Es 5 node cannot join the master node

Elastic Stack Elasticsearch
21

hi,

Your configuration must be :

NODE 1:

cluster.name: prod
node.name: "nodeone"
node.master: true
node.data: false
transport.tcp.port: 9300
http.port: 9200
network.host: 192.168.10.100
node.attr.rack: r1
discovery.zen.ping.unicast.hosts: ["192.168.10.100", "192.168.10.150"]

NODE 2:

cluster.name: prod
node.name: "nodetwo"
node.master: false
node.data: true
transport.tcp.port: 9300
http.port: 9200
network.host: 192.168.10.150
discovery.zen.ping.unicast.hosts: ["192.168.10.100", "192.168.10.150"]
node.attr.rack: r1

bye,
Xavier

22

And the reason why is right there in the logs, which is covered in the docs. I'm asking where you got stuck after reading the logs and going to the docs?

23

I have not yet configured the two elastic nodes. These last do not work together I do not know why, then I have to consult the log file I find this output

24

[2017-04-19T12:30:40,867][INFO ][o.e.n.Node ] [nodeone] starting ...
[2017-04-19T12:30:41,530][INFO ][o.e.t.TransportService ] [nodeone] publish_address {192.168.10.100:9300}, bound_addresses {192.168.10.100:9300}
[2017-04-19T12:30:41,558][INFO ][o.e.b.BootstrapChecks ] [nodeone] bound or publishing to a non-loopback or non-link-local address, enforcing bootstrap checks
[2017-04-19T12:30:41,570][ERROR][o.e.b.Bootstrap ] [nodeone] node validation exception
bootstrap checks failed
system call filters failed to install; check the logs and fix your configuration or disable system call filters at your own risk
[2017-04-19T12:30:41,606][INFO ][o.e.n.Node ] [nodeone] stopping ...
[2017-04-19T12:30:41,664][INFO ][o.e.n.Node ] [nodeone] stopped
[2017-04-19T12:30:41,665][INFO ][o.e.n.Node ] [nodeone] closing ...
[2017-04-19T12:30:41,761][INFO ][o.e.n.Node ] [nodeone] closed

25

that's what I did

26

Are you shure of the configured IP ? 192.168.10.100 ?

1 Like
27

Why do you keep showing these logs and not answering my question? The logs tell you the problem. Our docs for the bootstrap checks cover this scenario. If you read the logs, read the docs, and apply what you learn there, you can solve this problem. Where are you getting stuck?

5 Likes
28

i m sorry but i have this error the slave can't join the master

curl -XGET 192.168.10.100:9200/_cat/nodes?pretty
192.168.10.100 9 96 0 0.00 0.08 0.27 mi * nodeone

[root@nodeone Desktop]# curl 192.168.10.100:9200/_cluster/health?pretty
{
"cluster_name" : "prod",
"status" : "green",
"timed_out" : false,
"number_of_nodes" : 1,
"number_of_data_nodes" : 0,
"active_primary_shards" : 0,
"active_shards" : 0,
"relocating_shards" : 0,
"initializing_shards" : 0,
"unassigned_shards" : 0,
"delayed_unassigned_shards" : 0,
"number_of_pending_tasks" : 0,
"number_of_in_flight_fetch" : 0,
"task_max_waiting_in_queue_millis" : 0,
"active_shards_percent_as_number" : 100.0
}

29

Okay, so you started your node. Yay!

What error are you referring to? At this point you need to set up the second node as I indicated above. Please let me know specifically where you're stuck.

1 Like
30

Hi, my problem isn't related especially to one of the nodes and there is no errors , they are working perfectly the two of them. But the issue is that I am not able to interconnect them one to the other.

31

Can you show me what you have tried?

1 Like
32

Yes thank you very much I managed to deploy the cluster elasticsearch. I have another question how can I ensure the HA in this cluster?

33

Would you please define what you mean by HA?

34

If I have a problem with the master node.the system should not stop
I read in the documentation that if the master moeud is down the other node is elected as master

35

Okay, this is a very tricky problem. You see, with two master-eligible nodes you can not have HA on the master. The reason is this: if you have two nodes and you set them to both be master eligible, then you must set discovery.zen.minimum_master_nodes to 2. If you do not do this, you are at risk of split brain where both master-eligible nodes can think they are the master. However, if discovery.zen.minimum_master_nodes is 2, when you lose one of the master-eligible nodes the other master node will not be able to be elected as the master because it requires 2 master-eligible nodes (itself, and at least one other) to vote for it during a master election. Does that help?

36

Then I need to add another master node to have HA at masters level node.je want to put a solution that provides a high protection against infrastructure failures and intermittent network problems

37

That's right, to get master HA you have to have three master-eligible nodes, and set discovery.zen.minimum_master_nodes to 2.

1 Like
38

Thank you ,But when I do not specify a.master and node.data. There will be an election of master node and data node?
Can I fix an address for the entire cluster?

39

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