Hi All,
I am trying to run nexus(pro) in HA mode, for that I have created Nexus AMI using packer and ansible. I have done some customization in AMI in terms of changing location of sonatype-work directory, context path, changing default admin password etc.
I am launching 3 nodes cluster using ec2 autoscaling group, In nexus.log file I can see that nodes are able to communicate with each other
Members {size:3, ver:3} [
Member [<<node_1 IP>>]:5701 - 38ce35ca-7216-4285-9fa2-10f0b30e9c8b this
Member [<<node_2 IP>>]:5701 - 2b30a22f-42ce-457d-ac20-d78c03b535b0
Member [<<node_3 IP>>]:5701 - 391cbf14-832e-424a-ac22-c6a94826e8eb
]
However just after this I see below error in nexus.log file
ERROR [hz.nexus.event-1] *SYSTEM com.sonatype.nexus.hazelcast.internal.orient.SharedHazelcastPlugin - [D78F2A7F-9C0D18EC-B3E5B600-C3D0F3AB-EBE46CD1] Found a new node (Member [<<node_ip>>]:5701 - 391cbf14-832e-424a-ac22-c6a94826e8eb) with the same name as current: ‘D78F2A7F-9C0D18EC-B3E5B600-C3D0F3AB-EBE46CD1’. The node has been excluded. Change the name in its config/orientdb-dserver-config.xml file.
So, in this case all the cluster nodes are having same node identifier(D78F2A7F-9C0D18EC-B3E5B600-C3D0F3AB-EBE46CD1) and because of that only one node is acting as a master in the cluster(Ideally, It should be a active-active cluster). If I bring that node down another node becomes the master with same node ID.
I tried to locate “config/orientdb-dserver-config.xml” file in nexus installation as well as sonatype-work directory but it’s not there.
How can I remove this common Node ID from all the nodes and restart the nexus process on these nodes so that all the nodes would have their own unique node id?
Thanks,
Saurabh