Description of problem:
Starting in RHOL (Red Hat OpenShift Logging) 5.5 the collectors are restarted all at the same time. This change of behavior was not notified in the Release notes

Version-Release number of selected component (if applicable):
RHOL 5.5

*URL: *
https://docs.openshift.com/container-platform/4.11/logging/cluster-logging-release-notes.html

Description
In RHOL 5.5 the behavior when restarting the collectors was changed. In the past, they were restarted one by one, and currently, they are restarted all at the same time.

This change was not notified in the Release Notes for RHOL 5.5 and it's a change that it was agreed with engineering from Logging to be included on them after discussing this subject since the restart immediately in a big cluster for example of 200 nodes, then, 200 hundred of fluentd pods could cause a peak of load in the API and in the etcd.