I have 22 nodes in my cluster and i have written a batch program to take snapshot and delete 30 days older indexes. The batch program is preparing "curl -XDELETE" and executing it from Master node.
I can see index directories got delete from master node ($ESHOME/node/0/indices/topbeat-2016.09.01) but these index directories are present in other nodes. I was expecting it should delete and free disk space from all nodes.
Am i missing any configuration here?
Thanks & Regards,