Re: Bug#821793: fusermount should not attempt to use /etc/mtab

[Ximin Luo]

Nikolaus Rath:
> Can someone clarify if this is a bug in upstream fuse (which assumes
> existence of /etc/mtab) or in the software that creates the chroots
> without /etc/mtab?
> 
> If someone can give me a decent source that states that /etc/mtab does
> not need to exist, I'll be happy to change fusermount accordingly. 
> 

Hi Nikolaus,

It's hard to find an "authoritative" source for this but:

http://man.cat-v.org/unix_8th/5/mtab
"It does not matter to mount if there are duplicated entries nor to umount if a 
name cannot be found."

https://www.gnu.org/software/libc/manual/html_node/Mount-Information.html
"Note that the fstab and mtab files are maintained on a system by convention. 
It is possible for the files not to exist or not to be consistent with what is 
really mounted or available to mount, if the system’s administration policy 
allows it. But programs that mount and unmount filesystems typically maintain 
and use these files as described herein."

https://unix.stackexchange.com/questions/24182/how-to-get-the-complete-and-exact-list-of-mounted-filesystems-in-linux
"The definitive list of mounted filesystems in in /proc/mounts... [etc why 
/etc/mtab sometimes differs]".

In other words it's generally understood that /etc/mtab might not reflect 
"reality" and so fusermount shouldn't be relying on it to make decisions.

X

-- 
GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE
https://github.com/infinity0/pubkeys.git

___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds

Re: Bug#821793: fusermount should not attempt to use /etc/mtab

[Nikolaus Rath]

Hi,

Can someone clarify if this is a bug in upstream fuse (which assumes
existence of /etc/mtab) or in the software that creates the chroots
without /etc/mtab?

If someone can give me a decent source that states that /etc/mtab does
not need to exist, I'll be happy to change fusermount accordingly. 

Best,
-Nikolaus

-- 
GPG encrypted emails preferred. Key id: 0xD113FCAC3C4E599F
Fingerprint: ED31 791B 2C5C 1613 AF38 8B8A D113 FCAC 3C4E 599F

 »Time flies like an arrow, fruit flies like a Banana.«

___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds

Re: Bug#821793: fusermount should not attempt to use /etc/mtab

[Ximin Luo]

Control: affects -1 disorderfs
Control: affects -1 devscripts
Control: affects -1 reprotest

debrepro from devscripts, and reprotest, both use disorderfs. When cleaning up 
a build, they try to unmount the disorderfs filesystem using `fusermount -u` 
which doesn't work inside an schroot since it doesn't have /etc/mtab.

The workaround is to manually use "sudo umount" but this requires root access. 
Or symlink /etc/mtab to /proc/self/mounts.

X

On Mon, 23 May 2016 08:22:03 -0700 Nikolaus Rath  wrote:
> Hi Santiago,
> 
> On May 21 2016, Santiago Vila  wrote:
> > Hello Nikolaus and Laszlo.
> >
> > A recent build of python-llfuse does not fail to build anymore
> > in my /etc/mtab-less chroot.
> >
> > Has this bug been fixed in fusermount?
> >
> > This is likely but I'm not confindent enough to close this bug myself.
> > Could any of you check this?
> 
> The problem has not been fixed upstream. But maybe Laszlo has added a
> Debian patch?
> 
> Best,
> -Nikolaus
> 
> -- 
> GPG encrypted emails preferred. Key id: 0xD113FCAC3C4E599F
> Fingerprint: ED31 791B 2C5C 1613 AF38 8B8A D113 FCAC 3C4E 599F
> 
>  »Time flies like an arrow, fruit flies like a Banana.«
> 
> 

-- 
GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE
https://github.com/infinity0/pubkeys.git

___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds