Moving workloads onto self-hosted Apache Kafka from other messaging systems, from Confluent Platform, from ZooKeeper-era metadata details and from VMs to Kubernetes, with the tools each path documents.
What can still be undone while a cluster is in migration mode: the revert steps by migration phase, the controller epoch issue to work around, why finalization is one-way and the checks to run before removing ZooKeeper.
Carrying security across a ZooKeeper to KRaft migration: switching from AclAuthorizer to StandardAuthorizer, where ACLs and SCRAM credentials live in the metadata log, removed ZooKeeper settings and testing access after each phase.
Replacing RabbitMQ queues with Kafka topics: mapping exchanges and queues to topics and partitions, share groups for queue-style consumption, bridging with the Camel RabbitMQ source connector during cutover and rewriting consumers.
Moving JMS workloads from ActiveMQ Artemis to Kafka: JMS queue and topic semantics versus partitions and consumer groups, the Camel JMS Artemis source connector for a bridged migration, and ordering and redelivery differences.
Offloading IBM MQ traffic to Kafka with IBM's open-source MQ source and sink connectors: converters for binary, text, JSON and XML messages, exactly-once source mode requirements and a staged cutover of applications.
Leaving Event Hubs for your own Kafka: using the Event Hubs Kafka endpoint with MirrorMaker 2, SASL PLAIN connection strings on port 9093, moving consumer offsets and switching producers and consumers to the new cluster.
Moving from Pub/Sub to Kafka with Google's Pub/Sub Group Kafka Connector: the source connector for draining subscriptions, keys and partition schemes, ordering keys, the 10 MB message limit and running both during cutover.
Moving a self-managed Confluent Platform estate to Apache Kafka: what is Apache 2.0 and what is under the Confluent Community or Enterprise licenses, replacing Schema Registry, REST Proxy and ksqlDB, and connector license checks.
Re-platforming a VM-based Kafka cluster onto Strimzi: rebuilding brokers and controllers as node pools, recreating topics and users as custom resources, moving data with MirrorMaker 2 and switching clients.
Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.