Our cluster is configured as follows (This is a AWS managed Elasticservice)
- 8 data nodes
- 8 Shards
- 1 Replica shard
Here is an example of index:
green open service-logs-2017-12-10 ouyQ9Ew2SQQ 8 1 348663916 12687396 185.6gb 92.8gb
To optimize the search I wanted to merge the segments into one.
Here is the command I used to initiate force merge:
curl -XPOST <>/service-logs-2017-12-10/_forcemerge?max_num_segments=1
Now when I check the number of segments, I still see huge number of segments:
curl <>/_cat/segments | grep service-logs-2017-12-12 | wc -l
OUTPUT: 247
This includes primary and replica but this figure should be 2 after force merge.
Is there something I'm missing here?