High Availability Features

Dynamic Value Generation for server.id

The application instance is internally assigned a unique value for server.id.
When an application instance shuts down gracefully, the server.id value assigned to it is returned to the pool.
If an instance crashes, the server.id value is not immediately returned to the pool. A monitoring job will detect and recover the value, returning it to the pool. This regulates the continuous growth of values for server.id.
The values generated for server.id is in the range of 1 to 99.
The server.id value allocated from the pool is not bound to a specific node. For example, in a cluster of five nodes, if node1 is assigned server.id 1 and node2 is assigned server.id 2, a full restart of all nodes may result in the values being reassigned to different nodes. Although a node might coincidentally receive the same value after a restart, this behavior is not guaranteed.
The allocation and de-allocation of the server.id values are tracked centrally in a database.
Possibility of race-condition is anticipated when all the nodes can start together, and this situation is taken care to avoid allocation of duplicate value to multiple nodes.

Master Jobs High Availability

This feature provides a self-resiliency to the master jobs in the application against the failure of a master node on which these master jobs are running.
One of the active application instances is designated as the master and starts the master jobs on its node.
When a master node shuts down in planned or controlled manner, it deregisters itself and allows another active application instance to take over as the new master. The new master node then starts the master jobs.
In a crash scenario, a health check job running on all nodes designates one of the application instances as the master. The new master then starts the master jobs.
In both scenarios, planned/controlled shutdown or crash, the master node switchover occurs only during the next run of the health check job. As a result, a slight delay in the switchover may be expected.
Each application instance is internally assigned a unique server.id value. When an application instance shuts down gracefully, the assigned server.id is returned to the pool.

Factory Shipped Configuration Data with default values:

Configurations related to these new features are released as INC for table PMZM_INSTANCE_PROPERTIES.
It is recommended to run this feature with the default values. If any issues occur while using the default values, contact your implementation partner or the engineering team, depending on your support access.

2 High Availability Features