Elasticsearch Rolling Restart


I need to perform a rolling restart of my elasticsearch cluster which contains 3 master nodes and 7 data nodes. I believe a rolling restart is the best approach for patching and also cert renewals. From this documentation:

it does not state which sequence to perform the rolling restart. Do I start with the master nodes or data nodes first? Anyone has done this before or know what is the safe way?


This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.