Re: significance of "nobody" ownership permission?
From: Tim Haynes (usenet@stirfried.vegetable.org.uk)Date: 06/12/02
- Next message: : "iptables: state & forward confusion"
- Previous message: : "Re: significance of "nobody" ownership permission?"
- In reply to: : "Re: significance of "nobody" ownership permission?"
- Next in thread: : "Re: significance of "nobody" ownership permission?"
- Reply: : "Re: significance of "nobody" ownership permission?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
From: Tim Haynes <usenet@stirfried.vegetable.org.uk> Date: 12 Jun 2002 16:10:15 +0100
Kasper Dupont <kasperd@daimi.au.dk> writes:
[socket]
> > Why does that need to be executable?!
>
> It probably doesn't. AFAIK the executable bits on sockets has no
> significance. But it seems like sockets by default are created with
> permitions 777 and then just modified by the users umask. You don't
> specify any permitions when creating a socket through the bind system
> call.
Yeah, that makes sense. Wasn't there a phase in which the kernel produced a
default umask of 000 as well, thereby causing problems with initscripts?
(At least I vaguely remember this happening in Debian; a trawl of the bug
system there would probably reveal more detail.)
~Tim
-- The light of the world keeps shining, |piglet@stirfried.vegetable.org.uk Bright in the primal glow |http://spodzone.org.uk/
- Next message: : "iptables: state & forward confusion"
- Previous message: : "Re: significance of "nobody" ownership permission?"
- In reply to: : "Re: significance of "nobody" ownership permission?"
- Next in thread: : "Re: significance of "nobody" ownership permission?"
- Reply: : "Re: significance of "nobody" ownership permission?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]