Re: [mpls] A comment on the TTL in draft-dong-mpls-mna-encaps-00

["Dongjie (Jimmy)" <[e-mail]>]

Hi Stewart,

Thanks a lot for your review and comment.

I agree the design about the TTL field in the uniform model with PHP needs further consideration, and what you suggested can be one possible approach.

While I noticed that in RFC 6790 section 4.4, it says:

"a PHP LSR that recognizes the ELI MAY choose to pop the ELI and following label (which should be an entropy label) in addition to popping the tunnel label a PHP LSR "MAY choose to pop the ELI and following label (which should be an entropy label) in addition to popping the tunnel label, provided that doing so doesn't diminish its ability to load balance on the next hop."

Maybe some description about the behavior of the PHP LSR on the MNA Indicator could also be considered.

Best regards,
Jie

From: Stewart Bryant [mailto:[e-mail]]
Sent: Thursday, August 4, 2022 9:18 PM
To: mpls <[e-mail]>
Cc: Stewart Bryant <[e-mail]>; [e-mail]
Subject: A comment on the TTL in draft-dong-mpls-mna-encaps-00








3<https://datatracker.ietf.org/doc/html/draft-dong-mpls-mna-encaps#section-3>.  MNA Indicator



   The MNA Indicator is introduced to indicate the presence of any MNA

   information in the packet.  It can be used to indicate the existence

   of MNA actions and the optional associated data in the ISD, or the

   PSD or both.  Since this indicator is generic for all types of MPLS

   network actions, it is reasonable to allocate a basic Special Purpose

   MPLS label (bSPL) for it.  The TC and TTL fields of the MNA Indicator

   are redefined as flags.



   The format of the MNA Indicator is shown as below:



       0                   1                   2                   3

       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1

      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

      |       MNA Indicator=SPL (TBA)         |H|I|P|S|ISF|    RSV    |

      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

                  Figure 1. The format of MNA Indicator



   Where:



<snip>





   *  RSV: The rest 6 bits in the original TTL field are reserved for

      future use.  They MUST be set to 0 on transmit and MUST be ignored

      on receipt.

SB> There might be some merit in setting the RSV to 0b111111

SB> Then if some router receiving the SPL predecremented the TTL (not unknown in hardware) the upper bits of the TTL and hence the indicators would be unchanged

- Stewart