One question I have about ETCD back up and restore. If doing the back up from a . . .

Deirdre Rodgers:
One question I have about ETCD back up and restore. If doing the back up from a base node (not master node) then should I edit the endpoints flag to the address of the master/etcd server node?

ETCDCTL_API=3 etcdctl --endpoints=https://[masternode]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt \
     --cert=/etc/kubernetes/pki/etcd/server.crt --key=/etc/kubernetes/pki/etcd/server.key \
     snapshot save /tmp/snapshot-pre-boot.db

Munish Kumar Anand:
Hi @Deirdre Rodgers, yes,u need to use IP of master node in --endpoints if you are running these commands from base node.

Deirdre Rodgers:
@Munish Kumar Anand ok cool thanks! I knew I was doing something really wrong , is it the Internal IP from when you do a kubectl get nodes -o wide?

Munish Kumar Anand:
@Deirdre Rodgers That’s right. Even you can see the IP in etcd.yaml file. Look for --listen-client-urls in this file. If you run backup and restore from controlnode, then u can simply use 127.0.0.1 IP

Mohamed Ayman:
Also you need to ensure that the certificates path are correct and exists in the base node.