> Date: Thu, 6 Mar 2008 21:41:40 +1100 > From: Peter Jeremy <[EMAIL PROTECTED]> > Sender: [EMAIL PROTECTED] > > On Thu, Mar 06, 2008 at 09:51:11AM +0100, Ruben van Staveren wrote: > >The interesting thing, to stay on topic, is that people are willing to > >explore a feature called "SCTP" which to my knowledge is younger than > >"IPv6". This makes the whole discussion sort of moot, right ? > > In my case, I have a use for SCTP at work (we are using various > protocols that run on top of SCTP) but we don't have any IPv6 networks > in use. Personally, I find the IPv6 data reported in things like > netstat are annoying.
alias netstat netstat -f inet And the annoying part is NOT IPv6 sockets. There are not many. It's Unix sockets. But the alias takes care of both it they really annoy you. (I run IPv6, so I don't have such an issue.) > >had TCP/UDP for many years and they are still serving their purpose well, > >so why change ? > > TCP isn't sufficiently robust for some Telco purposes: They can't > accept the time it takes TCP to detect or recover from a link failure. > > >So give it a chance, only then there will be feedback and only then we can > >fix the problems. Otherwise it will stay just theoretical. > > Agreed. But at this stage I can't justify the effort to do anything > more than have a very cursory glance it at. What benefit would I > derive from setting up an IPv6 network and attempting to experiment > with it? My ISP won't support IPv6 and I'm reasonably certain my > cable-modem doesn't either so IPv6 connectivity would entail some > sort of tunnel. You don't set up an IPv6 network. You simply have end nodes that will use IPv6 when/if it is available by just making a one-line change in rc.conf as opposed to a kernel re-build. If you have a Windows Vista box (and I'm told several people do, even though I have never used one), it has IPv6, it is always enabled, and it REALLY, REALLY tries to use it using several mechanisms including Toredo tunnels (which are either very cool or the spawn of Satan, depending on who you talk to). -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: [EMAIL PROTECTED] Phone: +1 510 486-8634 Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751
pgp7JlnMbxRfR.pgp
Description: PGP signature
