Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
40 LinuxPartitions to USB external HDD?
235 views
Skip to first unread message
Avoi...@gmail.com
Nov 8, 2012, 5:32:19 PM11/8/12
to
After a disasterous attempt to <assemble> a MoBo to copy my
IDEs to SATA, I've decided to temporarily use an external USB
hard-drive.
But, as you know, the docos are all written for Micro$loth;
and I don't know about this "HPFS/NTFS" stuff?
$ fdisk -l == ...
Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1c0fd702
Device Boot Start End Blocks Id System
/dev/sda1 1 60802 488384536 7 HPFS/NTFS
---------------
$ cfdisk /dev/sda1 ==FATAL ERROR:
Bad primary partition 0: Partition begins after end-of-disk
Press any key to exit cfdisk
-> fdisk /dev/sda1 ==
The number of cylinders for this disk is set to 60801.
There is nothing wrong with that, but this is larger than 1024,
and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO)
2) booting and partitioning software from other OSs
(e.g., DOS FDISK, OS/2 FDISK)
Command (m for help): p
Disk /dev/sda1: 500.1 GB, 500105764864 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x6e697373
This doesn't look like a partition table
Probably you selected the wrong device.
Device Boot Start End Blocks Id System
/dev/sda1p1 ? 120528 234814 918008208 4f QNX4.x 3rd part
Partition 1 does not end on cylinder boundary.
/dev/sda1p2 ? 119381 153271 272218546+ 73 Unknown
Partition 2 does not end on cylinder boundary.
/dev/sda1p3 ? 113202 147075 272087568 2b Unknown
Partition 3 does not end on cylinder boundary.
/dev/sda1p4 ? 177064 177067 27487 61 SpeedStor
Partition 4 does not end on cylinde r boundary.
Partition table entries are not in disk order
-----------------
$ lsusb ==
Bus 003 Device 013: ID 0480:a007 Toshiba America Info. Systems, Inc.
...
------------------
Previously I had BIG problems with my Slakware-13 and /dev/sd*
with inode-size incompatibilities which I'd never even thought about.
Also, I want to have more than 16 partitions [for certin reasons]
and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
even it I don't YET anticipate booting from this USB hard-drive.
So, should I just try to partition it like the old IDEs ?
== TIA.
PS. my confidence was shattered by an expensive discovery that you
can't just interchange MOBOs, CPUs and power-supplies, like in the
old days.
IDEs to SATA, I've decided to temporarily use an external USB
hard-drive.
But, as you know, the docos are all written for Micro$loth;
and I don't know about this "HPFS/NTFS" stuff?
$ fdisk -l == ...
Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1c0fd702
Device Boot Start End Blocks Id System
/dev/sda1 1 60802 488384536 7 HPFS/NTFS
---------------
$ cfdisk /dev/sda1 ==FATAL ERROR:
Bad primary partition 0: Partition begins after end-of-disk
Press any key to exit cfdisk
-> fdisk /dev/sda1 ==
The number of cylinders for this disk is set to 60801.
There is nothing wrong with that, but this is larger than 1024,
and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO)
2) booting and partitioning software from other OSs
(e.g., DOS FDISK, OS/2 FDISK)
Command (m for help): p
Disk /dev/sda1: 500.1 GB, 500105764864 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x6e697373
This doesn't look like a partition table
Probably you selected the wrong device.
Device Boot Start End Blocks Id System
/dev/sda1p1 ? 120528 234814 918008208 4f QNX4.x 3rd part
Partition 1 does not end on cylinder boundary.
/dev/sda1p2 ? 119381 153271 272218546+ 73 Unknown
Partition 2 does not end on cylinder boundary.
/dev/sda1p3 ? 113202 147075 272087568 2b Unknown
Partition 3 does not end on cylinder boundary.
/dev/sda1p4 ? 177064 177067 27487 61 SpeedStor
Partition 4 does not end on cylinde r boundary.
Partition table entries are not in disk order
-----------------
$ lsusb ==
Bus 003 Device 013: ID 0480:a007 Toshiba America Info. Systems, Inc.
...
------------------
Previously I had BIG problems with my Slakware-13 and /dev/sd*
with inode-size incompatibilities which I'd never even thought about.
Also, I want to have more than 16 partitions [for certin reasons]
and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
even it I don't YET anticipate booting from this USB hard-drive.
So, should I just try to partition it like the old IDEs ?
== TIA.
PS. my confidence was shattered by an expensive discovery that you
can't just interchange MOBOs, CPUs and power-supplies, like in the
old days.
Grant
Nov 8, 2012, 7:24:47 PM11/8/12
to
a single vfat filesystem.
Try mounting /dev/sda if you want to recover the info, otherwise clear
the drive by writing zeroes to the first few MBs:
dd if=/dev/zero bs=1M count=4 of=/dev/sda
Leave out the count to clear the entire drive.
>
>Partition table entries are not in disk order
>-----------------
>
>$ lsusb ==
>Bus 003 Device 013: ID 0480:a007 Toshiba America Info. Systems, Inc.
>...
>------------------
>Previously I had BIG problems with my Slakware-13 and /dev/sd*
>with inode-size incompatibilities which I'd never even thought about.
>
>Also, I want to have more than 16 partitions [for certin reasons]
>and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>even it I don't YET anticipate booting from this USB hard-drive.
You may lose partitions >16. This happened to me way back, but
>Partition table entries are not in disk order
>-----------------
>
>$ lsusb ==
>Bus 003 Device 013: ID 0480:a007 Toshiba America Info. Systems, Inc.
>...
>------------------
>Previously I had BIG problems with my Slakware-13 and /dev/sd*
>with inode-size incompatibilities which I'd never even thought about.
>
>Also, I want to have more than 16 partitions [for certin reasons]
>and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>even it I don't YET anticipate booting from this USB hard-drive.
I don't remember the details, it used to work a very long time ago.
>
>So, should I just try to partition it like the old IDEs ?
>
>== TIA.
>
>PS. my confidence was shattered by an expensive discovery that you
>can't just interchange MOBOs, CPUs and power-supplies, like in the
>old days.
So many variations, once you get into multi-processing territory.
>So, should I just try to partition it like the old IDEs ?
>
>== TIA.
>
>PS. my confidence was shattered by an expensive discovery that you
>can't just interchange MOBOs, CPUs and power-supplies, like in the
>old days.
No, not with windows. Should work for Slackware, but then I need to
compile custom kernels to get a working screen. I'm about to dump the
particular hp compaq with that issue for a Core2Duo box that is better
behaved.
I upgraded CPUs recently, but still on Intel LGA775, cannot justify
going to latest CPU family for speed these days, I'll wait for something
to break.
Put an SSD in the Windows box I mostly type on :) No point for Linux,
unless once needs the disk buffering for movie editing or something big.
Grant.
Floyd L. Davidson
Nov 8, 2012, 8:21:45 PM11/8/12
to
Grant <o...@grrr.id.au> wrote:
>On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>
>>After a disasterous attempt to <assemble> a MoBo to copy my
>>IDEs to SATA, I've decided to temporarily use an external USB
>>hard-drive.
>>
>>But, as you know, the docos are all written for Micro$loth;
>>and I don't know about this "HPFS/NTFS" stuff?
>>
>>$ fdisk -l == ...
>>Disk /dev/sda: 500.1 GB, 500107862016 bytes
Note that this is looking at /dev/sda which is the whole disk.
>On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>
>>After a disasterous attempt to <assemble> a MoBo to copy my
>>IDEs to SATA, I've decided to temporarily use an external USB
>>hard-drive.
>>
>>But, as you know, the docos are all written for Micro$loth;
>>and I don't know about this "HPFS/NTFS" stuff?
>>
>>$ fdisk -l == ...
>>Disk /dev/sda: 500.1 GB, 500107862016 bytes
>>255 heads, 63 sectors/track, 60801 cylinders
>>Units = cylinders of 16065 * 512 = 8225280 bytes
>>Disk identifier: 0x1c0fd702
>>
>> Device Boot Start End Blocks Id System
>>/dev/sda1 1 60802 488384536 7 HPFS/NTFS
>>---------------
NTFS though of course it might not be formatted with an
NTFS filesystem, but it probably is.
>>
>>$ cfdisk /dev/sda1 ==FATAL ERROR:
/dev/sda1 partition, which of course does not have a
partition table.
All the rest of this is therefore bogus.
mount -t ntfs /dev/sda1 mount_point_directory
>dd if=/dev/zero bs=1M count=4 of=/dev/sda
>
>Leave out the count to clear the entire drive.
using fdisk:
fdisk /dev/sda
And then first used "d" to delete that one existing partition,
and "n" to add new partitions.
However, as noted, with the MSDOS PC style partition
table there is a limit of 16 partitions.
The solution there may be the use of the "Globally
Unique Identifier (GUID) Partition Table (GPT) format".
That probably has to be compiled into the kernel, and
then /gdisk/ is used instead of fdisk to create and
manipulate partitions.
I have no idea how well it works with USB devices that
are based on flash memory. I've only used it with 4TB
disk drives because it can access data on a drive
greater than 2TB.
--
Floyd L. Davidson http://www.apaflo.com/floyd_davidson
Ukpeagvik (Barrow, Alaska) fl...@apaflo.com
Kees Theunissen
Nov 9, 2012, 12:38:59 PM11/9/12
to
Grant wrote:
> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>> Also, I want to have more than 16 partitions [for certin reasons]
>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>> even it I don't YET anticipate booting from this USB hard-drive.
>
> You may lose partitions >16. This happened to me way back, but
> I don't remember the details, it used to work a very long time ago.
That could have happened when switching over from the now
> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>> Also, I want to have more than 16 partitions [for certin reasons]
>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>> even it I don't YET anticipate booting from this USB hard-drive.
>
> You may lose partitions >16. This happened to me way back, but
> I don't remember the details, it used to work a very long time ago.
deprecated old ide subsystem in the kernel to the newer libata
subsystem. Ide device names changed from /dev/hdXX to /dev/sdXX
with the libata subsystem.
There used to be a limit of 63 partitions on ide disks and
15 partitions on scsi disks.
From /usr/src/linux/Documentation/devices.txt:
...
3 block First MFM, RLL and IDE hard disk/CD-ROM interface
0 = /dev/hda Master: whole disk (or CD-ROM)
64 = /dev/hdb Slave: whole disk (or CD-ROM)
For partitions, add to the whole disk device number:
0 = /dev/hd? Whole disk
1 = /dev/hd?1 First partition
2 = /dev/hd?2 Second partition
...
63 = /dev/hd?63 63rd partition
For Linux/i386, partitions 1-4 are the primary
partitions, and 5 and above are logical partitions.
Other versions of Linux use partitioning schemes
appropriate to their respective architectures.
...
8 block SCSI disk devices (0-15)
0 = /dev/sda First SCSI disk whole disk
16 = /dev/sdb Second SCSI disk whole disk
32 = /dev/sdc Third SCSI disk whole disk
...
240 = /dev/sdp Sixteenth SCSI disk whole disk
Partitions are handled in the same way as for IDE
disks (see major number 3) except that the limit on
partitions is 15.
This limit doesn't exist anymore since kernel 2.6.28 (released in
December 2008). The device numbering scheme has been extended to
use major and minor device numbers greater than 255.
devices.txt continues:
...
---LARGE MAJORS!!!!!---
...
259 block Block Extended Major
Used dynamically to hold additional partition minor
numbers and allow large numbers of partitions per device
Slackware moved to the libata subsystem with the release of
slackware-13.1, kernel 2.6.33.4.
The limited number of minor scsi device numbers wasn't an issue
anymore with that kernel. But if you compile your own kernels
and switched over to libata with older kernels --or maybe with
the wrong combination of options-- this might have hit you.
Regards,
Kees.
--
Kees Theunissen.
Kees Theunissen
Nov 9, 2012, 1:11:09 PM11/9/12
to
Floyd L. Davidson wrote:
> However, as noted, with the MSDOS PC style partition
> table there is a limit of 16 partitions.
No there is no such limit with the MSDOS PC style partition
> However, as noted, with the MSDOS PC style partition
> table there is a limit of 16 partitions.
table. There used to be a limit of 16 with the assignment of minor
device numbers for scsi devices and a limit of 64 for ide devices.
Since the release of kernel 2.6.28 in December 2008 you can use an
unlimited number of partitions.
See my other message in this thread for more details.
Avoi...@gmail.com
Nov 9, 2012, 2:23:56 PM11/9/12
to
In article <871ug3o...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
> Grant <o...@grrr.id.au> wrote:
> >On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
> >
>>$ fdisk -l == ...
>>Disk /dev/sda: 500.1 GB, 500107862016 bytes
] Note that this is looking at /dev/sda which is the whole disk.
Sure.
> Grant <o...@grrr.id.au> wrote:
> >On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
> >
>>$ fdisk -l == ...
>>Disk /dev/sda: 500.1 GB, 500107862016 bytes
] Note that this is looking at /dev/sda which is the whole disk.
>>255 heads, 63 sectors/track, 60801 cylinders
>>Units = cylinders of 16065 * 512 = 8225280 bytes
>>Disk identifier: 0x1c0fd702
>>
>> Device Boot Start End Blocks Id System
>>/dev/sda1 1 60802 488384536 7 HPFS/NTFS
>>---------------
]And that says there is exactly 1 partition, marked as
]NTFS though of course it might not be formatted with an
]NTFS filesystem, but it probably is.
>>
>>$ cfdisk /dev/sda1 ==FATAL ERROR:
]And this is not. It is attempting to look ONLY at the
]/dev/sda1 partition, which of course does not have a
]partition table.
]
]All the rest of this is therefore bogus.
]I have no idea how well it works with USB devices that
]are based on flash memory.
Well no, google seems to say it's an IDE with a USB
interface. And the pdf-man talks about heating & liquid
as if it was a normal IDE.
===========
It's getting more crazy!
<><><>
root@darkstar:~# df
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 7692876 4968516 2333584 69% /
tmpfs 59696 0 59696 0% /dev/shm
/dev/hdc16 3903620 1653108 2250512 43% /mnt/ReiseBrdg
/dev/hdc6 2071384 1893592 72568 97% /mnt/hdc6
/dev/hdc11 980308 970760 0 100% /mnt/hdc11
/dev/hdc14 7692876 5857720 1444380 81% /mnt/FC1
/dev/hdc17 14421344 5077392 8611392 38% /mnt/DebDVD1bak
/dev/sda1 488384532 109516 488275016 1% /mnt/tmp
root@darkstar:~# mount /dev/sda /mnt/hd
mount: /dev/sda already mounted or /mnt/hd busy
root@darkstar:~# umount /mnt/tmp
root@darkstar:~# mount /dev/sda /mnt/hd
mount: No such file or directory
==============
-> mount -t ntfs /dev/sda /mnt/hd ==
mount: wrong fs type, bad option, bad superblock on /dev/sda,
missing codepage or helper program, or other error
In some cases useful info is found in syslog - try
dmesg | tail or so
-> dmesg | tail ==
usbserial_generic 3-1:1.1: device disconnected
FAT: invalid media value (0xf3)
VFS: Can't find a valid FAT filesystem on dev sda.
GFS2: gfs2 mount does not exist
NTFS volume version 3.1.
NTFS-fs error (device sda1): load_system_files(): Volume is dirty. Mounting read-only. Run chkdsk and mount in Windows.
NTFS-fs warning (device sda): is_boot_sector_ntfs(): Invalid boot sector checksum.
NTFS-fs error (device sda): read_ntfs_boot_sector(): Primary boot sector is invalid.
NTFS-fs error (device sda): read_ntfs_boot_sector(): Mount option errors=recover not used. Aborting without trying to recover.
NTFS-fs error (device sda): ntfs_fill_super(): Not an NTFS volume.
-> mount /dev/sda1 /mnt/hd
root@darkstar:~# mount -l
/dev/root on / type ext2 (rw,errors=continue) [Slak13full]
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
usbfs on /proc/bus/usb type usbfs (rw)
tmpfs on /dev/shm type tmpfs (rw)
/dev/hdc16 on /mnt/ReiseBrdg type reiserfs (rw)
/dev/hdc6 on /mnt/hdc6 type ext2 (rw) [Mandrake]
/dev/hdc11 on /mnt/hdc11 type ext2 (rw) [Data2010]
/dev/hdc14 on /mnt/FC1 type ext3 (rw) [-FC1-]
/dev/hdc17 on /mnt/DebDVD1bak type ext3 (rw) [DebDVD1bak]
/dev/sda1 on /mnt/hd type ntfs (rw) [TOSHIBA EXT]
-> date >> /mnt/hd/CRGlog
-bash: /mnt/hd/CRGlog: Read-only file system
=============
When I'm ready to lose all my settings and reboot, I'll try W98
if I can.
== Thanks.
PS. google suggests it's NOT semi-conductor.
It's IDE with a USB interface ?
