Skip to content

Conversation

@Shbinging
Copy link
Contributor

Ensure the number of installed ECMP next hops does not exceed zrouter.zav.multipath_num by stopping iteration once the limit is reached, matching OpenBSD behavior (see zebra_mpls_openbsd.c:250–251).

…in linux platform

  Ensure the number of installed ECMP next hops does not exceed zrouter.zav.multipath_num by stopping iteration once the limit is reached, matching OpenBSD behavior (see zebra_mpls_openbsd.c:250–251).

Signed-off-by: Bing Shui <[email protected]>
if (!nexthop)
continue;

if (nexthop_num >= zrouter.zav.multipath_num)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you have an example of this actually being wrong? Or is this just done from code inspection?

This is far too late in the install process. This is a Control plane enforcement issue not a dplane enforcement issue. This should be up in the zebra proper as part of one of the final steps/checks of a route that is about to be installed. This is because FRR supports a variety of data plane plugins/ architectures and each of them should not have to have this check.

I suspect that the zebra_mpls_openbsd.c check is a hold over from before the code became more stringent about checking this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants