I have a problem with configuring Elasticsearch with 2 nodes. First one lets say called "node1" will be master and the second "node2" will be slave. I created 2 virtual machines using VirtualBox. I set in configure that one is master(true) and the second is slave(master=false). It cant discovery automatically (I got an error no known master node, scheduling a retry ). I tried this Elasticsearch Cluster - No known master node, scheduling a retry but it doesnt help. I also changed MAC addresses of both virtual machines to be the same and got the same problem. But when I run both nodes on one virtual machine it works perfectly. I think the problem can be in virtualbox (network settings or something). Can somebody help me?
0
votes
1 Answers
1
votes
Review your reach between those nodes. firewall is the main problem in these cases. try using ufw
on the slave
sudo ufw allow from <ipMaster>
sudo ufw allow out from any to <ipMaster>
on the master
sudo ufw allow from <ipSlave>
sudo ufw allow out from any to <ipSlave>
take a look on discovery settings under the elasticsearch.yml file as well
under
/etc/elasticsearch/elasticsearch.yml
use your favorite text editor
nano /etc/elasticsearch/elasticsearch.yml
find the following lines:
discovery.zen.ping.unicast.hosts
AND SET:
discovery.zen.ping.unicast.hosts: ["<ipMaster>","<ipSlave>"]
ensure also that both nodes share the same "cluster.name" but different "node.name", under the same file find the lines:
cluster.name
node.name
AND then Set:
on the master
cluster.name: MyCluster
node.name: Master01
on the slave
cluster.name: MyCluster
node.name: Slave01
