http://blog.gmane.org/gmane.comp.file-systems.ntfs-3g.devel hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/900 <pre>Hi, Georg Bege wrote: Windows only uses a few percent of a 1TB drive, hence most of your data will still present after the setting to factory default, though uneasy to recover. You might prefer /dev/random before giving away your old computer. Jean-Pierre ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 </pre> Jean-Pierre André 2012-04-16T19:12:35 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/899 <pre>Hello guys I've a simple theoretical question, let's say I've got an 1TB HDD with an factory default set of NTFS Partitions. So a part for recovery and another part for data, the harddisk wont be occupied a lot just a single windows 7 inst. + few programs (and their data as well as some user data). Now someone accidently executes the recovery program which is found on many end-customer PCs (to renew the whole installation), so it comes that all that data is overwritten... maybe also user data... How high would be the chance (%) that one is able to recover a single file from it? Would such single-overwrite with an recovery installation render this quite impossible? cheers Georg ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 </pre> Georg Bege 2012-04-16T14:32:34 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/898 <pre>Hi, Brian wrote: The error you get results from a check done in a library which is supposed to be common to both situations. Maybe you check the libraries used : ldd ntfsundelete ldd ntfs-3g Also what are your partition parameters ? You should be able to get them from : ntfsinfo -fm /dev/sdc1 If this also fails, please post the output of : head -c 96 /dev/sdc1 | od -t x1 (For the above commands, and ntfsundelete, you must run as root). ------------------------------------------------------------------------------ Better than sec? Nothing is better than sec when it comes to monitoring Big Data applications. Try Boundary one-second resolution app monitoring today. Free. http://p.sf.net/sfu/Boundary-dev2dev </pre> Jean-Pierre André 2012-04-10T14:07:23 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/897 <pre>I have an NTFS partition -- probably created with Win2k which I can mount with no problems on linux using the ntfs-3g tools however when I try to run ntfsundelete on it: # ntfsundelete -s /dev/sdc1 NTFS signature is missing. Failed to mount '/dev/sdc1': Invalid argument The device '/dev/sdc1' doesn't have a valid NTFS. Maybe you selected the wrong device? Or the whole disk instead of a partition (e.g. /dev/hda, not /dev/hda1)? Or the other way around? It does mount fine: # mount -r /dev/sdc1 /mnt # ls /mnt 4d275a43cb65bc484b083111c98d FOUND.000 My Documents $AVG Google old bin Google Earth Recycled DrWatson Google Earth.lnk System Volume Information Any ideas why mount is working with it fine and ntfsundelete is not? I am using ntfs-3g version 2012.1.15AR.1 on Ubuntu Precise. ------------------------------------------------------------------------------ Better than sec? Nothing is better than sec when it comes to monitoring Big Data applications. Try Boundary one-second resolution app monitoring today. Free. http://p.sf.net/sfu/Boundary-dev2dev </pre> Brian 2012-04-10T12:11:40 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/896 <pre> There can be a couple of factors considered. But I do not yet see the real error reason for my NTFS surprises. Regards, Markus ------------------------------------------------------------------------------ Virtualization &amp; Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ </pre> SF Markus Elfring 2012-02-12T09:20:56 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/895 <pre>Hi Markus, SF Markus Elfring wrote: Good news. Do you have an idea of the circumstances ? Unclean unmounting, starting Linux with Windows hibernated, etc. ? Regards Jean-Pierre ------------------------------------------------------------------------------ Virtualization &amp; Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ </pre> Jean-Pierre André 2012-02-12T08:08:58 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/894 <pre> I let my computer check the affected partition a few more times. The missed directories have reappeared. I find it strange how often NTFS data structures became "temporarily" inconsistent on my disks recently. Regards, Markus ------------------------------------------------------------------------------ Virtualization &amp; Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ </pre> SF Markus Elfring 2012-02-12T06:55:30 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/893 <pre>Hi Markus, SF Markus Elfring wrote: This is indeed the root (inode 5), not a junction. Actually, the value to be expected was about 170GB, because du does not take the metadata into account, but 104GB is strange. ntfsclone does it (as root and device not mounted) 1) ntfsclone -m -t -O E.metadata /dev/sda6 (skip option -t if you have an old version) this creates a sparse file E.metadata with no user data. This appears as a big file. To make it a normal file, you can use tar with option -S or run ntfsclone again to build a special file : 2a) tar -Scf E.metadata.tar E.metadata 2b) ntfsclone -s -O E.metadata.img E.metadata This will result in a 20GB to 30GB file. Compress it (prefer bz2 or xz), and put it on a server where I can fetch it (and send its location in a private email). Regards Jean-Pierre ------------------------------------------------------------------------------ Virtualization &amp; Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ </pre> Jean-Pierre André 2012-02-09T08:14:51 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/892 <pre> 5 drwxr-xr-x 1 root users 4096 Feb 1 10:20 /windows/E 104G /windows/E 1.0K /windows/E/$RECYCLE.BIN How should the needed meta-data portion be extracted for my open issue? Regards, Markus ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> SF Markus Elfring 2012-02-08T22:07:17 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/891 <pre>Hi again, SF Markus Elfring wrote: I you showed the output of "ls -ladi /windows/E" I could be sure you are accessing the root of the file system. Get the total size of reachable files by "du -s /Windows/E" and check whether you get near 199GB. Check what is in the directory "$RECYCLE.BIN", which is the Windows trash bin. This might be difficult to interpret, but check its size "du -s '/windows/E/$RECYCLE.BIN'" (be sure to escape the '$') Nothing special with these names. This does not look like a permission issue. I assume you were accessing as root or as a user from the group users. This is very reliable. It outputs the space in use, so you can check whether your missing files are somewhere on the disk. Note that chkdsk may relocate orphaned files and directories. They are usually found in a special directory (there are none in your display). If you find evidence of unreachable files, and they are important to you, some forensic analysis is possible, you would have to extract the metadata and send them. Regards Jean-Pierre ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> Jean-Pierre André 2012-02-08T15:36:06 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/890 <pre> Version: 2011.4.12-3.1.2 Mount: /dev/disk/by-id/ata-SAMSUNG_SP2504C_S09QJ1UA109248-part6 /windows/E ntfs-3g users,gid=users,fmask=133,dmask=022,locale=de_DE.UTF-8 0 0 Linux Sonne.site 3.1.9-1.4-desktop #1 SMP PREEMPT Fri Jan 27 08:55:10 UTC 2012 (efb5ff4) x86_64 x86_64 x86_64 GNU/Linux Do you need any more details from my openSUSE 12.1 system? - geladen - Heim - Projekte If this would not be a permission issue, I guess that this file system was damaged somehow. This command has checked my disk for consistency recently. Regards, Markus ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> SF Markus Elfring 2012-02-08T14:15:14 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/889 <pre>Hi Markus, SF Markus Elfring wrote: Please indicate your ntfs-3g version, your mount options and your hardware configuration. Apparently /windows/E is a junction to a user's home directory, whereas /dev/sda6 is the full real partition hosting the junction target. Please post the output of "ls -ladi /windows/E" Please indicate the missing directory names. You did not display the hidden files or directories, you should display with "ls -la". This is however probably not the cause, your missing directories must be elsewhere on the partition. If you used special permission setting, please indicate which ones. Yes, usually, if the data exists and the partition is clean. If you have doubts, please start a "chkdsk /f" on a windows box. Regards Jean-Pierre ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> Jean-Pierre André 2012-02-08T13:01:27 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/888 <pre>Hello! I get the following information on my openSUSE 12.1 system. elfring&lt; at &gt;Sonne:~&gt; LANG=C &amp;&amp; df -h /dev/sda6 &amp;&amp; ls -l /windows/E Filesystem Size Used Avail Use% Mounted on /dev/sda6 205G 199G 5.9G 98% /windows/E total 4 drwxr-xr-x 1 root users 0 Jan 13 22:34 $RECYCLE.BIN drwxr-xr-x 1 root users 0 Feb 6 15:17 31aacd3a-93f7-4085-a2ba-5257e3adc132 drwxr-xr-x 1 root users 0 Feb 6 15:02 7714ab39-6974-4986-a5bc-74f23930bd10 drwxr-xr-x 1 root users 0 Jan 24 16:08 Config.Msi drwxr-xr-x 1 root users 0 Feb 24 2008 Corel drwxr-xr-x 1 root users 4096 Jul 27 2010 Dokumente drwxr-xr-x 1 root users 0 Apr 14 2006 bioscomp drwxr-xr-x 1 root users 0 Oct 9 2010 builds drwxr-xr-x 1 root users 0 Feb 6 15:00 c3dc762a-e2fe-4ee7-bb67-89e61f5e83f9 I am missing a few subdirectories in this directory display. They contain bigger files which contribute to the shown storage allocation considerably. I assume that special permission settings prevent a complete directory listing. Is the NTFS-3G