[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev rc save bug

From: Philip Webb
Subject: Re: lynx-dev rc save bug
Date: Wed, 7 Oct 1998 17:31:15 -0400 (EDT)

981007 Thomas Dickey wrote: 
> a link set up as described should be from a directory
> that isn't world writable.
> afaik, there's no reliable way to determine if that directory is owned
> by a privileged user: even '0' for root's is not universal
> in the quasi-Unix clones, but I decided to neglect those: they're very rare.
> But, like Bela, I'm curious who owns the directory where the link is
> (sounds like "/" - which should be owned by root -
> the convention afaik is to have root own all of the nonvolatile stuff,
> and privileged "users" such as "bin" and "news" own facilities).

users' directories are kept on a number of disk units,
whose physical names are replaced by symbolic links
--  /homes/purslow/  =  /homefs/u0/purslow/  -- ;
anything found under either must belong to the user  purslow :
why on earth would anyone want to write a Lynx function
which interpreted it in any other way?

this is a good university system with an overworked sysadmin,
who doesn't have the right  strace  & still has Lynx 2-6 as public version.
there's little reason for anyone to get neurotic about security,
as users are generally grown-up enough not to cause damage
even in some rare case of student hacking,
besides which i assume there is security software
to keep track of what users get upto & identify misbehaving individuals.

after having spent  2 days  trying to get the problem taken seriously
& doing the leg-work myself -- with good advice from BL & TD -- ,
i am not a little irritated to discover that i have been caused inconvenience
by something designed for people who choose to make their Lynx available
to all & sundry out in the unwashed & ill-behaved World:
people who run computer systems like that get what they deserve IMHO.
>> thanx lots to you for helping, tho' your fix doesn't work (below): 
> It should - the code goes through one point
> (maybe this is a different problem).
see my other message about how to really fix it for now:
in the end, with some help, i did it myself.
we should get to the bottom of the problem,
now it's been demonstrated it's not a piece of imagination.

SUPPORT     ___________//___,  Philip Webb : address@hidden
ELECTRIC   /] [] [] [] [] []|  Centre for Urban & Community Studies
TRANSIT    `-O----------O---'  University of Toronto

reply via email to

[Prev in Thread] Current Thread [Next in Thread]