mimaison commented on code in PR #17076: URL: https://github.com/apache/kafka/pull/17076#discussion_r1753841794
########## docs/ops.html: ########## @@ -3776,25 +3776,77 @@ <h5 class="anchor-heading"><a id="kraft_voter" class="anchor-link"></a><a href=" <p>A Kafka admin will typically select 3 or 5 servers for this role, depending on factors like cost and the number of concurrent failures your system should withstand without availability impact. A majority of the controllers must be alive in order to maintain availability. With 3 controllers, the cluster can tolerate 1 controller failure; with 5 controllers, the cluster can tolerate 2 controller failures.</p> - <p>All of the servers in a Kafka cluster discover the quorum voters using the <code>controller.quorum.voters</code> property. This identifies the quorum controller servers that should be used. All the controllers must be enumerated. Each controller is identified with their <code>id</code>, <code>host</code> and <code>port</code> information. For example:</p> + <p>All of the servers in a Kafka cluster discover the active controller using the <code>controller.quorum.bootstrap.servers</code> property. All the controllers should be enumerated in this property. Each controller is identified with their <code>host</code> and <code>port</code> information. For example:</p> - <pre><code class="language-bash">controller.quorum.voters=id1@host1:port1,id2@host2:port2,id3@host3:port3</code></pre> + <pre><code class="language-bash">controller.quorum.bootstrap.servers=host1:port1,host2:port2,host3:port3</code></pre> <p>If a Kafka cluster has 3 controllers named controller1, controller2 and controller3, then controller1 may have the following configuration:</p> <pre><code class="language-bash">process.roles=controller node.id=1 listeners=CONTROLLER://controller1.example.com:9093 [email protected]:9093,[email protected]:9093,[email protected]:9093</code></pre> +controller.quorum.bootstrap.servers=controller1.example.com:9093,controller2.example.com:9093,controller3.example.com:9093 +controller.listener.names=CONTROLLER</code></pre> - <p>Every broker and controller must set the <code>controller.quorum.voters</code> property. The node ID supplied in the <code>controller.quorum.voters</code> property must match the corresponding id on the controller servers. For example, on controller1, node.id must be set to 1, and so forth. Each node ID must be unique across all the servers in a particular cluster. No two servers can have the same node ID regardless of their <code>process.roles</code> values. + <p>Every broker and controller must set the <code>controller.quorum.bootstrap.servers</code> property. - <h4 class="anchor-heading"><a id="kraft_storage" class="anchor-link"></a><a href="#kraft_storage">Storage Tool</a></h4> + <h4 class="anchor-heading"><a id="kraft_storage" class="anchor-link"></a><a href="#kraft_storage">Provisioning Nodes</a></h4> <p></p> The <code>kafka-storage.sh random-uuid</code> command can be used to generate a cluster ID for your new cluster. This cluster ID must be used when formatting each server in the cluster with the <code>kafka-storage.sh format</code> command. <p>This is different from how Kafka has operated in the past. Previously, Kafka would format blank storage directories automatically, and also generate a new cluster ID automatically. One reason for the change is that auto-formatting can sometimes obscure an error condition. This is particularly important for the metadata log maintained by the controller and broker servers. If a majority of the controllers were able to start with an empty log directory, a leader might be able to be elected with missing committed data.</p> + <h5 class="anchor-heading"><a id="kraft_storage_standalone" class="anchor-link"></a><a href="#kraft_storage_standalone">Bootstrap a Standalone Controller</a></h5> + The recommended method for creating a new KRaft controller cluster is to bootstrap it with one voter and dynamically <a href="#kraft_reconfig_add">add the rest of the controllers</a>. Bootstrapping the first controller can be done with the following CLI command: + + <pre><code class="language-bash">kafka-storage format --cluster-id <cluster-id> --standalone --config controller.properties</code></pre> + + This command will 1) create a meta.properties file in metadata.log.dir with a randomly generated directory.id, 2) create a snapshot at 00000000000000000000-0000000000.checkpoint with the necessary control records (KRaftVersionRecord and VotersRecord) to make this Kafka node the only voter for the quorum. + + <h5 class="anchor-heading"><a id="kraft_storage_voters" class="anchor-link"></a><a href="#kraft_storage_voters">Bootstrap with Multiple Controllers</a></h5> + The KRaft cluster metadata partition can also be bootstrapped with more than one voter. This can be done by using the --initial-controllers flag: + + <pre><code class="language-bash">cluster-id=$(kafka-storage random-uuid) +controller-0-uuid=$(kafka-storage random-uuid) +controller-1-uuid=$(kafka-storage random-uuid) +controller-2-uuid=$(kafka-storage random-uuid) + +# In each controller execute +kafka-storage format --cluster-id ${cluster-id} \ + --initial-controllers "0@controller-0:1234:${controller-0-uuid},1@controller-1:1234:${controller-1-uuid},2@controller-2:1234:${controller-2-uuid}" \ + --config controller.properties</code></pre> + +This command is similar to the standalone version but the snapshot at 00000000000000000000-0000000000.checkpoint will instead contain a VotersRecord that includes information for all of the controllers specified in --initial-controllers. It is important that the value of this flag is the same in all of the controllers with the same cluster id. + +In the replica description 0@controller-0:1234:3Db5QLSqSZieL3rJBUUegA, 0 is the replica id, 3Db5QLSqSZieL3rJBUUegA is the replica directory id, controller-0 is the replica's host and 1234 is the replica's port. + + <h5 class="anchor-heading"><a id="kraft_storage_observers" class="anchor-link"></a><a href="#kraft_storage_observers">Formatting Brokers and New Controllers</a></h5> + When provisioning new broker and controller nodes that we want to add to an existing Kafka cluster, use the <code>kafka-storage.sh format</code> command without the --standalone or --initial-controllers flags. + + <pre><code class="language-bash">kafka-storage format --cluster-id <cluster-id> --config server.properties</code></pre> + + <h4 class="anchor-heading"><a id="kraft_reconfig" class="anchor-link"></a><a href="#kraft_reconfig">Controller membership changes</a></h4> + + <h5 class="anchor-heading"><a id="kraft_reconfig_add" class="anchor-link"></a><a href="#kraft_reconfig_add">Add New Controller</a></h5> + If the KRaft Controller cluster already exist, the cluster can be expanded by first provisioning a new controller using the <a href="#kraft_storage_observers">kafka-storage tool</a> and starting the controller. + + After starting the controller, the replication to the new controller can be monitored using the <code>kafka-metadata-quorum describe --replication</code> command. Once the new controller has caught up to the active controller, it can be added to the cluster using the <code>kafka-metadata-quorum add-controller</code> command. + + When using broker endpoints use the --bootstrap-server flag: + <pre><code class="language-bash">kafka-metadata-quorum --command-config controller.properties --bootstrap-server localhost:9092 add-controller</code></pre> + + When using controller endpoints use the --bootstrap-controller flag: + <pre><code class="language-bash">kafka-metadata-quorum --command-config controller.properties --bootstrap-controller localhost:9092 add-controller</code></pre> + + <h5 class="anchor-heading"><a id="kraft_reconfig_remove" class="anchor-link"></a><a href="#kraft_reconfig_remove">Remove Controller</a></h5> + If the KRaft Controller cluster already exist, the cluster can be shrunk using the <code>kafka-metadata-quorum remove-controller</code> command. Until KIP-996: Pre-vote has been implemented and released, it is recommended to shutdown the controller that will be removed before running the remove-controller command. Review Comment: `exist` -> `exists` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
