Re: IGMPv4

Hi

I think there are issues with IGMPv3. I suggested IGMPv4 some time ago.

I think that some issues are:

1. Mixing include and exclude sources -> information of sources of SSM (source specific multicast) is lost when mixing and PIM router can not use SSM
2. Separating include / exclude -> group timer and exclude list are not neccessary
3. No exclude list -> all the timers of exclude sources =0 -> no need for timers in exclude sources
4. No need for timers in Include sources when the routers keeps information of every host and channel -> no include timers (the reason is in GMI)
5. no exclude timers + no include timers -> timers are not necessary -> everything is much simpler, 

1,2,3,4,5 -> the state machine of the router is decribed with 8 lines (without source timers, group timers, exclude lists and Q(G,S))

regards

Alvaro

PS. BCP79/ more than 10 applications filed in USPTO / first one will be isued in 12 days.


-----Mensaje original-----
De: [email protected] en nombre de Indranil Bhattacharya
Enviado el: jue 17/12/2009 8:16
Para: Bharat Joshi
CC: [email protected]
Asunto: Re: [magma] IGMPv4
 
Thanks Brian and Bharat. No, I do not see any issue. I just asked.

Regards,
Indranil

On Thu, Dec 17, 2009 at 8:42 AM, Bharat Joshi <[email protected]>wrote:

> Hi,
>
>     Though there are no proposals for IGMPv4 right now but a good effort to
> simplify the IGMPv3/MLDv2 protocol is going on in MBONED. You can take a
> look at tools.ietf.org/html/draft-ietf-mboned-lightweight-igmpv3-mldv2
>
>     BTW, do you see any issues in current version of IGMP/MLD because of
> which you are asking for the next version?
>
> Thanks,
> Bharat
> ________________________________________
> From: [email protected] [[email protected]] On Behalf Of Brian
> Haberman [[email protected]]
> Sent: Thursday, December 17, 2009 7:21 AM
> To: Indranil Bhattacharya
> Cc: [email protected]
> Subject: Re: [magma] IGMPv4
>
> Indranil Bhattacharya wrote:
> > Hello,
> >
> >         Is there work going on IGMPV4/MLDV3? Where can I find documents
> > on this?
> >
>
> There are no proposals for additional work on IGMP or MLD.
>
> Regards,
> Brian
>
> _______________________________________________
> magma mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/magma
>

_______________________________________________
magma mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/magma