Mohamed Boucadair has entered the following ballot position for
draft-ietf-pce-sr-bidir-path-21: Discuss

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to 
https://www.ietf.org/about/groups/iesg/statements/handling-ballot-positions/ 
for more information about how to handle DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
https://datatracker.ietf.org/doc/draft-ietf-pce-sr-bidir-path/



----------------------------------------------------------------------
DISCUSS:
----------------------------------------------------------------------

Hi Cheng, Mach, Weiqiang, Rakesh, and Quan,

Thank you for the effort put into this specification. The theory of operation
is well described and key events are identified/logged.

Thanks to Carlos Pignataro for the OPSDIR review. Thanks to the authors for
engaging and addressing the points raised by Carlos. Appreciated.

Please find below some few DISCUSSion points:

# Multipath dependency

CURRENT:
   This extension also utilizes the
   procedure defined in [I-D.ietf-pce-multipath] to carry the multiple
   EROs and the associated reverse path EROs for an SR LSP.

As there is a dependency on that one, are we confident that one is stable
enough to progress this one?

I appreciate that the writeup says:

   “This I-D uses a PCEP Object/TLV defined in the normatively referenced I-D
   but
    makes no changes to it.”

but the question is also the other way around: is there any change in
I-D.ietf-pce-multipath that would impact this spec?

IMO, maybe wait I-D.ietf-pce-multipath to be approved before sending both
documents to the RFC Editor? This is more convenient rather than pulling the
doc from the RFC Editor queue.

# Too generic

CURRENT:
   The PCEP YANG module [RFC9826] defines a data model for LSP
   associations.  However, it does not include information for
   associated bidirectional SR LSPs.

ACK.

The question is: Are there key parameters that would be needed for the feature
defined in this doc. For example, would the following be needed:

* Indicate whether bidir paths are supported by a node?

* Control activate/deactivation of bidir association?

* Counters of active bidir paths?

* Counters of failed bidir paths?

# Liveness detection

CURRENT:
   Mechanisms defined in this document do not imply any new liveness
   detection and monitoring requirements.

I’m afraid some elaboration is needed here.

The intro claimed that some apps requires both direction, monitoring should be
covering both directions in order to assess the serviceability.

We need to say at least so (and require that both directions are correlated at
the manager level).


----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

# Stateless

The abstract says:

   The mechanisms defined in this document are
   applicable to both stateless and stateful PCEs for PCE-initiated and
   PCC-initiated LSPs

I’m not sure where stateless is discussed in the doc.

If so, can we make that explicit for readers? I’m not asking to change the
abstract but to have that in the introduction, for example.

# Mobile case

CURRENT:
   There are some applications that require bidirectional paths in SR
   networks, for example, such as in mobile backhaul transport networks.

## Great that you provided an example.

## As readers may not be familiar with specifics of that case: I wonder whether
you can include a pointer where one can further dig into this specific case.

# (Fully) Symmetric paths

I wonder whether we can clarify early in the doc that bidir is not necessarily
about having fully symmetric paths.

# Overview

Can we say that the description assumes that no errors are encountered and all
operations are successful.

# Section 8

I know the context of that naming in PCE, but the considerations are beyond
management. Also, in order to be consistent with latest discussion in OPS, I
suggest the following change:

OLD: 8.  Manageability Considerations

NEW: 8.  Operational Considerations

# nits

## paradigm

Source routing was there before SR. Can we please consider this change through
the doc:

OLD: source routing paradigm

NEW: source routing

## There isn’t only no one network

Please consider this change in the abstract

OLD:  in each direction in the network)

NEW: OLD:  in each direction in a network)

## Expand PCC in the introduction

## Terminology

OLD: Protocol (PCEP), PCEP Peer, PCEP speaker.

NEW: OLD: Protocol (PCEP), PCEP Peer, and PCEP speaker.

## Section 4

OLD: the the procedures defined in [RFC9059]

NEW: the procedures defined in [RFC9059]

## Section 5.2

OLD: The PST

NEW: The PCEP Path Setup Type (PST)

## Section 8.6: These are not new anymore

OLD: New tools such

NEW: Tools such

Cheers,
Med



_______________________________________________
Pce mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to