High Availability Clustering (HA-C) is a feature designed to improve uptime by having a cluster of redundant NXRM "nodes" (instances) within a single data center. This feature allows you to maintain availability to your Nexus Repository Manager in the event one of the nodes becomes unavailable. A typical NXRM HA-C cluster is shown below:
Software and Hardware Requirements
Here’s a list of the most important things you will need:
- A test environment to evaluate your cluster configuration and processes
- Nexus Repository Manager Professional license and the latest version
- A load balancer, such as NGINX or Apache HTTP or AWS ELB
- 3 separate instances of NXRM installed on 3 different systems (e.g. 3 different EC2 instances) in a single datacenter
- Network connectivity between the 3 different systems so the NXRM can communicate with each other on several ports
- Separate local and shared storage (the difference will be described below)
- Each node needs to be configured according to our general requirements.
In addition to these items, this guide will clarify the key concepts of Node, Blob Store, and Hazelcast and outline the operational considerations specific to HA-C.
Known Issues and Limitations
- The initial functionality intends to support a three-node cluster in a single datacenter. Cross-datacenter (WAN) replication is not supported at this time. For AWS, all nodes must be in the same availability zone.
- Upgrading an already established cluster to a new version of NXRM requires shutting down all nodes. Support for rolling upgrades is not yet available.
- Apt and Go formats are currently disabled.
Topics in this section
- Configuring Nodes
- Configuring Blob Stores
- Configuring Hazelcast
- Designing your Cluster Backup/Restore Process
- Initial Setup - High Availability
- Operating your cluster
- Load Balancing