[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: defaulting to net.ipv6.bindv6only=1 for squeeze



On Sat, Oct 24, 2009 at 08:24:31PM +0200, Marco d'Itri wrote:
> 
> Applications can change the behaviour for their sockets using setsockopt
> and the IPV6_V6ONLY option[2], and many already do this to prevent the
> need of adjusting their configuration depending on how the system is
> configured.

Why don't you fix the applications not doing that?  As I see
it this only has an effect on what I would call broken
applications.  I see 2 possible things:
- The application is broken and doesn't try to open 2 sockets
  at all.  This means now you break their ipv4 support.
- The application already tries to open the 2 sockets, but
  didn't call setsockopt() so the 2nd one just fails.  Adding
  the setsockopt() is a very simple change and doesn't break
  anything and has the effect you want.  This is a very
  simple change to make.

It looks like the only application I'm running that isn't calling
the setsockopt is samba.


Kurt


Reply to: