Query time increases on a particular index out of many indexes in es. This happens for 6-7 hours on some random days. The start time always is from 10:30 PM IST(occurs at same time on the day it happens), but there is no pattern on days it occurs. Out of 8-10 indices, only this one suffers. This index never gets updated but other indices in the same cluster are daily replaced by a new index.
Hello @Mohit_Ojha
I would suggest to provide some additional information such as:
- Elasticsearch version
- Type or example of query
- Index size & shard numbers (
GET <index name>,GET _cat/indices?v,GET _cat/shards)
It might be related to the amount of concurrent searches: if the shard(s) are hosted on a single machine, only this machine will be able to serve the search requests, creating a bottleneck.
You might increase the number of replicas to scale the reads.
Other suggestions might be found in our documentation
If you're on Elasticsearch 7.3.0 and you experience slowdown in Kibana when date histograms, you might be hitting a regression:
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.