[FixBug] Fix Exception on Convert.toInt of YarnGateway to track the new JobManager address
Purpose of the pull request
Fix dinky cannot track the JobManager address in yarn HA mode https://github.com/DataLinkDC/dinky/issues/4363
Brief change log
getSeconds()*1000 has been added to Convert.toInt in YarnGateway
Verify this pull request
-
Configure high availability mode and ZooKeeper quorum in Flink configuration file: high-availability.type: zookeeper high-availability.storageDir: hdfs:///flink/recovery high-availability.zookeeper.quorum: address1:2181[,...],addressX:2181
-
Dinky submits the application mode job to yarn
-
Get Jobmanager address from FLINKwebui
-
Stop JobManager(kill Jobmanager Process),After that , JobManager address will change kill Jobmanager Process:
JobManager address will change Automatically:
spotless apply
public static int getZookeeperSessionTimeout(Configuration configuration) { return Convert.toInt(configuration.get(HighAvailabilityOptions.ZOOKEEPER_SESSION_TIMEOUT)); }