Hemant,
On 30/01/2024 17:38, Hemant Sharma wrote:
Hi Ahmed,
Thanks for clearing that up.
Under section, 11. Advantages of using the expected post-convergence
path during FRR, at the beginning of page 17, it says
As an example, if the operator has implemented a protection
against a node failure, the expected post-convergence path used
during FRR will be the one considering that the node has failed.
However, even if a single link is failing or a set of links is
failing (instead of the full node), the node-protecting post-
convergence path will be used.
Now, Cisco IOS-XR follows this rule, with no issues. But the Nokia
SROS does things a bit differently. It calculates the post-convergence
path first, then trims down the SRLG and nodes. The problem is,
sometimes it ends up with no protection path at all.
They're claiming that using Cisco's approach to TI-LFA can cause
issues with seamless BFD during an incident. Unfortunately, I don't
have the details on what kind of trouble they're talking about.
Would it be accurate to assert that their implementation is non-compliant?
As far as the implementation finds the backup path if it exists, all is
good, no mater how it is done. If it fails to find an existing backup
path, it's a bug in the implementation.
thanks,
Peter
Regards,
Hemant
On Tue, Jan 30, 2024 at 3:43 AM Ahmed Bashandy
<[email protected]> wrote:
Thanks for the comment
Regarding the paragraph from section 6 that you are referring to,
this paragraph was part of an example and not a recommendation.
RFC5286 and RFC7490 are referred to in
draft-ietf-rtgwg-segment-routing-ti-lfa. So IMO it would be
redundant to add any recommendation in
draft-ietf-rtgwg-segment-routing-ti-lfa that already exist in the
RFCs.
Thanks
Ahmed
On 12/30/23 2:46 PM, Hemant Sharma wrote:
Hello,
The RFC 7490, 5.2.2. Selecting Repair Paths, provides a
recommendation on how to choose a PQ node when multiple options
are available.
such as,
"For consistency in behavior, it is RECOMMENDED that the member
of the set of routers {PQ} with the lowest cost S->P be the
default choice for P."
or
"As described in [RFC5286], always selecting a PQ node that is
downstream to the destination with respect to the repairing node
prevents the formation of loops when the failure is worse than
expected. The use of downstream nodes reduces the repair
coverage, and operators are advised to determine whether adequate
coverage is achieved before enabling this selection feature."
whereas, in the current draft, 6. TI-LFA Repair path, there is
only a subtle reference to the selection of a specific P node, as
shown below
* P(S, N1) intersection with PCPath is [N2, R1], *R1 being
the deeper
downstream node in PCP, it can be assumed to be used as P node*
(this is an example and an implementation could use a different
strategy to choose the P node).
However, in Section 6.2 on FRR path using a PQ node, there is no
explicit mention of the scenario when there are multiple PQ nodes
along the post-convergence path.
I believe it is worthwhile to include similar recommendations
here for the sake of completeness, either in the same section or
creating another one, copying this part from above.
"For consistency in behavior, it is RECOMMENDED that the member
of the set of routers {PQ} with the lowest cost S->P be the
default choice for P."
or
"As described in [RFC5286], always selecting a PQ node that is
downstream to the destination with respect to the repairing node
prevents the formation of loops when the failure is worse than
expected. The use of downstream nodes reduces the repair
coverage, and operators are advised to determine whether adequate
coverage is achieved before enabling this selection feature."
Please confirm if I have correctly understood this.
Thanks!
Regards,
Hemant Sharma
_______________________________________________
rtgwg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/rtgwg
_______________________________________________
rtgwg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/rtgwg
