lua-resty-upstream-healthcheck icon indicating copy to clipboard operation
lua-resty-upstream-healthcheck copied to clipboard

Published 20 hours ago • verified publisher icon openresty

Why Is the Health Check Status Inherited by PeerID During Reload?

Open c1390528 opened this issue 2 years ago • 0 comments

I have a cluster that contains two servers. The health check of the first server is down and the second server is OK. When I delete the first server and retain the second server, after reload , the second server is immediately set to Down.This is because the peer status is updated based on the peer ID.See: https://github.com/openresty/lua-resty-upstream-healthcheck/blob/346f3a642b6717a26382e4032160aa766af0188c/lib/resty/upstream/healthcheck.lua#L378

To solve this problem, can we use the name of the peer to identify a peer instead of the ID?

c1390528 avatar Jan 29 '22 03:01 c1390528