Re: [REPOST][PATCH][RFC] vfs: add message print mechanism for the mount/umount into the VFS layer

[Mike Mestnik <cheako911@gmail.com>]

On Thu, Jan 14, 2010 at 2:14 AM, Al Viro <viro@zeniv.linux.org.uk> wrote:
> On Thu, Jan 14, 2010 at 03:48:37PM +0900, Toshiyuki Okajima wrote:
>> Hi.
>>
>> Now, on the VFS-layer any messages aren't printed at mount/umount time
>> (when the operation normally terminates).
>> But there are some filesystems which print their own specific messages at those
>> times.
>>
>> For the purpose of the system management and so on, users (especially,
>> enterprise users) want to observe their system operations from the system logs.
>> We may manage to observe some filesystems' operations (mount/umount) from the
>> logs. But the filesystems of which we can observe the operations are not all.
>>
>> Therefore to achieve the observations for all filesystems is to add the common
>> message mechanism into the VFS layer. If any filesystems' specific messages at
>> mount/umount time are added into this, we can distinguish more easily each
>> message among the system logs for our systems' observations.
>>
>> This patch provides the following:
>> - message output of common format from VFS at mount/umount time
>> - the functions of printing the filesystem's specific information at
>>   mount/umount time
>
> I am not going to apply that.  For one thing, printk is improper mechanism
> for "observing [normal] operations".  Vague references to "enterprise
> users" wanting that do not constitute a valid rationale.
>
> What's more, there is absolutely no point in taking care to have mount(2)
> spew something in log when explicitly asked to do so; caller can bloody
> well do it itself.  From userland.  And on umount side your logics is
> simply b0rken, since the presence of report depends on the order of vfsmount
> demise in case when there are several vfsmounts over given superblock.
>
There is something to the kernel's verbose report of a file system
mount.  For example only the kernel will have access to the feature
set that was successfully initiated.  Userspace can guess by examining
the superblock and the kernels list of supported features, but that's
a long shot to start with.

> I can see a value in having something e.g. in /proc that would report the
> list of active filesystems (or active filesystems of given type).  poll()able,
> a-la /proc/mounts.  Format of such thing would have to be considered
> carefully, but at least it's something that might make sense.

This information is useful long after the file system is mounted and
the current mounts is a-la /etc/fstab|mtab, no need to fix that ;).
If there was a project to implement a poll()able mounts I'd also
suggest an extended /proc/mounts that showed the filesystem's
available features, highlighting the ones that impact userland(like
the missing functionality that exists in fat and ntfs).  I know these
examples are rather static, but the idea would apply to file systems
/w or w/o extent for example as ext4 can have it both ways.  The
difference from mounts is that these are not user tunable parameters
and the goal is to express the kernels usefulness/effectiveness
against a given filesystem.

One project that could build on this would be an fuse project that
creates meta-files(optionally on another filesystem) to plug these
holes providing the umsdos functionality.  I myself have used split to
place large files on vfat, perhaps the future is an fuse application
that does this on the users behalf and can be extended to manage
sparse files as well.
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html