Re: Testing netapplet

[Matthew Garrett <mgarrett@chiark.greenend.org.uk>]

Thomas Hood <jdthood@yahoo.co.uk> wrote:

> That is a problem.  Configuration tools should not break when they
> encounter configurations that they do not understand.

Netapplet is not a configuration tool. It will launch one for you - at
present, it gives you the gnome-system-tools network program. There's no
reason it couldn't use anything else.

> All I can infer from this paragraph is that you don't like ifupdown,
> possibly because you do not understand how it works.

That's not the case. I like ifupdown a great deal - it's just very
difficult to provide something that works alongside it nicely.

> In mapping, ifupdown calls an external program that makes the decision
> about which interface definition to choose for a given physical
> interface to be brought up.  A front end for ifup/ifdown (which is
> what I gather netapplet is supposed to be) either uses mapping
> (if it does "ifup IFACE") or overrides mapping (if it does
> "ifup IFACE=LIFACE"); it does not need to know how the external
> program does the mapping.  All the information needed by the front
> end is available.  E.g., to find out how a physical interface is
> currently configured the program can look in /etc/network/ifstate.

Ah, but it does. For instance, to make guessnet work nicely with
/etc/network/interfaces you need to use something like:

mapping eth0
       script /usr/sbin/guessnet-ifupdown
       map work
       map default: dhcp
       map timeout: 3
       map verbose: true

iface dhcp inet dhcp

iface work inet static
        address 192.168.128.204
        netmask 255.255.255.0
        broadcast 192.168.128.255
        gateway 192.168.128.240

which is all well and good, but how do I know which of these things
should be presented as a list of interfaces to the user and which are
purely logical? Without knowing how guessnet-ifupdown works, there's no
way I can know that dhcp and work have no relation to real interfaces
and shouldn't be presented to the user by default.

> I don't know what you mean by this.  What kind of policy would
> you like to see?

Sorry, that wasn't clear. The guessnet case is an example of the lack of
policy. There's no way of telling whether something actually refers to
an interface (and so is of interest to netapplet) or not.

-- 
Matthew Garrett | mjg59-chiark.mail.debian.devel@srcf.ucam.org