diff mbox series

[3/3] xfs: don't allow most setxattr to immutable files

Message ID 155379545404.24796.5019142212767521955.stgit@magnolia (mailing list archive)
State New, archived
Headers show
Series vfs: make immutable files actually immutable | expand

Commit Message

Darrick J. Wong March 28, 2019, 5:50 p.m. UTC
From: Darrick J. Wong <darrick.wong@oracle.com>

The chattr manpage has this to say about immutable files:

"A file with the 'i' attribute cannot be modified: it cannot be deleted
or renamed, no link can be created to this file, most of the file's
metadata can not be modified, and the file can not be opened in write
mode."

However, we don't actually check the immutable flag in the setattr code,
which means that we can update project ids and extent size hints on
supposedly immutable files.  Therefore, reject a setattr call on an
immutable file except for the case where we're trying to unset
IMMUTABLE.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_ioctl.c |    8 ++++++++
 1 file changed, 8 insertions(+)

Comments

Amir Goldstein March 28, 2019, 9:24 p.m. UTC | #1
On Thu, Mar 28, 2019 at 7:51 PM Darrick J. Wong <darrick.wong@oracle.com> wrote:
>
> From: Darrick J. Wong <darrick.wong@oracle.com>
>
> The chattr manpage has this to say about immutable files:
>
> "A file with the 'i' attribute cannot be modified: it cannot be deleted
> or renamed, no link can be created to this file, most of the file's
> metadata can not be modified, and the file can not be opened in write
> mode."
>
> However, we don't actually check the immutable flag in the setattr code,
> which means that we can update project ids and extent size hints on
> supposedly immutable files.  Therefore, reject a setattr call on an
> immutable file except for the case where we're trying to unset
> IMMUTABLE.
>

I think if preventing modification of projid and extent size hints is what you
are after you should place the check in xfs_ioctl_setattr() and not in
xfs_ioctl_setattr_xflags().

Yes, it sounds tempting to block changes of xfs_ioc_setxflags(),
but it leads you to a trap of 2nd time chattr +i fails on -EPERM,
because chattr(1) doesn't optimize out the SETFLAGS ioctl
in the case of unmodified flags.
I think if you try to fix that, code will get ugly, so I suggest that
you let SETFLAGS slide.

Thanks,
Amir.

> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  fs/xfs/xfs_ioctl.c |    8 ++++++++
>  1 file changed, 8 insertions(+)
>
>
> diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
> index 2bd1c5ab5008..9cf0bc0ae2bd 100644
> --- a/fs/xfs/xfs_ioctl.c
> +++ b/fs/xfs/xfs_ioctl.c
> @@ -1067,6 +1067,14 @@ xfs_ioctl_setattr_xflags(
>             !capable(CAP_LINUX_IMMUTABLE))
>                 return -EPERM;
>
> +       /*
> +        * If immutable is set and we are not clearing it, we're not allowed
> +        * to change anything else in the inode.
> +        */
> +       if ((ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) &&
> +           (fa->fsx_xflags & FS_XFLAG_IMMUTABLE))
> +               return -EPERM;
> +
>         /* diflags2 only valid for v3 inodes. */
>         di_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags);
>         if (di_flags2 && ip->i_d.di_version < 3)
>
Dave Chinner March 28, 2019, 9:29 p.m. UTC | #2
On Thu, Mar 28, 2019 at 10:50:54AM -0700, Darrick J. Wong wrote:
> From: Darrick J. Wong <darrick.wong@oracle.com>
> 
> The chattr manpage has this to say about immutable files:
> 
> "A file with the 'i' attribute cannot be modified: it cannot be deleted
> or renamed, no link can be created to this file, most of the file's
> metadata can not be modified, and the file can not be opened in write
> mode."
> 
> However, we don't actually check the immutable flag in the setattr code,
> which means that we can update project ids and extent size hints on
> supposedly immutable files.  Therefore, reject a setattr call on an
> immutable file except for the case where we're trying to unset
> IMMUTABLE.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  fs/xfs/xfs_ioctl.c |    8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> 
> diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
> index 2bd1c5ab5008..9cf0bc0ae2bd 100644
> --- a/fs/xfs/xfs_ioctl.c
> +++ b/fs/xfs/xfs_ioctl.c
> @@ -1067,6 +1067,14 @@ xfs_ioctl_setattr_xflags(
>  	    !capable(CAP_LINUX_IMMUTABLE))
>  		return -EPERM;
>  
> +	/*
> +	 * If immutable is set and we are not clearing it, we're not allowed
> +	 * to change anything else in the inode.
> +	 */
> +	if ((ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) &&
> +	    (fa->fsx_xflags & FS_XFLAG_IMMUTABLE))
> +		return -EPERM;
> +
>  	/* diflags2 only valid for v3 inodes. */
>  	di_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags);
>  	if (di_flags2 && ip->i_d.di_version < 3)