Grant
Nov 9, 2012, 4:19:03 PM11/9/12
to
On Fri, 09 Nov 2012 18:38:59 +0100, Kees Theunissen <theu...@rijnh.nl> wrote:
>Grant wrote:
>> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>
>>> Also, I want to have more than 16 partitions [for certin reasons]
>>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>>> even it I don't YET anticipate booting from this USB hard-drive.
>>
>> You may lose partitions >16. This happened to me way back, but
>> I don't remember the details, it used to work a very long time ago.
>
>That could have happened when switching over from the now
>deprecated old ide subsystem in the kernel to the newer libata
>subsystem. Ide device names changed from /dev/hdXX to /dev/sdXX
>with the libata subsystem.
>There used to be a limit of 63 partitions on ide disks and
>15 partitions on scsi disks.
That could explain it :)
>Grant wrote:
>> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>
>>> Also, I want to have more than 16 partitions [for certin reasons]
>>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>>> even it I don't YET anticipate booting from this USB hard-drive.
>>
>> You may lose partitions >16. This happened to me way back, but
>> I don't remember the details, it used to work a very long time ago.
>
>That could have happened when switching over from the now
>deprecated old ide subsystem in the kernel to the newer libata
>subsystem. Ide device names changed from /dev/hdXX to /dev/sdXX
>with the libata subsystem.
>There used to be a limit of 63 partitions on ide disks and
>15 partitions on scsi disks.
>
...
>This limit doesn't exist anymore since kernel 2.6.28 (released in
>December 2008). The device numbering scheme has been extended to
>use major and minor device numbers greater than 255.
Yes, I haven't tried lots of partitions for years since I had some go
>December 2008). The device numbering scheme has been extended to
>use major and minor device numbers greater than 255.
missing -- no data loss, just return to older system and merge data,
reduce partitions. I agree it likely was during that hda -> sda period.
I didn't stop to work out why, just limited max partitions.
Seems idea of separate read-only /usr is fading with some of the new
stuff the large distros are pushing.
...
>Slackware moved to the libata subsystem with the release of
>slackware-13.1, kernel 2.6.33.4.
>The limited number of minor scsi device numbers wasn't an issue
>anymore with that kernel. But if you compile your own kernels
>and switched over to libata with older kernels --or maybe with
>the wrong combination of options-- this might have hit you.
Yes, I've switched over to libata for a while yet, holding the
>slackware-13.1, kernel 2.6.33.4.
>The limited number of minor scsi device numbers wasn't an issue
>anymore with that kernel. But if you compile your own kernels
>and switched over to libata with older kernels --or maybe with
>the wrong combination of options-- this might have hit you.
view that the slackware kernel is there to let one boot the machine
and compile one's own kernel to suit the hardware ;)
Right now I have a 'new' machine to play with, it has a spare HDD
raided from another box, not yet in use. I'll have a look at lots
of logical partitions to see how it goes.
Grant.
>
>
>Regards,
>
>Kees.
Floyd L. Davidson
Nov 9, 2012, 5:09:15 PM11/9/12
to
Avoi...@gmail.com wrote:
>In article <871ug3o...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
>]I have no idea how well it works with USB devices that
>]are based on flash memory.
>
>Well no, google seems to say it's an IDE with a USB
>interface. And the pdf-man talks about heating & liquid
>as if it was a normal IDE.
The controller for the flash memory is an IDE controller,
>In article <871ug3o...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
>]I have no idea how well it works with USB devices that
>]are based on flash memory.
>
>Well no, google seems to say it's an IDE with a USB
>interface. And the pdf-man talks about heating & liquid
>as if it was a normal IDE.
which then uses USB to commuicate with the computer.
>===========
>It's getting more crazy!
><><><>
>root@darkstar:~# df
>Filesystem 1K-blocks Used Available Use% Mounted on
>/dev/sda1 488384532 109516 488275016 1% /mnt/tmp
>root@darkstar:~# mount /dev/sda /mnt/hd
>mount: /dev/sda already mounted or /mnt/hd busy
>root@darkstar:~# umount /mnt/tmp
>root@darkstar:~# mount /dev/sda /mnt/hd
>mount: No such file or directory
But what you want to mount is /dev/sda1, which was
>root@darkstar:~# mount /dev/sda /mnt/hd
>mount: /dev/sda already mounted or /mnt/hd busy
>root@darkstar:~# umount /mnt/tmp
>root@darkstar:~# mount /dev/sda /mnt/hd
>mount: No such file or directory
already mounted when you began that sequency of
commands.
Again, there is an NTFS filesystem on /dev/sda1, but
there is no filesystem on /dev/sda.
>==============
>-> mount -t ntfs /dev/sda /mnt/hd ==
>mount: wrong fs type, bad option, bad superblock on /dev/sda,
> missing codepage or helper program, or other error
> In some cases useful info is found in syslog - try
> dmesg | tail or so
>-> dmesg | tail ==
>
>-> mount /dev/sda1 /mnt/hd
And that works, as it should.
>root@darkstar:~# mount -l
...
>/dev/sda1 on /mnt/hd type ntfs (rw) [TOSHIBA EXT]
There is it, just as it was to start with.
>-> date >> /mnt/hd/CRGlog
>-bash: /mnt/hd/CRGlog: Read-only file system
writing to an NTFS filesystem. That is a compile time
option for the kernel.
>=============
>
>When I'm ready to lose all my settings and reboot, I'll try W98
>if I can.
by rebooting, but you should be able to configure things
to come back up just about exactly as they were when you
shut the system down.
Check to see if your NTFS file system support is a
module or if it is built into the kernel. If it is a
module, you can remove it, compile the module again with
write support added, and put it back into place and this
should all work. Otherwise you'll have to recompile the
entire kernel with the writable options selected, and
the of course you actually will have to reboot.
>PS. google suggests it's NOT semi-conductor.
>It's IDE with a USB interface ?
of course a semi-conductor device. Note that the IDE
controller can use and of USB, PCI, SATA or SCSI busses
for communications.
Grant
Nov 9, 2012, 8:29:25 PM11/9/12
to
On Sat, 10 Nov 2012 08:19:03 +1100, Grant <o...@grrr.id.au> wrote:
>On Fri, 09 Nov 2012 18:38:59 +0100, Kees Theunissen <theu...@rijnh.nl> wrote:
>
>>Grant wrote:
>>> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>>
>>>> Also, I want to have more than 16 partitions [for certin reasons]
>>>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>>>> even it I don't YET anticipate booting from this USB hard-drive.
>>>
>>> You may lose partitions >16. This happened to me way back, but
>>> I don't remember the details, it used to work a very long time ago.
>>
>>That could have happened when switching over from the now
>>deprecated old ide subsystem in the kernel to the newer libata
>>subsystem. Ide device names changed from /dev/hdXX to /dev/sdXX
>>with the libata subsystem.
>>There used to be a limit of 63 partitions on ide disks and
>>15 partitions on scsi disks.
>
>That could explain it :)
root@deimos:~# fdisk -l
>On Fri, 09 Nov 2012 18:38:59 +0100, Kees Theunissen <theu...@rijnh.nl> wrote:
>
>>Grant wrote:
>>> On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
>>
>>>> Also, I want to have more than 16 partitions [for certin reasons]
>>>> and eg. AFAIK Slakware's <lilo script> assumes < 17 partitions;
>>>> even it I don't YET anticipate booting from this USB hard-drive.
>>>
>>> You may lose partitions >16. This happened to me way back, but
>>> I don't remember the details, it used to work a very long time ago.
>>
>>That could have happened when switching over from the now
>>deprecated old ide subsystem in the kernel to the newer libata
>>subsystem. Ide device names changed from /dev/hdXX to /dev/sdXX
>>with the libata subsystem.
>>There used to be a limit of 63 partitions on ide disks and
>>15 partitions on scsi disks.
