Why shard relocating is happening all the time

Elastic Stack Elasticsearch
1

Hi All,

Greeting !!

*** Need URGENT HELP ***

Why shard relocating happening all the time? Due to that heap percentage is getting increasing

"cluster_name" : "BPOConnectElasticSearch",
"status" : "green",
"timed_out" : false,
"number_of_nodes" : 7,
"number_of_data_nodes" : 6,
"active_primary_shards" : 17586,
"active_shards" : 35172,
"relocating_shards" : 2,
"initializing_shards" : 0,
"unassigned_shards" : 0,
"delayed_unassigned_shards" : 0,
"number_of_pending_tasks" : 4253,
"number_of_in_flight_fetch" : 0,
"task_max_waiting_in_queue_millis" : 21983473,
"active_shards_percent_as_number" : 100.0

heap.percent ram.percent load node.role master name
73 61 0.48 d - BPOConnectDataNode04
19 61 0.56 d m BPOConnectDataNode02
60 49 0.15 - m BPOConnectManagementNode
17 63 0.35 d m BPOConnectDataNode01
99 56 2.64 d * BPOConnectDataNode03
98 59 1.95 d m BPOConnectDataNode06
11 64 0.80 d m BPOConnectDataNode05

Note: All Node having 50% of heap of RAM. All node have more or equal to 30GB RAM

2

You have way too many shards. We recommend 600-700 per node, not nearly 6000.

You need to reduce that dramatically.

3

Yes i agree., but past 3 years we are using this setup. How to reduce this shard because all the logs are important and we dont want to increase any additional servers.

Kindly suggest what best we can do?

4

Take a look at - How do I increase or reduce the shard count of an existing index?

5

Did the suggestions I provided In this post not work?

Also please do not open multiple threads about the same issue.

2 Likes
6

Now its up and running fine:)

8

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