Check Topic Lag

Check

New

Check the consumer lag for a given topic (lag is calculated by the difference between topic offset and consumer offset).

Install now

Check Topic Lag

Check the consumer lag for a given topic (lag is calculated by the difference between topic offset and consumer offset).

Check

Install now

Check Topic Lag

Check

New

Check the consumer lag for a given topic (lag is calculated by the difference between topic offset and consumer offset).

Install now

Check Topic Lag

Check the consumer lag for a given topic (lag is calculated by the difference between topic offset and consumer offset).

Check

Install now

Go back to list

The action "Check Topic Lag" within the experiment editor.

Introduction

With this action, Steadybit can check the lag of a consumer for a topic for a given time.

Use Cases

You can check the topic lag of your consumer during your experiments.
You can fail the experiment if the lag is over a certain threshold.

Parameters

Parameter	Description	Default
Duration	The duration of the attack.	10s
Topic to track lag	Required. The topic to check the lag
Lag alert threshold	Required. The acceptable threshold of offset difference between the topic offset and the consumer offset

Statistics

-Stars

Simulate Kafka brokers downtime and see how consumers are handling topic lag

An experiment to block access to brokers for consumers while producing records in the topic. The consumers should get back to consume when the kafka brokers are available again and the accumulated lag must be deal with quickly.

Message Queue

Kafka

Recoverability

Containers

Kafka consumers

Kafka topics

Validate consumer's behavior when new leader is elected

Verify that your application handles a change of the leader properly.

Motivation

By testing your system's resilience to Kafka leader changes in a partition, you can identify potential weaknesses and take appropriate measures to improve its performance.

Structure

To conduct this experiment, we will ensure that all Kafka topics and producers are ready and that the consumer receives and processes messages correctly. We will then elect a new leader for one partition in Kafka and expect the system to work fine.