High Availability

There are two options to ensure that services are uninterrupted: cold standby and clustering.

Cold standby

The cold standby option is simple to implement, reduces downtime near to zero, but does not improve server performance. In this setup, there is a single active server (primary server) connected to a database and a remote file system. There is also a secondary server that is configured to connect to the same database and file system, but this server is not running. If the active node fails, the secondary server is started and network traffic is routed to it. This event is called failover. IBM® UrbanCode Deploy has no automatic process to failover, but it can be automated.

A diagram of a cold standby system, showing the normal operation and the failover state

You can convert a cold standby system to a clustered system by putting shared files on network storage and connecting the servers with network relays. See Configuring clustered servers for high availability.

Clustered

The high availability (HA) feature increases scalability and availability by distributing processing across a cluster of servers. Each server is an independent node that cooperates in common processing. The goal is to be as fault-tolerant as possible while requiring little or no manual intervention.

The IBM UrbanCode Deploy servers create a JMS mesh (via ActiveMQ); all servers know about each other. All services are active on each server.

A diagram of a clustered system, showing two servers that share a database

For installation instructions, see Configuring clustered servers for high availability.


Feedback