>
>That could explain it :)
Disk /dev/sda: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders, total 390721968 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
Disk identifier: 0xece5bc25
Device Boot Start End Blocks Id System
/dev/sda2 41945088 62916607 10485760 83 Linux
/dev/sda3 62916608 83888127 10485760 83 Linux
/dev/sda4 83888128 390721967 153416920 5 Extended
/dev/sda5 83890176 88084479 2097152 82 Linux swap
/dev/sda6 88086528 222304255 67108864 83 Linux
/dev/sda7 222306304 390721967 84207832 7 HPFS/NTFS/exFAT
Disk /dev/sdb: 120.0 GB, 120034123776 bytes
255 heads, 63 sectors/track, 14593 cylinders, total 234441648 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
Disk identifier: 0x7fb864cc
Device Boot Start End Blocks Id System
/dev/sdb2 4196352 8390655 2097152 83 Linux
/dev/sdb3 8390656 12584959 2097152 83 Linux
/dev/sdb4 12584960 234441647 110928344 5 Extended
/dev/sdb5 12587008 16781311 2097152 83 Linux
/dev/sdb6 16783360 20977663 2097152 83 Linux
/dev/sdb7 20979712 25174015 2097152 83 Linux
/dev/sdb8 25176064 29370367 2097152 83 Linux
/dev/sdb9 29372416 33566719 2097152 83 Linux
/dev/sdb10 33568768 37763071 2097152 83 Linux
/dev/sdb11 37765120 41959423 2097152 83 Linux
/dev/sdb12 41961472 46155775 2097152 83 Linux
/dev/sdb13 46157824 50352127 2097152 83 Linux
/dev/sdb14 50354176 54548479 2097152 83 Linux
/dev/sdb15 54550528 58744831 2097152 83 Linux
/dev/sdb16 58746880 62941183 2097152 83 Linux
/dev/sdb17 62943232 67137535 2097152 83 Linux
/dev/sdb18 67139584 71333887 2097152 83 Linux
/dev/sdb19 71335936 75530239 2097152 83 Linux
/dev/sdb20 75532288 79726591 2097152 83 Linux
/dev/sdb21 79728640 83922943 2097152 83 Linux
/dev/sdb22 83924992 88119295 2097152 83 Linux
/dev/sdb23 88121344 92315647 2097152 83 Linux
/dev/sdb24 92317696 96511999 2097152 83 Linux
/dev/sdb25 96514048 100708351 2097152 83 Linux
/dev/sdb26 100710400 104904703 2097152 83 Linux
/dev/sdb27 104906752 109101055 2097152 83 Linux
/dev/sdb28 109103104 113297407 2097152 83 Linux
/dev/sdb29 113299456 117493759 2097152 83 Linux
/dev/sdb30 117495808 121690111 2097152 83 Linux
/dev/sdb31 121692160 125886463 2097152 83 Linux
/dev/sdb32 125888512 130082815 2097152 83 Linux
/dev/sdb33 130084864 134279167 2097152 83 Linux
/dev/sdb34 134281216 138475519 2097152 83 Linux
/dev/sdb35 138477568 142671871 2097152 83 Linux
/dev/sdb36 142673920 146868223 2097152 83 Linux
/dev/sdb37 146870272 151064575 2097152 83 Linux
/dev/sdb38 151066624 155260927 2097152 83 Linux
/dev/sdb39 155262976 159457279 2097152 83 Linux
/dev/sdb40 159459328 163653631 2097152 83 Linux
/dev/sdb41 163655680 167849983 2097152 83 Linux
/dev/sdb42 167852032 172046335 2097152 83 Linux
root@deimos:~#
So there :o)
Grant.
Avoi...@gmail.com
Nov 10, 2012, 1:07:32 AM11/10/12
to
> Grant <o...@grrr.id.au> wrote:
> >On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
> >
I said I'd try under W98. It wanted a USB-driver, and failed.
> >On Thu, 8 Nov 2012 22:32:19 +0000 (UTC), Avoi...@gmail.com wrote:
> >
So I've got a CD of Kongoni: slackware derivative, [like slakx, because
slackware if perhaps the most OPEN distribution], who's knl is just a
few months older than my slackware-13.
Kogi auto-mounts all partitions at boot-time.
It had a big delay near <*udev*>.
Here's some logs that I pasted to the problematic sda1, under kogi,
and which I can now READ-only under slackware-13:----
Fri Nov 9 21:51:20 SAST 2012
==============================
==> booted kogoni ==
aufs on / type aufs (rw,si=cc3b8a23,nowarn_perm)
/proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
usbfs on /proc/bus/usb type usbfs (rw)
tmpfs on /dev/shm type tmpfs (rw)
/dev/hdb on /mnt/hdb type iso9660 (ro,noatime)
usbfs on /proc/bus/usb type usbfs (rw)
tmpfs on /dev/shm type tmpfs (rw)
/dev/hdc1 on /mnt/hdc1 type vfat (rw,noatime,quiet,umask=0,check=s,shortname=mixed)
/dev/hdc2 on /mnt/hdc2 type vfat (rw,noatime,quiet,umask=0,check=s,shortname=mixed)
/dev/hdc3 on /mnt/hdc3 type ext2 (rw,noatime)
/dev/hdc5 on /mnt/hdc5 type ext2 (rw,noatime)
/dev/hdc6 on /mnt/hdc6 type ext2 (rw,noatime)
/dev/hdc7 on /mnt/hdc7 type ext2 (rw,noatime)
/dev/hdc11 on /mnt/hdc11 type ext2 (rw,noatime)
/dev/hdc12 on /mnt/hdc12 type ext2 (rw,noatime)
/dev/hdc14 on /mnt/hdc14 type ext3 (rw,noatime)
/dev/hdc15 on /mnt/hdc15 type ext2 (rw,noatime)
/dev/hdc16 on /mnt/hdc16 type reiserfs (rw,noatime)
/dev/hdc17 on /mnt/hdc17 type ext3 (rw,noatime)
/dev/hdc53 on /mnt/hdc53 type ext3 (rw,noatime)
/dev/hdc54 on /mnt/hdc54 type ext3 (rw,noatime)
/dev/hdc55 on /mnt/hdc55 type vfat (rw,noatime,quiet,umask=0,check=s,shortname=mixed)
/dev/hdc57 on /mnt/hdc57 type ext2 (rw,noatime)
/dev/sda1 on /mnt/sda1 type fuseblk (rw,noatime,allow_other,blksize=4096)
=============================================
-> cfdisk /dev/sda ==
FATAL ERROR: Bad primary partition 0: Partition ends in the final partial
cylinder
Press any key to exit cfdisk
=> also: cfdisk /dev/sda1
-> fdisk /dev/sda
The number of cylinders for this disk is set to 60801.
There is nothing wrong with that, but this is larger than 1024,
and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO)
2) booting and partitioning software from other OSs
(e.g., DOS FDISK, OS/2 FDISK)
Command (m for help): p
Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1c0fd702
Device Boot Start End Blocks Id System
/dev/sda1 1 60802 488384536 7 HPFS/NTFS
Command (m for help):
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1c0fd702
Device Boot Start End Blocks Id System
/dev/sda1 1 60802 488384536 7 HPFS/NTFS
-------------------------------------------------
The above was pasted/written under kogi to the problematic disk.
Now I'm back in slackware-13, and the disk can read-only:-
root@darkstar:/home/Hardware/USBtoshiba# df
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 7692876 4968928 2333172 69% /
tmpfs 59696 0 59696 0% /dev/shm
/dev/hdc16 3903620 1653052 2250568 43% /mnt/ReiseBrdg
/dev/hdc6 2071384 1893592 72568 97% /mnt/hdc6
/dev/hdc11 980308 970760 0 100% /mnt/hdc11
/dev/hdc14 7692876 5857720 1444380 81% /mnt/FC1
/dev/hdc17 14421344 5077392 8611392 38% /mnt/DebDVD1bak
/dev/sda1 488384532 109524 488275008 1% /mnt/hd <-- takes 100 secs!
/dev/hdc11 980308 970760 0 100% /mnt/hdc11
/dev/hdc14 7692876 5857720 1444380 81% /mnt/FC1
/dev/hdc17 14421344 5077392 8611392 38% /mnt/DebDVD1bak
root@darkstar:/home/Hardware/USBtoshiba# ls -l /mnt/hd
total 5524
-r-------- 1 root root 5296 2012-11-09 23:14 CRGlog <-- written by Kongoni
-r-------- 2 root root 16022 2012-01-27 08:28 Software Offer.hta
-r-------- 1 root root 9518 2011-05-16 08:40 TMP.ico
-r-------- 2 root root 5599808 2011-09-15 04:20 TOSHIBA STOR.E BASICS.pdf
-r-------- 2 root root 1244 2011-05-16 08:40 Toshiba Places.html
-r-------- 1 root root 80 2011-09-15 06:12 autorun.inf
dr-x------ 1 root root 8192 2012-03-24 02:00 images
dr-x------ 1 root root 4096 2012-03-24 02:00 lang
-----------------------
I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
has M$- EOL, but my CRGlog has *nix-EOL.
IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
can?
== TIA.
Floyd L. Davidson
Nov 10, 2012, 3:29:44 AM11/10/12
to
Avoi...@gmail.com wrote:
>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> has M$- EOL, but my CRGlog has *nix-EOL.
> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
>
>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
>can?
