Description of problem:
etcd container is always marked as unhealthy in RHOSP 17.1 deployments because "etcdctl endpoint health" returns status via stderr. As a result, grep command is unable to find "is healthy" in the output and marks service as unhealthy.

Reproducer:
bash-5.1$ export ETCDCTL_API=3
bash-5.1$ export ETCD_LISTEN_CLIENT_URLS="http://172.17.1.32:2379"
bash-5.1$ ETCDCTL_ENDPOINTS=$ETCD_LISTEN_CLIENT_URLS etcdctl endpoint health
http://172.17.1.32:2379 is healthy: successfully committed proposal: took = 2.298206ms
bash-5.1$ echo $?
0
bash-5.1$ ETCDCTL_ENDPOINTS=$ETCD_LISTEN_CLIENT_URLS etcdctl endpoint health | grep "is healthy"
http://172.17.1.32:2379 is healthy: successfully committed proposal: took = 4.146423ms
bash-5.1$ echo $?
1
bash-5.1$ ETCDCTL_ENDPOINTS=$ETCD_LISTEN_CLIENT_URLS etcdctl endpoint health 2>/dev/null | grep "is healthy"
bash-5.1$

Version-Release number of selected component (if applicable): RHOSP 17.1

How reproducible: see description