Hi Kowsalya,

On Tue, Nov 25, 2025 at 4:55 PM Kowsalya Dhevi <
[email protected]> wrote:

> Hi All,
>
> Kindly share your insights regarding the following query.
>
> RFC 5440, section 7.8, states that "If no path satisfying the constraints
> could be found by the PCE, the METRIC objects MAY also be present in the
> PCRep message with the NO-PATH object to indicate the constraint metric
> that could be satisfied."
>
> However, I have encountered a scenario where the PCE sent a PCRep message
> with an Explicit Route Object (ERO), even when it could not locate a path
> meeting the constraints. Specifically, the Path Computation Client (PCC)
> sent a PCReq message with a Metric Object for "Hop Counts" set to "1" with
> the B flag enabled. Nevertheless, the PCE responded with a PCRep message
> containing an ERO object but with a different Metric Object "TE Metric".
>
>
Dhruv: Just for my understanding - what is the content of the ERO?

https://datatracker.ietf.org/doc/html/rfc5440#section-7.5 is clear that
"When a PCE cannot find a path satisfying a set of constraints, it MUST
include a NO-PATH object in the PCRep message."

The presence of the ERO in PCRep indicates successful path computation.



> In such an instance, I wonder if it is appropriate for the PCC to send a
> PCErr message. If so, I would appreciate it if you could provide the
> appropriate Error-type and Error-value.
>
>
Dhruv: If there is no NO-PATH object present in the PCRep message, the PCC
does not have a way to infer that the path computation failed. This is why
RFC 5440 does not have an explicit error message specified.

For comparison, Stateful PCE implementations use an empty ERO to signal "no
path" in the PCUpd message (not PCRep). That is why I asked about ERO
content - to understand whether the PCE is sending a path that clearly
violates the hop-count constraint or something else is going on.

Thanks!
Dhruv
(As a WG participant)



> Regards,
> Kowsalya Dhevi
>
> --
>
_______________________________________________
Pce mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to