Metric On Shard Reads and Writes

We have a huge cluster with 250 nodes.
Multiple nodes go high load at various times, so want to know which shards are actually having higher READs and WRITEs so that I can see any hot shards are there.

How can log count of READs and WRITEs per shard? or is this already available?

What version are you on?
What is the output from the _cluster/stats?pretty&human API?

Which version of Elasticsearch are you using? Do you have monitoring enabled? Is there anything in the logs around the time this occurs?

using Elasticsearch 6.8

the output of _cluster/stats returns the node info, segment info, docs info & the plugins used

cant share everything since its company internal. I can share specifics if want anything specific.

I want to setup a metric says me which shard has most reads and writes

anyway to add the below metrics

  1. no of requests per Index & shard
  2. latency per index & shard

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