bindfs - Project Hosting on Google Code

Last 7 days Dec 28, 2009 issue 10 (Cannot mount from one drive to another under MacOSX) changed by martin.partel - Sorry for the delay, I didn't have access to a mac earlier. Unfortunately I wasn't able to reproduce the problem. I don't see how bindfs could cause errors such as yours, since it accesses the source directory just like any other program. Perhaps there's something wrong with your MacFUSE version. The latest release of MacFUSE seems to be 2.0.3, but you claim to use 2.1.5, which is probably not an official stable release. Could you try a stable MacFUSE? You can also enable debug messages on most FUSE filesystems by using the -d option. Status: Unreproducible Owner: martin.partel Sorry for the delay, I didn't have access to a mac earlier. Unfortunately I wasn't able to reproduce the problem. I don't see how bindfs could cause errors such as yours, since it accesses the source directory just like any other program. Perhaps there's something wrong with your MacFUSE version. The latest release of MacFUSE seems to be 2.0.3, but you claim to use 2.1.5, which is probably not an official stable release. Could you try a stable MacFUSE? You can also enable debug messages on most FUSE filesystems by using the -d option. Status: Unreproducible Owner: martin.partel Last 30 days Dec 21, 2009 issue 10 (Cannot mount from one drive to another under MacOSX) reported by henrik.aronsen - What steps will reproduce the problem? bindfs /Applications /Volumes/external_drive/Applications ls -l /Volumes/external_drive/Applications ls: : Device not configured bindfs /Volumes/external_drive/stuff /tmp/stuff ls -l /tmp/stuff ls: /tmp/stuff: Input/output error What version of bindfs and FUSE are you using? On what operating system? bindfs 1.8.3 MacFUSE 2.1.5 MacOSX 10.6.2 What steps will reproduce the problem? bindfs /Applications /Volumes/external_drive/Applications ls -l /Volumes/external_drive/Applications ls: : Device not configured bindfs /Volumes/external_drive/stuff /tmp/stuff ls -l /tmp/stuff ls: /tmp/stuff: Input/output error What version of bindfs and FUSE are you using? On what operating system? bindfs 1.8.3 MacFUSE 2.1.5 MacOSX 10.6.2 Older Oct 30, 2009 issue 9 (Missing dependency on fuse-utils in Debian package) changed by martin.partel - It's good that this was brought up. The next version of Debian's bindfs package will depend on fuse-utils. Summary: Missing dependency on fuse-utils in Debian package Status: Fixed Owner: martin.partel Labels: Type-Other Priority-Low OpSys-Linux −Type-Defect −Priority-Medium It's good that this was brought up. The next version of Debian's bindfs package will depend on fuse-utils. Summary: Missing dependency on fuse-utils in Debian package Status: Fixed Owner: martin.partel Labels: Type-Other Priority-Low OpSys-Linux −Type-Defect −Priority-Medium Oct 30, 2009 issue 9 (bindfs and fstab > can't mount) commented on by coudysk - my fault, problem solved with apt-get install fuse-utils my fault, problem solved with apt-get install fuse-utils Oct 30, 2009 issue 9 (bindfs and fstab > can't mount) reported by coudysk - What steps will reproduce the problem? 1. add to /etc/fstab bindfs#/home/share/serviscs /home/www/serviscs fuse perms=0000:u+rD 0 0 2. mount -a What is the expected result? What do you see instead? mount device What version of bindfs and FUSE are you using? On what operating system? dpkg -l \| grep -i fuse ii libfuse2 2.7.4-1.1 Filesystem in USErspace library dpkg -l \| grep -i bindfs ii bindfs 1.8-1 mirrors or overlays a local directory with altered permissions uname -a Linux server 2.6.26-2-686 #1 SMP Sat Oct 17 17:59:23 UTC 2009 i686 GNU/ Linux I'm using Debian Lenny 5.0 Please provide any additional information below. mount -a gives me mount: wrong fs type, bad option, bad superblock on bindfs#/home/share/ serviscs, missing codepage or helper program, or other error (for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount.<type> helper program) In some cases useful info is found in syslog - try dmesg \| tail or so and nothing is in dmesg or /var/log/messages What steps will reproduce the problem? 1. add to /etc/fstab bindfs#/home/share/serviscs /home/www/serviscs fuse perms=0000:u+rD 0 0 2. mount -a What is the expected result? What do you see instead? mount device What version of bindfs and FUSE are you using? On what operating system? dpkg -l \| grep -i fuse ii libfuse2 2.7.4-1.1 Filesystem in USErspace library dpkg -l \| grep -i bindfs ii bindfs 1.8-1 mirrors or overlays a local directory with altered permissions uname -a Linux server 2.6.26-2-686 #1 SMP Sat Oct 17 17:59:23 UTC 2009 i686 GNU/ Linux I'm using Debian Lenny 5.0 Please provide any additional information below. mount -a gives me mount: wrong fs type, bad option, bad superblock on bindfs#/home/share/ serviscs, missing codepage or helper program, or other error (for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount.<type> helper program) In some cases useful info is found in syslog - try dmesg \| tail or so and nothing is in dmesg or /var/log/messages Apr 20, 2009 issue 8 (Support multiple directories joining) commented on by angelo.c...@gmail.com - Right. I haven't found them by myself. Thanks Right. I haven't found them by myself. Thanks Apr 20, 2009 issue 8 (Support multiple directories joining) changed by martin.partel - There already exist file systems that do this in both userspace (http://apps.sourceforge.net/mediawiki/fuse/index.php?title=UnionFileSystems) and in the Linux kernel (http://www.filesystems.org/project-unionfs.html). You should be able to mount bindfs on top of unionfs and vice versa. I don't see the benefit of duplicating the functionality of unionfs in bindfs. Status: WontFix Labels: Type-Enhancement −Type-Defect There already exist file systems that do this in both userspace (http://apps.sourceforge.net/mediawiki/fuse/index.php?title=UnionFileSystems) and in the Linux kernel (http://www.filesystems.org/project-unionfs.html). You should be able to mount bindfs on top of unionfs and vice versa. I don't see the benefit of duplicating the functionality of unionfs in bindfs. Status: WontFix Labels: Type-Enhancement −Type-Defect Apr 20, 2009 issue 8 (Support multiple directories joining) reported by angelo.castellani - IMHO will be very interesting joining multiple directory contents through bindfs and fuse. You can select various folders containing similar content (network folders, removable devices folders) and see all them in the same.. You may list directory to be joined on mount and with intelligent detection missing folders are ignored until they are present (with collisions handling). I think these feature will be very interesting to be integrated in NG desktop environments. Angelo IMHO will be very interesting joining multiple directory contents through bindfs and fuse. You can select various folders containing similar content (network folders, removable devices folders) and see all them in the same.. You may list directory to be joined on mount and with intelligent detection missing folders are ignored until they are present (with collisions handling). I think these feature will be very interesting to be integrated in NG desktop environments. Angelo Mar 30, 2009 issue 7 (Patch to change ctime to be equal to mtime) commented on by shinys...@yahoo.co.uk - nice one cheers! Shez nice one cheers! Shez Mar 28, 2009 issue 7 (Patch to change ctime to be equal to mtime) changed by martin.partel - A bit of an obscure feature, but why not. I added it to 1.8.3 and named it --ctime-from-mtime because IMO that makes the direction of the copying more clear. Status: Fixed Owner: martin.partel Labels: Type-Enhancement −Type-Defect A bit of an obscure feature, but why not. I added it to 1.8.3 and named it --ctime-from-mtime because IMO that makes the direction of the copying more clear. Status: Fixed Owner: martin.partel Labels: Type-Enhancement −Type-Defect Mar 28, 2009 News Wiki page edited by martin.partel Mar 28, 2009 bindfs-1.8.3.tar.gz (bindfs 1.8.3 source) file uploaded by martin.partel - Labels: Featured Type-Source OpSys-Linux OpSys-OSX Labels: Featured Type-Source OpSys-Linux OpSys-OSX Mar 28, 2009 bindfs-1.8.3.tar.gz (bindfs 1.8.3 source) file uploaded by martin.partel - Labels: Featured Type-Source OpSys-Linux OpSys-OSX Labels: Featured Type-Source OpSys-Linux OpSys-OSX Mar 27, 2009 issue 7 (Patch to change ctime to be equal to mtime) reported by shinys...