Handling IPv6 neighbor entering STALE state prematurely
This occasionally occurred in some linux devices which contain lots of neighbor.
My solution is down below:
net.ipv6.neigh.default.gc_thresh1 should be 30 or 50 in default set,
we should extend it and
net.ipv6.neigh.default.gc_thresh3 as well.
net.ipv6.route.max_size set ipv6 route size to more then default size in case that route incomplete from kernel route table.
net.ipv6.neigh.default.gc_thresh1 = 512 net.ipv6.neigh.default.gc_thresh2 = 1024 net.ipv6.neigh.default.gc_thresh3 = 2048 net.ipv6.route.max_size = 2100000000