> In that case I misunderstood you, sorry. What would be a better name?
Each option should describe what it does. If you are adding an option that
sets up routing in a way not described by -g, then give that option a name
saying what it actually does.
__
At Fri, 15 Apr 2005 10:44:00 +0200,
Marco Gerards wrote:
>
> Roland McGrath <[EMAIL PROTECTED]> writes:
>
> >> Roland McGrath <[EMAIL PROTECTED]> writes:
> >>
> >> >> The most important part of the patch is setting up the route, for
> >> >> which no interface or utility exists.
> >> >
> >> > The
Roland McGrath <[EMAIL PROTECTED]> writes:
>> Right, but -g does not set the route like we need it for DHCP. I
>> understand if you do not like the name of the option, but that does
>> not make it useless.
>
> I never said it was useless. I said it was poorly named. It probably also
> overloads
> Right, but -g does not set the route like we need it for DHCP. I
> understand if you do not like the name of the option, but that does
> not make it useless.
I never said it was useless. I said it was poorly named. It probably also
overloads too many settings.
__
Roland McGrath <[EMAIL PROTECTED]> writes:
>> Roland McGrath <[EMAIL PROTECTED]> writes:
>>
>> >> The most important part of the patch is setting up the route, for
>> >> which no interface or utility exists.
>> >
>> > There is -g.
>>
>> Which sets the gateway, how would that help?
>
> You said t
> Roland McGrath <[EMAIL PROTECTED]> writes:
>
> >> The most important part of the patch is setting up the route, for
> >> which no interface or utility exists.
> >
> > There is -g.
>
> Which sets the gateway, how would that help?
You said there was no existing facility for setting routes, which
Roland McGrath <[EMAIL PROTECTED]> writes:
>> The most important part of the patch is setting up the route, for
>> which no interface or utility exists.
>
> There is -g.
Which sets the gateway, how would that help?
--
Marco
___
Bug-hurd mailing list
> Because without this patch, such things are not possible. Setting the
> address to 0.0.0.0 was not possible, but now it is.
Fine, so make that -a 0.0.0.0.
> The most important part of the patch is setting up the route, for
> which no interface or utility exists.
There is -g.
__
On Thu, Apr 14, 2005 at 07:28:07PM +0200, Marco Gerards wrote:
> Michael Banck <[EMAIL PROTECTED]> writes:
> > Your patch is very useful of course, I just think we do not need an
> > explicit --dhcp function but rather should make -a 0.0.0.0 [...] work
> > equivalently (if this is not already the c
Michael Banck <[EMAIL PROTECTED]> writes:
> On Thu, Apr 14, 2005 at 01:51:53PM +0200, Marco Gerards wrote:
>> Michael Banck <[EMAIL PROTECTED]> writes:
>>
>> > As the DHCP script sets things to 0.0.0.0 anyway, I am a bit puzzled why
>> > this has to be a user-visable option.
>>
>> Because withou
On Thu, Apr 14, 2005 at 02:15:44PM +0200, Marco Gerards wrote:
> Joachim Nilsson <[EMAIL PROTECTED]> writes:
> > However, is there any way of making the patch a little bit more
> > generic? If anything, just to be able to remove the "dhcp"
> > connection that I think Roland was against.
> This me
On Thu, Apr 14, 2005 at 01:51:53PM +0200, Marco Gerards wrote:
> Michael Banck <[EMAIL PROTECTED]> writes:
>
> > As the DHCP script sets things to 0.0.0.0 anyway, I am a bit puzzled why
> > this has to be a user-visable option.
>
> Because without this patch, such things are not possible. Settin
Joachim Nilsson <[EMAIL PROTECTED]> writes:
> I'd like to try it out but I'm still in the rebuilding phase of
> my "pick-up-the-hurd-hobby". However, is there any way of making
> the patch a little bit more generic? If anything, just to be able
> to remove the "dhcp" connection that I think Rola
Marco Gerards wrote:
"Alfred M. Szmidt" <[EMAIL PROTECTED]> writes:
Can you please tell me if the patch is ok like it is now or if I
should change anything?
Does it actually work? I recall that I tried it (might have been a
older patch), and it didn't work for me.
It does for me. And if peopl
On Thu, Apr 14, 2005 at 10:10:24AM +0200, Alfred M. Szmidt wrote:
>Can you please tell me if the patch is ok like it is now or if I
>should change anything?
>
> Does it actually work? I recall that I tried it (might have been a
> older patch), and it didn't work for me.
You need Marco's
"Alfred M. Szmidt" <[EMAIL PROTECTED]> writes:
>Can you please tell me if the patch is ok like it is now or if I
>should change anything?
>
> Does it actually work? I recall that I tried it (might have been a
> older patch), and it didn't work for me.