@yahoo.co.uk - Hey, We're using a small patch to bindfs-1.8.2 to help us around an issue with using NetVaults to backup rsnapshot directories. NetVaults by checks the ctime which is constantly changing due to hardlinks being created and removed when rsnapshot runs. To work around this we've implemented the following patch. If its got any legs please feel free to incorporate into the tree. Cheers Shez Hey, We're using a small patch to bindfs-1.8.2 to help us around an issue with using NetVaults to backup rsnapshot directories. NetVaults by checks the ctime which is constantly changing due to hardlinks being created and removed when rsnapshot runs. To work around this we've implemented the following patch. If its got any legs please feel free to incorporate into the tree. Cheers Shez Mar 01, 2009 Installation (Installation instructions.) Wiki page edited by martin.partel Jan 24, 2009 Installation (Installation instructions.) Wiki page edited by martin.partel Jan 24, 2009 Installation (Installation instructions.) Wiki page added by martin.partel Dec 14, 2008 issue 6 (License clarification) commented on by opensource@till.name - Thanks for mentioning the URL, I will update it. Thanks for mentioning the URL, I will update it. Dec 14, 2008 issue 6 (License clarification) Status changed by martin.partel - Great. I'll close this then. Thanks for packaging bindfs! I noticed your .spec file still points to the old home page. You should probably update that. Status: Fixed Great. I'll close this then. Thanks for packaging bindfs! I noticed your .spec file still points to the old home page. You should probably update that. Status: Fixed Dec 14, 2008 issue 6 (License clarification) commented on by opensource@till.name - IANAL, but this is how I understand the sentence from the GPL license I quoted and this is what the Fedora License Guidelines say, search for GPL+ there: https://fedoraproject.org/wiki/Licensing Nevertheless it would be probably best to ask fuse to clarify this in their source files to avoid any confusion. The bindfs license header look good to me now. IANAL, but this is how I understand the sentence from the GPL license I quoted and this is what the Fedora License Guidelines say, search for GPL+ there: https://fedoraproject.org/wiki/Licensing Nevertheless it would be probably best to ask fuse to clarify this in their source files to avoid any confusion. The bindfs license header look good to me now. Dec 13, 2008 News Wiki page edited by martin.partel Dec 13, 2008 issue 6 (License clarification) Status changed by martin.partel - Thanks for reporting. I released 1.8.2 specifying GPL v2+ in each source file and the README file. Is it OK now? I started coding from a FUSE example file, which, like all other FUSE files, only had the short notice you quoted. Does this mean that for FUSE one "may choose any version ever published by the Free Software Foundation"? Status: Accepted Thanks for reporting. I released 1.8.2 specifying GPL v2+ in each source file and the README file. Is it OK now? I started coding from a FUSE example file, which, like all other FUSE files, only had the short notice you quoted. Does this mean that for FUSE one "may choose any version ever published by the Free Software Foundation"? Status: Accepted Dec 13, 2008 bindfs-1.8.2.tar.gz (bindfs 1.8.2 source) file uploaded by martin.partel - Labels: Featured Type-Source OpSys-Linux OpSys-OSX Labels: Featured Type-Source OpSys-Linux OpSys-OSX Dec 12, 2008 issue 6 (License clarification) reported by opensource@till.name - Currently the license headers say: \| This program can be distributed under the terms of the GNU GPL. \| See the file COPYING. COPYING says this: \| If the Program does not specify