Table of Contents |
---|
Status
Current State: Under DiscussionAccepted
Discussion Thread: here here
JIRA: Jira server ASF JIRA columns key,summary,type,created,updated,due,assignee,reporter,priority,status,resolution serverId 5aa69414-a9e9-3523-82ec-879b028fb15b key KAFKA-5461
Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).
Motivation
There is currently no metric that captures the total number of topics in a Kafka cluster. It would be useful to maintain a total topic count metric on the Kafka Controller
the Kafka Controller.
Public Interface
There are no programming public interface changes. We are only adding a new metric.
Proposed Changes
The metric name would be:
- GlobalTopicCount
|
|
|
GlobalTopicCount is updated when the ControllerEventThread gets a TopicChange and TopicDeletion request and is purged on Controller failover.
- GlobalPartitionCount
|
...
GlobalPartitionCount is updated when the ControllerEventThread gets a TopicChange, TopicDeletion, and PartitionReassigment request and is purged
...
on Controller failover.
Compatibility
...
and Migration Plan
The changes are completely backward compatible.
...
Rejected Alternatives
TotalTopicCount
metric is collected from the MetadataCache
for each Broker
. It Aggregate GlobalTopicCount metric from the MetadataCache for each Broker. This approach was rejected because it is not the source of truth (as opposed to the controller metric). The metric is good to have and also helps identify delays in propagation of metadata updates This approach nevertheless is potentially useful in identifying delays in metadata propagation to the brokers.
kafka.server:type=MetadataCache,name= |
GlobalTopicCount
|
|