When using a cassandra cluster, DB.xml config file will have multiple IPs ( for each node). How can we know which is the current cassandra node used by the DMA for troubleshooting purpose.
And, what are the rules to use/switch to another node ? roundrobin ? random ?
Marieke Goethals [SLC] [DevOps Catalyst] Selected answer as best 31st May 2024
Hi Abel,
The IPs in db.xml are used to setup contact with Cassandra, once connection is established a connectionpool is created by the Cassandra driver towards all nodes within the connected datacenter (even nodes not defined in the db.xml).
This means that there is no singular connection towards Cassandra. DataMiner will send queries to the different nodes in a RoundRobin fashion.
Marieke Goethals [SLC] [DevOps Catalyst] Selected answer as best 31st May 2024