable.
I did not use it for a very long time.

What do you think?

My Best.
===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 3:47 AM
> From: "Patrick Dupre" 
> To: [e-mail]
> Cc: [e-mail]
> Subject: Re: /dev/loop
>
> With another disk, I have,
>
> Disk /dev/sdc: 465.8 GiB, 500107862016 bytes, 976773168 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> Disklabel type: dos
> Disk identifier: 0x00000000
>
> Device     Boot     Start       End   Sectors   Size Id Type
> /dev/sdc1  *     27265024  39847935  12582912     6G  7 HPFS/NTFS/exFAT
> /dev/sdc2        39847936  50333695  10485760     5G  7 HPFS/NTFS/exFAT
> /dev/sdc3        50335744  62918655  12582912     6G 82 Linux swap / Solaris
> /dev/sdc4        62918656 976773119 913854464 435.8G  f W95 Ext'd (LBA)
> /dev/sdc5        62920704  63944703   1024000   500M 83 Linux
> /dev/sdc6       310751232 399998975  89247744  42.6G 83 Linux
>
> or
>
> Disk /dev/sdc - 500 GB / 465 GiB - CHS 60801 255 63
> Current partition structure:
>      Partition                  Start        End    Size in sectors
>
> 1 * HPFS - NTFS           1697  43 11  2480 106 58   12582912
>  2 P HPFS - NTFS           2480 106 59  3133  32 35   10485760
>  3 P Linux Swap            3133  65  5  3916 128 52   12582912
>  4 E extended LBA          3916 128 53 60801  80 15  913854464
>  5 L Linux                 3916 161 22  3980  95 19    1024000
>    X extended             19343  93  1 24898 200  6   89247822
>  6 L Linux                19343  94 16 24898 200  6   89247744 [backup]
>
> Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63
> Current partition structure:
>      Partition                  Start        End    Size in sectors
>
> Partition sector doesn't have the endmark 0xAA55
>
> The most important partition is the sdc6 where I have my backups
> Actually, only a few numbers of files (but big).
> There are gaps again, lvm partitions.
>
> sdc5 is a /boot, I can mount it, but there are few information.
>
> Any tools that you could suggest?
>
> testdisk on /dev/sd6 gives:
> Disk /dev/sdc6 - 45 GB / 42 GiB - CHS 5555 255 63
>
> The harddisk (45 GB / 42 GiB) seems too small! (< 65 GB / 60 GiB)
> Check the harddisk size: HD jumpers settings, BIOS detection...
>
> The following partitions can't be recovered:
>      Partition               Start        End    Size in sectors
> > Linux                 1409 145 17  6964 251  7   89247744 [backup]
>
> Linux                 2404 112 13  7959 218  3   89247744 [backup]
>
> It looks like that the LVM partitions are the most of issues.
>
> I really do not understand how 3 HD can fail simultaneously.
>
> ===========================================================================
>  Patrick DUPRÉ                                 | | email: [e-mail]
>  Laboratoire de Physico-Chimie de l'Atmosphère | |
>  Université du Littoral-Côte d'Opale           | |
>  Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
>  189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
> ===========================================================================
> > Sent: Friday, July 14, 2017 at 3:19 AM
> > From: "Patrick Dupre" 
> > To: [e-mail]
> > Subject: Re: /dev/loop
> >
> > Yes, I did not partition the disk with gaps.
> >
> > ===========================================================================
> >  Patrick DUPRÉ                                 | | email: [e-mail]
> >  Laboratoire de Physico-Chimie de l'Atmosphère | |
> >  Université du Littoral-Côte d'Opale           | |
> >  Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
> >  189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
> > ===========================================================================
> > > Sent: Friday, July 14, 2017 at 2:59 AM
> > > From: "Samuel Sieb" 
> > > To: [e-mail]
> > > Subject: Re: /dev/loop
> > >
> > > On 07/13/2017 05:11 PM, Patrick Dupre wrote:
> > > > I tried to recover the partition table for 1 disk.
> > > > I worked, but there are plenty of issues.
> > > >
> > > > Device     Boot     Start       End   Sectors   Size Id Type
> > > > /dev/sdd1  *     24981075  36258704  11277630   5.4G  7 HPFS/NTFS/exFAT
> > > > /dev/sdd2        36258705  46508174  10249470   4.9G  7 HPFS/NTFS/exFAT
> > > > /dev/sdd3        60030976 206895103 146864128    70G 8e Linux LVM
> > > > /dev/sdd4       206895104 488396799 281501696 134.2G  f W95 Ext'd (LBA)
> > > > /dev/sdd5       206899200 290785279  83886080    40G 8e Linux LVM
> > > > /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
> > > >
> > > > As you can see this table is wrong
> > >
> > > You don't say why you think it's wrong.  I do see that there are some 
> > > strange gaps.   Particularly the first partition is nowhere near the 
> > > start of the disk.
> > > > The harddisk (45 GB / 41 GiB) seems too small! (< 60 GB / 56 GiB)
> > > > Check the harddisk size: HD jumpers settings, BIOS detection...
> > > > Disk /dev/sdd: 232.9 GiB, 250059350016 bytes, 488397168 sectors
> > >
> > > Where are you getting that number?  Fdisk says that the harddisk size is 
> > > 233 GB, or 250GB in marketing numbers.
> >
> > testdisk says
> > > users mailing list -- [e-mail]
> > > To unsubscribe send an email to [e-mail]
> >
> > users mailing list -- [e-mail]
> > To unsubscribe send an email to [e-mail]
>
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[stan]

