Re: [trill] [rbridge] Query on draft-ietf-trill-esadi-00

Hi, chatterjee.

Thanks for your comments

Please see my comment in inline.

>Hi Authors,

>I have few comments and a query about  draft-ietf-trill-esadi-00.

1> Section 4.2 Receipt of ESADI PDUs

>   "Because all ESADI instance for VLAN-x are adjacent, when RB1
>receives an ESADI-CSNP from RB2 and detects that it has ESADI-LSPs
>that RB1 is missing, it sets the transmission flag only for its own
>ESADI-LSPs that RB1 is missing. Missing ESADI-LSPs originated by other
> ESADI instances will be detected by those other instances."

>   I am not able to understand the quoted sentences.Could you please 
clarify?

[hfw] I think this is simiar with the ISIS flooding mechanism.

RB1 receives ESADI-CSNP from RB2, RB2 may DRB or not. ESADI-LSP contains 
LSP list.

If the LSP list can not contain the LSP originated by RB1, or the LSP 
stored in RB1 locally is newer than the received list,
RB1 sets the transimission flag for its own originated LSP. If the 
transimission time out, RB1 should multicast its originated LSP.

If the LSP in the ESADI-CSNP LSP list is newer than the its local, a 
ESADI-PSNP will be schedule to send in the network.

>2> The figure 2 and 3 in Section 2. ESADI Protocol Overview have the
>Inner MAC DA as "Next Hop Dest. Address"   which should be
>"All-Egress-RBridges" (formerly called "All-ESADI-RBridges")

 [hfw]You are right. There is some editor's error in the figure.

In figure 2 & 3 in ESADI draft, the Inner MAC DA should be 
"All-Egress-RBridges". 

In figure 2, the Outer MAC DA as "All-RBrides Multicast Address" should be 
"Next Hop Dest. Address", because both unicast and multicast ESADI frames 
are supported by ESADI.  

>3> In Section 6. ESADI-LSP Contents
>   "ESADI-LSP number zero MUST NOT exceed 1470 minus 24 bytes in
>length (1446 bytes) but if received longer, it is still processed
>normally."
>
>   should be changed to
>
>   "The ESADI payload for ESADI-LSP number zero MUST NOT exceed 1470
>minus 24 bytes in length  (1446 bytes) but if received longer, it is
>still processed normally."
>Also should it be 1470-24 or Sz-24 ? Sz being the campus wide MTU.

[hfw] Ok

somnath chatterjee <somnath.chatterjee01@gmail.com> 
发件人:  trill-bounces@ietf.org
2012-07-12 20:32

收件人
trill@ietf.org
抄送

主题
[trill] [rbridge] Query on draft-ietf-trill-esadi-00

Hi Authors,

I have few comments and a query about  draft-ietf-trill-esadi-00.

1> Section 4.2 Receipt of ESADI PDUs

   "Because all ESADI instance for VLAN-x are adjacent, when RB1
receives an ESADI-CSNP from RB2 and detects that it has ESADI-LSPs
that RB1 is missing, it sets the transmission flag only for its own
ESADI-LSPs that RB1 is missing. Missing ESADI-LSPs originated by other
ESADI instances will be detected by those other instances."

   I am not able to understand the quoted sentences.Could you please 
clarify?

2> The figure 2 and 3 in Section 2. ESADI Protocol Overview have the
Inner MAC DA as "Next Hop Dest. Address"   which should be
"All-Egress-RBridges" (formerly called "All-ESADI-RBridges")

3> In Section 6. ESADI-LSP Contents
   "ESADI-LSP number zero MUST NOT exceed 1470 minus 24 bytes in
length (1446 bytes) but if received longer, it is still processed
normally."

   should be changed to

   "The ESADI payload for ESADI-LSP number zero MUST NOT exceed 1470
minus 24 bytes in length  (1446 bytes) but if received longer, it is
still processed normally."
Also should it be 1470-24 or Sz-24 ? Sz being the campus wide MTU.

Thank you,
Regards,
Somnath
_______________________________________________
trill mailing list
trill@ietf.org
https://www.ietf.org/mailman/listinfo/trill