Hi all,

We may add on top of the reasons below the strong impact PCE-initiation
has from the implementation's perspective (which is somehow linked to
the first 2 items).
Up to I-D.ietf-pce-stateful-pce, it was all about configuration on PCCs,
just allowing PCEs to trigger actions. I-D.ietf-pce-pce-initiated-lsp is
a significant change: it enables PCEs to create on PCCs states which are
not part of operator-specified PCCs' configuration. As a result, there
has already been some commercial implementations at the stage described
by the 2nd item.

My 2 cents,

Julien


Oct. 04, 2017 - [email protected]:
> 1) I'm wondering why this spec is not part of I-D.ietf-pce-stateful-pce as it 
> is also not published yet...?
> 
> Jon> It has been published now.  The main reasons were
> - it took longer for PCE-initiated LSPs to be accepted into the PCE WG, and 
> the authors did not want to hold up work on the base draft
> - I-D.ietf-pce-stateful-pce was a self-contained set of function and it was 
> envisioned that there was a class of device that would not implement the LSP 
> initiation extensions
> - I-D.ietf-pce-stateful-pce was already fairly long and complex, and merging 
> them would have been an editing / reviewing headache.

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

Reply via email to