Re: pessimistic NetworkManager and off-line mode (was: Diskless clients and NetworkManager)
- From: John Mahoney <jmahoney waav com>
- To: Marc Herbert <Marc Herbert gmail com>
- Cc: networkmanager-list gnome org
- Subject: Re: pessimistic NetworkManager and off-line mode (was: Diskless clients and NetworkManager)
- Date: Thu, 30 Apr 2009 12:51:00 -0400
For instance this most obvious false positive: you are connected but not
to the internet. Then why would pidgin try to reach MSN servers in this
case?
post from two days ago.
http://www.nabble.com/-WISH--Make-NM-autodetect-if-it-can-reach-the-internet-td23286488.html
Currently, its seems to be that if Network Manager does not handle the default route, you probably want to disable it and its back to scripts and command line. The design of NM seems to heavily rely on this default route caveat. It is kinda like using NM you become an indentured servant. Ease of use often comes at a compromise on flexibility. I guess the real question would be, is this default route caveat a necessity?
--
John
.
On Thu, Apr 30, 2009 at 12:25 PM, Marc Herbert
<Marc Herbert gmail com> wrote:
David Sundqvist a écrit :
> Perhaps. Taking a quick look at the NM project webpage, it's not
> really that farfetched: "Using the awesome power and flexibility of
> dbus and hal, NetworkManager provides facilities for other
> applications like browsers, email clients, or system services to be
> aware of the network's state and adjust their operation accordingly."
>
> That certainly could sound fairly authoritative to a presumptive
> software developer, so I can understand why they make that assumption.
WOW! I did not realize NM was that ambitious, thanks for enlightening me.
> 2) Disabuse application developers of the notion that NetworkManager
> has an accurate status. The first step of which would be to change the
> project page a bit to say "be aware of NM's perhaps not entirely
> accurate idea of the network's state"
You can indeed easily imagine tons of "accidental" network
configurations where the idea might be "not entirely accurate"; either
false positive (not actually connected) or false negative.
For instance this most obvious false positive: you are connected but not
to the internet. Then why would pidgin try to reach MSN servers in this
case?
- Granted false positives are not that bad. After all it is just back to
the old TCP time-outs days people were used to and seldom complaining about.
- On the other false negatives are much much more annoying, like seen
here: http://www.google.com/search?q=firefox+starts+offline (500,000
hits, thks David)
So maybe...
3) NM could simply lean towards the optimistic side and report
connectivity as soon as in doubt. Maybe NM should trigger off-line modes
ONLY when it is 200% sure that it OWNS EVERY single interface in the
system and is positively sure they are ALL absolutely out of order.
So instead of (too) ambitiously trying to be definitive and reporting
network "OFF" versus "ON", NM would rather be cautious + optimistic and
report network "really OFF" versus "probably ON". That would fix these
500,000 google hits without involving the application writers at all.
Comments?
_______________________________________________
NetworkManager-list mailing list
NetworkManager-list gnome org
http://mail.gnome.org/mailman/listinfo/networkmanager-list
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]