Is this fix this issue too?
#2404 (comment)

when used

bind_ipv6_link_local=0

I am not sure. The whole identification code is too convoluted for a casual analysis. If you can apply and test, we'll get the answer back.

As I said, this whole part will benefit from a rewrite. It'll be smaller, easier to analyze, more efficient and less error prone - for example, right now it's performing almost the same exchange with netlink twice and it can't handle dropped netlink messages.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.