Hi,
I am trying to add a new node to an already existing cluster but it seems the discovery won’t do its job. Even after restarting the master node the discovery process won’t start or at least not detect the new node.
- network & firewall settings are open and connection on ports 9200 and 9300 are working
- the elasticsearch.yml contains the new node, resp. the new node contains the master in the discovery settings
- the elasticsearch.service is up&running
Any tips how to enforce it or what I can do about it?
Best regards
Please share the followings and let me try to help you.
- telnet from new node to one of the existing cluster node
telnet ip:9300
- elasticsearch logs from new node and existing cluster elected master node
the problem can be because of the new node build up it's own cluster or it's joined another cluster before which can prevent to join to the existing cluster
Hi,
Thanks for your questions.
In fact, I did not use the telnet command but rather tried via “openssl s_client…” and here are the results:
new node -> master node: works for 9200 but not for 9300
master -> new node: works for both 9200 and 9300 just fine
Thus, I would expect this to be a FW issue. locally, on both nodes the same configurations are active and ports 9200/tcp and 9300/tcp are open. Therefore I would assume there is an issue somewhere in the network FW.
Edit: Just came back from discussing with the network team again. There was in fact something missing. Topic can be closed.