[pkg-wpa-devel] What's wrong with the new configuration scheme, IMO

[Felix Homann]

On Thursday 20 April 2006 12:45, Kel Modderman wrote:
> Felix Homann wrote:
> > Moreover, ifup should not even try to configure the interface unless it
> > is connected to an AP. Otherwise ensuring correct (or intended) behaviour
> > of scripts in /e/n/if-up.d is very difficult. [2]
>
> This is something the planned wifiroamd plans to address.

Sure, but wifiroamd does not exist right now. 

> > Since you can't easily start them
> > seperately - especially in a mode 1 setup - debugging gets more
> > difficult.
>
> I agree, 

Great!

> there is already a bug open with a patch to wpa_supplicant to 
> allow logging, maybe you or I could revive that thread on the hostap
> mailing list?

Would you kindly do so, you already know where the thread is.

> >
> > (i)  Having one stanza to specify the ssid from wireless-tools and one
> > from wpa-supplicant is not the best solution.
>
> , providing a means of accepting either one would be nice,

Agreed.

> but technically wrong.

OK, I can accept that.

> This is something I battled with for a long time when I was "debating"
> with the ubuntu crowd. The decision to make no option require quotes
> caused a lot of time and effort to be spent on my part. I would love to
> hear other opinions on this topic.

At least, as I've mentioned earlier, it should be documented.

> > If you really want to have the capability of connecting to different APs
> > within the new config scheme, put this feature entirely into /e/n/i, by
> > e.g. introducing a "wpa-config-block" stanza, or suggest using logical
> > interfaces in mode 1 (does that work?).
>
> wifiroamd.

wifiroamd does not exist!

> Well, the stanza provided by /e/n/i are appended to the ctrl_interface ,
> which has already read in the network blocks from the .conf file.
> Therefore, the conf file will always override.

Put this into the documentation then.

> > I think mixing wpa-conf with other wpa-*-stanzas (apart from
> > maybe wpa-action related stanzas) should lead to an error. [9]
>
> No, I disagree. There is no harm done here. Additionally, I can use the
> conf file, but still create a new network via /e/n/i, and the new
> network block can be used. There is no conflict or problem here that I
> can see.

Maybe not a real problem. But it's rather confusing.

> > Although the documentation claims that "With action scripts, you can
> > develop your own advanced roaming configuration." I don't think you can.
>
> Yes it can, although it would require an advanced and highly customised
> action script.

Don't claim it, prove it! (Sure, it is possible somehow, but as mentioned in 
my original mail the action script would have to more or less reimplement 
ifupdown).

> I guess it is obvious that we should scrap reference to the action
> scripts from the documentation, at least not suggest them as a preferred
> mode.

Right.

> Dbus interface in the new wpa_supplicant coupled with a dbus aware
> dhcp-client would be a nice solution too ; )

Maybe, but why not first look what can be done with tools that already exist? 

Kind regards,

Felix