Re: How to turn off IPV6 (link local)

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

 



On Thu, Jun 17, 2004 at 08:13:47AM +0300, Pekka Savola wrote:
> On Wed, 16 Jun 2004, Bernd Eckenfels wrote:
> > In article <1087370535.24446.2981.camel@segesta.zurich.ibm.com> you wrote:
> > >> If the application prefers ipv6 it will try to use them, and only on connect
> > >> it receives an error. Some do then ipv4 fallback (lynx) some dont (mozilla).
> > > 
> > > Fix your routing as that is the problem.
> > 
> > No it is not, a Host with only link-local addresses is perfectly well
> > configured. The application can actually bind to the family and try to
> > connect to the internet destination, it will just have to deal with the net
> > not reachable error. And some applications dont. (but this is actually only
> > the reson why i ave looked at the sysctl options, even if all aplications
> > are fixed the options still behave wrong)
> [...]
> 
> I think we may be having a slightly different problem here (as what
> the others are discussing).  When you have enabled IPv6 but the
> network does not support IPv6 (i.e., you have link-local addresses),
> there are certain problems which may lead to longer timeouts.  This
> depends principally on (at least) two things:
> 
>  1) whether the node has a "on-link assumption", i.e., a default route 
> to your interface.
> 
>  2) whether TCP implementation aborts connection when it receives a 
> "soft" ICMP error (against the host requirements RFC), 
> 
> These are both described at some length in 
> http://www.ietf.org/internet-drafts/draft-ietf-v6ops-v6onbydefault-02.txt
> 
> The key point here is that in these cases, even if the application was 
> properly programmed, you might need to wait for a long time (dozens of 
> seconds, even minutes) before the application would fall back to IPv4.
> 
> IMHO, Linux should do both:
>  1) remove the IPv6 default routes pointing on each interface when the 
> interface is created,

They key is, there is no default route, unless its autoconfigured, in
whcih it should have one....

>  2) abort TCP connections which are in SYN-SENT state when an ICMP 
> error is received.
> 
> This would help the robustness a lot in the scenarios where you want 
> to enable IPv6, and make sure it works, even if you didn't have IPv6 
> connectivity.
> 
> I.e., this is a critical thing for vendors which might want to ship 
> with IPv6 enabled by default.
> 
> -- 
> Pekka Savola                 "You each name yourselves king, yet the
> Netcore Oy                    kingdom bleeds."
> Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
> 
> 
> 
> -
> : send the line "unsubscribe linux-net" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Trent Lloyd <lathiat@bur.st>
Bur.st Networking Inc.
-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Netdev]     [Ethernet Bridging]     [Linux 802.1Q VLAN]     [Linux Wireless]     [Kernel Newbies]     [Security]     [Linux for Hams]     [Netfilter]     [Git]     [Bugtraq]     [Yosemite News and Information]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux PCI]     [Linux Admin]     [Samba]

  Powered by Linux