wrote:

Hi all,

We can forward logs to the external EFK using fluentd forward protocol, syslog protocol and log forwarding API.
 If we forward logs the the external EFK then we do NOT use internal elasticsearch log store.
Then is it possible to scale to 0 for elasticsearch, kibana and curator pods?
Or is it possible to configure only logging collector(fluentd)? If yes, how to configure it?

 
You don't need to scale any pods to zero for the internal ES, kibana, curator, if the only thing you wish to do is collection only. The ClusterLogging CR is quite open to your choices, e.g. collector only:
 
apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
metadata:
  name: "instance"
  namespace: "openshift-logging"
spec:
  managementState: "Managed"
  collection:
    logs:
      type: "fluentd"
      fluentd: {}
 
Basically, this is the bare minimum you can get from Cluster-Logging if the only concern is collection. Of course you need a LogForwarding CR for forwarding the logs.
You can be picky on which components you want to deploy, e.g. logStore, visualization, curation.