Hi Haomian,
Thank you again for the review and suggestions. We believe we have addressed
your comments.
BR, Dan.
-Original Message-
From: Pce On Behalf Of Zhenghaomian (Zhenghaomian,
Optical Technology Research Dept)
Sent: 01 April 2019 08:39
To: adr...@olddog.co.uk; pce@ietf.org
Subject: [Pce] 答复: Working Group last call on draft-ietf-pce-stateful-hpce-06
Hi, WG,
I have read this document and believe this work is very useful. Stateful H-PCE
will enable a more efficient way to do the computation with a group of PCEs.
The document is in a good shape as well.
I support to move forward on this document, some minor comments are provided to
be fixed after the LC:
- The page number in ToC is not consistent, maybe an update on word would be
needed;
- PCEP stateful extension in RFC8231, and PCEP initiation extension in RFC8281,
are usually considered as two separate works. Given the fact we have merged the
gmpls extension with two features, it is reasonable to have these two features
in the h-pce work as well. I noticed there is corresponding descriptions in
section 3.3, and I think it would be useful if one sentence can be summarized
in the abstract.
OLD:
A Stateful Path Computation Element (PCE) maintains information on
the current network state, including: computed Label Switched Path
(LSPs), reserved resources within the network, and pending path
computation requests. This information may then be considered when
computing new traffic engineered LSPs, and for associated
and dependent LSPs, received from Path Computation Clients (PCCs).
NEW:
A Stateful Path Computation Element (PCE) maintains information on
the current network state, including: computed Label Switched Path
(LSPs), reserved resources within the network, and pending path
computation requests. This information may then be considered when
computing new traffic engineered LSPs, and for associated
and dependent LSPs, received from Path Computation Clients (PCCs).
Initialize the result of path computation from PCE is also helpful for
the PCC to gracefully establish the computed LSP.
- As mentioned in IETF 104 PCE WG session (draft-ietf-pce-enhanced-errors),
error handling issues need to be mentioned for inter-pce works, and this work
exactly fits into the scope. It is suggested to add one small section about
this. How about the following?
7.7. Error Handling between PCEs
Error types specified in PCEP should be properly propagate between parent and
child PCEs. The propagation, notification and criticality level defined in
[I-D. ietf-pce-enhanced-errors] are recommended.
- The idnits report an unused reference 'pcep-yang', probably because of the
line in section 7.2 for citation is not properly broken in the middle. Editing
will be helpful to fix it.
We are looking forward to see the improvement after the WG LC, thank you.
Best wishes,
Haomian
-邮件原件-
发件人: Pce [mailto:pce-boun...@ietf.org] 代表 Adrian Farrel
发送时间: 2019年3月14日 6:01
收件人: pce@ietf.org
主题: [Pce] Working Group last call on draft-ietf-pce-stateful-hpce-06
Hi working group,
This email starts a working group last call for draft-ietf-pce-stateful-hpce-06.
I would like to hear messages of support or concern about this draft.
If you support its progression towards publication as an RFC, please let us
know that you have read the latest revision, and explain why you think the work
is important. Indications of implementation would also be welcome - although
this document is informational, I believe some people may have built stateful
hierarchical PCEs for experimentation or deployment.
If you are opposed to the progression or have concerns please articulate them.
As always, review comments and nits are most welcome.
Because of the effort that is going in to preparing for IETF-104 and because of
the time spent away, this last call will run for three weeks and end on April
4th.
Thanks,
Adrian
___
Pce mailing list
Pce@ietf.org
https://www.ietf.org/mailman/listinfo/pce
___
Pce mailing list
Pce@ietf.org
https://www.ietf.org/mailman/listinfo/pce
___
Pce mailing list
Pce@ietf.org
https://www.ietf.org/mailman/listinfo/pce