a version number of this License, you may \| choose any version ever published by the Free Software Foundation. Is it intended to license bindfs under the terms of any GPL license? This question came up in the Fedora Review of this package: https://bugzilla.redhat.com/show_bug.cgi?id=469027#c4 There is also a nice howto about how to license sofware under the GPL available at: http://www.gnu.org/licenses/gpl-howto.html It also includes the default headers for GPL licensed software. Currently the license headers say: \| This program can be distributed under the terms of the GNU GPL. \| See the file COPYING. COPYING says this: \| If the Program does not specify a version number of this License, you may \| choose any version ever published by the Free Software Foundation. Is it intended to license bindfs under the terms of any GPL license? This question came up in the Fedora Review of this package: https://bugzilla.redhat.com/show_bug.cgi?id=469027#c4 There is also a nice howto about how to license sofware under the GPL available at: http://www.gnu.org/licenses/gpl-howto.html It also includes the default headers for GPL licensed software. Dec 12, 2008 issue 5 (install -m 0644 fails always on bindfs filesystem) commented on by opensource@till.name - Thank you very much for your fast updates, this works perfectly now. Thank you very much for your fast updates, this works perfectly now. Dec 11, 2008 News Wiki page edited by martin.partel Dec 11, 2008 issue 5 (install -m 0644 fails always on bindfs filesystem) changed by martin.partel - It turned out install wants to set an xattr named "system.posix_acl_access". bindfs 1.8 (stupidly) defaults to making xattrs read-only, so that gives install a "permission denied". Workaround: give the xattr-rw option. I released bindfs 1.8.1 to change the default to xattr-rw. Status: Fixed Owner: martin.partel Labels: Type-Enhancement −Type-Defect It turned out install wants to set an xattr named "system.posix_acl_access". bindfs 1.8 (stupidly) defaults to making xattrs read-only, so that gives install a "permission denied". Workaround: give the xattr-rw option. I released bindfs 1.8.1 to change the default to xattr-rw. Status: Fixed Owner: martin.partel Labels: Type-Enhancement −Type-Defect Dec 11, 2008 bindfs-1.8.1.tar.gz (bindfs 1.8.1 source) file uploaded by martin.partel - Labels: Featured Type-Source OpSys-Linux OpSys-OSX Labels: Featured Type-Source OpSys-Linux OpSys-OSX Dec 11, 2008 issue 5 (install -m 0644 fails always on bindfs filesystem) reported by opensource@till.name - What steps will reproduce the problem? 1. in my /etc/fstab I have this: bindfs#/home/rpmbuilder/rpmbuild-orig/ /home/rpmbuilder/rpmbuild fuse mirror=rpmbuilder:till,perms=0644:a+X,create-for-user=rpmbuilder,create-for-group=rpmbuilder,create-with-perms=0664:a+X,chmod-normal 0 0 2. The filesystem is mounted at bootup, i.e. by root 3. In the directory where the bindfs is mounted, this fails 4. touch foo; install -m 0664 foo foo2 install: setting permissions for `foo2': Permission denied What is the expected output? What do you see instead? I would like to have install to suceed somehow, whatever it takes. Given that the bindfs is mounted as root and this sentence from the manpage, this should not fail: \| This will succeed if the user has the appropriate mirrored permissions \| to chmod the mirrored file AND the mounter has enough permissions \| to chmod the real file. Also "chmod 0664 foo" does not fail directly, but it also does not really chmod the file. What version of the product are you using? On what operating system? bindfs: 1.8 on Fedora 8 Kernel: 2.6.25.14-69.fc8 Fuse: 2.7.3-3.fc8 What steps will reproduce the problem? 