What do we need IPv6 really for? (Re: IPv6 traffic data in Asian networks?)

[Daniel G. Kluge]

Am 22.03.2007 um 17:05 schrieb Kevin Loch:

> Rémi Denis-Courmont wrote:
>> I can only agree here. For any service that works fine through  
>> IPv4 NATs, if not through proxies, there is really no business  
>> advantage to offering IPv6... and there will definitely be a lower  
>> perceived quality of service on average: IPv6 has much more  
>> transient (or not) reachability problems, and typically bigger  
>> latencies and lower bandwidth, particularly if a transition  
>> mechanisms are involved.
>> As such, IPv6 for any commercial website is, I am afraid to say, a  
>> big NO NO at the moment. I am pretty much afraid this is a vicious  
>> circle, to which I could find no exit strategy.
>
> With the exception of the ARIN website itself, I have not seen "much
> more" transient reachability problems on IPv6.  I have seen IPv6  
> enabled
> on commercial websites without any problems.  I'm not saying
> it's perfect but it's alot better than "NO NO NO".
>

Well, I wouldn't call it transient, but there was an ISP that had a  
complete dual-stack setup that was missing 2002::/32 for several  
months and nobody seemed to have noticed or bothered.

On a side note, I stopped using IPv6 because it gives me:

panic(cpu 0 caller 0x0035D913): freeing free mbuf
Backtrace, Format - Frame : Return Address (4 potential args on stack)
0x14043da8 : 0x128d08 (0x3cb134 0x14043dcc 0x131de5 0x0)
0x14043de8 : 0x35d913 (0x3eb914 0xc 0x14043e48 0x128f65)
0x14043e28 : 0x6d44a4 (0x25cbc500 0x0 0x1 0x2)
0x14043e48 : 0x98d47c (0x127e5000 0x25cbc500 0x0 0x6)
0x14043e78 : 0x98b6f8 (0x127e5000 0x25cbc500 0x0 0x28c2f80)
0x14043e98 : 0x98f8c2 (0x127e5000 0x0 0x1000000 0x279e640)
0x14043f08 : 0x39a463 (0x127e5000 0x28c2f80 0x1 0x279e640)
0x14043f58 : 0x399635 (0x28c2f80 0x135eb4 0x0 0x279e640)
0x14043f88 : 0x39936b (0x28bc4c0 0x4c 0x0 0x0)
0x14043fc8 : 0x19ac1c (0x28bc4c0 0x0 0x19e0b5 0x29014a8) Backtrace  
terminated-in
valid frame pointer 0x0
       Kernel loadable modules in backtrace (with dependencies):
          com.apple.iokit.AppleYukon(1.0.11b2)@0x989000
             dependency: com.apple.iokit.IONetworkingFamily(1.5.1) 
@0x6cd000
             dependency: com.apple.iokit.IOPCIFamily(2.2)@0x572000
             dependency: com.apple.iokit.IOACPIFamily(1.2.0)@0x64c000
          com.apple.iokit.IONetworkingFamily(1.5.1)@0x6cd000

Kernel version:
Darwin Kernel Version 8.9.1: Thu Feb 22 20:55:00 PST 2007;  
root:xnu-792.18.15~1/RELEASE_I386

And yes, it also crashed on 10.4.8 too, that's three months, one  
minor version and counting...

Cheers,
-daniel