Ubuntu
network-manager package

  1. Bug #183303
  2. Comment #14

Comment 14 for bug 183303

Revision history for this message
Motin (motin) wrote : Re: Can't create new wireless network

After receiving a very informative comment from NetworkManager Dan Williams (see the upstreams linked bugreport: http://bugzilla.gnome.org/show_bug.cgi?id=541816#c2 ), it is very likely this bug is really a combination of one or many mac80211-based driver bugs as well as bugs in wpa_supplicant.

@Brian Pitts, @rihasant, @dnel,@dlgandalf:
You all seem to be using mac80211-based cards/drivers, which makes it even more viable.

@Francesco:
Since you are using ipw2200, Bug #75358 may be more appropriate for you. Dan Williams also replied in a similar manner here, but mentions a ipw2200 patch suggestion as well: http://www.nabble.com/NM-0.6.5-Great-Problem-td18027324.html

kernel (wireless-testing.git):
872ba53395b2a8be08c3ea2d39e225e5b4a8cb40 mac80211: decrease IBSS creation
latency
507b06d0622480f8026d49a94f86068bb0fd6ed6 mac80211: send association event on
IBSS create

Two following four patches was suggested to be able to create Ad Hoc networks using 80211-based drivers:
wpa_supplicant:
[PATCH] wpa_supplicant: give adhoc associations a bit more time - http://lists.shmoo.com/pipermail/hostap/2008-June/017908.html
[PATCH v2] wext: handle mode switches correctly for mac80211 - http://lists.shmoo.com/pipermail/hostap/2008-June/017907.html

According to the following changelog, the two mac80211 patches has been incorporated into the mainline kernel by 2.6.25.6-1:
http://lists.rpath.org/pipermail/foresight-commits/2008-June/024368.html

The wpa_supplicant patches seemed to have been applied since version 0.6.3-2 (https://launchpad.net/ubuntu/+source/wpasupplicant):
  * Add 52_handle_mac80211_mode_switch.patch to enhance handling of mode
    switching for mac80211 using interfaces.
  * Add 53_give_adhoc_assoc_more_time.patch to give adhoc associations a bit
    more time.

Thus, it would be very interesting to hear if this problem persists in the current Intrepid development version (it oughtn't...)! Do we have any brave Intrepid tester that can confirm or approve that these patches have fixed the issue?