On Fri, 14 Jul 2017 13:22:50 +0200
"Patrick Dupre"  wrote:
> 2 days ago, I received a brand new UBS key (Lexar S75).
> I inserted it inside the UBS port of my Linux machine.
> I open the key (it mounted automatically) and read the instructions
> about the encryption coming with the key (by opening a file).
> I did not pay too much attention because I was not interested,

This can be a recipe for disaster when working with computers.
> I just changed the size of the 
> first partition and I added a ext4 partition.

Is it possible you made a mistake and actually changed a different
partition table?  The partition table for one of your disks?
> Later I turned off the PC.
> The following day, I tried to restart te computer, but I was
> unable to boot.
> Thus, I try to read the key from another computer and gparted did not
> like (fdisk was giving wrong partitions). Thus, I had to entirely
> repartition the stick.

Again, a good indication that you didn't actually partition the
USB stick when you thought you did.
> Anyway, I started to investigate the issues with the PC which could
> not boot. And I arrived at the point that the partition tables of the
> 3 HD were mess up.
>
> Now, I am making the link.
> Can this stick have a virus which may have mess up the partition
> tables.

I don't think so.
> Please, note that one of the HD probably still had a XP OS bootable.
> I did not use it for a very long time.
>
> What do you think?

Unless you ran a program from the USB stick, the fact XP was on the
disk would have no effect.  And it would have to be a linux program
that modified the partition tables, so it would have to run as root.
Opening a file would not do this.  Hmm, if you used an editor with a
script interpreter in it, and the file had a script for that editor,
and the script was set to run on open, and you were root, and Jupiter
aligned with Mars, and ...

Far more likely you partitioned the wrong disk.

Or, does that PC have power issues?  Or could one of the disks be
failing?

[Patrick Dupre]

Thank,

However, how could I have destroy the partition table of 3 disks
using gparted only?

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 4:49 PM
> From: stan 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> On Fri, 14 Jul 2017 13:22:50 +0200
> "Patrick Dupre"  wrote:
> > 2 days ago, I received a brand new UBS key (Lexar S75).
> > I inserted it inside the UBS port of my Linux machine.
> > I open the key (it mounted automatically) and read the instructions
> > about the encryption coming with the key (by opening a file).
> > I did not pay too much attention because I was not interested,
>
> This can be a recipe for disaster when working with computers.
> > I just changed the size of the 
> > first partition and I added a ext4 partition.
>
> Is it possible you made a mistake and actually changed a different
> partition table?  The partition table for one of your disks?
> > Later I turned off the PC.
> > The following day, I tried to restart te computer, but I was
> > unable to boot.
> > Thus, I try to read the key from another computer and gparted did not
> > like (fdisk was giving wrong partitions). Thus, I had to entirely
> > repartition the stick.
>
> Again, a good indication that you didn't actually partition the
> USB stick when you thought you did.
> > Anyway, I started to investigate the issues with the PC which could
> > not boot. And I arrived at the point that the partition tables of the
> > 3 HD were mess up.
> >
> > Now, I am making the link.
> > Can this stick have a virus which may have mess up the partition
> > tables.
>
> I don't think so.
> > Please, note that one of the HD probably still had a XP OS bootable.
> > I did not use it for a very long time.
> >
> > What do you think?
>
> Unless you ran a program from the USB stick, the fact XP was on the
> disk would have no effect.  And it would have to be a linux program
> that modified the partition tables, so it would have to run as root.
> Opening a file would not do this.  Hmm, if you used an editor with a
> script interpreter in it, and the file had a script for that editor,
> and the script was set to run on open, and you were root, and Jupiter
> aligned with Mars, and ...
>
> Far more likely you partitioned the wrong disk.
>
> Or, does that PC have power issues?  Or could one of the disks be
> failing?
> _______________________________________________
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[stan]