1. in my /etc/fstab I have this: bindfs#/home/rpmbuilder/rpmbuild-orig/ /home/rpmbuilder/rpmbuild fuse mirror=rpmbuilder:till,perms=0644:a+X,create-for-user=rpmbuilder,create-for-group=rpmbuilder,create-with-perms=0664:a+X,chmod-normal 0 0 2. The filesystem is mounted at bootup, i.e. by root 3. In the directory where the bindfs is mounted, this fails 4. touch foo; install -m 0664 foo foo2 install: setting permissions for `foo2': Permission denied What is the expected output? What do you see instead? I would like to have install to suceed somehow, whatever it takes. Given that the bindfs is mounted as root and this sentence from the manpage, this should not fail: \| This will succeed if the user has the appropriate mirrored permissions \| to chmod the mirrored file AND the mounter has enough permissions \| to chmod the real file. Also "chmod 0664 foo" does not fail directly, but it also does not really chmod the file. What version of the product are you using? On what operating system? bindfs: 1.8 on Fedora 8 Kernel: 2.6.25.14-69.fc8 Fuse: 2.7.3-3.fc8 Aug 17, 2008 News Wiki page edited by martin.partel Aug 17, 2008 issue 4 (segmentation fail on x86_64) Status changed by martin.partel - Found it and fixed it in 1.8. Thanks again for reporting! Status: Fixed Found it and fixed it in 1.8. Thanks again for reporting! Status: Fixed Aug 04, 2008 issue 4 (segmentation fail on x86_64) commented on by jphokz - Hello, i'll try it, in meanwhile please send me your ssh key and desired username and I'll give you some account on 64bit machine. My fuse version is fuse-2.7.3-1.el5.rf I'll also try the same problem on ubuntu machine, which has more recent packages. Hello, i'll try it, in meanwhile please send me your ssh key and desired username and I'll give you some account on 64bit machine. My fuse version is fuse-2.7.3-1.el5.rf I'll also try the same problem on ubuntu machine, which has more recent packages. Aug 01, 2008 issue 4 (segmentation fail on x86_64) changed by martin.partel - Thanks for reporting! Unfortunately I've no idea what could cause that, and I won't have the opportunity to try bindfs on an x86_64 machine for several weeks. It does work on a PPC, though. Could you try running that through valgrind and post any errors it prints? Compile bindfs with --enable-debug for better backtraces in both gdb and valgrind. Which version of FUSE do you have? It's a possibility that the bug is in libfuse's option parser. Status: Accepted Owner: martin.partel Thanks for reporting! Unfortunately I've no idea what could cause that, and I won't have the opportunity to try bindfs on an x86_64 machine for several weeks. It does work on a PPC, though. Could you try running that through valgrind and post any errors it prints? Compile bindfs with --enable-debug for better backtraces in both gdb and valgrind. Which version of FUSE do you have? It's a possibility that the bug is in libfuse's option parser. Status: Accepted Owner: martin.partel Aug 01, 2008 issue 4 (segmentation fail on x86_64) commented on by jphokz - hardcoding new uid and gid in bindfs.c worked. In my case settings.new_uid=getuid(); and commented out block around line 1015: /* Parse new owner and group / / if (od.user) { if (!user_uid(od.user, &settings.new_uid)) { fprintf(stderr, "Not a valid user ID: %s\n", od.user); return 1; } } / hardcoding new uid and gid in bindfs.c worked. In my case settings.new_uid=getuid(); and commented out block around line 1015: / Parse new owner and group / / if (od.user) { if (!user_uid(od.user, &settings.new_uid)) { fprintf(stderr, "Not a valid user ID: %s\n", od.user); return 1; } } */ Aug 01, 2008 issue 4 (segmentation fail on x86_64) reported by jphokz - What steps will reproduce the problem? 1. on Centos 5.1 x86_64, /root/bindfs-1.7/src/bindfs --owner=root --create-for-user=root --create-with-perms=u+rw,a+rD --perms=0000:u=rwD --chmod-ignore /mnt/nas /mnt/nasfix 2. with only --owner=root it works fine 3. used ulimit -c unlimited and backtraced in gdb: (gdb) bt #0 0x00000030292341ca in ____strtoll_l_internal () from /lib64/libc.so.6 #1 0x00000000004035a2 in user_uid ( username=0xd000000 <Address 0xd000000 out of bounds>, ret=0x605e70) at /usr/include/stdlib.h:336 #2 0x00000000004026b1 in main (argc=<value optimized out>, argv=0x7fffa2dda7a8) at bindfs.c:1015 (gdb) quit What is the expected output? What do you see instead? - Expected behavior is same as on i686 system. - Segfault. What version of the product are you using? On what operating system? Centos 5.1 Linux xenhost 2.6.18-53.1.13.el5xen #1 SMP Tue Feb 12 13:33:07 EST 2008 x86_64 x86_64 x86_64 GNU/Linux bindfs 1.7 gcc: Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --enable-plugin --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux Thread model: posix gcc version 4.1.2 20071124 (Red Hat 4.1.2-42) Please provide any additional information below. What steps will reproduce the problem? 