ntfs-3g crashed with SIGABRT in get_node()

[userdce]

ntfs-3g crashed with SIGABRT in raise()

ProblemType: Crash
DistroRelease: Ubuntu 12.10
Package: ntfs-3g 1:2012.1.15AR.5-2ubuntu1
ProcVersionSignature: Ubuntu 3.5.0-2.2-generic 3.5.0-rc4
Uname: Linux 3.5.0-2-generic x86_64
NonfreeKernelModules: wl
ApportVersion: 2.2.5-0ubuntu1
Architecture: amd64
Date: Sun Jul 1 02:40:34 2012
ExecutablePath: /bin/ntfs-3g
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Beta amd64 (20120408)
ProcCmdline: /sbin/mount.ntfs-3g /dev/sda1 /media/data1 -o rw,noexec,nosuid,nodev,user,fmask=0111,dmask=0000,uid=1000
ProcEnviron:
 TERM=linux
 LANG=en_US.UTF-8
Signal: 6
SourcePackage: ntfs-3g
StacktraceTop:
 raise () from /lib/x86_64-linux-gnu/libc.so.6
 abort () from /lib/x86_64-linux-gnu/libc.so.6
 ?? () from /lib/x86_64-linux-gnu/libfuse.so.2
 ?? () from /lib/x86_64-linux-gnu/libfuse.so.2
 ?? () from /lib/x86_64-linux-gnu/libfuse.so.2
Title: ntfs-3g crashed with SIGABRT in raise()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

[Apport retracing service (apport)]

StacktraceTop:
 get_node (nodeid=11, f=<optimized out>) at fuse.c:498
 get_node (f=<optimized out>, nodeid=11) at fuse.c:492
 forget_node (f=f@entry=0x7ffcd2761d90, nodeid=11, nlookup=565) at fuse.c:1241
 do_forget (f=f@entry=0x7ffcd2761d90, ino=<optimized out>, nlookup=<optimized out>) at fuse.c:2619
 fuse_lib_forget_multi (req=0x7ffcd2789fc0, count=987, forgets=<optimized out>) at fuse.c:2635

[Jean-Pierre (jean-pierre-andre)]

This looks like a problem in fuse, not in ntfs-3g. The fuse library is getting a request from the fuse kernel module, and this is a request for forgetting an inode with invalid parameters. The fuse library which is being used is not the one bundled with ntfs-3g.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Jean-Pierre (jean-pierre-andre)]

As I have already said for bug 104944, the patch to fuse for this bug is in http://fuse.git.sourceforge.net/git/gitweb.cgi?p=fuse/fuse;a=commit;h=1061a0a2d90148bd2e7f32e1e694399db2dbe087

Is there no management of duplicated reports ?

[aa-hcl (aa-hcl)]

I have the same report: for unknown reason ntfs-3g crashed with same title "ntfs-3g srashed wtih SGIABRT in raise()". Ubuntu report was sent out automatically.

Is there any way to get it fixed?

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Laurent GRÉGOIRE (laurent-gregoire)]

For information, I installed while ago a patched (http://fuse.git.sourceforge.net/git/gitweb.cgi?p=fuse/fuse;a=commitdiff;h=1061a0a2d90148bd2e7f32e1e694399db2dbe087) fuse package on a system having this issue, and this fixed it. Backporting this patch to the fuse package is easy, and by reading the code changes from the patch this should not be too risky.

[Dimitri John Ledkov (xnox)]

I've uploaded the mentioned patch into raring.
To upload to quantal/precise, I'd need a test case.
Does anyone know an easy way to reproduce this which is also easy to setup? (e.g. not requiring windows or anything heavy like that)

[Launchpad Janitor (janitor)]

This bug was fixed in the package fuse - 2.9.0-1ubuntu3

---------------
fuse (2.9.0-1ubuntu3) raring; urgency=low

  * Fix "fuse internal error: node NNN not found" (LP: #1019806)
 -- Dmitrijs Ledkovs <email address hidden> Thu, 21 Mar 2013 10:39:45 +0000

[Laurent GRÉGOIRE (laurent-gregoire)]

This may reproduce the issue:
- Create a NTFS partition (mkfs.ntfs)
- Copy lots of random files to it
- Run n parallels global md5sum on all the files: "find . -type f | xargs md5sum &"
AFAIR I managed to reliably reproduce the issue with this method.

[Rolf Leggewie (r0lf)]

quantal has seen the end of its life and is no longer receiving any updates. Marking the quantal task for this ticket as "Won't Fix".

[Steve Langasek (vorlon)]

The Precise Pangolin has reached end of life, so this bug will not be fixed for that release