Re: buglet in ext2 sticky bit?

Jan Vroonhof (vroonhof@math.ethz.ch)
25 Oct 1999 18:26:48 +0200


Alexander Viro <viro@math.psu.edu> writes:

> Exactly. So _when_ do you need an ability to delete file in /tmp and
> ability to replace its contents doesn't satisfy you?

When you want to allow the non-owners to clean up. i.e. "please delete
this when you are finished with it".

> b) most of the existing UNIX[1] installations does exactly that.

Unless the distribution of Unix systems in the States is completely
different form here most of the UNIX installations that actually have
a non-trivial number of real users per system are SYSV. This feature
only matters in that case.

> d) changing it is unlikely to buy you anything - I'm yet to see a
> program that relies on the alternative behaviour.

To me the SYSV semantics seem much more useful.[2] So the question in my
eyes should be "Are there important programs that rely on the current
behavior?". Or is there disagreement about whether you would like to
have this feature if you were designing the sticky bit from scratch?

> [1] Including Linux. And in the next couple of years the share is
> going to increase, _both_ from Linux and *BSD sides.

Yes so be before the other UNIX'en get run over by the World
Domination steamroller we better make sure we copy the good ideas from
there. Thus now seems like a good time to discuss thing like this.

I won't shed any tears about things remaining the way they are, but I
just wanted you to know there are people who consider the SYSV
behavior a feature. Please don't dismiss is just on the grounds of
"that is the way it always been done."

Jan

Footnotes:
[2] So if you want to be anal about it, the program is 'rm'. Who cares
about programs, I as a user want (and have used) this feature.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/