Hi Experts,
I have an ELK setup of 7.7 version with 3 nodes of Elasticsearch, 2 kibana and 2 logstash on a windows server 64 bit.
A. Below is the existing cluster information:-
Configuration of Elasticsearch servers:
16 GB Ram, 500 GB disk space, 8 core
I have a daily data ingestion of 5 beats i.e., Packetbeat, Metricbeat, Filebeat, Winlogbeat and Heartbeat from different servers.
The daily average volume of data is around 17Gb with 7 days of retention period.
The Elasticsearch index policy is 3 primary with 1 replica, so each index and shards has a very less volume of data in it.
B. Below is the new cluster plan :-
- Rollover at 30 GB with 7 days of retention period.
- I will create 1 index for every beats.
- I will reduce the number of shards to 1 with 1 replica because the daily average data for 5 beats is 17 GB it means each shards will have data of around 3.2 GB with 7 days retention period.
- Upgrade the existing environment to 7.15 version
Is this approach correct? Or should I upgrade the existing cluster to 7.15 version and later do the changes?
Or should I first do the changes in existing cluster later upgrade it?
Please help me in this what plan should I have to follow.
Regards,
Tahseen