NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/comp/unix/aix/8128

Text File | 1992-07-23 | 1.6 KB | 34 lines

Newsgroups: bull.sys.aix,comp.unix.aix
Path: sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!yktnews!prener
From: [e-mail] (Dan Prener)
Subject: Re: Changing name associated with a process
Sender: [e-mail] (NNTP News Poster)
Message-ID: <[e-mail]>
In-Reply-To: rbh@ganymede's message of 24 Jul 92 01:56:47 GMT
Date: Fri, 24 Jul 1992 07:27:55 GMT
Disclaimer: This posting represents the poster's views, not necessarily those of IBM
References: <[e-mail]>
    <[e-mail]>
    <[e-mail]>
    <[e-mail]>
Nntp-Posting-Host: prener.watson.ibm.com
Organization: IBM T.J. Watson Research Center, Hawthorne, New York
Lines: 16

In article <[e-mail]> rbh@ganymede (Ronald Harvey) writes:

>Being able to change argv[0] et al for the whole world to see is a
>blatant 'covert channel'.  The machine will not get a B2 security
>rating...

Could you explain how this conveys more information than I would by
invoking an executable (say that does a "sleep" for a while and nothing else)
by a name I have chosen, then renaming the executable and running it again,
and so on?  It is true that there are differences -- if I just change argv[0]
it will keep the same process number and start time, etc.  But I don't see
how to send any more information because of those differences.  After all,
I could encode the equivalent of a process number and a start time in
my names.
--
                                   Dan Prener ([e-mail])