Release 80: Kafka Streams, Topic Regex Search, Freshness Metrics

kPow v80 provides a brand new Kafka Streams UI with the GA release of the open-source kPow Streams Agent.

Also included in this release is multi-topic inspect by regex or consumer group, new ‘freshness’ metrics for topics and consumer groups, configurable columns for all tabular data, and a number of minor bug fixes and improvements.

Note: kPow now creates one more internal topic oprtr.compute.snapshots.v2-oprtr_snaphot_state_v2-changelog, if you have ACLs configured check your kPow permissions.

Kafka Streams UI

kPow provides a top-level UI for Kafka Streams applications.

Configure the open-source kPow Streams Agent and quickly see summary metrics on stream activity, reads, and lag. Use the kPow Workflows UI to dive into topology visualisation.

Topic Search by Regex or Group

kPow’s powerful Data Inspect functionality allows you to inspect multiple topics at once, and now you can specify topics by regex, or choose to inspect the same topics being consumed by a consumer group.

Freshness Metrics

See how long it has been since a topic was produced to or a consumer group read a message – right down to a topic partition or group assignment level.

Three new metrics are exposed via Prometheus for alerting purposes:

  • topic_production_inactive_mins
  • topic_consumption_inactive_mins
  • group_consumption_inactive_mins

Configurable Columns

As we introduce even more power to the kPow UI, some of the tabular data has become a little crowded – so now you can choose the columns to display for each table, and see a simple description of what each column represents.

Full Changelog

  • New Kafka Streams UI
  • GA open-source kPow Streams Agent
  • Multi-topic search by topic regex
  • Multi-topic search by group
  • Topic and consumer group freshness metrics
  • Exposed topic and consumer inactivity via Prometheus
  • Configurable columns for tabular data
  • Topic read/s metric
  • Under-replicated partitions count in summary data
  • Fixed ‘in-sync?’ and controller broker UI regressions
  • Open client websocket session count metrics
  • Reduced kPow internal topic disk usage

Get this release on Dockerhub or download as a JAR file.

docker pull operatr/kpow:80

Java 8 JAR

kPow uses the latest available version of Jetty10 which requires Java 11+

If you must run kPow with Java 8+ you can use this JAR instead, it uses Jetty9. You should update to Java 11+ and prefer the main JAR where possible.