do I need open firewall for ports 2379, 2380 to each etcd node itself?

[dulhaver]

I am building a 3 node etcd cluster that should handle postgres HA via patroni and haproxy.
Due to company policies I need to building everything manually. Getting this to work is quite a challenge but I am making baby steps towards success.

current road blocker ist getting etcd to recognise all 3 members

etcd@rd5430:/opt/custom/etcd> etcdctl member list
{"level":"warn","ts":"2025-05-26T08:11:52.638644+0200","logger":"etcd-client","caller":"v3@v3.5.17/retry_interceptor.go:63","msg":"retrying of unary invoker failed","target":"etcd-endpoints://0xc00021a5a0/127.0.0.1:2379","attempt":0,"error":"rpc error: code = DeadlineExceeded desc = latest balancer error: last connection error: connection error: desc = \"transport: Error while dialing: dial tcp 127.0.0.1:2379: connect: connection refused\""}

I had organized for port 2379, 2380 being accessible on the respective other 2 nodes in the cluster, but the above smells a bit like as if I may need to open a route to each node itself also?

Or is there something wrong with my environment

ETCD_NAME="etcd_2"
ETCD_LISTEN_PEER_URLS="http://0.0.0.0:2380"
ETCD_LISTEN_CLIENT_URLS="http://0.0.0.0:2379"
ETCD_INITIAL_ADVERTISE_PEER_URLS="http://10.254.29.93:2380"
ETCD_INITIAL_CLUSTER_STATE="new"
ETCD_ADVERTISE_CLIENT_URLS="http://10.254.29.93:2379"
ETCD_INITIAL_CLUSTER_TOKEN="Start1234"
ETCD_INITIAL_CLUSTER="etcd_0=10.254.29.91:2380,etcd_1=10.254.29.92:2380,etcd_2=10.254.29.93:2380"
ETCD_HOME='/opt/custom/etcd'
ETCD_BIN="$ETCD_HOME/bin"

... i.e. using 0.0.0.0 instead of the ip addresse (10.254.29.93 in my current example) ?

[vivekpatani]

@dulhaver I'm surprised that indeed does not work, since the resolution of your listen client URLs should resolve locally never bouncing anywhere but your own machine. Are you still stuck? I can maybe help look into this, if you need a hand.