[Dnsmasq-discuss] 2.42 release candidate available.

[Simon Kelley]

Currently, all is quiet in dnsmasq development. Since it's the start of 
summer in my part of the world, I'm hoping that things stay that way so 
that I can get out and enjoy the sunshine.

Therefore, even though the changeset from 2.41 is not extensive, I'm 
planning to push out a 2.42 release soonish, to get the small number of 
bugfixes and enhancements I do have into circulation. So I've made a 
2.42rc1 release, available at

http://www.thekelleys.org.uk/dnsmasq/release-candidates/dnsmasq-2.42rc1.tar.gz

Please try it out and report any problems or regressions here.

The changelog from 2.41 is appended below.


Cheers,

Simon.



--------------------------------------------------------------------------------
             Define _GNU_SOURCE to avoid problems with later glibc
             headers. Thanks to Jima for spotting the problem.

             Add --dhcp-alternate-port option. Thanks to Jan Psota for
             the suggestion.

             Fix typo in code which is only used on BSD, when Dbus and
             IPv6 support is enabled. Thanks to Roy Marples.

             Updated Polish translations - thank to Jan Psota.

             Fix OS detection logic to cope with GNU/FreeBSD.

             Fix unitialised variable in DBus code - thanks to Roy
             Marples.

             Fix network enumeration code to work on later NetBSD -
             thanks to Roy Marples.

             Provide --dhcp-bridge on all BSD variants.

             Define _LARGEFILE_SOURCE which removes an arbitrary 2GB
             limit on logfiles. Thanks to Paul Chambers for spotting
             the problem.

             Fix RFC3046 agent-id echo code, broken for many
             releases. Thanks to Jeremy Laine for spotting the problem
             and providing a patch.

             Added Solaris 10 service manifest from David Connelly in
             contrib/Solaris10

             Add --dhcp-scriptuser option.

             Support new capability interface on suitable Linux
             kernels, removes "legacy support in use" messages. Thanks
             to Jorge Bastos for pointing this out.