software capable to make the other data accessible again from this partition? Regards, Markus ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> SF Markus Elfring 2012-02-08T11:44:44 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/887 <pre>Hi Diego Diego Elio Pettenò wrote: Yes, we had an exceptionally long freeze period since the release candidate. Your patch is already included in the special advanced version 2011.10.9AR.2 with workarounds for OpenIndiana (when used on Linux the workarounds are disabled of course). See http://b.andre.pagesperso-orange.fr/openindiana-ntfs-3g.html Regards Jean-Pierre ------------------------------------------------------------------------------ Try before you buy = See our experts in action! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-dev2 _______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel&lt; at &gt;lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Jean-Pierre André 2012-01-22T20:26:03 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/886 <pre>Il giorno dom, 22/01/2012 alle 18.29 +0200, Szabolcs Szakacsits ha scritto: Any reason why this does not include my patch from December 11 for ntfsprobe? Thanks, </pre> Diego Elio Pettenò 2012-01-22T17:22:25 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/885 <pre> Greetings, We are happy to announce the next stable release of NTFS-3G and ntfsprogs. The package contains the following improvements, changes and fixes: * ntfs-3g: fixed device path canonicalization for use by devmapper * ntfs-3g: fixed setting DOS names when defined with lower-case chars * ntfs-3g: fixed attribute flags controlling recursive writes * ntfs-3g: fixed compilation on OpenIndiana * ntfs-3g: fixed options parsing on OSes with no extended attributes * ntfs-3g: fixed relatime as a default mount option * ntfs-3g: fixed endless recursion when MFT extents are described by themselves * ntfs-3g: fixed the description of inherit option * ntfs-3g: fixed overwriting a truncated file * ntfs-3g: fixed truncation of DOS file names (12 ntfschars, not 12 utf8 chars) * ntfs-3g: fixed the setting of attributes by secaudit (index not synced) * ntfs-3g: faster compression * ntfs-3g: new option delay_mtime to delay updates of mtime+ctime * ntfsfix: new option -d to clear the dirty flag if fix is successful * ntfsfix: fixed volume dirty flag test * ntfsfix: new option to clear the list of bad sectors * ntfsfix: fixed compilation on Sparc * ntfsfix: fixed a bug causing a segmentation violation * ntfsfix: repair self-located MFT data bug * ntfscp: fix free space calculation * ntfscp: support compression * ntfsresize: implemented expanding runlists * ntfsresize: updated the description of the -f option * ntfsresize: expand an NTFS volume downwards * ntfsclone: backup bootsector not be to accounted for * ntfsclone: creating/restoring a metadata image * ntfsundelete: try to recover the file name when undeleting * ntfsundelete: use inode number to name unnamed files * ntfswipe: Big endian and other fixes * secaudit: prefixed owner and group SID in ACL display * library: fixed big cluster support using 4kB sector disk * library: fixed huge data writes * library: use transparent compression by default * library: fixed several bad returns in error conditions * library: enabled getting the sector size on FreeBSD and MacOSX * build: fixed “make libs” to only build libraries The stable release is available at http://www.tuxera.com/ The git repository source code is available at git clone git://ntfs-3g.git.sourceforge.net/gitroot/ntfs-3g/ntfs-3g Many thanks to Jean-Pierre Andre, Erik Larsson, Anton Altaparmakov and many other contributors! Best regards, Tuxera Open Source Team ------------------------------------------------------------------------------ Try before you buy = See our experts in action! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-dev2 _______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel&lt; at &gt;lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Szabolcs Szakacsits 2012-01-22T16:29:51 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/884 <pre>Thinking about the duplicates I added a check to my script to not accidentially overwrite an existing file. Result (plus some minor improvements) attached.