Bug#669314: NFS: kernel forces trailing slash for export in /proc/self/mounts
I've applied the patch to Linus' master HEAD (43c422eda99b894f18d1cca17bcd2401efaf7bd0, at the time) and the patch seems to work fine. /proc/self/mounts correctly reflects whether or not the user specified to use a trailing slash - and nothing is obviously broken. If the patch looks fine to others, I propose we move this patch into wherever is the next step for testing it more broadly: e.g. Debian experimental, and wherever it needs to go to end up in Linus' master branch. Thanks, Chris smime.p7s Description: S/MIME cryptographic signature
Bug#669314: NFS: kernel forces trailing slash for export in /proc/self/mounts
I haven't had a chance to test it yet thanks. I'll be too busy until at least mid-next week, but I will test it then if nobody beats me to it. -Chris On Sep 30, 2012, at 2:23 PM, Jonathan Nieder jrnie...@gmail.com wrote: Thanks for looking it over. Did you get a chance to test Ben's patch? Curious, Jonathan smime.p7s Description: S/MIME cryptographic signature
Bug#669314: NFS: kernel forces trailing slash for export in /proc/self/mounts
I had a couple friends over today and we made a trivial patch to remove trailing slashes. We do not know C and have never created a patch for the kernel before, so there is undoubtedly a better way to do it. However we hope this helps in your efforts. 0001-Fixes-trailing-slash-in-nfs-devname.patch Description: Binary data This patch was created from the offending commit (c7f404b40a366). But I've also applied it to to Linus Torvalds' master HEAD (3f0c3c8fe30c7) with success. -Chris smime.p7s Description: S/MIME cryptographic signature
Bug#669314: NFS: kernel forces trailing slash for export in /proc/self/mounts
On Sep 16, 2012, at 7:00 AM, Ben Hutchings wrote: This was my first thought - but what if userland provides a device name with a slash on the end? I think we have to report it back with the slash in that case. That is a great point. We could not find any standard as to whether or not there should be a trailing slash. And from all the examples I could find, there seems to be a convention of omitting a trailing slash. However, if userland provides a trailing slash, it would seem appropriate to retain it. As a point of comparison, matching the given input is the behavior of 2.6.32-5 in Debian Squeeze. So I think your approach is better. -Chris smime.p7s Description: S/MIME cryptographic signature
Bug#669314: No Reply From Alexander Viro
Just an update, I emailed Alexander Viro twice and I haven't heard back from Alexander Viro or anyone on the kernel development mailing lists. I am surprised that this issue is being ignored. -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/9cd0814e-7d63-4c11-8154-b67715428...@salk.edu
Bug#669314: NFS: kernel forces trailing slash for export in /proc/self/mounts
Hi Alexander Viro et al,
This is an escalation of Debian Bug #669314 http://bugs.debian.org/669314,
which I will
re-elaborate in this email for your convenience.
You committed a change to the way the linux kernel reports NFS mounts - now
with a
trailing slash for the remote export (among other changes). The change happened
here:
commit c7f404b40a3665d9f4e9a927cc5c1ee0479ed8f9
Author: Al Viro v...@zeniv.linux.org.uk
Date: Wed Mar 16 06:59:40 2011 -0400
vfs: new superblock methods to override /proc/*/mount{s,info}
a) -show_devname(m, mnt) - what to put into devname columns in mounts,
mountinfo and mountstats
b) -show_path(m, mnt) - what to put into relative path column in
mountinfo
Leaving those NULL gives old behaviour. NFS switched to using those.
Signed-off-by: Al Viro v...@zeniv.linux.org.uk
The problematic behavior is that NFS exports now have a trailing slash in
/proc/self/mounts.
This still seems to be the case in newer kernels, for example in Debian's
3.3.2-1~experimental.1.
and HEAD in Linus Torvalds' master branch, presently commit:
02edf6abe01610a5fb379df442de3c837ad99467
I believe it is/was convention to leave a trailing slash off of the nfs export
in /etc/fstab, e.g.:
nfsserver:/srv/ubuntu-32 /mnt/ubuntu-32 nfs
ro,nfsvers=3,soft,intr,tcp,nodev,noatime,nosuid,rsize=32768,wsize=32768
and I'd also expect the same in /proc/self/mounts
Expected Result:
I expect /proc/self/mounts to show (notice no trailing slash on the export):
nfsserver:/srv/ubuntu-32 /mnt/ubuntu-32 nfs
ro,nosuid,nodev,noatime,vers=3,rsize=32768,wsize=32768,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,mountaddr=198.202.1.1,mountvers=3,mountport=41576,mountproto=tcp,local_lock=none,addr=198.202.1.1
0 0
Actual Result:
But instead in /proc/self/mounts I get (notice the trailing slash):
nfsserver:/srv/ubuntu-32/ /mnt/ubuntu-32 nfs
ro,nosuid,nodev,noatime,vers=3,rsize=32768,wsize=32768,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,mountaddr=198.202.1.1,mountvers=3,mountport=41576,mountproto=tcp,local_lock=none,addr=198.202.1.1
0 0
Rammifications:
This simple change has a lot of implications because lots of things parse
/etc/fstab and /proc/self/mounts, for example system configuration tools
and mount.nfs and umount.nfs.
If you use the former convention and try to umount an export on a newer
kernel it will fail:
user@hostname:/proc/self$ sudo umount.nfs nfsserver:/srv/ubuntu-32
umount.nfs: nfsserver:/srv/ubuntu-32: not found
And if you run sudo mount -va, mount will not recognize that the fstab mounts
have already been mounted; mounting all mounts twice on the same mount point.
This quickly gets messy.
If there is a new convention to display the trailing slash, we need to update
our tools to handle this change. If there is not a new convention, I'd argue
this is a bug.
So is this a new convention or not? What is the appropriate way for Debian to
move forward?
Thanks,
Chris Hiestand
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8f8193bd-84c6-4905-8789-de7eb2579...@salk.edu
Bug#669314: [squeeze - wheezy regression] nfs: kernel forces trailing slash for export in /proc/self/mounts
On Jun 17, 2012, at 3:31 PM, Jonathan Nieder wrote:
I should have said v2.6.39-rc1~460^2^ (nfs: switch NFS from -get_sb()
to -mount(), 2011-03-16). So:
Thanks Jonathan,
Based on this tip I tested that commit and neighboring commits and was
able to find the commit that introduced the problem:
c7f404b40a3665d9f4e9a927cc5c1ee0479ed8f9
vfs: new superblock methods to override /proc/*/mount{s,info}
The last commit to be problem free was the previous one:
f8ad9c4bae99854c961ca79ed130a0d11d9ab53c
nfs: nfs_do_{ref,sub}mount() superblock argument is redundant
I think this bug should be fixed before Wheezy becomes stable.
Thanks a lot,
Chris
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8f7b9848-6664-4ae6-83ca-4f363be14...@salk.edu
Bug#669314: linux-source-3.2: After mounting nfs export, kernel forces trailing slash for export in /proc/self/mounts
Package: linux-source-3.2 Version: 3.2.15-1 Severity: normal Tags: upstream Dear Maintainer, I believe it is convention to leave a trailing slash off of the nfs export in /etc/fstab, e.g.: nfsserver:/srv/ubuntu-32 /mnt/ubuntu-32 nfs ro,nfsvers=3,soft,intr,tcp,nodev,noatime,nosuid,rsize=32768,wsize=32768 I expect /proc/self/mounts to show (notice no trailing slash on the export): nfsserver:/srv/ubuntu-32 /mnt/ubuntu-32 nfs ro,nosuid,nodev,noatime,vers=3,rsize=32768,wsize=32768,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,mountaddr=198.202.1.1,mountvers=3,mountport=41576,mountproto=tcp,local_lock=none,addr=198.202.1.1 0 0 But instead in /proc/self/mounts I get (notice the trailing slash): nfsserver:/srv/ubuntu-32/ /mnt/ubuntu-32 nfs ro,nosuid,nodev,noatime,vers=3,rsize=32768,wsize=32768,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,mountaddr=198.202.1.1,mountvers=3,mountport=41576,mountproto=tcp,local_lock=none,addr=198.202.1.1 0 0 This simple change has a lot of implications because lots of things parse /proc/self/mounts, for example mount.nfs and umount.nfs user@hostname:/proc/self$ sudo umount.nfs nfsserver:/srv/ubuntu-32 umount.nfs: nfsserver:/srv/ubuntu-32: not found And if you run sudo mount -va, mount will not recognize that the fstab mounts have already been mounted; mounting all mounts twice on the same mount point. This quickly gets messy. If there is a new convention to force the trailing slash in linux-3, we need to update our tools to handle this change. If there is not a new convention, this is a bug. As a workaround, I can boot into the squeeze kernel, linux-image-2.6.32-5-amd64, and the problem no longer reproduces. This is why I believe it is a problem in the kernel and not in nfs-common. Thanks, Chris -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (900, 'testing'), (350, 'testing-proposed-updates') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-2-amd64 (SMP w/12 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages linux-source-3.2 depends on: ii binutils 2.22-6 ii bzip2 1.0.6-1 Versions of packages linux-source-3.2 recommends: ii gcc 4:4.6.2-4 ii libc6-dev [libc-dev] 2.13-27 ii make 3.81-8.1 Versions of packages linux-source-3.2 suggests: ii libncurses5-dev [ncurses-dev] 5.9-4 ii libqt4-dev 4:4.7.4-2 -- no debconf information -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120418224506.5139.4861.report...@sagan.snl.salk.edu