On Fri, 14 Jul 2017 17:22:13 +0200
"Patrick Dupre"  wrote:
> However, how could I have destroy the partition table of 3 disks
> using gparted only?

It is certainly possible to destroy the partition table of 3 disks
with gparted.  I can't say how you would have done it, or even if you
did.  I only say that it seems unlikely it was a virus from the USB
stick that was responsible for the destruction.

[Patrick Dupre]

OK,

Now, what would you do with this
When I try to mount this partition:
/dev/sdd6       317001728 405065727  88064000    42G 83 Linux
(>L Linux                19732 113 30 25214  44 46   88064000 [Backup])
The size is OK
But:
mount /dev/sdd6 /mnt/tmp
mount: wrong fs type, bad option, bad superblock on /dev/sdd6,
       missing codepage or helper program, or other error

[27212.278322] EXT4-fs (sdd6): error loading journal
[28134.857250] JBD2: no valid journal superblock found
[28134.857254] EXT4-fs (sdd6): error loading journal
[29533.339281] JBD2: no valid journal superblock found
[29533.339286] EXT4-fs (sdd6): error loading journal

Should I run fsck?

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 5:47 PM
> From: stan 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> On Fri, 14 Jul 2017 17:22:13 +0200
> "Patrick Dupre"  wrote:
> > However, how could I have destroy the partition table of 3 disks
> > using gparted only?
>
> It is certainly possible to destroy the partition table of 3 disks
> with gparted.  I can't say how you would have done it, or even if you
> did.  I only say that it seems unlikely it was a virus from the USB
> stick that was responsible for the destruction.
> _______________________________________________
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[stan]

On Fri, 14 Jul 2017 18:01:16 +0200
"Patrick Dupre"  wrote:
> Now, what would you do with this
> When I try to mount this partition:
> /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
> (>L Linux                19732 113 30 25214  44 46   88064000
> [Backup]) The size is OK
> But:
> mount /dev/sdd6 /mnt/tmp
> mount: wrong fs type, bad option, bad superblock on /dev/sdd6,
>        missing codepage or helper program, or other error
>
> [27212.278322] EXT4-fs (sdd6): error loading journal
> [28134.857250] JBD2: no valid journal superblock found
> [28134.857254] EXT4-fs (sdd6): error loading journal
> [29533.339281] JBD2: no valid journal superblock found
> [29533.339286] EXT4-fs (sdd6): error loading journal
>
> Should I run fsck?

Yes.  The first run I would use

e2fsck -c -v /dev/sdd6

which is verbose and read only, so it should be faster.  It doesn't
make any changes to the disk.  The disk is already unmounted, which is
what you want when doing these diagnostics and repairs.

Once you have that output, it will tell you what kind of shape the
filesystem is in, and you can decide your next step.

You will probably need to use the -b option to point
to a valid superblock, and the -p option to fix any errors, at some
point.  Depending on how much time you have, you could also run
e2fsck -c -c -v /dev/sdd6
before that, which does a non-destructive read-write test.  This is
slower, but will tell you if the disk is bad; there will be lots of
write failures if it is bad.

Slow and deliberate is the way to go, measuring 3 times before you cut
once.

man e2fsck for more information.

If you install smartmontools, you can use smartctl to get data from the
disks themselves, if they have it.
man smartctl

[Samuel Sieb]

On 07/14/2017 10:16 AM, stan wrote:
> e2fsck -c -v /dev/sdd6
>
> which is verbose and read only, so it should be faster.  It doesn't
> make any changes to the disk.  The disk is already unmounted, which is
> what you want when doing these diagnostics and repairs.

