At 02:26 AM 2/17/99 -0500, Brian Reichert wrote:
>On Wed, Feb 17, 1999 at 01:18:33PM +1100, Mark Delany wrote:
>> >Is the inode just a handy unique number? Or are there file access
>> >speed tricks, e.g. opening files directly using inode.
>>
>> Handy unique filename. Vastly superior to tmpnam() and all the lame
>> variants that go with it.
>>
>> It's not for speed - excepting speed of creating a unique filename.
>
>I seem to recall the inode numbers biting people who were copying
>filesystems (backups, changing disks). If your assertions are
>correct, wouln't it make some sense to come up with some other
>cheap unique filename?
Possibly. What do you propose? The current method guarantees a unique file
name first time, every time. Since it's needed for every new mail, you want
it to be efficient, right?
An alternative to your suggestion is to alert people to this dependency and
write a few tools/scripts to assist on those rare occassions that people need
to migrate or restore.
Once alerted though it's mostly a non-problem. For changing disks I found it
just as easy to set a global smtproutes to forward the queue to another
system and I'd be very surprised if many people restore a mailq from backup.
Regards.
