Elasticsearch 跨集群复制 - 追随者索引“index_not_found_exception”

Question

I have 2 elastic clusters (Cluster1 and Cluster2) and I am trying to configure a follower index in Cluster2 from a leader index of Cluster1.

I have followed the next steps:

1. Add Cluster1 as remote cluster in Cluster2.

Configuration image

2. Configure the next users:

• In Cluster1 user "cross-cluster-user" with the role "remote-replication". cross-cluster-user configuration image

• In Cluster2 user "cross-cluster-user" with the role "remote-replication". cross-cluster-user configuration image

3. When I try to create a follower index of "newblogs" index, I have the next error:

Can't create follower index no such index [newblogs] index_not_found_exception: no such index [newblogs]

Error image

The newblogs index exists in Cluster1: Get index result

My elasticsearch version is 8.3.3.

Any help will be appreciated.

Best regards.

elasticsearch.yml (Cluster1)

cluster.name: elastic-lab
node.name: ${HOSTNAME}
network.host: _eth1_
cluster.initial_master_nodes: ["node1"]

#----------------------- BEGIN SECURITY AUTO CONFIGURATION -----------------------
#
# The following settings, TLS certificates, and keys have been automatically
# generated to configure Elasticsearch security features on 28-08-2022 15:46:47
#
# --------------------------------------------------------------------------------

# Enable security features
xpack.security.enabled: true

xpack.security.enrollment.enabled: true

# Enable encryption for HTTP API client connections, such as Kibana, Logstash, and Agents
xpack.security.http.ssl:
  enabled: true
  keystore.path: certs/http.p12

# Enable encryption and mutual authentication between cluster nodes
xpack.security.transport.ssl:
  enabled: true
  verification_mode: certificate
  keystore.path: certs/transport.p12
  truststore.path: certs/transport.p12
#----------------------- END SECURITY AUTO CONFIGURATION -------------------------

elasticsearch.yml (Cluster2)

cluster.name: elastic-lab2
node.name: ${HOSTNAME}
network.host: _eth1_
cluster.initial_master_nodes: ["node1"]

#----------------------- BEGIN SECURITY AUTO CONFIGURATION -----------------------
#
# The following settings, TLS certificates, and keys have been automatically
# generated to configure Elasticsearch security features on 28-08-2022 16:07:28
#
# --------------------------------------------------------------------------------

# Enable security features
xpack.security.enabled: true

xpack.security.enrollment.enabled: true

# Enable encryption for HTTP API client connections, such as Kibana, Logstash, and Agents
xpack.security.http.ssl:
  enabled: true
  keystore.path: certs/http.p12

# Enable encryption and mutual authentication between cluster nodes
xpack.security.transport.ssl:
  enabled: true
  verification_mode: certificate
  keystore.path: certs/transport.p12
  truststore.path: certs/transport.p12
#----------------------- END SECURITY AUTO CONFIGURATION -------------------------

Answer 1

It was a very silly problem...I have 2 Clusters (Cluster1 and Cluster2) and each consists in 1 elasticsearch node (hostname: node1) and 1 kibana virtual machine (hostname: node4). "Node1" has different IP addresses in each cluster, but when I configured node1 (of Cluster1) as a seed node, the resolution of the name "node1" was the IP address of the node1 from Cluster2. This was the reason the remote cluster appeared as connected, it was connected to its own node1!

I have configured the seed node by IP (instead of hostname) and it seems to work. I had to change the option "verification_mode" in elasticsearch.yml of all nodes to the value "none" (because I was having SSL issues and this is only a lab).

Note: I didn´t have to configure any user or role for the clusters to connect, even though the documentation says so.

Best regards.