[tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Guy Harris
It's 2008. Should we enable IPv6 support by default in libpcap and tcpdump (as long as the OS supports IPv6 to a sufficient extent that we can compile the support in), and let users do --disable-ipv6 if, for whatever reason, they don't want it? - This is the tcpdump-workers list. Visit

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Gert Doering
Hi, On Wed, Feb 06, 2008 at 03:37:05AM -0800, Guy Harris wrote: It's 2008. Should we enable IPv6 support by default in libpcap and tcpdump (as long as the OS supports IPv6 to a sufficient extent that we can compile the support in), and let users do --disable-ipv6 if, for whatever reason,

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Rick Jones
Gert Doering wrote: Hi, On Wed, Feb 06, 2008 at 03:37:05AM -0800, Guy Harris wrote: It's 2008. Should we enable IPv6 support by default in libpcap and tcpdump (as long as the OS supports IPv6 to a sufficient extent that we can compile the support in), and let users do --disable-ipv6 if, for

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread sthaug
It's 2008. Should we enable IPv6 support by default in libpcap and tcpdump (as long as the OS supports IPv6 to a sufficient extent that we can compile the support in), and let users do --disable-ipv6 if, for whatever reason, they don't want it? Yes please. Steinar Haug, Nethelp

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Gert Doering
Hi, On Wed, Feb 06, 2008 at 10:09:33AM -0800, Rick Jones wrote: What is the reason for having optional IPv6 in the first place (besides OSes that don't provide all necessary header files)? Memory savings? ISTR there were some funnies on some OSes where IPv6 was pre-enabled but not

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Carter Bullard
Are you going to make it so that the routines in inet.c, such as add_addr_to_iflist() and pcap_lookupnet() can work with IPv6 addresses? Carter On Feb 6, 2008, at 3:41 PM, Gert Doering wrote: Hi, On Wed, Feb 06, 2008 at 10:09:33AM -0800, Rick Jones wrote: What is the reason for having

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Guy Harris
Carter Bullard wrote: Are you going to make it so that the routines in inet.c, such as add_addr_to_iflist() and pcap_lookupnet() can work with IPv6 addresses? add_addr_to_iflist() appears to work with IPv6 addresses on my OS X 10.4.11 machine, as pcap_findalldevs() finds the IPv6 addresses

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Rick Jones
Gert Doering wrote: Hi, On Wed, Feb 06, 2008 at 10:09:33AM -0800, Rick Jones wrote: What is the reason for having optional IPv6 in the first place (besides OSes that don't provide all necessary header files)? Memory savings? ISTR there were some funnies on some OSes where IPv6 was

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Gert Doering
Hi, On Wed, Feb 06, 2008 at 12:58:49PM -0800, Guy Harris wrote: Gert Doering wrote: tcpdump is not actually *doing* any IPv6. It's doing IPv6 name resolution. OK, good argument. Some IPv6 related stuff without actually sending IPv6 packets (unless the resolver library does it internally)

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Carter Bullard
I think the point is that you shouldn't have routines that are 'inherently' limited to IPv4 addresses, if you support IPv6. Tcpdump is still using pcap_lookupnet(). On Feb 6, 2008, at 4:48 PM, Guy Harris wrote: Carter Bullard wrote: Are you going to make it so that the routines in inet.c,

Re: [tcpdump-workers] Should we enable IPv6 support by default?

2008-02-06 Thread Guy Harris
Carter Bullard wrote: I think the point is that you shouldn't have routines that are 'inherently' limited to IPv4 addresses, if you support IPv6. Removing pcap_lookupnet() would break source and binary compatibility. I suppose libpcap 1.0 could, in theory, break both, so we could perhaps get