Looks fine - catches both FS_IOC_SETFLAGS and FS_IOC_FSSETXATTR
for XFS.

Do the other filesystems that implement FS_IOC_FSSETXATTR have
the same bug?

Cheers,

Dave.
Darrick J. Wong March 29, 2019, 4:02 a.m. UTC | #3
On Fri, Mar 29, 2019 at 08:29:48AM +1100, Dave Chinner wrote:
> On Thu, Mar 28, 2019 at 10:50:54AM -0700, Darrick J. Wong wrote:
> > From: Darrick J. Wong <darrick.wong@oracle.com>
> > 
> > The chattr manpage has this to say about immutable files:
> > 
> > "A file with the 'i' attribute cannot be modified: it cannot be deleted
> > or renamed, no link can be created to this file, most of the file's
> > metadata can not be modified, and the file can not be opened in write
> > mode."
> > 
> > However, we don't actually check the immutable flag in the setattr code,
> > which means that we can update project ids and extent size hints on
> > supposedly immutable files.  Therefore, reject a setattr call on an
> > immutable file except for the case where we're trying to unset
> > IMMUTABLE.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> >  fs/xfs/xfs_ioctl.c |    8 ++++++++
> >  1 file changed, 8 insertions(+)
> > 
> > 
> > diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
> > index 2bd1c5ab5008..9cf0bc0ae2bd 100644
> > --- a/fs/xfs/xfs_ioctl.c
> > +++ b/fs/xfs/xfs_ioctl.c
> > @@ -1067,6 +1067,14 @@ xfs_ioctl_setattr_xflags(
> >  	    !capable(CAP_LINUX_IMMUTABLE))
> >  		return -EPERM;
> >  
> > +	/*
> > +	 * If immutable is set and we are not clearing it, we're not allowed
> > +	 * to change anything else in the inode.
> > +	 */
> > +	if ((ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) &&
> > +	    (fa->fsx_xflags & FS_XFLAG_IMMUTABLE))
> > +		return -EPERM;
> > +
> >  	/* diflags2 only valid for v3 inodes. */
> >  	di_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags);
> >  	if (di_flags2 && ip->i_d.di_version < 3)
> 
> Looks fine - catches both FS_IOC_SETFLAGS and FS_IOC_FSSETXATTR
> for XFS.
> 
> Do the other filesystems that implement FS_IOC_FSSETXATTR have
> the same bug?

Yes.  I'm not even 100% sure I've finished playing xfs whackamole yet.

--D

> Cheers,
> 
> Dave.
> -- 
> Dave Chinner
> david@fromorbit.com
diff mbox series

Patch

diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index 2bd1c5ab5008..9cf0bc0ae2bd 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -1067,6 +1067,14 @@  xfs_ioctl_setattr_xflags(
 	    !capable(CAP_LINUX_IMMUTABLE))
 		return -EPERM;
 
+	/*
+	 * If immutable is set and we are not clearing it, we're not allowed
+	 * to change anything else in the inode.
+	 */
+	if ((ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) &&
+	    (fa->fsx_xflags & FS_XFLAG_IMMUTABLE))
+		return -EPERM;
+
 	/* diflags2 only valid for v3 inodes. */
 	di_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags);
 	if (di_flags2 && ip->i_d.di_version < 3)