Re: [kitten] Cancel message Re: Alexey's comments Re: WGLC of draft-ietf-kitten-sasl-oauth-18

[Alexey Melnikov <[e-mail]>]

Hi Bill,

> On 8 Jan 2015, at 16:05, Bill Mills <[e-mail]> wrote:
> 
> You're talking about 3.5 in 4422?

Yes.

> To me it looks like the single kvsep message looks exactly like the abort message that needs to be defined by the protocol?

No, it is a SASL protocol mapping construct (defined once for all mechanisms), you defined a SASL mechanism construct.
> 
> 
> On Thursday, January 8, 2015 1:49 AM, Alexey Melnikov <[e-mail]> wrote:
> 
> 
> Hi Bill,
> 
>> On 6 Jan 2015, at 21:36, Bill Mills <[e-mail]> wrote:
>> 
>> " The client MUST then send either an additional client response consisting of a single %x01 (control A) character to the server in order to allow the server to finish the exchange or send a SASL cancellation token as defined in ACAP[RFC2244] section 6.3.1."
> 
> Actually I meant RFC 4422 (SASL). ACAP has cancellation token, but so does IMAP, SMTP, LDAP,...
> 
> If you like, I send you an IMAP example.
> 
>> 
>> 
>> On Sunday, January 4, 2015 3:37 AM, Alexey Melnikov <[e-mail]> wrote:
>> 
>> 
>> Hi Bill,
>> 
>> > On 3 Jan 2015, at 00:56, Bill Mills <[e-mail]> wrote:
>> > 
>> > 3.2.3 and an explicit message:  Long ago in the life of this doc I was told that some implementations may not support an empty message, so we put the single character message there to have an explicit payload.  I'm a bit leery of changing this now since there are implementations in play that use it this way.
>> 
>> 
>> I didn't suggest you should be sending empty message. I said you should be using SASL cancellation token, which is a mandatory RFC 4422 feature.
>> 
>> Any implementation would have to support this mode of operation anyway, because a SASL client can cancel any exchange.
> 
>