* Re: [fuse-devel] fusermount reports "Device or resource busy" but unmounts it anyway
[not found] <20150512093413.GA19964@redhat.com>
@ 2015-05-13 8:27 ` Miklos Szeredi
2015-05-13 8:29 ` Richard W.M. Jones
0 siblings, 1 reply; 3+ messages in thread
From: Miklos Szeredi @ 2015-05-13 8:27 UTC (permalink / raw)
To: Richard W.M. Jones; +Cc: fuse-devel, Eric W. Biederman, Linux-Fsdevel
[CCs added]
On Tue, May 12, 2015 at 11:34 AM, Richard W.M. Jones <rjones@redhat.com> wrote:
>
> Has something changed in FUSE / the kernel in the past few months so
> that 'fusermount -u /mp' reports the error:
>
> fusermount: failed to unmount /mp: Device or resource busy
>
That's from the umount2() syscall, and most likely not a fuse issue.
> and returns a non-zero exit code, but the mountpoint is unmounted
> anyway?
Is this in 4.1-rc? If so, then the recent fs/namespace.c patches from
Eric Biederman would be suspect.
Thanks,
Miklos
>
> We have a fusermount wrapper which relies on the fact that fusermount
> either succeeds & unmounts; or reports an error & doesn't unmount --
> the wrapper retries in this case.
>
> Rich.
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [fuse-devel] fusermount reports "Device or resource busy" but unmounts it anyway
2015-05-13 8:27 ` [fuse-devel] fusermount reports "Device or resource busy" but unmounts it anyway Miklos Szeredi
@ 2015-05-13 8:29 ` Richard W.M. Jones
2015-05-13 14:50 ` Eric W. Biederman
0 siblings, 1 reply; 3+ messages in thread
From: Richard W.M. Jones @ 2015-05-13 8:29 UTC (permalink / raw)
To: Miklos Szeredi; +Cc: fuse-devel, Eric W. Biederman, Linux-Fsdevel
On Wed, May 13, 2015 at 10:27:09AM +0200, Miklos Szeredi wrote:
> [CCs added]
>
> On Tue, May 12, 2015 at 11:34 AM, Richard W.M. Jones <rjones@redhat.com> wrote:
> >
> > Has something changed in FUSE / the kernel in the past few months so
> > that 'fusermount -u /mp' reports the error:
> >
> > fusermount: failed to unmount /mp: Device or resource busy
> >
>
> That's from the umount2() syscall, and most likely not a fuse issue.
>
> > and returns a non-zero exit code, but the mountpoint is unmounted
> > anyway?
>
> Is this in 4.1-rc? If so, then the recent fs/namespace.c patches from
> Eric Biederman would be suspect.
Yes, this is reproducible with:
kernel-4.1.0-0.rc2.git3.1.fc23.x86_64 (from Fedora Rawide)
fuse-2.9.3-4.fc22.x86_64
I will see if I can make a small reproducer and get back.
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-p2v converts physical machines to virtual machines. Boot with a
live CD or over the network (PXE) and turn machines into KVM guests.
http://libguestfs.org/virt-v2v
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [fuse-devel] fusermount reports "Device or resource busy" but unmounts it anyway
2015-05-13 8:29 ` Richard W.M. Jones
@ 2015-05-13 14:50 ` Eric W. Biederman
0 siblings, 0 replies; 3+ messages in thread
From: Eric W. Biederman @ 2015-05-13 14:50 UTC (permalink / raw)
To: Richard W.M. Jones; +Cc: Miklos Szeredi, fuse-devel, Linux-Fsdevel
"Richard W.M. Jones" <rjones@redhat.com> writes:
> On Wed, May 13, 2015 at 10:27:09AM +0200, Miklos Szeredi wrote:
>> [CCs added]
>>
>> On Tue, May 12, 2015 at 11:34 AM, Richard W.M. Jones <rjones@redhat.com> wrote:
>> >
>> > Has something changed in FUSE / the kernel in the past few months so
>> > that 'fusermount -u /mp' reports the error:
>> >
>> > fusermount: failed to unmount /mp: Device or resource busy
>> >
>>
>> That's from the umount2() syscall, and most likely not a fuse issue.
>>
>> > and returns a non-zero exit code, but the mountpoint is unmounted
>> > anyway?
>>
>> Is this in 4.1-rc? If so, then the recent fs/namespace.c patches from
>> Eric Biederman would be suspect.
>
> Yes, this is reproducible with:
>
> kernel-4.1.0-0.rc2.git3.1.fc23.x86_64 (from Fedora Rawide)
> fuse-2.9.3-4.fc22.x86_64
>
> I will see if I can make a small reproducer and get back.
I will be happy to look at this if there is a reasonably small reproducer.
I don't think I changed anything that could cause that kind of failure,
but I have touched a fair amount of that code recently and at the very
least I will be familiar with it.
The only case I can imagine is the mount point being deleted.
Eric
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2015-05-13 14:54 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20150512093413.GA19964@redhat.com>
2015-05-13 8:27 ` [fuse-devel] fusermount reports "Device or resource busy" but unmounts it anyway Miklos Szeredi
2015-05-13 8:29 ` Richard W.M. Jones
2015-05-13 14:50 ` Eric W. Biederman
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).