1. on Centos 5.1 x86_64, /root/bindfs-1.7/src/bindfs --owner=root --create-for-user=root --create-with-perms=u+rw,a+rD --perms=0000:u=rwD --chmod-ignore /mnt/nas /mnt/nasfix 2. with only --owner=root it works fine 3. used ulimit -c unlimited and backtraced in gdb: (gdb) bt #0 0x00000030292341ca in ____strtoll_l_internal () from /lib64/libc.so.6 #1 0x00000000004035a2 in user_uid ( username=0xd000000 <Address 0xd000000 out of bounds>, ret=0x605e70) at /usr/include/stdlib.h:336 #2 0x00000000004026b1 in main (argc=<value optimized out>, argv=0x7fffa2dda7a8) at bindfs.c:1015 (gdb) quit What is the expected output? What do you see instead? - Expected behavior is same as on i686 system. - Segfault. What version of the product are you using? On what operating system? Centos 5.1 Linux xenhost 2.6.18-53.1.13.el5xen #1 SMP Tue Feb 12 13:33:07 EST 2008 x86_64 x86_64 x86_64 GNU/Linux bindfs 1.7 gcc: Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --enable-plugin --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux Thread model: posix gcc version 4.1.2 20071124 (Red Hat 4.1.2-42) Please provide any additional information below.

issue 4 (segmentation fail on x86_64) reported by jphokz - What steps will reproduce the problem? 1. on Centos 5.1 x86_64, /root/bindfs-1.7/src/bindfs --owner=root --create-for-user=root --create-with-perms=u+rw,a+rD --perms=0000:u=rwD --chmod-ignore /mnt/nas /mnt/nasfix 2. with only --owner=root it works fine 3. used ulimit -c unlimited and backtraced in gdb: (gdb) bt #0 0x00000030292341ca in ____strtoll_l_internal () from /lib64/libc.so.6 #1 0x00000000004035a2 in user_uid ( username=0xd000000 <Address 0xd000000 out of bounds>, ret=0x605e70) at /usr/include/stdlib.h:336 #2 0x00000000004026b1 in main (argc=<value optimized out>, argv=0x7fffa2dda7a8) at bindfs.c:1015 (gdb) quit What is the expected output? What do you see instead? - Expected behavior is same as on i686 system. - Segfault. What version of the product are you using? On what operating system? Centos 5.1 Linux xenhost 2.6.18-53.1.13.el5xen #1 SMP Tue Feb 12 13:33:07 EST 2008 x86_64 x86_64 x86_64 GNU/Linux bindfs 1.7 gcc: Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --enable-plugin --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux Thread model: posix gcc version 4.1.2 20071124 (Red Hat 4.1.2-42) Please provide any additional information below.

What steps will reproduce the problem? 1. on Centos 5.1 x86_64, /root/bindfs-1.7/src/bindfs --owner=root --create-for-user=root --create-with-perms=u+rw,a+rD --perms=0000:u=rwD --chmod-ignore /mnt/nas /mnt/nasfix 2. with only --owner=root it works fine 3. used ulimit -c unlimited and backtraced in gdb: (gdb) bt #0 0x00000030292341ca in ____strtoll_l_internal () from /lib64/libc.so.6 #1 0x00000000004035a2 in user_uid ( username=0xd000000 <Address 0xd000000 out of bounds>, ret=0x605e70) at /usr/include/stdlib.h:336 #2 0x00000000004026b1 in main (argc=<value optimized out>, argv=0x7fffa2dda7a8) at bindfs.c:1015 (gdb) quit What is the expected output? What do you see instead? - Expected behavior is same as on i686 system. - Segfault. What version of the product are you using? On what operating system? Centos 5.1 Linux xenhost 2.6.18-53.1.13.el5xen #1 SMP Tue Feb 12 13:33:07 EST 2008 x86_64 x86_64 x86_64 GNU/Linux bindfs 1.7 gcc: Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --enable-plugin --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux Thread model: posix gcc version 4.1.2 20071124 (Red Hat 4.1.2-42) Please provide any additional information below.

Last 7 days

Last 30 days

Older