To start data nodes and compute nodes, you can:
(1) configure the parameter datanodeRestartInterval in controller.cfg.
For detailed instructions, please refer to the Cluster Mode — DolphinDB 2.0 documentation.
(2) connect to the controller you started and then use the startDataNode
function.
(3) execute a script like the following one:
./dolphindb -home /data/ddb/server/clusterDemo/data/P1-node1 -logFile /data/ddb/server/clusterDemo/log/P1-node1.log -jobLogFile /data/ddb/server/clusterDemo/log/P1-node1_job.log -keyPath /data/ddb/server/clusterDemo/data/agent1/keys -publicName 127.0.0.1 -console 0 -mode datanode -localSite 10.5.0.4:8975:P1-node1 -webLoginRequired 0 -perfMonitoring 1 -sites 10.5.0.7:8979:P3-node1,10.5.0.6:8977:P2-node1,10.5.0.4:8975:P1-node1,10.5.0.2:8972:controller2:controller,10.5.0.3:8973:controller3:controller,10.5.0.5:8971:controller1:controller -streamingHAMode raft -stdoutLog 0 -newValuePartitionPolicy add -config /noconfig -controllerSite 10.5.0.2:8972:controller2 -chunkCacheEngineMemSize 8 -maxConnections 512 -streamingRaftGroups 2:P1-node1:P2-node1:P3-node1 -maxMemSize 8 -dfsReplicationFactor 2 -remoteExecutors 2 -workerNum 8 -localExecutors 7 -dataSync 1 -maxPubConnections 64 -lanCluster 0 -subExecutors 2 -enableHTTPS 0 -subPort 8001