Try this and see what you get:
>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> has M$- EOL, but my CRGlog has *nix-EOL.
> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
>
>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
>can?
$zgrep NTFS /proc/config.gz
CONFIG_NTFS_FS=y
# CONFIG_NTFS_DEBUG is not set
CONFIG_NTFS_RW=y
I suspect the last line is different, and might look like this:
# CONFIG_NTFS_RW is not set
Avoi...@gmail.com
Nov 10, 2012, 12:34:53 PM11/10/12
to
In article <87mwypn...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
> Avoi...@gmail.com wrote:
> >I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> > has M$- EOL, but my CRGlog has *nix-EOL.
> > IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
> >
> >I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
> >can?
>
> Try this and see what you get:
>
> $zgrep NTFS /proc/config.gz
> CONFIG_NTFS_FS=y
> # CONFIG_NTFS_DEBUG is not set
> CONFIG_NTFS_RW=y
>
> I suspect the last line is different, and might look like this:
>
> # CONFIG_NTFS_RW is not set
>
> --
-> whatis zgrep == zgrep []
> Avoi...@gmail.com wrote:
> >I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> > has M$- EOL, but my CRGlog has *nix-EOL.
> > IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
> >
> >I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
> >can?
>
> Try this and see what you get:
>
> $zgrep NTFS /proc/config.gz
> CONFIG_NTFS_FS=y
> # CONFIG_NTFS_DEBUG is not set
> CONFIG_NTFS_RW=y
>
> I suspect the last line is different, and might look like this:
>
> # CONFIG_NTFS_RW is not set
>
> --
(1) - search possibly compressed files for a regular expression
-> zgrep --help ==
Usage: /usr/bin/zgrep [OPTION]... [-e] PATTERN [FILE]...
Look for instances of PATTERN in the input FILEs, using their
uncompressed contents if they are compressed.
...
-> zgrep NTFS /proc/config.gz ==
CONFIG_NTFS_FS=y
# CONFIG_NTFS_DEBUG is not set
CONFIG_NTFS_RW=y <-- looks OK ?
# CONFIG_NTFS_DEBUG is not set
==> move 'bad-USB-connection' == see LED activate
-> mount /dev/sda1 -t ntfs /mnt/tmp
-> mount == ...
/dev/sda1 on /mnt/tmp type ntfs (rw) <-*! 'rw'
BUT then I can't save/update my log ?!
-> ls -l /mnt/tmp ==
-r-------- 1 root root 5296 2012-11-09 23:14 CRGlog
-> mount == ...
/dev/sda1 on /mnt/tmp type ntfs (rw)
---------
`mount` contradicting `ls -l` is bad.
Apparently it wasn't properly connected at boot this time.
But that's no excuse. It must be mountable AFTER bootup.
Don't tell any of the tribal-hooligans on this forum, who
may go hysterical, but I'm going to test under Debian:
Lenny & Etch [3-4 yr olds] . I'm promiscious.
Thanks. I'll report back.
Grant
Nov 10, 2012, 4:37:05 PM11/10/12
to
On Fri, 09 Nov 2012 23:29:44 -0900, fl...@apaflo.com (Floyd L. Davidson) wrote:
>Avoi...@gmail.com wrote:
>>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
>> has M$- EOL, but my CRGlog has *nix-EOL.
>> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
>>
>>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
>>can?
>
>Try this and see what you get:
>
> $zgrep NTFS /proc/config.gz
> CONFIG_NTFS_FS=y
> # CONFIG_NTFS_DEBUG is not set
> CONFIG_NTFS_RW=y
>
>I suspect the last line is different, and might look like this:
>
> # CONFIG_NTFS_RW is not set
Kernel ntfs write is not complete, nor safe.
>Avoi...@gmail.com wrote:
>>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
>> has M$- EOL, but my CRGlog has *nix-EOL.
>> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
>>
>>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
>>can?
>
>Try this and see what you get:
>
> $zgrep NTFS /proc/config.gz
> CONFIG_NTFS_FS=y
> # CONFIG_NTFS_DEBUG is not set
> CONFIG_NTFS_RW=y
>
>I suspect the last line is different, and might look like this:
>
> # CONFIG_NTFS_RW is not set
These days it is much better to use ntfs-3g instead. The ntfs-3g suite is
very reliable though slow due to being run in user space (FUSE).
Grant.
Floyd L. Davidson
Nov 10, 2012, 6:13:08 PM11/10/12
to
question of it being RO applies specifically to NTFS.
Clearly that was not the problem though.
Now I'm wondering if he has something wierd in /etc/fstab, because
I can't think of any other way to get the permission the way he
showed them.
Avoi...@gmail.com
Nov 11, 2012, 4:08:27 AM11/11/12
to
/dev/hdc15 / ext2 defaults 1 1
#/dev/hdc1 /mnt/fat1 vfat defaults 1 0
#/dev/hdc2 /mnt/fat2 vfat defaults 1 0
#/dev/cdrom /mnt/cdrom auto noauto,owner,ro 0 0
/dev/fd0 /mnt/floppy auto noauto,owner 0 0
devpts /dev/pts devpts gid=5,mode=620 0 0
proc /proc proc defaults 0 0
tmpfs /dev/shm tmpfs defaults 0 0
/dev/hdc45 swap swap defaults 0 0
# copied from `df` & mtab
/dev/hdc16 /mnt/ReiseBrdg reiserfs rw 0 0
/dev/hdc6 /mnt/hdc6 ext2 rw 0 0
/dev/hdc11 /mnt/hdc11 ext2 rw 0 0
/dev/hdc14 /mnt/FC1 ext3 rw 0 0
/dev/hdc17 /mnt/DebDVD1bak ext3 rw 0 0
/dev/sda1 /mnt/tmp ntfs rw 0 0
-----------------------
This is getting deeper than I need at this stage of life.
M$ is screwing us via new hardware drivers.
Especially after all the hassle with trying to install slakware to a CF:sda*
and running into inode-size incompatibility ...etc.
Because in that case, the on-hand Debian distros where more successfull, I've
'probed' them too, for this case. But the logs below seem to show similar
problems [they do, however isolate the /etc/fstab suspicion].
I'm really wondering HOW kogi manages to mount it RW?
<><><> now boot debian Etch ? Lenny <><><>
-> rdev == /dev/.static/dev/hda3 /
-> man fsck.nfs ==
Reformatting fsck.nfs(8), please wait...
FSCK.NFS(8) FSCK.NFS(8)
NAME
fsck.nfs - Dummy fsck.nfs script that always returns success.
SYNOPSIS
fsck.nfs
DESCRIPTION
Debian GNU/Linux need this for when the root file system is on NFS: there is no way to
find out if root is NFS mounted and we really want to do a "fsck -a /".
EXIT CODE
The exit code returned by mount.nfs is always zero, meaning successful completion.
SEE ALSO
fsck(8), fstab(5).
Initscripts May 2004 FSCK.NFS(8)
----------------------------------
-> man fsck == ...
SEE ALSO
fstab(5), mkfs(8), fsck.ext2(8) or fsck.ext3(8) or e2fsck(8), cramfsck(8), fsck.minix(8),
fsck.msdos(8), fsck.jfs(8), fsck.nfs(8), fsck.vfat(8), fsck.xfs(8), fsck.xiafs(8),
reiserfsck(8).
-> fsck -a /dev/sda == ...
fsck.ext2: Bad magic number in super-block while trying to open /dev/sda
-> fsck -a /dev/sda1 ==
fsck 1.39 (29-May-2006)
fsck: fsck.ntfs: not found <--so it recognised the correct type
fsck: Error 2 while executing fsck.ntfs for /dev/sda1
-> which fsck == /sbin/fsck
-> which fsck.ntfs == na
--> ls -l /proc/scsi | grep c ==
-r--r--r-- 1 root root 0 2012-11-11 06:33 device_info
-r--r--r-- 1 root root 0 2012-11-11 06:33 scsi
-> cat /proc/scsi/scsi ==
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: TOSHIBA Model: External USB 3.0 Rev: 0001
Type: Direct-Access ANSI SCSI revision: 02
--------------------------------> DebianLenny shows:----
-> dmesg | grep usb == many
-> dmesg | grep shiba ==
[ 7.947896] usb 3-2: Manufacturer: Toshiba
-> lsusb ==
Bus 003 Device 002: ID 0480:a007 Toshiba America Info. Systems, Inc.
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
-> cd /mnt
-> mount /dev/sda1 /mnt/sda1 ==
mount USBstik sda1
[ 716.286203] NTFS driver 2.1.29 [Flags: R/W MODULE].
