Elastic agents kubernetes manifest keeps a set of persistent volumes for the agent. It might make sense to try deleting that volume or connecting to the container and deleting the state directory and deploying again.
In particular, if the agent on that node had been deployed against a different fleet or a different Elasticsearch cluster, changing the enrollment token or the fleet server in the manifest won't have an effect because the fleet server and ES token are already cached on disk
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.