Dhruv, Hi!

Thanks for the quick clarification!

The mandate that I was missing is that the PCC MUST not include the
<actual-attributes-list> in the PCRpt if it does not include the RRO. I
thought the SR-RRO was optional (even if the LSP is UP/ACTIVE), but it
doesn't seem to be the case.

Regards,
-Pavan


On Mon, Feb 21, 2022 at 10:21 AM Dhruv Dhody <[email protected]> wrote:

> Hi Pavan,
>
> An important factor to distinguish between the actual and intended
> attribute list is the presence of RRO (i.e. <actual-path>) and the order
> of objects in the PCRpt message.
>
> If the RRO is present, any attributes encoded before it, are to be
> considered as part of <actual-attribute-list> and those after it, as part
> of <intended-attribute-list>.
>
> If the RRO is absent, all attributes are part
> of <intended-attribute-list>.
>
> Am I missing something?
>
> Thanks!
> Dhruv
>
> On Mon, Feb 21, 2022 at 8:11 PM Vishnu Pavan Beeram <[email protected]>
> wrote:
>
>> WG,
>>
>>
>> As per RFC8231, a PCRpt message can carry a list of <state-report>
>> entries, where:
>>
>>   <state-report> ::= [<SRP>]
>>                      <LSP>
>>                      <path>
>>   where:
>>       <path>::= <intended-path>
>>                 [<actual-attribute-list><actual-path>]
>>                 <intended-attribute-list>
>>
>>       <actual-attribute-list>::=[<BANDWIDTH>]
>>                                 [<metric-list>]
>>
>> When the report carries both "actual attributes" and "intended
>> attributes", there needs to be a way to unambiguously distinguish between
>> the attributes in each list. With BANDWIDTH, the distinction is provided by
>> the ObjectType (1 vs 2). The METRIC object, however, does not have that
>> distinction provided via ObjectType. When intended metric objects and actual
>> metric objects are contiguous (other objects between these are optional),
>> there doesn't seem to be a clear way to do the demarcation.
>>
>>
>> Section 6.1 of RFC8231 seems to hint at the use of the C flag for this.
>> **
>>
>>    When included as part of the actual-attribute-list,
>>    Object-Type 2 [RFC5440 <https://datatracker.ietf.org/doc/html/rfc5440>] 
>> SHOULD be used for the BANDWIDTH object, and
>>    the C flag SHOULD be set in the METRIC object [RFC5440 
>> <https://datatracker.ietf.org/doc/html/rfc5440>].
>>
>>
>> **
>>
>>
>> But this section does not preclude the use of the C flag in the METRIC
>> object listed in the "intended-attribute-list" block of the PCRpt message.
>> An implementation may use the C flag in the intended-metric carried in the
>> PCRpt as a means to request the PCE to send the "computed metric" in the
>> PCUpd message. If it is mandated that the C flag should not be set in
>> the METRIC object when listed as part of the "intended-attribute-list",
>> then the PCC loses the ability to request the PCE to send the "computed
>> metric".
>>
>>
>> Looking through packet traces from various interoperability reports, I do
>> see that there is no consistent interpretation of this flag. Any thoughts
>> on what should be the best practice for this?
>>
>>
>> Regards,
>>
>> -Pavan
>> _______________________________________________
>> Pce mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/pce
>>
>
_______________________________________________
Pce mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/pce

Reply via email to