Did you get the wrong parameter?  -c is for checking for bad blocks, -n 
is to do a readonly check.

[stan]

On Fri, 14 Jul 2017 11:17:09 -0700
Samuel Sieb  wrote:
> Did you get the wrong parameter?  -c is for checking for bad blocks,
> -n is to do a readonly check.

No, I jumped ahead and suspected that there is something wrong with the
disks.  If all Patrick did is what he said he did, like you, I don't
see how he could have affected all three disks with gparted.  So, I'm
thinking disk malfunction of some kind.  The -c will check if there are
a bunch of bad blocks on the disk (blocks that can't be read), an
indication of a failing drive.  It is read only.

You're right, though, that specifying -n as well would be safer.

Something just doesn't add up here, this is way too much damage for a
simple single command error.

[Samuel Sieb]

On 07/14/2017 09:01 AM, Patrick Dupre wrote:
> When I try to mount this partition:
> /dev/sdd6       317001728 405065727  88064000    42G 83 Linux
> (>L Linux                19732 113 30 25214  44 46   88064000 [Backup])
> The size is OK
> But:
> mount /dev/sdd6 /mnt/tmp
> mount: wrong fs type, bad option, bad superblock on /dev/sdd6,
>         missing codepage or helper program, or other error
>
> [27212.278322] EXT4-fs (sdd6): error loading journal
> [28134.857250] JBD2: no valid journal superblock found
> [28134.857254] EXT4-fs (sdd6): error loading journal
> [29533.339281] JBD2: no valid journal superblock found
> [29533.339286] EXT4-fs (sdd6): error loading journal
>
> Should I run fsck?

I doubt that would be helpful at all since there are strong indications 
that your partition table is not correct.  And if you let it make 
changes, you will just be corrupting things even more.

[Samuel Sieb]

On 07/14/2017 04:22 AM, Patrick Dupre wrote:
> I just changed the size of the
> first partition and I added a ext4 partition.

Did you unmount the drive before doing these changes?  gparted won't let 
you modify a partition that is mounted.  You say you changed the size of 
the first partition, was there more than 1?
> Thus, I try to read the key from another computer and gparted did not
> like (fdisk was giving wrong partitions). Thus, I had to entirely
> repartition the stick.

Were you using gparted or fdisk or both?  So the USB drive had a mangled 
partition table as well?  That would be 4 drives messed up then.
> Anyway, I started to investigate the issues with the PC which could
> not boot. And I arrived at the point that the partition tables of the
> 3 HD were mess up.

I could kind of understand one disk getting messed up if you picked the 
wrong disk, but I have no idea how all of them would be.
> Now, I am making the link.
> Can this stick have a virus which may have mess up the partition tables.

Extremely unlikely since you didn't actually run anything from it.
> Please, note that one of the HD probably still had a XP OS bootable.
> I did not use it for a very long time.

That should not be relevant other than you should see a FAT32 partition 
in the table.

This has been a very long email thread, so I might have dropped some 
info, but are you using a live image right now for investigating this? 
Have you tried looking at these hard drives using a different computer?

I would recommend that you try a program called "gpart" to try to find 
the partitions.  If the first try doesn't come up with a working layout, 
you can add the "-f" parameter to make it do a full scan of the disk 
instead of skipping when it thinks it found a valid filesystem.

[Patrick Dupre]

Hello,

Thank,

THis is the result of gpart (no -f)

My problem is to guess where was the extended partition.
In addition there was a swap partition is sdd5

Begin scan...
Possible partition(Windows NT/W2K FS), size(13mb), offset(6696mb)
Possible extended partition at offset(10106mb)
   Possible partition(Windows NT/W2K FS), size(12197mb), offset(12197mb)
   Possible partition(Linux ext2), size(500mb), offset(28811mb)
   Possible partition(Linux LVM2 physical volume), size(71711mb), offset(29312mb)
   Possible partition(Linux LVM2 physical volume), size(40960mb), offset(101025mb)
   Possible partition(Linux ext2), size(43000mb), offset(154786mb)
End scan.

Checking partitions...

