Shard allocation not working if on different subnet?

hi,
I was doing some testing where for e.g
node1 IP : 192.168.168.X
node2 IP : 192.168.168.X
node3 IP : 10.91.10.X

They are able to discover one another, but shards are only allocated among node1 and node2, while node3 remains empty.
ES does not support this?

discovery.zen.ping.unicast.hosts ip is add?

Hi, yes, the problem is not about the discovery of the nodes.
i can see node3 is discovered and it joins the cluster, but the shards are not being pushed to this node3

Hi,
its working now.. i am not sure if its the ES version problem, i just align the ES version and it seems fine now.

which version as now?

Hi, we are using an old version 2.4.X.

if the new cluster can use 5.x or 6.x is better..

Hi thanks for the advise, does 5.X and 6.X offer more stuff in terms of replication, multiple data centers etc.?

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