Bug 20527

Summary: redundant binaries for similar IPv4 and IPv6 tools
Product: Base System Reporter: Robert Watson <rwatson>
Component: binAssignee: Hajimu UMEMOTO <ume>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: 5.0-CURRENT   
Hardware: Any   
OS: Any   

Description Robert Watson freebsd_committer freebsd_triage 2000-08-10 21:40:00 UTC
ping6 and traceroute6 are seperate binaries from ping and traceroute.
For other IPv6 tools (telnet, rsh, ...) the binaries are the same.
This inconsistency is unfortunate, and should (one would imagine) be
correctable, with an optional "-4" or "-6" to force the use of AF_INET
or AF_INET6 to retrieve a particular protocol.  It would be nice to
reduce the number of setuid binaries in the system, as well as,
wherever possible, use the same tools for both IP protocols to minimize
unnecessary differences.

Fix: 

Patches not available.
How-To-Repeat: 
ls /usr/sbin/traceroute6 /usr/sbin/traceroute
ls /sbin/ping6 /sbin/ping
Comment 1 Johan Karlsson freebsd_committer freebsd_triage 2000-08-10 21:47:06 UTC
Responsible Changed
From-To: freebsd-bugs->ume

Let our IPv6 maintainer have a look at this.
Comment 2 Sheldon Hearn 2000-08-11 10:55:58 UTC
On Thu, 10 Aug 2000 16:35:24 -0400, rwatson@FreeBSD.ORG wrote:

> >Number:         20527
> >Category:       bin
> >Synopsis:       redundant binaries for similar IPv4 and IPv6 tools

Hi Robert,

Did someone in particular ask you to send this "wish-list" class PR? :-)

Ciao,
Sheldon.
Comment 3 Robert Watson freebsd_committer freebsd_triage 2000-08-11 14:39:20 UTC
On Fri, 11 Aug 2000, Sheldon Hearn wrote:

> On Thu, 10 Aug 2000 16:35:24 -0400, rwatson@FreeBSD.ORG wrote:
> 
> > >Number:         20527
> > >Category:       bin
> > >Synopsis:       redundant binaries for similar IPv4 and IPv6 tools
> 
> Hi Robert,
> 
> Did someone in particular ask you to send this "wish-list" class PR? :-)

Heh. :-)  Actually, I made an off-hand comment in a place with many
FreeBSD coding types, and the response was, ``yes, you're right, it should
be'' and related types of things.  One rationale is reducing the number of
setuid binaries, since in the base OS, setuid root is required for raw
socket access, another is consistency.  In any case, it would be nice,
although presumably there are strong rationales for having two binaries in
the first place, which might be nice to discuss and see if they still hold
true in the light of support framework improvements, integration into the
base OS, etc.

  Robert N M Watson 

robert@fledge.watson.org              http://www.watson.org/~robert/
PGP key fingerprint: AF B5 5F FF A6 4A 79 37  ED 5F 55 E9 58 04 6A B1
TIS Labs at Network Associates, Safeport Network Services
Comment 4 ume 2000-08-17 18:56:39 UTC
I think there are few merits to merge binaries, and we shouldn't do it
at least for the present.  The reason are:

  - There are few shareable code between ping and ping6, or traceroute
    and traceroute6.  This is because, raw socket related APIs are
    quite different between IPv4 and IPv6.  So, there are few benefit
    to merge these.

  - ping6 and traceroute6 are maintained by KAME Project, and we have
    merging issue.  Once we have such merged binaries, merge or
    maintainance will be harder.

--
Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan
ume@mahoroba.org  ume@bisd.hitachi.co.jp  ume@FreeBSD.org
http://www.imasy.org/~ume/
Comment 5 Hajimu UMEMOTO freebsd_committer freebsd_triage 2001-03-13 12:46:33 UTC
State Changed
From-To: open->closed

The author of ping6 added the reason why ping and ping6 are exist 
separately to BUGS section of ping6 manpage.