* Warning: more than 4 primary partitions: 6.
Partition(OS/2 HPFS, NTFS, QNX or Advanced UNIX): primary 
Partition(OS/2 HPFS, NTFS, QNX or Advanced UNIX): primary 
Partition(Linux ext2 filesystem): primary 
Partition(Linux LVM physical volume): primary 
Partition(Linux LVM physical volume): invalid primary 
Partition(Linux ext2 filesystem): invalid primary 
Ok.

Guessed primary partition table:
Primary partition(1)
   type: 007(0x07)(OS/2 HPFS, NTFS, QNX or Advanced UNIX)
   size: 13mb #s(28261) s(13714360-13742620)
   chs:  (853/173/17)-(855/111/53)d (853/173/17)-(855/111/53)r

Primary partition(2)
   type: 007(0x07)(OS/2 HPFS, NTFS, QNX or Advanced UNIX)
   size: 12197mb #s(24981011) s(24981074-49962084)
   chs:  (1023/254/63)-(1023/254/63)d (1554/254/63)-(3109/253/61)r

Primary partition(3)
   type: 131(0x83)(Linux ext2 filesystem)
   size: 500mb #s(1024000) s(59004928-60028927)
   chs:  (1023/254/63)-(1023/254/63)d (3672/226/11)-(3736/160/8)r

Primary partition(4)
   type: 142(0x8E)(Linux LVM physical volume)
   size: 71711mb #s(146864128) s(60030976-206895103)
   chs:  (1023/254/63)-(1023/254/63)d (3736/192/41)-(12878/159/17)r

* HPFS - NTFS           1555   0  1  2256 254 63   11277630
 P HPFS - NTFS           2257   0  1  2894 254 63   10249470
 D Linux Swap            2895   1  1  3672 254 63   12498507
> D Linux                 3672 226 11  3736 160  8    1024000 [Boot1]

P Linux LVM             3736 192 41 12878 159 17  146864128
 L Linux LVM            12878 224 19 18100 139 23   83886080
 L Linux                19732 113 30 25214  44 46   88064000 [Backup]

I cannot make the partition table coherent.

===========================================================================
 Patrick DUPRÉ                                 | | email: [e-mail]
 Laboratoire de Physico-Chimie de l'Atmosphère | |
 Université du Littoral-Côte d'Opale           | |
 Tel.  (33)-(0)3 28 23 76 12                   | | Fax: [phone]
 189A, avenue Maurice Schumann                 | | 59140 Dunkerque, France
===========================================================================
> Sent: Friday, July 14, 2017 at 8:33 PM
> From: "Samuel Sieb" 
> To: [e-mail]
> Subject: Re: /dev/loop
>
> On 07/14/2017 04:22 AM, Patrick Dupre wrote:
> > I just changed the size of the
> > first partition and I added a ext4 partition.
>
> Did you unmount the drive before doing these changes?  gparted won't let 
> you modify a partition that is mounted.  You say you changed the size of 
> the first partition, was there more than 1?
> > Thus, I try to read the key from another computer and gparted did not
> > like (fdisk was giving wrong partitions). Thus, I had to entirely
> > repartition the stick.
>
> Were you using gparted or fdisk or both?  So the USB drive had a mangled 
> partition table as well?  That would be 4 drives messed up then.
> > Anyway, I started to investigate the issues with the PC which could
> > not boot. And I arrived at the point that the partition tables of the
> > 3 HD were mess up.
>
> I could kind of understand one disk getting messed up if you picked the 
> wrong disk, but I have no idea how all of them would be.
> > Now, I am making the link.
> > Can this stick have a virus which may have mess up the partition tables.
>
> Extremely unlikely since you didn't actually run anything from it.
> > Please, note that one of the HD probably still had a XP OS bootable.
> > I did not use it for a very long time.
>
> That should not be relevant other than you should see a FAT32 partition 
> in the table.
>
> This has been a very long email thread, so I might have dropped some 
> info, but are you using a live image right now for investigating this? 
> Have you tried looking at these hard drives using a different computer?
>
> I would recommend that you try a program called "gpart" to try to find 
> the partitions.  If the first try doesn't come up with a working layout, 
> you can add the "-f" parameter to make it do a full scan of the disk 
> instead of skipping when it thinks it found a valid filesystem.
> _______________________________________________
> users mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]