| 1 |
On Sat, May 31, 2003 at 11:29:27PM -0400, Peter Johanson wrote: |
| 2 |
> Solutions: |
| 3 |
> * Seperate net.ethX and net6.ethX scripts. |
| 4 |
> Problems: When something fails in current net.ethX script, |
| 5 |
> we do 'ifconfig ethX down'. If we do this with co-existing |
| 6 |
> v4+v6, we bring down the other form of networking. |
| 7 |
> |
| 8 |
> Advantages: Lets us have a "provide net" and "provide net6". |
| 9 |
> Other scripts can independantly depend on one or the other, or both, |
| 10 |
> or have "need net \ use net6", etc. |
| 11 |
> |
| 12 |
> More easily allows for IPv6 only networks if in an IPv6 only |
| 13 |
> environment. (example: LAN using 6to4 tools) |
| 14 |
I think this would be the better solution in general, but I have one |
| 15 |
thing to add with it. |
| 16 |
ifconfig should really get deprecated. The 'ip' tool of sys-apps/iproute |
| 17 |
has long been noted as mostly superior in functionality (there are some |
| 18 |
obscure, obsolete commands in ifconfig that I have never heard of |
| 19 |
being used in recent times). |
| 20 |
|
| 21 |
'ip' also provides a single tool for easily doing things that would take |
| 22 |
a few other commands. |
| 23 |
|
| 24 |
One particullar distinction is the seperation of the link and address |
| 25 |
layers, which are mostly synonymous in ifconfig. |
| 26 |
|
| 27 |
One other thing to consider in this, is possible integration of |
| 28 |
zeroconf (http://www.zeroconf.org/). |
| 29 |
|
| 30 |
On any system, IPv4 network will practially always exist, even if it is |
| 31 |
only on the loopback interface. |
| 32 |
|
| 33 |
-- |
| 34 |
Robin Hugh Johnson |
| 35 |
E-Mail : robbat2@××××××××××××××.net |
| 36 |
Home Page : http://www.orbis-terrarum.net/?l=people.robbat2 |
| 37 |
ICQ# : 30269588 or 41961639 |
| 38 |
GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 |
Archives