[Bug 499773] Re: Race with ureadahead can mean that /var/lib/ureadahead/debugfs appears in /etc/mtab

[QuentinHartman]

This is mainly for the benefit of people who search google for this
problem and end up here.

I've seen this on several servers. A simple "umount
/var/lib/ureadahead/debugfs" removes the errant mtab entry and cleans up
df output if it's bothering you.

It's highly unlikely that this mount lying around is causing any actual
problems. The only people who have reported disk full problems
legitimately had full disks, and the performance degredation just
doesn't make sense. It's entirely possible that there is some other
coincident problem that is causing the performance problem that the
above user reported.

If you are having problems with a full disk issues but df is reporting
you have free space, check the output from du. It's possible for a file
to be deleted, but held open by a long-running process which prevents
that space from being actually freed. This will make df report free
space when there actually isn't any. To see if this might be affecting
you, run "lsof | grep deleted". If you see any files listed there,
especially large ones, see if you can stop the process holding them open
and that space will become usable again. It's likely you will see at
least a few files in the list, probably log files, temp files, and lock
files. Those are likely not worth worrying about.

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to mountall in Ubuntu.
https://bugs.launchpad.net/bugs/499773

Title:
  Race with ureadahead can mean that /var/lib/ureadahead/debugfs appears
  in /etc/mtab

Status in “mountall” package in Ubuntu:
  Triaged

Bug description:
  When the system is booted, the mountall process may notice a
  temporarily-mounted file system at /var/lib/ureadahead/debugfs and
  record that fact in /etc/mtab.  Later in the boot process, this file
  system is unmounted, but mountall doesn't record unmounts, so it never
  "forgets" this path when it's unmounted.

  Mountall should if necessary remove this mount from mtab.

  To confirm on your system that it is indeed not really mounted, look for it in /proc/self/mountinfo
  Otherwise this may be due to a crash of the ureadahead process during boot.

  ---- original report ----

  Binary package hint: ureadahead

  $ mount
  /dev/sda1 on / type ext3 (rw,relatime,errors=remount-ro)
  proc on /proc type proc (rw)
  none on /sys type sysfs (rw,noexec,nosuid,nodev)
  none on /sys/fs/fuse/connections type fusectl (rw)
  none on /sys/kernel/debug type debugfs (rw)
  none on /sys/kernel/security type securityfs (rw)
  udev on /dev type tmpfs (rw,mode=0755)
  none on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620)
  none on /dev/shm type tmpfs (rw,nosuid,nodev)
  none on /var/run type tmpfs (rw,nosuid,mode=0755)
  none on /var/lock type tmpfs (rw,noexec,nosuid,nodev)
  none on /lib/init/rw type tmpfs (rw,nosuid,mode=0755)
  none on /var/lib/ureadahead/debugfs type debugfs (rw)
  /dev/sda6 on /home type ext4 (rw,relatime)
  binfmt_misc on /proc/sys/fs/binfmt_misc type binfmt_misc (rw,noexec,nosuid,nodev)
  gvfs-fuse-daemon on /home/hildeb/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev,user=hildeb)

  /var/lib/ureadahead/debugfs is left lingering around. It's use is
  somewhat limited, since most Ubuntu users are not kernel developers --
  if it's used as intended: http://kerneltrap.org/node/4394

  ProblemType: Bug
  Architecture: amd64
  Date: Wed Dec 23 11:04:11 2009
  DistroRelease: Ubuntu 9.10
  NonfreeKernelModules: nvidia
  Package: ureadahead 0.90.3-2
  ProcEnviron:
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  ProcVersionSignature: Ubuntu 2.6.31-17.54-generic
  SourcePackage: ureadahead
  Uname: Linux 2.6.31-17-generic x86_64

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/mountall/+bug/499773/+subscriptions