[pkg-wpa-devel] Bug#779007: Bug#779007: closed by Stefan Lippers-Hollmann <s.l-h at gmx.de> (Re: Bug#779007: iw: Implement hook script for ifupdown, please?)

[Elliott Mitchell]

On Tue, Feb 24, 2015 at 05:11:14AM +0100, Stefan Lippers-Hollmann wrote:
> On 2015-02-23, Elliott Mitchell wrote:
> > On Mon, Feb 23, 2015 at 03:03:17PM +0100, Stefan Lippers-Hollmann wrote:
> > > On 2015-02-22, Elliott Mitchell wrote:
> [...]
> > > [...] On the other hand iw does not provide 
> > > any kind of connection supervision, it's just a one-shot configuration
> > > tool. This means a wireless connection purely configured by iw won't
> > > recover from connection losses (getting out of range, temporary 
> > > interference, etc.) and it can't roam between different access points
> > > either.
> > 
> > The latter is exactly what I expected from reading the package
> > description, behavior identical to `ifconfig` and `iwconfig`.
> 
> No, it can't. iw isn't a daemon, if your connection drops, your 
> connection is gone - mac80211 drivers won't try to reconnect. That's
> a task for userspace, a supervising daemon - wpasupplicant in the most
> trivial case or more sophisticated ones if you like.
> 
> Of course one could add create yet-another-networking-daemon and add
> it to the iw package, but this would be far from trivial - and be
> beyond its scope. After all you don't expect /bin/ls to suddenly grow
> an inotify backend daemon.

How is this relevant to the original request, adding an ifupdown script?

It wasn't explicit, but I was thinking of something along the lines of
what wireless-tools provides.  Simply something that brought up the
interface to a minimally functional state.  I had no expectation of any
of the tasks you bring up being handled.


> iw can work without wpa_supplicant, it can do lots of things without 
> wpa_supplicant[0] - what it can't do is handling WPA2 encryption (as 
> mac80211 pushes that into user-space) or to actually manage connection
> and reconnect when needed. In a perfect environment, without 
> interference and always a strong signal being available, where links 
> never drop and encryption are tales from the sci-fi universe - you can 
> use iw to configure an unencrypted link. But you can't to that in 
> practice, as nothing is ever perfect and connections will drop 
> frequently - and non-WPA2 encrypted networks (which is where you 
> hard-depend on wpa_supplicant either way) are the exception, not the 
> norm - especially according to IEEE 802.11n. But I'm repeating
> myself...

Not a single one of these things was requested.


> iw is not the primary tool to connect to a wireless network - and it
> doesn't depend on wpa_supplicant at all (the primary use-cases for
> connecting to a wifi network do, but not iw). iw is totally optional[1]
> when it comes to connecting to a wireless network, its use starts when
> you want to debug on the low-level - or scan for APs, or add multiple
> VIFs or do other low-level interface manipulations - or just to provide
> a sample implementation of the nl80211 API. Compare it to iwlist, not 
> iwconfig (unless you know what you're doing and/ or do quite advanced 
> stuff, which is where you might compare its use to that of ethtool,
> unnecessary of $Aunt_Tilly's notebook, but essential if you're doing
> advanced stuff and know what you're doing). The most one could argue 
> for, would be adding wpa_supplicant as a suggests - but this would be 
> quite far-fetched[2].

And this is what one might expect an iw ifupdown script to do.
Crucially, that item of handling multiple VIFs.


> > I believe the correct spelling is "nyetwork-munger" (hopefully it has
> > improved, but right now that isn't the issue).  Meanwhile I haven't come
> > to a conclusion (or concussion, yet) on systemd, but I'm rather concerned
> > about that situation.  More notably though, those aren't the right tool
> > for the particular situation.
> 
> Independent of your statement itself, this language doesn't really help
> your case.

Sorry, I did some experimentation with it a longish while ago and it
wasn't pleasant trying to get it to do simple tasks.  Perhaps it is
better now, but things were easily solved without it.


-- 
(\___(\___(\______          --=> 8-) EHM <=--          ______/)___/)___/)
 \BS (    |         EHeM+sigmsg at m5p.com  PGP 87145445         |    )   /
  \_CS\   |  _____  -O #include <stddisclaimer.h> O-   _____  |   /  _/
8A19\___\_|_/58D2 7E3D DDF4 7BA6 <-PGP-> 41D1 B375 37D0 8714\_|_/___/5445