2005-04-20 20:29:37

by Roman Zippel

[permalink] [raw]
Subject: Re: [PATCH] hfsplus: add an option to force r/w mount

Hi,

On Wed, 20 Apr 2005, Colin Leroy wrote:

> for some reason yet unknown, fsck.hfsplus doesn't correctly set the
> HFSPLUS_VOL_UNMNT flag here.

If fsck doesn't mark it clean, there must be a reason and that also means
you shouldn't mount it writable.

bye, Roman


2005-04-21 01:34:29

by Colin Leroy

[permalink] [raw]
Subject: Re: [PATCH] hfsplus: add an option to force r/w mount

On 20 Apr 2005 at 22h04, Roman Zippel wrote:

Hi,

> Hi,
>
> On Wed, 20 Apr 2005, Colin Leroy wrote:
>
> > for some reason yet unknown, fsck.hfsplus doesn't correctly set the
> > HFSPLUS_VOL_UNMNT flag here.
>
> If fsck doesn't mark it clean, there must be a reason and that also
> means you shouldn't mount it writable.

that's what i thought too, but adding debug printf() to it, i see that
it marks it clean at the end. However that doesn't seem to effectively
land on the disc. imho the patch has an interest anyway...

--
Colin

2005-04-21 02:18:38

by Colin Leroy

[permalink] [raw]
Subject: Re: [PATCH] hfsplus: add an option to force r/w mount

On 20 Apr 2005 at 22h04, Roman Zippel wrote:

Hi,

> > for some reason yet unknown, fsck.hfsplus doesn't correctly set the
> > HFSPLUS_VOL_UNMNT flag here.
>
> If fsck doesn't mark it clean, there must be a reason

By the way, the reason is that this stupid utility opens the device
read-only (hence it can't fix nothing).

from hfsplusutils/libhfsp/src/fscheck.c:

result = fscheck_volume_open(&vol, device, HFSP_MODE_RDONLY);

Seeing that it has been untouched since 2002, I guess it'll be hard to
get it fixed. Or maybe it's just me being idiot and I should use
another fsck ?

--
Colin