Skip to content

Latest commit

 

History

History
43 lines (30 loc) · 1.37 KB

README.md

File metadata and controls

43 lines (30 loc) · 1.37 KB

Verifying MirrorMaker2

To verify that messages are stored in the target Apache Kafka cluster, we will deploy a set of applications to check that we have data coming from the source cluster:

oc apply -f check-apps/

These applications are deployed as:

❯ oc get deployment
NAME                                            READY   UP-TO-DATE   AVAILABLE   AGE
hello-world-consumer-check-greetings            1/1     1            1           61m
hello-world-consumer-check-greetings-reversed   1/1     1            1           61m

Checking the logs of the pods created we could confirm that we are consuming data in the target cluster with data produced in the source cluster.

Migration Consumers

Now we could stop de original consumer from source cluster and start a new one in the target. As the consumers offset are sync, the new instances only process the latest records not processed in the source cluster.

Execute in the source cluster:

oc scale --replicas=0 deployment/hello-world-streams
oc scale --replicas=0 deployment/hello-world-consumer

Deploy in the target cluster:

oc apply -f ./consumer-apps/

Now you could check that the new consumers in the target cluster are consuming the data starting from the latest offset processed in the source cluster.

Now, the last step is move your producer apps to this new Kafka cluster.