Avahi possible regression in 0.6.10-0ubuntu3.2

I am also affected by this bug on i386

Will check that out, thanks for the bug.

I'm also affected in exactly the same way.

This is known upstream, see Avahi bug #72 (linked)
I hope this will be fixed shortly.

Adding backport tasks, USN-380-1 introduced this regression.

Proposed upstream patch works wonderfully, thanks Trent!

 avahi (0.6.15-1ubuntu5) feisty; urgency=low
 .
   * Add debian/patches/00upstream_01-check-netlink-creds.patch:
     - Check validity of netlink packets by verifying that they come from uid
       0; drop the previous PID-based check, since it does not work.
     - This should fix avahi for operation with network-manager again.
     - Closes: LP#72728
   * debian/patches/ubuntu-dhclient-check-unicast-local.patch:
     - Completely disabling Avahi for all times is a bit harsh, given that this
       situation can happen dynamically in any random WLAN. If a unicast
       .local domain is found, merely stop avahi, but do not disable it in the
       default file.
     - Also, create a flag /var/run/disabled-for-unicast-local when stopping
       Avahi, so that user's sessions can pick this up and display a
       notification.

Any idea how long it will take to release the corrected edgy binary package?

I need a running system at friday, is it worth waiting or should I better change the sources according to the patch and build it by hand?

At this stage I would be doing up my own package, the feisty package likely wont compile on edgy due to the IFLA_RTA patch being somewhat incorrect.

You could grab SVN, or grab the feisty package and modify avahi-core/iface-linux.c and avahi-autoipd/iface-linux.c according to these patches

http://www.avahi.org/changeset/1338
http://www.avahi.org/changeset/1339

I will prepare an update now, so it should actually make it until tomorrow. No firm promises, though.

Fixes uploaded, awaiting build and publishing.

Excellent thx.

http://www.ubuntu.com/usn/usn-380-2