Earlier in the clustering environment, whenever a node went down, the same node only could recover the process flows of that node. To overcome this, Adeptia Connect now supports Cluster failover, which means that if a node goes down then the primary node would recover the process flows of that node and would do load balancing.
For clustering failover support environment, you need to do the following:
- Create a shared recovery folder that all the nodes in the cluster can access
- All the nodes need to have full permission to access this shared recovery folder
Cluster failover support can have two scenarios:
Secondary node goes down
When the secondary node goes down then after a retry time of three minutes, the primary node detects the failure and starts recovering the process flows, which were running on the secondary node. All the recovered process flows are also load balanced among the available nodes. If during the retry time, the secondary node becomes alive, then it initiates a self-recovery.
You can configure the retry time in the server-configure.properties file.
Primary node goes down
If the primary node goes down then after a retry time of three minutes, the secondary node becomes the new primary node and starts recovering the process flows, which were running on the old primary node. If during the retry time, the old primary node become alive, then it initiates a self-recovery.
You can configure the retry time in the server-configure.properties file.
You can also set up receiving mail notification from Adeptia Connect in case any node in a cluster goes down. To receive notification email, ensure that you have enabled the following properties:
- Configure time interval in property abpm.diagnostics.failure.notification.interval to trigger a check for threshold values.
- Provide email in property abpm.diagnostics.notification.receiver.email on which you want to receive the notification.
By default, these properties are disabled. Click here to know how to enable these properties.
