mmap / mtime updates

[Marek Michalkiewicz]

Hello,

What is the current status of the bug where modifications to file
contents made via mmap fail to update mtime of the file?

This was discussed a few months ago in this thread:

http://lkml.org/lkml/2006/5/17/138

where patches have been posted, but it seems that no final solution
has been agreed on and applied to the kernel tree. Updating ctime
and mtime appears to be required by the standard:

http://www.opengroup.org/onlinepubs/009695399/functions/mmap.html

The st_ctime and st_mtime fields of a file that is mapped with
MAP_SHARED and PROT_WRITE shall be marked for update at some point
in the interval between a write reference to the mapped region and
the next call to msync() with MS_ASYNC or MS_SYNC for that portion
of the file by any process.

and failing to do it can lead to potential data loss as well, if
modified files are not backed up (I'm seeing the problem with Samba
tdb files not being backed up by rsnapshot, for example).

Thanks,
Marek
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

[Peter Staubach]

Marek Michalkiewicz wrote:
> Hello,
>
> What is the current status of the bug where modifications to file
> contents made via mmap fail to update mtime of the file?
>
> This was discussed a few months ago in this thread:
>
> http://lkml.org/lkml/2006/5/17/138
>
> where patches have been posted, but it seems that no final solution
> has been agreed on and applied to the kernel tree. Updating ctime
> and mtime appears to be required by the standard:
>
> http://www.opengroup.org/onlinepubs/009695399/functions/mmap.html
>
> The st_ctime and st_mtime fields of a file that is mapped with
> MAP_SHARED and PROT_WRITE shall be marked for update at some point
> in the interval between a write reference to the mapped region and
> the next call to msync() with MS_ASYNC or MS_SYNC for that portion
> of the file by any process.
>
> and failing to do it can lead to potential data loss as well, if
> modified files are not backed up (I'm seeing the problem with Samba
> tdb files not being backed up by rsnapshot, for example).

I am working on porting the patches to the current upstream kernel.
As soon as I complete this work, then I will repost the patch and
we'll see where things go then.

Thanx...

ps