Skip Menu |
 

Download (untitled) / with headers
text/plain 3.4KiB
From tytso@MIT.EDU Tue Nov 26 20:46:25 1996
Received: from MIT.EDU (SOUTH-STATION-ANNEX.MIT.EDU [18.72.1.2]) by rt-11.MIT.EDU (8.7.5/8.7.3) with SMTP id UAA20026 for <bugs@RT-11.MIT.EDU>; Tue, 26 Nov 1996 20:46:25 -0500
Received: from DCL.MIT.EDU by MIT.EDU with SMTP
id AA05216; Tue, 26 Nov 96 20:46:24 EST
Received: by dcl.MIT.EDU (5.x/4.7) id AA08031; Tue, 26 Nov 1996 20:46:23 -0500
Message-Id: <9611270146.AA08031@dcl.MIT.EDU>
Date: Tue, 26 Nov 1996 20:46:23 -0500
From: tytso@MIT.EDU
Reply-To: tytso@MIT.EDU
To: krb5-bugs@MIT.EDU
Subject: need INADDR_NONE cleanup
X-Send-Pr-Version: 3.99

Show quoted text
>Number: 256
>Category: telnet
>Synopsis: need INADDR_NONE cleanup
>Confidential: no
>Severity: serious
>Priority: low
>Responsible: hartmans
>State: closed
>Class: sw-bug
>Submitter-Id: unknown
>Arrival-Date: Tue Nov 26 20:47:01 EST 1996
>Last-Modified: Sat Dec 28 21:01:29 EST 1996
>Originator: Theodore Y. Ts'o
>Organization:
mit
Show quoted text
>Release: 1.0-development
>Environment:
DEC/ALPHA, OSF/1

Show quoted text
>Description:

Telnet to a hostname causes an attempt to opena connection to
255.255.255.255, which fails. This makes telnet completely useless.

Show quoted text
>How-To-Repeat:
Try to telnet to a host

Show quoted text
>Fix:
The bug was introduced by a mycroft-requested change for NetBSD/Alpha,
which broke things under Digital Unix. The problem is that inet_addr is
declared to be an unsigned long for NetBSD, and as an unsigned int for OSF/1.
To make things worse, telnet/commands.c is doing a lot of really dodgy things
like declaring inet_addr() in its .c file, instead of relying on the
system header files.

Tom is working on a change, that will hopefully be a bit more
platform indepdent. This change will have to be tested on a wide
number of platforms.

Show quoted text
>Audit-Trail:

State-Changed-From-To: open-feedback
State-Changed-By: tlyu
State-Changed-When: Fri Nov 29 20:12:31 1996
State-Changed-Why:

Not closing this yet; I've checked in changes to fix stuff for the 1.0
release, but jhawk points out that there should be no need to mask of
the lower 32 bits. 4.3BSD derived systems are likely to return
((unsigned_long) -1) anyway, so INADDR_NONE probably wants to be
defined as -1 unless the system provides it. I'll probably deal with
this post-1.0. Removing the bogus prototype for inet_addr is probably
all we needed in the first place, but I would rather not disturb
frozen code again before the release since it does seem to work now.

State-Changed-From-To: feedback-analyzed
State-Changed-By: tlyu
State-Changed-When: Tue Dec 17 18:52:29 1996
State-Changed-Why:

We need to look at this again to clean it up a bit.


From: Sam Hartman <hartmans@MIT.EDU>
To: Tom Yu <tlyu@MIT.EDU>
Cc: Sam Hartman <hartmans@MIT.EDU>, tytso@MIT.EDU, krb5-bugs@MIT.EDU
Subject: Re: telnet/256: need INADDR_NONE cleanup
Date: Wed, 18 Dec 1996 12:10:27 -0500 (EST)

From: Sam Hartman <hartmans@luminous.mit.edu>
Date: 18 Dec 1996 12:10:27 -0500
In-Reply-To: Tom Yu's message of Tue, 17 Dec 1996 18:53:21 -0500
Message-ID: <tslengnrd6k.fsf@luminous.MIT.EDU>
Lines: 7
X-Mailer: Gnus v5.2.37/Emacs 19.34


I looked at commands.c, and as far as I can tell, the current
state of the code is reasonable. It probably is not necessary to mask
off 0xffffffff, but that should never hurt in an IPV4 world.

--Sam

State-Changed-From-To: analyzed-closed
State-Changed-By: hartmans
State-Changed-When: Sat Dec 28 21:01:09 1996
State-Changed-Why:
Seeing no disagreement, I'm closing this out.


Show quoted text
>Unformatted: