Erik Kringen created KAFKA-5656:
-----------------------------------
Summary: Support bulk attributes request on KafkaMbean where some
attributes do not exist
Key: KAFKA-5656
URL: https://issues.apache.org/jira/browse/KAFKA-5656
Project: Kafka
Issue Type: Bug
Components: clients
Reporter: Erik Kringen
Priority: Minor
According to Oracle documentation on [Implementing a Dynamic
MBean|http://docs.oracle.com/cd/E19698-01/816-7609/6mdjrf83d/index.html]
bq. The bulk getter and setter methods usually rely on the generic getter and
setter, respectively. This makes them independent of the management interface,
which can simplify certain modifications. In this case, their implementation
consists mostly of error checking on the list of attributes. However, all bulk
getters and setters must be implemented so that an error on any one attribute
does not interrupt or invalidate the bulk operation on the other attributes.
bq. If an attribute cannot be read, then its name-value pair is not included in
the list of results. If an attribute cannot be written, it will not be copied
to the returned list of successful set operations. As a result, if there are
any errors, the lists returned by bulk operators will not have the same length
as the array or list passed to them. In any case, the bulk operators do not
guarantee that their returned lists have the same ordering of attributes as the
input array or list.
The current implementation of
{code}org.apache.kafka.common.metrics.JmxReporter.KafkaMbean#getAttributes{code}
returns an empty list if any of the the requested attributes are not found.
This method should instead log the exception but allow all requested attributes
that are present to be returned, as prescribed via the DynamicMBean interface.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
