shardingsphere-elasticjob
shardingsphere-elasticjob copied to clipboard
Use getAllServersBriefInfo method .Why can I still get the old service instance information?
Since each instance of a task needs to be registered once, we will use the getAllServersBriefInfo method to obtain all the service instances in the cluster. However, all the old service instances will be obtained every time. As a result, we often cannot register the task with other instances in our cluster. How can I solve it?
Server nodes are persist nodes. It won't be removed after ZooKeeper client disconnected.
How can I ensure that all machines in the cluster are registered with the latest ip on zk? Because our service is inevitably need to restart, and each restart will be the history of all the registered ip information is obtained. Is there a way to solve this problem?