[ 716.877715] NTFS volume version 3.1.
[ 716.886281] NTFS-fs error (device sda1): load_system_files(): Volume is dirty.
Mounting read-only. Run chkdsk and mount in Windows.
-> df == Toshiba LED flashing 100 secs ? to initialise ?
...
/dev/sda1 488384532 109524 488275008 1% /mnt/sda1
-> ls /mnt/sda1/CRGlog -l ==
-r-------- 1 root root 5296 2012-11-09 21:14 /mnt/sda1/CRGlog <-- read-only!
-> umount /dev/sda1
==> fsck - check and repair a Linux file system
-> fsck /dev/sda1 ==
fsck 1.41.3 (12-Oct-2008)
fsck: fsck.ntfs: not found
fsck: Error 2 while executing fsck.ntfs for /dev/sda1
Floyd L. Davidson
Nov 11, 2012, 5:05:13 AM11/11/12
to
Avoi...@gmail.com wrote:
>
>-> cd /mnt
>-> mount /dev/sda1 /mnt/sda1 ==
>mount USBstik sda1
>[ 716.286203] NTFS driver 2.1.29 [Flags: R/W MODULE].
>[ 716.877715] NTFS volume version 3.1.
>[ 716.886281] NTFS-fs error (device sda1): load_system_files(): Volume is dirty.
> Mounting read-only. Run chkdsk and mount in Windows.
Okay, there is the problem, and the solution.
>
>-> cd /mnt
>-> mount /dev/sda1 /mnt/sda1 ==
>mount USBstik sda1
>[ 716.286203] NTFS driver 2.1.29 [Flags: R/W MODULE].
>[ 716.877715] NTFS volume version 3.1.
>[ 716.886281] NTFS-fs error (device sda1): load_system_files(): Volume is dirty.
> Mounting read-only. Run chkdsk and mount in Windows.
>-> df == Toshiba LED flashing 100 secs ? to initialise ?
>...
>/dev/sda1 488384532 109524 488275008 1% /mnt/sda1
>
>-> ls /mnt/sda1/CRGlog -l ==
>-r-------- 1 root root 5296 2012-11-09 21:14 /mnt/sda1/CRGlog <-- read-only!
>
>-> umount /dev/sda1
>==> fsck - check and repair a Linux file system
>-> fsck /dev/sda1 ==
>fsck 1.41.3 (12-Oct-2008)
>fsck: fsck.ntfs: not found
>fsck: Error 2 while executing fsck.ntfs for /dev/sda1
NTFS file system. Given that we are lucky that people
could even figure out enough of how it works to write
code to mount it, that isn't surprising. Unfortunate
though...
It isn't being mounted RO. It is corrupted, and that
is causing the problems you see.
The error message says that CHKDSK, on a Windows system,
should be used to correct the problem.
Avoi...@gmail.com
Nov 11, 2012, 9:37:49 AM11/11/12
to
In article <877gpsm...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
> Avoi...@gmail.com wrote:
> >
> >-> cd /mnt
> >-> mount /dev/sda1 /mnt/sda1 ==
> >mount USBstik sda1
> >[ 716.286203] NTFS driver 2.1.29 [Flags: R/W MODULE].
> >[ 716.877715] NTFS volume version 3.1.
> >[ 716.886281] NTFS-fs error (device sda1): load_system_files(): Volume is dirty.
> > Mounting read-only. Run chkdsk and mount in Windows.
>
> Okay, there is the problem, and the solution.
>
That's only one of 4 'opinions'. See below.
> Avoi...@gmail.com wrote:
> >
> >-> cd /mnt
> >-> mount /dev/sda1 /mnt/sda1 ==
> >mount USBstik sda1
> >[ 716.286203] NTFS driver 2.1.29 [Flags: R/W MODULE].
> >[ 716.877715] NTFS volume version 3.1.
> >[ 716.886281] NTFS-fs error (device sda1): load_system_files(): Volume is dirty.
> > Mounting read-only. Run chkdsk and mount in Windows.
>
> Okay, there is the problem, and the solution.
>
> >-> df == Toshiba LED flashing 100 secs ? to initialise ?
> >...
> >/dev/sda1 488384532 109524 488275008 1% /mnt/sda1
> >
> >-> ls /mnt/sda1/CRGlog -l ==
> >-r-------- 1 root root 5296 2012-11-09 21:14 /mnt/sda1/CRGlog <-- read-only!
> >
> >-> umount /dev/sda1
> >==> fsck - check and repair a Linux file system
> >-> fsck /dev/sda1 ==
> >fsck 1.41.3 (12-Oct-2008)
> >fsck: fsck.ntfs: not found
> >fsck: Error 2 while executing fsck.ntfs for /dev/sda1
>
> Note that there simply is no fsck option to work on an
> NTFS file system. Given that we are lucky that people
> could even figure out enough of how it works to write
> code to mount it, that isn't surprising. Unfortunate
> though...
>
out-of-the-water?
> It isn't being mounted RO. It is corrupted, and that
> is causing the problems you see.
>
> The error message says that CHKDSK, on a Windows system,
> should be used to correct the problem.
>
Of the 4 krnl 2.6* installations: slak13, kogi, debLenny, debEtch;
the Debs are pointing to WINDOWS.
Perhaps, by the time that I got to test with the 2 Debs, I had already
bumped the bad-connector. But when I initially started, I was very
careful, and slak13 didn't mount it.
To retest if kogi can still R/W, which I expect, I need to swap one
IDE for the CD, and then I'll report back.
-------------
But since I've got a well informed user of `Gnus` from the
diagonally oposite side of the globe, if you also use emacs, you
may be able to advise on this more difficult problem:
I've got a tested/working awk script, which does:
<deleteFromInFile allFurther TextBlocks WhichRegexMatch dFile>,
which I call by:
`gawk -f awk.code dFile inFile > cleanedFile`
And then I check how the inFile has been cleaned of repeated
garbage, by comparing:
`cat inFile | wc -l`
`cat cleaned | wc -l`
As you know, these days, if you accumulate 6 pages from a
'knowledge site', by appending the http-ed text to a file,
using eg. lynx, like building a book, you get each 'page',
having the same repeated <header material>...etc.
These repeats is what the awk-script deletes.
Normal editors CAN <search & replace> all further text,
but usually only one line - max.
emacs can <RegexSearch & replace> all further text of
multi-line size/buffer.
As a previous emacs-slight-experimenter I checked
IF/HOW emacs would do it.
There seems to be problems with multiple RE-syntax in emacs?
So I need to use the already proven awk-script.
Presently, it's very AWKward, since I need to cycle the text
between 2 files, copying the <RepeatingTextBlocks> to dFile.
Obviously, you'd rather have it like a normal editor , where
the deletions, become visible on the single file being edited.
Do you think emacs can use the awk-script to repeatedly
update the 'view' of the same text-file while it's changing,
repeatedly, like a normal editor would show the search
and replaceAll?
== TIA.
PS. when I do a `df` now the Toshiba LED starts flashing for
about 100secs, which is hardly surprising, given the 4-line USB2?
And now the LED is ON. Probably it was previously not rotating?
Oh no, it's started flashing for 30secs, and 'waiting' the command.
Floyd L. Davidson
Nov 11, 2012, 11:47:10 AM11/11/12
to
Avoi...@gmail.com wrote:
>Do you think emacs can use the awk-script to repeatedly
>update the 'view' of the same text-file while it's changing,
>repeatedly, like a normal editor would show the search
>and replaceAll?
I'm sure that emacs can do that, but I mess with eLisp about
>Do you think emacs can use the awk-script to repeatedly
>update the 'view' of the same text-file while it's changing,
>repeatedly, like a normal editor would show the search
>and replaceAll?
every 10 years, and generally forget everything I learned before
I am forced to do it again. Right now its been almost that long,
and it would be like pulling teeth for me to get up to speed on
it. However (and note that I use XEmacs rather than GNU Emacs),
this is a part of my gnus.el file, and is an email scrubbing
routine. It doesn't do much, but it might give you an idea
how to accomplish what you want using eLisp.
;;;
;;; Fix email buffers.
