Fluentd pod fails to start on a RHOCP Node

Issue

• Thelogging-fluentd-xxxxx pod in openshift-logging namespace fails to start on a node.
• Restarting the fluentd pod does not resolve the issue.
• Fluentd pod shows the below error :

Warning  Failed  78s (x3 over 108s)  kubelet, node1.example.com  (combined from similar events): Error: container create failed: time="2021-02-26T10:07:31Z" level=error msg="container_linux.go:349: starting container process caused \"exec: \\\"scl\\\": executable file not found in $PATH\""
container_linux.go:349: starting container process caused "exec: \"scl\": executable file not found in $PATH"
  Normal  Pulled  66s (x13 over 3m40s)  kubelet, node1.example.com  Container image "registry.redhat.io/openshift4/ose-logging-fluentd@sha256:054daee22016641e266ad86db488fe63d235f513da8068c017bbc2ff79dda974" already present on machine

• Only one pod is in a pending state:

# oc get pods -l component=fluentd
fluentd-27zgj                                  1/1    Running    0         1h55m
fluentd-4l2h8                                  1/1    Running    0         1h55m
fluentd-5gvx7                                  0/1    Pending    0         32m
fluentd-5h7tq                                  1/1    Running    0         1h55m
fluentd-64qpk                                  1/1    Running    0         1h55m
fluentd-74l2r                                  1/1    Running    0         1h55m
fluentd-8h67t                                  1/1    Running    0         1h55m
fluentd-8zdzh                                  1/1    Running    0         1h55m
fluentd-97pmp                                  1/1    Running    0         1h55m
fluentd-9rfn2                                  1/1    Running    0         1h55m