linux-ext4 - Re: [PATCH] fuse2fs: updates for message reporting journal is not supported

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20251017232521.GI6170@frogsfrogsfrogs>
Date: Fri, 17 Oct 2025 16:25:21 -0700
From: "Darrick J. Wong" <djwong@...nel.org>
To: Dave Dykstra <dwd@...n.ch>
Cc: linux-ext4@...r.kernel.org,
	Dave Dykstra <2129743+DrDaveD@...rs.noreply.github.com>
Subject: Re: [PATCH] fuse2fs: updates for message reporting journal is not
 supported

On Fri, Oct 17, 2025 at 03:09:57PM -0500, Dave Dykstra wrote:
> On Fri, Oct 17, 2025 at 12:18:00PM -0700, Darrick J. Wong wrote:
> > On Thu, Oct 16, 2025 at 03:09:03PM -0500, Dave Dykstra wrote:
> > > This makes two changes to the message that is shown saying that fuse2fs
> > > does not support the journal.  First is that it reverts the check to
> > > what it was before 3875380 to look at the ro option not being set
> > > instead of checking the RW flag.  That's because I don't think this
> > > message needs to be shown when the ro option is set even when it was
> > > opened RW; there should be nothing to corrupt when it is ro.
> > > 
> > > Second, it changes the message to say that writing is not supported
> > > rather than using the journal is not supported.  The current message is
> > > confusing because in fact the journal is used for recovery when needed
> > > and possible.
> > > 
> > > Also submitted as PR https://github.com/tytso/e2fsprogs/pull/251
> > > 
> > > Signed-off-by: Dave Dykstra <2129743+DrDaveD@...rs.noreply.github.com>
> > > ---
> > >  misc/fuse2fs.c | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/misc/fuse2fs.c b/misc/fuse2fs.c
> > > index cb5620c7..c46cc03b 100644
> > > --- a/misc/fuse2fs.c
> > > +++ b/misc/fuse2fs.c
> > > @@ -4774,10 +4774,10 @@ int main(int argc, char *argv[])
> > >  		}
> > >  	}
> > >  
> > > -	if (global_fs->flags & EXT2_FLAG_RW) {
> > > +	if (!fctx.ro) {
> > 
> > Again, rw != EXT2_FLAG_RW.
> > 
> > The ro and rw mount options specify if the filesystem mount is writable.
> > You can mount a filesystem in multiple places, and some of the mounts
> > can be ro and some can be rw.
> > 
> > EXT2_FLAG_RW specifies that the filesystem driver can write to the block
> > device.  fuse2fs should warn about incomplete journal support any time
> > the **filesystem** is writable, independent of the write state of the
> > mount.
> 
> Are you saying that is indeed possible for a read-only mount to cause
> file corruption or data loss if there's not a graceful unmount?  If so,

No, I'm saying that filesystem drivers can *themselves* write metadata
to a filesystem mounted ro.  ro means that user programs can't write to
the files under a particular mountpoint.  This has long been the case
for the kernel implementations of ext*, XFS, btrfs, etc.

I've said this three times now, and this is the last time I'm going to
say it.

> it sure seems like that should be avoided if possible!  Since fuse2fs
> does not support writing the journal, perhaps its behavior should be
> different than the kernel's behavior for this too.  Perhaps once the
> journal is recovered it should be remounted without EXT2_FLAG_RW.

No.  Ted and I are trying to minimize the differences between the kernel
and fuse2fs.

> In any case, during further testing I did find a serious problem with
> this change in that it changes more than just the message; it also skips
> reading in the inode bitmap, which causes a problem later.  So at
> minimum this patch should only affect the message, not the rest of the
> stuff in that if statement.

Are you running fstests QA on these patches before you send them out?

--D

> > Filesystems are allowed to write to the block device even if the mount
> > itself is readonly, e.g. kernel ext4 recovering the journal on an ro
> > mount.
> > 
> > NAK.
> > 
> > --D
> > 
> > >  		if (ext2fs_has_feature_journal(global_fs->super))
> > >  			log_printf(&fctx, "%s",
> > > - _("Warning: fuse2fs does not support using the journal.\n"
> > > + _("Warning: fuse2fs does not support writing the journal.\n"
> 
> What do you think about this message change?
> 
> Dave
> 
> > >     "There may be file system corruption or data loss if\n"
> > >     "the file system is not gracefully unmounted.\n"));
> > >  		err = ext2fs_read_inode_bitmap(global_fs);
> > > -- 
> > > 2.43.5
> > > 
> > > 
> > 
>