;;;
(defvar fld-fix-mail-status t)
(defun fld-fix-mail ()
"Automatically edit email buffers"
(interactive)
(if fld-fix-mail-status
(save-excursion
(goto-char (point-min))
(replace-regexp "[Bb][Aa][Cc][Kk][Gg][Rr][Oo][Uu][Nn][dD].*:.*#.*;" "background: #030303;")
(goto-char (point-min))
(replace-regexp "[Bb][Gg][Cc][Oo][Ll][Oo][Rr].*:.*#.*;" "bgcolor: #040404;")
(goto-char (point-min))
(replace-regexp "[Bb][Gg][Cc][Oo][Ll][Oo][Rr].*=.*>" "bgcolor=#050505 text=#f5f5f5>")
(goto-char (point-min))
(replace-regexp "\"[Cc][Oo][lL][Oo][rR]:.*\"" "\"color: #f8f8f8\"")
(goto-char (point-min))
(replace-regexp " [Cc][Oo][lL][Oo][rR]:.*:" " color: #fbfbfb:")
(goto-char (point-min))
(replace-regexp " [Cc][Oo][lL][Oo][rR]:.*[!f][!a].* " " color: #fafafa ")
(goto-char (point-min))
(replace-string "\r\n" "\n")
(set-mark nil)
(when (featurep 'xemacs)
(zmacs-deactivate-region)) ; clears highlighted region
(message "Email **FIXED**"))))
(add-hook 'nnmail-prepare-incoming-hook 'fld-fix-mail)
Avoi...@gmail.com
Nov 11, 2012, 1:36:42 PM11/11/12
to
> The error message says that CHKDSK, on a Windows system,
> should be used to correct the problem.
>
== these logged tests seem interesting===========
> should be used to correct the problem.
>
Sunday: Immediately after booting kogi, `df` shows sda1 mounted
ie. it was auto-mounted [with the other 19 'entries' of `df`].
=> now let's save this /dev/sda1 text.
==> and exit `pico` and re-enter `pico`
===> So kogi CAN R/W the otherwise problematic ntfs !
=> let's try `sfdisk` ? After umounting.
=?=> why reported as 'busy' ?
# lsof | grep sda
xfsdatad/ 412 root cwd DIR 0,14 300 2 /
xfsdatad/ 412 root rtd DIR 0,14 300 2 /
xfsdatad/ 412 root txt unknown /proc/412/exe
mount.ntf 6330 root 3u BLK 8,1 7104 /dev/sda1
bash 7138 root cwd DIR 8,1 4096 5 /mnt/sda1
=?=> booter called `mount.ntf`
-> which mount.ntf ==
which: no mount.ntf in (/usr/local/sbin:/usr/sbin:/sbin:/usr/local/bin:/usr/bin:/bin:/usr/games:/usr/lib/qt/bin)
-> pgrep mount.ntf == 6330
-> pstree -p
init(1)-+-acpid(6473)
|-agetty(7143)
|-agetty(7144)
|-agetty(7145)
|-atd(6816)
|-avahi-daemon(6937)---avahi-daemon(6939)
|-bash(7138)---pico(12197)
|-bash(7141)
|-bash(7142)---pstree(13511)
|-crond(6814)
|-cupsd(6778)
|-dbus-daemon(6421)
|-gpm(10837)
|-hald(6511)---hald-runner(6512)-+-hald-addon-acpi(6547)
| |-hald-addon-inpu(6538)
| `-hald-addon-stor(6548)
|-klogd(6382)
|-mount.ntfs-3g(6330) <-- next one after udevd(5056) *!
|-ntpd(6465)---ntpd(6471)
|-rpc.portmap(6436)
|-rpc.statd(6441)
|-sshd(6456)
|-syslogd(6378)
|-udevd(5056)
|-wicd(6427)
`-wicd-monitor(6524)
=> since $PATH seems restricted to /usr:
-> find /bin -name "*ntfs*"
/bin/ntfs-3g <-- *?
/bin/ntfs-3g.probe
/bin/ntfscat
/bin/ntfscluster
/bin/ntfscmp
/bin/ntfsfix
/bin/ntfsinfo
/bin/ntfsls
-> ls -l /bin/ntfs*
-rwxr-xr-x 1 root root 35616 2009-06-29 20:20 /bin/ntfs-3g*
-rwxr-xr-x 1 root root 8592 2009-06-29 20:20 /bin/ntfs-3g.probe*
-rwxr-xr-x 1 root root 22328 2009-06-29 20:20 /bin/ntfscat*
-rwxr-xr-x 1 root root 27628 2009-06-29 20:20 /bin/ntfscluster*
-rwxr-xr-x 1 root root 26512 2009-06-29 20:20 /bin/ntfscmp*
-rwxr-xr-x 1 root root 24400 2009-06-29 20:20 /bin/ntfsfix*
-rwxr-xr-x 1 root root 47080 2009-06-29 20:20 /bin/ntfsinfo*
-rwxr-xr-x 1 root root 24384 2009-06-29 20:20 /bin/ntfsls*
==> via `less`: these seem all to be binaries.
PS. I'm pasting this all into Toshiba under kogi.
Now I'll reconfigure the hardware to boot slak13 to post-Inet.
PSS. Even under kogi, Umount /dev/sda1 is a PROBLEM
I'll exit here and first try to kill:
|-mount.ntfs-3g(6330)
|-ntpd(6465)---ntpd(6471)
and then `umount /dev/sda1`
---
After killing 6330, I was able to `umount /dev/sda1`
But then could NOT see the dir, despite `REmount it` reporting ok.
Which suggests that it's the special procedure that kogi does
during boot, that enables R/W mounting of the ntfs; and probably
a related special command is needed to <umount>.
============
OMG, after rebooting slakware13, and immediately doing a `df`,
which confirmed that the /etc/fstab mount had failed, and noticing
out of the corner of my eye that Toshiba's LED was 'active',
I started reading PV's story about <Protestant Europe & IBM>,
and before I'd finished, `df` returned, showing sda1 mounted!!
I hadn't realised that `df` was still busy.
So I imediately open this log, but Oh-crap, I can't save it
because slak13 mounted it as RO. So I'll just paste it out.
------------------------
It would be interesting to know what tricks kogi is using,
to mount the file R/W? But I'm out of ideas.
Perhaps the Toshiba partition *IS* corrupted as Debian reports,
but I suspect that Debian just can't manage it.
Avoi...@gmail.com
Nov 12, 2012, 3:09:18 AM11/12/12
to
In article <878va86...@pix.apaflo.com>, fl...@apaflo.com (Floyd L. Davidson) wrote:
-- snip --
The required command: `ntfs-3g /dev/sda1 /mnt/tmp`
is less interesting than the ROUTE to find it:--
1. a slakware derived distribution named kogoni auto-mounted
all visible partitions, including the problematic sda1, at boot time.
2. How/what was kogi doing ?
3. `pstree -p` together with the new key-word "ntfs", led to the
docos via, `which`, `--help`, `man`.
Understandably the response delay [what do we expect with
4 lines instead of 44] gives unexpected behaviour, eg.
-> ntfs-3g /dev/sda1 /mnt/tmp ==
Failed to write lock '/dev/sda1': Resource temporarily unavailable
Error opening '/dev/sda1': Resource temporarily unavailable
Failed to mount '/dev/sda1': Resource temporarily unavailable
==> see LED flashing & try again
-> df | grep sda ==
/dev/sda1 488384532 109524 488275008 1% /mnt/tmp
-> ls /mnt/tmp == <stuff>
-> mount == ...
/dev/sda1 on /mnt/tmp type fuseblk (rw,allow_other,blksize=4096)
I must admit that I saw the key-phrase "ntfs-3g" some where at
the beginning of this long investigation, but failed to follow it.
===========================
> Avoi...@gmail.com wrote:
> >Do you think emacs can use the awk-script to repeatedly
> >update the 'view' of the same text-file while it's changing,
> >repeatedly, like a normal editor would show the search
> >and replaceAll?
>
> I'm sure that emacs can do that, but I mess with eLisp about
> every 10 years, and generally forget everything I learned before
> I am forced to do it again. Right now its been almost that long,
> and it would be like pulling teeth for me to get up to speed on
> it. However (and note that I use XEmacs rather than GNU Emacs),
> this is a part of my gnus.el file, and is an email scrubbing
> routine. It doesn't do much, but it might give you an idea
> how to accomplish what you want using eLisp.
>
> ;;;
> ;;; Fix email buffers.
> ;;;
> (defvar fld-fix-mail-status t)
> (defun fld-fix-mail ()
> "Automatically edit email buffers"
>
> (interactive)
> (if fld-fix-mail-status
....
Thanks, I'll experiment with that.
-- snip --
The required command: `ntfs-3g /dev/sda1 /mnt/tmp`
is less interesting than the ROUTE to find it:--
1. a slakware derived distribution named kogoni auto-mounted
all visible partitions, including the problematic sda1, at boot time.
2. How/what was kogi doing ?
3. `pstree -p` together with the new key-word "ntfs", led to the
docos via, `which`, `--help`, `man`.
