Skip to main content

Getting Started

This guide takes you from the first start to the records of your first topic. It takes about ten minutes and applies to Kadeck Platform and Kadeck Desktop; where the two differ, the step says so.

The first steps in Kadeck: sign in, connect a cluster, find a topic, look at its records, narrow them down, keep and share the result

Kadeck Platform or Kadeck Desktop​

Kadeck PlatformKadeck Desktop
Made forTeams and organizationsOne person
RunsOn a server, used in the browserAs an application on your computer
ConnectionsSet up once by an administrator, shared by rightsYour own, on your computer
Users and rightsUsers, groups, roles, data protection, audit logNone: it is your computer
To try things outAny cluster you connectAlso a local Apache Kafka cluster, started from Kadeck

Everything from Connect a cluster on works the same way in both.

Sign in​

Kadeck Platform: open the address of your Kadeck Platform in the browser and sign in with the account your administrator gave you. If you are the one installing it, start with the deployment guide.

Kadeck Desktop: start the application.

On the first start of a free edition, Kadeck asks for two things:

  1. License agreement: scroll to the end of the agreement and click Confirm Agreement.
  2. About you: enter your name, email, company and role, then click Start now.

You land on the Connections page.

Connect a cluster​

A connection is one Apache Kafka cluster, together with the services that belong to it: Schema Registry, Kafka Connect and ksqlDB. Kadeck does not copy your data. It reads and writes where the data lives.

Kadeck connects to Apache Kafka clusters of any provider and reads the data in place

On Kadeck Platform, connections are created by administrators. If a connection is already listed on the Connections page, continue with Find your topic.

  1. On the Connections page, click Add Connection.

  2. Pick where your cluster lives. Kadeck asks only for what that provider hands out and fills in the rest:

    ProviderWhat you enter
    Confluent CloudBootstrap servers, API key and secret; optionally the Schema Registry URL and its API key and secret
    Amazon MSKBootstrap servers; IAM access control, or the username and password for SASL/SCRAM
    AivenService URI, user and password; optionally the CA certificate (ca.pem)
    RedpandaBootstrap servers, user and password
    Azure Event HubsNamespace and connection string
    IBM Event StreamsBootstrap servers and API key
    Apache KafkaNothing yet: you set up the connection by hand in the next step
    Client propertiesA configuration as you would put it in a .properties file
  3. Click Continue. The connection opens with everything filled in. Give it a Name you will recognize, for example prod-eu, and pick a Color.

  4. Click Test connection.

  5. Click Create.

The connection appears as a card on the Connections page. If it is not connected yet, click Connect on its card.

Good to know:

  • The tabs Schema Registry, Kafka Connect, ksqlDB and JMX of the connection are optional. You can add them later via the menu of the card.
  • To change your choice of provider, click Change at the top of the dialog. What you entered stays.
  • Amazon Kinesis is available as well: click Amazon Kinesis instead in step 2.
  • Kadeck Desktop only: if you have no cluster at hand, start the Local Apache Kafka Cluster from its card on the Connections page.

Find your topic​

There are two ways to a topic:

  • The Catalog lists the topics of all your connections with their owner, labels and domain. Use it when you know what you are looking for, but not where it is. Click a topic to see its details, and open its records from there.
  • The Explorer lists the topics of the connection you are working on, next to your favorites, views, topic sets and pipes. Use it when you want to look at records.

To create a topic, click Create topic in the Explorer or the Catalog. Enter a name, set the partitions and replicas, and click Create topic. Cleanup policy, retention and common settings such as compression are one click away in the same dialog.

Look at records​

Open the Explorer and click a topic. Kadeck reads its newest records and shows them as a list.

A map of the Data Explorer: the topic list, the tools of the view, the settings and filters, the record list and the details of a record, and how you move between them
  1. Topic list: your favorites, topic sets, pipes, views and all topics of the connection.
  2. Tools of the view: Description, Lineage (who produces to the topic and who consumes from it), Time diagram (when the records shown were written) and the JavaScript Processor.
  3. Settings and filters: where reading starts (a point in time or an offset), which partitions are read, how many records are shown, and the filters.
  4. Record list: one line per record. Switch between Table and Json, and click a column header to sort.
  5. Record details: click a record to see its Meta data and headers, its Key and its Value.

Things to try first:

  • Go back in time: set the start to a time window such as the last hour, or to an offset, and read again.
  • Watch the topic live: click Consume live data. New records appear as they are written.
  • Check the codec: Apache Kafka stores keys and values as bytes. Kadeck detects the format (JSON, Avro, Protobuf and others) on its own; if a value looks wrong, click Change codecs.
  • Make a column: open a record, click an attribute in its Value and add it as a column to the list.

Note: Kadeck Free limits how many records a view shows.

Narrow it down​

There are three ways to get from all records to the ones you need, from quick to powerful:

  • Search: type into Search displayed results to search the records that are shown.

  • Attribute filters: open a record, click an attribute and filter by its value. The filter is applied while reading, so it also finds records that are not shown yet.

  • JavaScript Processor: a small script that runs for every record. Return false to drop a record, or change the record and return it:

    // only payments above 1000, with the amount rounded
    if (rec.value.amount <= 1000) return false;
    rec.value.amount = Math.round(rec.value.amount);
    return rec;

    See JavaScript Processor for everything a script can do.

Keep and share​

What you have set up (start position, filters, codecs, columns, processor) is a view. A view is worth keeping when you will need it again or someone else should see the same thing.

  • Save the view: open the menu of the view and click Save view. Saved views appear beneath their topic in the Explorer. On Kadeck Platform a view is private or shared with everyone who may read the topic.
  • Look at several topics together: add a second topic or view to form a Topic Set, and arrange the set on a canvas with the Flow Visualizer.
  • Write the result to another topic: click Pipe to turn the view into a Kadeck Pipe. The pipe writes what the view shows to a target topic, once or around the clock.
  • Create or correct records: click Create new records, or select records in the list and produce them to a topic again.
  • Take records with you: select records and click Export records.

Where to go next​

AreaWhat it is for
ConsumersSee the lag of every consumer group, find slow or orphaned consumers, and reset offsets
SchemasBrowse and edit the subjects of your Schema Registry
Kafka ConnectCheck, pause and restart connectors
ACLs and QuotasManage the access rules and quotas of the cluster
MonitoringFollow the health of clusters, brokers and topics
AlertsGet notified when a topic or the infrastructure needs attention
KSQL LabQuery your streams with SQL
AuditSee who did what, on Kadeck Platform

Need help?​

Click the speech bubble in the header of Kadeck to send us feedback, or write to support@xeotek.com.