>> This process continues so that retransmissions occur with an exponentially
>>  increasing interval that caps at T2.

I think this actually means

> - Initially => T1 = 0.5s
> - Fires    => MIN(2*T1, T2) = 1s
> - Fires    => MIN(4*T1, T2) = 2s
> - Fires    => MIN(8*T1, T2) = 4s
> - Fires    => MIN(16*T1, T2) = 4s

2,4, 8, 16... which is exponential.

On Wed, Apr 6, 2011 at 10:06 PM, Iñaki Baz Castillo <[email protected]> wrote:
> Hi, RFC 3261 section 17.1.2.2 states that:
>
>   If an
>   unreliable transport is in use, the client transaction MUST set timer
>   E to fire in T1 seconds.  If timer E fires while still in this state,
>   the timer is reset, but this time with a value of MIN(2*T1, T2).
>   When the timer fires again, it is reset to a MIN(4*T1, T2).  This
>   process continues so that retransmissions occur with an exponentially
>   increasing interval that caps at T2.  The default value of T2 is 4s,
>   and it represents the amount of time a non-INVITE server transaction
>   will take to respond to a request, if it does not respond
>   immediately.  For the default values of T1 and T2, this results in
>   intervals of 500 ms, 1 s, 2 s, 4 s, 4 s, 4 s, etc.
>
> I've calculated those intervals and can't get same values:
>
> Timer E:
> - Initially => T1 = 0.5s
> - Fires    => MIN(2*T1, T2) = 1s
> - Fires    => MIN(4*T1, T2) = 2s
> - Fires    => MIN(4*T1, T2) = 2s
> - Fires    => MIN(4*T1, T2) = 2s
> ...
>
> So the resulting intervals are 500ms, 1s, 2s, 2s, 2s, 2s, etc.
> In short, it's obvious that MIN(4*T1, T2) will always be 2 seconds! it
> can never reach 4s.
>
>
> IMHO the text should say:
>
>   If timer E fires while still in this state,
>   the timer is reset, but this time with a value of MIN(2*TIMER_E, T2).
>
> And remove the line about "When the timer fires again, it is reset to
> a MIN(4*T1, T2)
>
> So we get:
>
> Timer E:
> - Initially => T1 = 0.5s
> - Fires    => MIN(2*TIMER_E, T2) = 1s
> - Fires    => MIN(2*TIMER_E, T2) = 2s
> - Fires    => MIN(2*TIMER_E, T2) = 4s
> - Fires    => MIN(2*TIMER_E, T2) = 4s
>
>  Am I wrong? Thanks a lot.
>
>
> PS: Sorry for the cross-posting, I don't know which maillist is better
> to report it.
>
>
> --
> Iñaki Baz Castillo
> <[email protected]>
>
> _______________________________________________
> Sip-implementors mailing list
> [email protected]
> https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to