Description
There is a very noticeable slow-down that is more obvious on slower systems/clusters when more and more HDFS processors exist or when more and more HBase processors exist. If they are enabled or running then it can cause tremendous slowdowns in response time by a multiple of 2-10x. This in turn leads the UI to be unresponsive.
Stacktraces such as this one are commonly present:
"Clustering Tasks Thread-3" Id=172 BLOCKED on java.lang.Class@600aa5f8
at org.apache.nifi.hadoop.SecurityUtil.isSecurityEnabled(SecurityUtil.java:83)
at org.apache.nifi.hadoop.KerberosProperties.validatePrincipalAndKeytab(KerberosProperties.java:121)
at org.apache.nifi.processors.hadoop.AbstractHadoopProcessor.customValidate(AbstractHadoopProcessor.java:154)
at org.apache.nifi.components.AbstractConfigurableComponent.validate(AbstractConfigurableComponent.java:123)
at org.apache.nifi.controller.StandardProcessorNode.isValid(StandardProcessorNode.java:912)
at org.apache.nifi.controller.FlowController.getProcessorStatus(FlowController.java:2529)
at org.apache.nifi.controller.FlowController.getGroupStatus(FlowController.java:2146)
at org.apache.nifi.controller.FlowController.getGroupStatus(FlowController.java:2162)