How is ES data distributed in each data nodes

I am using Elasticsearch 7.8.0 in k8s environment. I have 20 ES ingest nodes for serving traffic, 25 data nodes , 3 master nodes. How is the data stored getting distributed across all ES data nodes because I see on some of the data nodes data stored is ununiform.
Example- In some data nodes it has stored upto 175GB. But in some data nodes it has stored some MB.

ingest pod -
pod limit, request: 30Gb
cpu - 1
 -Xms15g -Xmx15g
data pod -
pod limit, request: 30Gb
 cpu - 1
 -Xms15g -Xmx15g
master pod -
pod limit, request - 2Gb
cpu - 1
 -Xms1g -Xmx1g
shards disk.indices disk.used disk.avail disk.percent node          
    45        8.4gb     8.5gb      1.1tb      1.1tb            0  elasticsearch-data-18
    45         67gb    67.1gb        1tb      1.1tb            5  elasticsearch-data-9
    45      687.8mb   784.5mb      1.1tb      1.1tb            0  elasticsearch-data-10
    45       91.4gb    91.8gb        1tb      1.1tb            7  elasticsearch-data-21
     5      115.4gb   161.7gb        1tb      1.1tb           13  elasticsearch-data-12
    45        1.1gb     1.2gb      1.1tb      1.1tb            0  elasticsearch-data-0
    45      117.7gb   175.7gb   1013.1gb      1.1tb           14  elasticsearch-data-13
    45      158.7gb     159gb        1tb      1.1tb           13  elasticsearch-data-17
    45        1.2gb     1.3gb      1.1tb      1.1tb            0  elasticsearch-data-20
    45          1gb     3.6gb      1.1tb      1.1tb            0  elasticsearch-data-11
    44      127.5gb   127.6gb        1tb      1.1tb           10  elasticsearch-data-16
    45       15.5gb    15.9gb      1.1tb      1.1tb            1  elasticsearch-data-15
    45        1.9gb       2gb      1.1tb      1.1tb            0  elasticsearch-data-1
    45        9.1gb     9.2gb      1.1tb      1.1tb            0  elasticsearch-data-19
    45        2.3gb     2.4gb      1.1tb      1.1tb            0  elasticsearch-data-3
    32       36.8gb    36.9gb      1.1tb      1.1tb            3  elasticsearch-data-24
     2       10.7gb      17gb      1.1tb      1.1tb            1  elasticsearch-data-14
    45       58.9gb    59.2gb      1.1tb      1.1tb            4  elasticsearch-data-4
    45       74.5gb   139.3gb        1tb      1.1tb           11  elasticsearch-data-6
    45         12gb    12.1gb      1.1tb      1.1tb            1  elasticsearch-data-5
    45      665.5mb   762.3mb      1.1tb      1.1tb            0  elasticsearch-data-8
    45       11.7gb    11.8gb      1.1tb      1.1tb            0  elasticsearch-data-7
    45        1.3gb     1.4gb      1.1tb      1.1tb            0  elasticsearch-data-23
    45          8gb       8gb      1.1tb      1.1tb            0  elasticsearch-data-2
    73                                                            UNASSIGNED

Elasticsearch balances by shard count, which you can see is pretty even for the most part.

You do have some anomalies though, such as those nodes with only a few shards. What does your elasticsearch.yml look like?

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