It does for me. And if people don't
Michael Banck <[EMAIL PROTECTED]> writes:
> As the DHCP script sets things to 0.0.0.0 anyway, I am a bit puzzled why
> this has to be a user-visable option.
Because without this patch, such things are not possible. Setting the
address to 0.0.0.0 was not possible, but now it is.
The most importa
On Thu, Apr 14, 2005 at 01:29:43AM +0200, Joachim Nilsson wrote:
> Alfred M. Szmidt wrote:
> > Roland McGrath wrote:
> > > Marco Gerards wrote:
> >> > What this patch does is preparing the state of the interfaces so
> >> > it can use 0.0.0.0 as address and broadcast.
> >> That has nothing to do wit
Can you please tell me if the patch is ok like it is now or if I
should change anything?
Does it actually work? I recall that I tried it (might have been a
older patch), and it didn't work for me.
___
Bug-hurd mailing list
Bug-hurd@gnu.org
http:
Roland McGrath <[EMAIL PROTECTED]> writes:
> Setting the addresses and the route to match them is indeed what it does,
> and that is not specific to DHCP just because that's the motivation for it.
Right. And I would like to see this patch applied soon. Can you
please tell me if the patch is ok
Setting the addresses and the route to match them is indeed what it does,
and that is not specific to DHCP just because that's the motivation for it.
___
Bug-hurd mailing list
Bug-hurd@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-hurd
Alfred M. Szmidt wrote:
> Roland McGrath wrote:
> > Marco Gerards wrote:
> > What this patch does is preparing the state of the interfaces so
> > it can use 0.0.0.0 as address and broadcast.
> That has nothing to do with DHCP per se and shouldn't give the
> impression that it does.
Actually, it doe
> What this patch does is preparing the state of the interfaces so
> it can use 0.0.0.0 as address and broadcast.
That has nothing to do with DHCP per se and shouldn't give the
impression that it does.
Actually, it does. The patch also sets up the routing table. Which
is impossible
> What this patch does is preparing the state of the interfaces so it
> can use 0.0.0.0 as address and broadcast.
That has nothing to do with DHCP per se and shouldn't give the impression
that it does.
___
Bug-hurd mailing list
Bug-hurd@gnu.org
htt
Roland McGrath <[EMAIL PROTECTED]> writes:
>> What this patch does is preparing the state of the interfaces so it
>> can use 0.0.0.0 as address and broadcast.
>
> That has nothing to do with DHCP per se and shouldn't give the impression
> that it does.
What do you suggest?
--
Marco
__
> What's the rationale for a built-in hack instead of just using
> dhclient?
dhclient is still required.
What this patch does is preparing the state of the interfaces so it
can use 0.0.0.0 as address and broadcast. When using this patch it
is possible to port dhclient.
s/port/
Roland McGrath <[EMAIL PROTECTED]> writes:
> What's the rationale for a built-in hack instead of just using dhclient?
dhclient is still required.
What this patch does is preparing the state of the interfaces so it
can use 0.0.0.0 as address and broadcast. When using this patch it is
possible to
What's the rationale for a built-in hack instead of just using dhclient?
___
Bug-hurd mailing list
Bug-hurd@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-hurd
Marco,
Excellent work.
As far as I am concerned, once someone else has tested this, this
should be committed.
Thanks for the work,
Neal
___
Bug-hurd mailing list
Bug-hurd@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-hurd
Has my blessing--for what it is worth--if you fix the code to use
memset() instead of bzero().
___
Bug-hurd mailing list
Bug-hurd@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-hurd
"Neal H. Walfield" <[EMAIL PROTECTED]> writes:
> The patch looks basically good, however, as Alfred already pointed
> out, even if this is a hack we should still support dhcp per-device
> and not just hardcode eth0.
[...]
> Also, a small typo: it should be `started *with* the'.
Both issues are
The patch looks basically good, however, as Alfred already pointed
out, even if this is a hack we should still support dhcp per-device
and not just hardcode eth0.
If you do this, I am willing to support something like this for
inclusion.
> * options.c (options): Add the option `dhcp'.
>
Here is a patch to make pfinet use 0.0.0.0 as IP address and make
broadcasting possible when pfinet is started with `--dhcp'.
Perhaps it would be nice to include this in the Hurd, or just in
Debian if it is too much of a hack. I don't want to put too much
time in it if it is complet
Hi,
Here is a patch to make pfinet use 0.0.0.0 as IP address and make
broadcasting possible when pfinet is started with `--dhcp'. Perhaps
it would be nice to include this in the Hurd, or just in Debian if it
is too much of a hack. I don't want to put too much time in it if it
is completely wrong
Hi,
DHCP wasn't possible yet because of two reasons:
- The routing tabled need to be set up. Because of lacking routing
interface this was not possible.
- The IP address of 0.0.0.0 was not valid.
I've added the RPCs so adding/deleting routes is possible now. If this
patch is committed support
35 matches
Mail list logo
