Test the fault tolerant mechanism for ES cluster nodes

Elastic Stack Elasticsearch
1

Hello,

I have configured 3 node ES cluster.
All 3 nodes are master eligible. So if one goes down other should be eligible to elect itself as a master node and will form a 2 node cluster.
How i can check this with my cluster?
Can anyone guide me on this more?

Regards,
Priyanka

2

This is correct.just kill one the master node and you'll see that happening.

3

Hello @dadoonet,

And where i can see exactly this means one master is down and other node elect itself as master node and remaining 2 nodes are communicating with each other.
How i can check this is happening? like any UI is there?

will kibana help me in this?

Regards,
Priyanka

4 
GET /_cat/nodes?v
5

Hello @dadoonet,

I have tried mentioned command.
First i have killed one master node. and after that ran this command.
I can see only 2 nodes after that.

How to check in that 2 nodes which one is master node?
and how to confirm that those remaining 2 nodes are serving queries for us?

Regards,
Priyanka

6

The command I shared gives you that information. What is its output?

7

Hello @dadoonet,

That command showing remaining 2 nodes names and some values for heap.percent, ram.percent, cpu ,load_1m, load_5m, load_15m ,node.role.

Not able to identify which node is master node.

Regards,
Priyanka

8

The identity of the master node is not relevant. It may change at any time. It is sufficient to check that GET _cluster/health reports green (or yellow).

9

If you had share the full output, I could have tell.

Anyway, look at node.role.

10

Strangely i got an issue with this. If i kill my current master node the role does not get switch and my cluster become unavailable while having 3 master eligible node on my cluster

11

Hello @dadoonet,

Please find below output:
ip master name : * data node 1
heap.percent:16
ram.percent:27
cpu:5
load_1m: null
load_5m: null
load_15_m: null
node.master: dilm

ip master name : - data node 2
heap.percent:21
ram.percent:53
cpu:2
load_1m: null
load_5m: null
load_15_m: null
node.master: dilm

master is denoted by * and data node with -.
I can see the difference. but node.master is showing dilm. what is that?

Reagrds,
Priyanka

12

I believe you are talking about Master role does not switch automaticaly. Let's keep your question there.

13

Please don't post images of text as they are hard to read, may not display correctly for everyone, and are not searchable.

Instead, paste the text and format it with </> icon or pairs of triple backticks (```), and check the preview window to make sure it's properly formatted before posting it. This makes it more likely that your question will receive a useful answer.

It would be great if you could update your post to solve this.

Here is what I can see on my side:

ip         heap.percent ram.percent cpu load_1m load_5m load_15m node.role master name
172.23.0.2            8          94   2    0.16    0.07     0.06 dilm      *      c4c5acf93d1e

As you can see there is a master field with a * for the current master node.
That's what you're looking for I think.

14

Hello @dadoonet,

I do not want to display ip addresses as company policy. so that is why i was updating it in excel.
Now i have edited the post.
Yes i can find now which is the master node by symbol *.
In node.master, i can see dilm is mentioned.
what does that mean?

Regards,
Priyanka

15

It's documented here: https://www.elastic.co/guide/en/elasticsearch/reference/current/cat-nodes.html

16

Hello @dadoonet,

Thanks for help!!
I got it now.

Regards,
Priyanka

17

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