Understandably the response delay [what do we expect with
4 lines instead of 44] gives unexpected behaviour, eg.
-> ntfs-3g /dev/sda1 /mnt/tmp ==
Failed to write lock '/dev/sda1': Resource temporarily unavailable
Error opening '/dev/sda1': Resource temporarily unavailable
Failed to mount '/dev/sda1': Resource temporarily unavailable
==> see LED flashing & try again
-> df | grep sda ==
/dev/sda1 488384532 109524 488275008 1% /mnt/tmp
-> ls /mnt/tmp == <stuff>
-> mount == ...
/dev/sda1 on /mnt/tmp type fuseblk (rw,allow_other,blksize=4096)
I must admit that I saw the key-phrase "ntfs-3g" some where at
the beginning of this long investigation, but failed to follow it.
===========================
> Avoi...@gmail.com wrote:
> >Do you think emacs can use the awk-script to repeatedly
> >update the 'view' of the same text-file while it's changing,
> >repeatedly, like a normal editor would show the search
> >and replaceAll?
>
> I'm sure that emacs can do that, but I mess with eLisp about
> every 10 years, and generally forget everything I learned before
> I am forced to do it again. Right now its been almost that long,
> and it would be like pulling teeth for me to get up to speed on
> it. However (and note that I use XEmacs rather than GNU Emacs),
> this is a part of my gnus.el file, and is an email scrubbing
> routine. It doesn't do much, but it might give you an idea
> how to accomplish what you want using eLisp.
>
> ;;;
> ;;; Fix email buffers.
> ;;;
> (defvar fld-fix-mail-status t)
> (defun fld-fix-mail ()
> "Automatically edit email buffers"
>
> (interactive)
> (if fld-fix-mail-status
Thanks, I'll experiment with that.
Avoi...@gmail.com
Nov 12, 2012, 10:08:49 AM11/12/12
to
In article <m3it985gn5kl2nrqn...@4ax.com>, Grant <o...@grrr.id.au> wrote:
> On Fri, 09 Nov 2012 23:29:44 -0900, fl...@apaflo.com (Floyd L. Davidson) wrote:
>
> >Avoi...@gmail.com wrote:
> >>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> >> has M$- EOL, but my CRGlog has *nix-EOL.
> >> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
> >>
> >>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
> >>can?
>
> On Fri, 09 Nov 2012 23:29:44 -0900, fl...@apaflo.com (Floyd L. Davidson) wrote:
>
> >Avoi...@gmail.com wrote:
> >>I've never before used HPFS/NTFS, and I note that 'their file:'autorun.inf
> >> has M$- EOL, but my CRGlog has *nix-EOL.
> >> IIRC vfat defaults-mounts with execute-enabled, but not this HPFS/NTFS.
> >>
> >>I wonder why slackware-13 can't handle it, whereas slak-derived-kogi
> >>can?
>
> Kernel ntfs write is not complete, nor safe.
>
> These days it is much better to use ntfs-3g instead. The ntfs-3g suite is
> very reliable though slow due to being run in user space (FUSE).
>
> Grant.
Hi Grant,
>
> These days it is much better to use ntfs-3g instead. The ntfs-3g suite is
> very reliable though slow due to being run in user space (FUSE).
>
> Grant.
in tracing back your post, I'm wondering why I didn't follow your pointer
to ntfs-3g. "ntfs-3g suite" sounds intimidating, since I considered the
linux-capable fs/S to be settled.
After the exhausing research, I've been writing to the single partitioned
disk. But I don't want to write into physical areas which will be needed
for later partitions. Perhaps it just scatters the used nodes?
Previously I just used `cfdisk` to partition.
I guess you know that, that fails for this 500GB sda.
Should I try fdisk, sfdisk or read all the mans for *ntfs* ?
== TIA.
Chick Tower
Nov 12, 2012, 2:02:04 PM11/12/12
to
On 2012-11-11, Avoi...@gmail.com <Avoi...@gmail.com> wrote:
> I started reading PV's story about <Protestant Europe & IBM>,
Patrick Volkerding wrote such a story? Because, in any Slackware forum
> I started reading PV's story about <Protestant Europe & IBM>,
or newsgroup, PV can mean one, and only one, person.
--
Chick Tower
For e-mail: aols2 DOT sent DOT towerboy AT xoxy DOT net
Keith Keller
Nov 12, 2012, 4:11:13 PM11/12/12
to
On 2012-11-12, Chick Tower <c.t...@deadspam.com> wrote:
> On 2012-11-11, Avoi...@gmail.com <Avoi...@gmail.com> wrote:
>> I started reading PV's story about <Protestant Europe & IBM>,
>
> Patrick Volkerding wrote such a story? Because, in any Slackware forum
> or newsgroup, PV can mean one, and only one, person.
If the OP's posting history is any guideline, this comment is just as
> On 2012-11-11, Avoi...@gmail.com <Avoi...@gmail.com> wrote:
>> I started reading PV's story about <Protestant Europe & IBM>,
>
> Patrick Volkerding wrote such a story? Because, in any Slackware forum
> or newsgroup, PV can mean one, and only one, person.
full of crap as the rest of his posts.
--keith
--
kkeller...@wombat.san-francisco.ca.us
(try just my userid to email me)
AOLSFAQ=http://www.therockgarden.ca/aolsfaq.txt
see X- headers for PGP signature information
Grant
Nov 12, 2012, 8:23:07 PM11/12/12
to
On Mon, 12 Nov 2012 13:11:13 -0800, Keith Keller <kkeller...@wombat.san-francisco.ca.us> wrote:
>On 2012-11-12, Chick Tower <c.t...@deadspam.com> wrote:
>> On 2012-11-11, Avoi...@gmail.com <Avoi...@gmail.com> wrote:
>>> I started reading PV's story about <Protestant Europe & IBM>,
>>
>> Patrick Volkerding wrote such a story? Because, in any Slackware forum
>> or newsgroup, PV can mean one, and only one, person.
>
>If the OP's posting history is any guideline, this comment is just as
>full of crap as the rest of his posts.
You got that right ;)
>On 2012-11-12, Chick Tower <c.t...@deadspam.com> wrote:
>> On 2012-11-11, Avoi...@gmail.com <Avoi...@gmail.com> wrote:
>>> I started reading PV's story about <Protestant Europe & IBM>,
>>
>> Patrick Volkerding wrote such a story? Because, in any Slackware forum
>> or newsgroup, PV can mean one, and only one, person.
>
>If the OP's posting history is any guideline, this comment is just as
>full of crap as the rest of his posts.
Grant.
>
>--keith
Unknown
Dec 15, 2012, 6:06:07 AM12/15/12
to
And IF/HOW I was fooled, was that the USB-Samsung DVD W/R of mine has got
a
normal IDE device inside the box. Which was nice, since I could then also
plug it to the old-PC which had no USB. Quite a while back you advised me
to get a <double what ever you're using now>. Since I lag behind the
curve,
I'd just get the common/proven/old/going-cheap which seems like 500GB now.
--
Following my disasterous experience, trying to get an IDE + SATA
motherboard,
where I didn't realise that it/they have a new 24-pin PSU connectors, I.
decided to go USB, before one of my old IDEs crash and I'm really screwed.
Is Alaska also like an island, where they only sell coke & KFC and mirrors
and beads for the natives; so you can't buy and IDE + SATA MoBo, only the
current fashion? I eventually found a 2nd hand one which I suspect I've
destroyed, although I didn't see any smoke.
If I'd know that this Toshiba was just semiconductor, I'd have rather.
bought a 1T bigger-physical device. But the USB2 AND USB3 is what fooled
me, because I didn't want to find that I'm out of date with USB also.
Perhaps semiconductor is not too bad, but you don't want to buy a teaspoon
expecting it to be of stainless-steel and find its plastic when you open
the bag.
What's with the limited R/W cycles with flash-mem these days?
The LED stays off unless it's accessed.
BTW, I failed to save part of the single/first NTFS partition, because
I couldn't get W98 to see it. It's nice to have a recent M$ on the
1st partition to be able to do exectly tasks like that.
Thanks for the help.
----------------------------------
I read about 'peer pressure on school kids'. Does this persist as they
age,and lead to tribalism: don't look outside of your 'home-town' linux
distro?
And don't report any experiments outside of the normal/popular, because
unfamiliar information causes insecure readers to become agressive.
Unknown
Dec 15, 2012, 6:06:07 AM12/15/12
to
On Fri, 09 Nov 2012 13:09:15 -0900, Floyd L. Davidson wrote:
0 new messages