netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* ip -6 addr flush dev <device> flushes also the autogenerated link-local address
@ 2005-03-02 21:59 Peter Bieringer
  2005-03-02 22:09 ` jamal
  0 siblings, 1 reply; 6+ messages in thread
From: Peter Bieringer @ 2005-03-02 21:59 UTC (permalink / raw)
  To: Maillist USAGI-users; +Cc: Maillist netdev

Hi,

sure one know that using 

       ip -6 addr flush dev <device>

flushes (removes) all IPv6 addresses on a device.

Unfortunately, it also removes the autogenerated link-local address.

It reappears only if interface was cycled down and up, not so nice.


What about an feature that the automagically generated link-local address
will not be removed on "flush"?

        Peter
-- 
Dr. Peter Bieringer                     http://www.bieringer.de/pb/
GPG/PGP Key 0x958F422D               mailto: pb at bieringer dot de 
Deep Space 6 Co-Founder and Core Member  http://www.deepspace6.net/

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: ip -6 addr flush dev <device> flushes also the autogenerated link-local address
  2005-03-02 21:59 ip -6 addr flush dev <device> flushes also the autogenerated link-local address Peter Bieringer
@ 2005-03-02 22:09 ` jamal
  2005-03-02 23:14   ` Thomas Graf
  0 siblings, 1 reply; 6+ messages in thread
From: jamal @ 2005-03-02 22:09 UTC (permalink / raw)
  To: Peter Bieringer; +Cc: Maillist USAGI-users, Maillist netdev, Thomas Graf


Is this a user space problem? I think we had something along these lines
fixed recently. Thomas?

cheers,
jamal

On Wed, 2005-03-02 at 16:59, Peter Bieringer wrote:
> Hi,
> 
> sure one know that using 
> 
>        ip -6 addr flush dev <device>
> 
> flushes (removes) all IPv6 addresses on a device.
> 
> Unfortunately, it also removes the autogenerated link-local address.
> 
> It reappears only if interface was cycled down and up, not so nice.
> 
> 
> What about an feature that the automagically generated link-local address
> will not be removed on "flush"?
> 
>         Peter

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: ip -6 addr flush dev <device> flushes also the autogenerated link-local address
  2005-03-02 22:09 ` jamal
@ 2005-03-02 23:14   ` Thomas Graf
  2005-03-03  0:00     ` YOSHIFUJI Hideaki / 吉藤英明
  0 siblings, 1 reply; 6+ messages in thread
From: Thomas Graf @ 2005-03-02 23:14 UTC (permalink / raw)
  To: jamal; +Cc: Peter Bieringer, Maillist USAGI-users, Maillist netdev

* jamal <1109801372.1098.228.camel@jzny.localdomain> 2005-03-02 17:09
> 
> Is this a user space problem? I think we had something along these lines
> fixed recently. Thomas?

I didn't fix any of these but it's definitely a userspace problem.
One can easly work around this by iterating over all scopes
except for link local and provide the scope to flush. We'll have
to put additional filters into print_addrinfo() if we want this
to be default but I'm not even sure what the official policy should
be like. Any standards on this?

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: ip -6 addr flush dev <device> flushes also the autogenerated link-local address
  2005-03-02 23:14   ` Thomas Graf
@ 2005-03-03  0:00     ` YOSHIFUJI Hideaki / 吉藤英明
  2005-03-03  7:25       ` Peter Bieringer
  0 siblings, 1 reply; 6+ messages in thread
From: YOSHIFUJI Hideaki / 吉藤英明 @ 2005-03-03  0:00 UTC (permalink / raw)
  To: tgraf; +Cc: hadi, pb, usagi-users, netdev, yoshfuji

In article <20050302231400.GU31837@postel.suug.ch> (at Thu, 3 Mar 2005 00:14:00 +0100), Thomas Graf <tgraf@suug.ch> says:

> * jamal <1109801372.1098.228.camel@jzny.localdomain> 2005-03-02 17:09
> > 
> > Is this a user space problem? I think we had something along these lines
> > fixed recently. Thomas?
> 
> I didn't fix any of these but it's definitely a userspace problem.
> One can easly work around this by iterating over all scopes
> except for link local and provide the scope to flush. We'll have
> to put additional filters into print_addrinfo() if we want this
> to be default but I'm not even sure what the official policy should
> be like. Any standards on this?

I believe that flush should remove all addresses including link-local.
So, current behavior is correct.

--yoshfuji

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: ip -6 addr flush dev <device> flushes also the autogenerated link-local address
  2005-03-03  0:00     ` YOSHIFUJI Hideaki / 吉藤英明
@ 2005-03-03  7:25       ` Peter Bieringer
  2005-03-03 12:53         ` Thomas Graf
  0 siblings, 1 reply; 6+ messages in thread
From: Peter Bieringer @ 2005-03-03  7:25 UTC (permalink / raw)
  To: usagi-users, netdev



--On Thursday, March 03, 2005 09:00:09 AM +0900 "YOSHIFUJI Hideaki /
=?iso-2022-jp?B?GyRCNUhGIzFRTEAbKEI=?=" <yoshfuji@linux-ipv6.org> wrote:

> In article <20050302231400.GU31837@postel.suug.ch> (at Thu, 3 Mar 2005
> 00:14:00 +0100), Thomas Graf <tgraf@suug.ch> says:
> 
>> * jamal <1109801372.1098.228.camel@jzny.localdomain> 2005-03-02 17:09
>> > 
>> > Is this a user space problem? I think we had something along these
>> > lines fixed recently. Thomas?
>> 
>> I didn't fix any of these but it's definitely a userspace problem.
>> One can easly work around this by iterating over all scopes
>> except for link local and provide the scope to flush. We'll have
>> to put additional filters into print_addrinfo() if we want this
>> to be default but I'm not even sure what the official policy should
>> be like. Any standards on this?
> 
> I believe that flush should remove all addresses including link-local.
> So, current behavior is correct.

My pitfall was that I didn't know (but Pekka told me) that "ip -6 flush"
also understand scopes.

So I use the suggestion given by Thomas to run through all scopes of the
addresses and flush dedicated like

ip -6 addr flush dev <device> scope global
ip -6 addr flush dev <device> scope site

BTW: how many scopes are currently defined? 

ip help shows me:
 SCOPE-ID := [ host | link | global | NUMBER ]

What means NUMBER and why is "site" understood but not in online help?

        Peter
-- 
Dr. Peter Bieringer                     http://www.bieringer.de/pb/
GPG/PGP Key 0x958F422D               mailto: pb at bieringer dot de 
Deep Space 6 Co-Founder and Core Member  http://www.deepspace6.net/

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: ip -6 addr flush dev <device> flushes also the autogenerated link-local address
  2005-03-03  7:25       ` Peter Bieringer
@ 2005-03-03 12:53         ` Thomas Graf
  0 siblings, 0 replies; 6+ messages in thread
From: Thomas Graf @ 2005-03-03 12:53 UTC (permalink / raw)
  To: Peter Bieringer; +Cc: usagi-users, netdev

> BTW: how many scopes are currently defined? 
> 
> ip help shows me:
>  SCOPE-ID := [ host | link | global | NUMBER ]
> 
> What means NUMBER and why is "site" understood but not in online help?

The kernel hardcodes the following scopes internally:
0 {global | universe}
200 site
253 link
254 host
255 nowhere

Additional names may be assigned to numbers in /etc/iproute2/rt_scopes

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2005-03-03 12:53 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-03-02 21:59 ip -6 addr flush dev <device> flushes also the autogenerated link-local address Peter Bieringer
2005-03-02 22:09 ` jamal
2005-03-02 23:14   ` Thomas Graf
2005-03-03  0:00     ` YOSHIFUJI Hideaki / 吉藤英明
2005-03-03  7:25       ` Peter Bieringer
2005-03-03 12:53         ` Thomas Graf

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).