------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev_______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f&lt; at &gt;public.gmane.org https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Dennis Schridde 2011-12-27T14:09:59 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/883 <pre>Hi Jean-Pierre! On Tuesday 27 December 2011 09:29:58 Jean-Pierre André wrote: Thanks for this tip! It seems I can recover my files afterall. :) I attached the script I wrote to recover my files. Hope it is useful to somebody. Searching for that file I also found a _restore{UUID} directory. What is that about? To me this would be a huge improvement. I would not have to resort to chkdsk logs and parsing stat to regularly fix my filenames, if the Windows name would persist. You mean potential name clashes? To piss off Linux users? Thanks for your detailed answer, Dennis------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev_______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f&lt; at &gt;public.gmane.org https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Dennis Schridde 2011-12-27T13:48:40 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/882 <pre>Hi Dennis, Dennis Schridde wrote: The chkdsk for Windows 7 creates a log in "/System Volume Information/Chkdsk/Chkdsk*.log" which contains the old name and the inode number, such as : "Deleted invalid filename foo:bar (83) in directory 30." You can query the inode number of "file1234.chk" with stat or option -i of ls ("ls -li file1234.chk"), and associate the old name according to inode number. But of course what is legal on Windows may be changed by Microsoft without prior notice... Deletion of file names containing some reserved characters is new to chkdsk for Windows 7. Older variants did not delete the file names which were legal in the Posix name space. This is probably possible in general, using hard links, but it would be complex : - when deleting such a file, a search for a possible alternate name would have to be done. - when doing a directory list, returning both names would have to be avoided. In such a case, the Posix name is deleted by chkdsk and the Windows name is kept (no file1234.chk is created), so there is no real improvement. Replacing reserved characters by underscores would create duplicates. The option "windows_names" forbids the creation of file names not compatible with Windows (reserved characters, names ending with a space or a dot). Not possible. My opinion is that the deletion of such file names has been added intentionally by Microsoft. That would be a bug in chkdsk for Windows 7. None that I know of. A possible way is to build a shell script based on find(1), but you will have to deal with duplicates. Currently the only way is to use the option "windows_names". This has no effect on existing files. Regards Jean-Pierre ------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev </pre> Jean-Pierre André 2011-12-27T08:29:58 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/881 <pre>Hello! I just ran Windows 7 SP1's chkdsk on an NTFS partition that I usually work on from Linux using NTFS-3G 2011.4.12. chkdsk decided to delete a lot of filenames because they were "invalid" (that seems to mean they contain characters like : or ? which Windows does not like). Now I am in a state where I have a few dozen file1234.chk filenames and dir1234.chk directories, have to figure out what they contain and properly rename them. (At least it did not entirely delete the files...) The NTFS-3G manual says about this in the "Windows Filename Compatibility" section: "[ntfs-3g] always creates new files in the POSIX namespace for maximum portability and interoperability reasons. [...] This is perfectly legal on Windows, [...]" Now my questions/ideas to prevent such mess in the future: * Is it possible to instruct ntfs-3g to create both, a Windows and a POSIX filename? * Can it map "invalid" characters of the POSIX filename to e.g. underscores _ in the Windows filename? Does "windows_names" do this? * Is it possible to put the whole partition into POSIX mode, so that Windows and tools like chkdsk would let those filenames alone? * Is it possible that there are filenames on my disk, which are not in the POSIX namespace, but not conforming to the Windows namespace, either? How could I fix this? * Is there a free (preferably FOSS) chkdsk tool that behaves more sane, e.g. by renaming the filenames to such with invalid characters replaced by underscores _? * Is there any other way I can prevent the problem in the future? Kind regards, Dennis------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev_______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f&lt; at &gt;public.gmane.org https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Dennis Schridde 2011-12-27T02:18:48 http://permalink.gmane.org/gmane.comp.file-systems.ntfs-3g.devel/880 <pre>Hi Diego, Patch merged for future versions. Thanks. Jean-Pierre ------------------------------------------------------------------------------ Learn Windows Azure Live! Tuesday, Dec 13, 2011 Microsoft is holding a special Learn Windows Azure training event for developers. It will provide a great way to learn Windows Azure and what it provides. You can attend the event by watching it streamed LIVE online. Learn more at http://p.sf.net/sfu/ms-windowsazure _______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel&lt; at &gt;lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel </pre> Jean-Pierre ANDRE 2011-12-11T07:39:41 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.file-systems.ntfs-3g.devel