From e2bbf175a2184bd76f6c54ccf8456babeb1a46fc Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Tue, 9 Apr 2024 15:16:35 +0200 Subject: Adding upstream version 9.1. Signed-off-by: Daniel Baumann --- doc/user/nexthop_groups.rst | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) create mode 100644 doc/user/nexthop_groups.rst (limited to 'doc/user/nexthop_groups.rst') diff --git a/doc/user/nexthop_groups.rst b/doc/user/nexthop_groups.rst new file mode 100644 index 0000000..45f64ee --- /dev/null +++ b/doc/user/nexthop_groups.rst @@ -0,0 +1,29 @@ +.. _nexthop-groups: + +Nexthop Groups +============== + +Nexthop groups are a way to encapsulate ECMP information together. It's a +listing of ECMP nexthops used to forward packets. + +.. clicmd:: nexthop-group NAME + + Create a nexthop-group with an associated NAME. This will put you into a + sub-mode where you can specify individual nexthops. To exit this mode type + exit or end as per normal conventions for leaving a sub-mode. + +.. clicmd:: nexthop [A.B.C.D|X:X::X:XX] [interface [onlink]] [nexthop-vrf NAME] [label LABELS] + + Create a v4 or v6 nexthop. All normal rules for creating nexthops that you + are used to are allowed here. The syntax was intentionally kept the same as + creating nexthops as you would for static routes. + +.. clicmd:: resilient buckets (1-256) idle-timer (1-4294967295) unbalanced-timer (1-4294967295) + + Create a resilient Nexthop Group with the specified number of buckets, and + associated timers. Instead of using the normal kernel hashing methodology + this specifies that X buckets will be created for the nexthop group and + when a nexthop is lost the buckets forwarding that particular nexthop + will be automatically re-assigned. This cli command must be the first + command entered currently. Additionally this command only works with linux 5.19 + kernels or newer. -- cgit v1.2.3