Re: [Ice] I submitted and plan to present draft-thatcher-ice-remove-candidate in Berlin

Peter said:

"The real question is whether the removing agent needs to know whether the
receiving agent is going to do anything with the signal.  And since
everything will continue to work even if the receiving agent ignores the
signal (simply with some perhaps wasted resources), then it's not necessary
for negotiation at all."

[BA] The removing agent only needs to know if the receiving agent
understands removal if that will make a difference in subsequent behavior.
If a removed candidate will never be re-added, I don't think it matters,
because a receiving agent not understanding removal will eventually figure
out that the removed candidate is inactive, and that pairs involving the
removed candidate are unusable. But if it is desired to be able to re-add a
removed candidate without an ICE restart, then it might make a difference.

On Tue, Jul 12, 2016 at 10:21 AM, Peter Thatcher <pthatcher@google.com>
wrote:

>
>
> On Tue, Jul 12, 2016 at 10:07 AM, Christer Holmberg <
> christer.holmberg@ericsson.com> wrote:
>
>> Hi,
>>
>>
>>
>> The text says:
>>
>>
>>
>> “The Receiving Agent MAY keep the existing candidate pairs that use
>>
>>               the Removed Candidates and behave as though the Removed
>> Candidates
>>
>>               had not been removed for those candidate pairs.”
>>
>>
>>
>> What is meant by “behave”? Sending keep-alives? Sending media? That will
>> obviously not work.
>>
>
> Both keep-alives and media.  If the removed candidate is a TURN candidate
> that the removing agent wishes to deallocate, it may work for a short
> period of time before the removing agent actually deallocates it.  So the
> Receiving Agent may choose to keep sending media until it finds a different
> candidate pair to use instead.   Or it may not.
>
>
>>
>>
>> I think it would be good to state wether the usage/support of the
>> mechanism must be negotiated before it is used.
>>
> 
> 
> Signaling the removal of candidate never does any harm, so I don't see a
> need in negotiating that your going to signal removal (modulo a specific
> signaling protocol needing to figure out how to send new messages if it
> chooses to do so).  The real question is whether the removing agent needs
> to know whether the receiving agent is going to do anything with the
> signal.  And since everything will continue to work even if the receiving
> agent ignores the signal (simply with some perhaps wasted resources), then
> it's not necessary for negotiation at all.
>
> You are correct that this should be clear in the draft.  I will add it.
>
>
>
>>
>>
>> Regards,
>>
>>
>>
>> Christer
>>
>>
>>
>>
>>
>>
>>
>> *From:* Ice [mailto:ice-bounces@ietf.org] *On Behalf Of *Peter Thatcher
>> *Sent:* 12 July 2016 17:35
>> *To:* ice@ietf.org
>> *Subject:* [Ice] I submitted and plan to present
>> draft-thatcher-ice-remove-candidate in Berlin
>>
>>
>>
>> I submitted draft-thatcher-ice-remove-candidate and would like to speak
>> about in my allotted time on the agenda when I will also be speaking about draft-thatcher-ice-network-cost
>> and draft-thatcher-ice-renomination.
>>
>>
>>
>> It's a very simple draft. Basically it says you can "remove" candidates
>> just like trickle-ice allows you to add candidates. You could probably read
>> it in 3 minutes. Please do so:
>>
>>
>>
>> https://www.ietf.org/id/draft-thatcher-ice-remove-candidate-00.txt
>>
>>
>>
>> Thanks,
>>
>> Peter (Chair hat off)
>>
>
>
> _______________________________________________
> Ice mailing list
> Ice@ietf.org
> https://www.ietf.org/mailman/listinfo/ice
>
>