Configuration¶
The Confluent Platform Docker images support passing configuration variables dynamically using environment variables. More specifically, we use the Docker -e
or --env
flags for setting various settings in the respective images when starting up the images.
Some configuration variables are required when starting up the Docker images. We have outlined those variables below for each component along with an example of how to pass them. For a full list of all available configuration options for each Confluent Platform component, you should refer to their respective documentation.
Note
The configuration variable names are prefixed with the name of the component. For example, the Kafka image will take variables prefixed with
KAFKA_
.
ZooKeeper¶
The ZooKeeper image uses variables prefixed with ZOOKEEPER_
with the variables expressed exactly as they would appear in the zookeeper.properties
file. As an example, to set clientPort
, tickTime
, and syncLimit
run the command below:
docker run -d \ --net=host \ --name=zookeeper \ -e ZOOKEEPER_CLIENT_PORT=32181 \ -e ZOOKEEPER_TICK_TIME=2000 \ -e ZOOKEEPER_SYNC_LIMIT=2 confluentinc/cp-zookeeper:3.2.1
Required Settings¶
ZOOKEEPER_CLIENT_PORT
This field is always required. Tells ZooKeeper where to listen for connections by clients such as Kafka.
ZOOKEEPER_SERVER_ID
Only required when running in clustered mode. Sets the server ID in themyid
file, which consists of a single line containing only the text of that machine’s id. Somyid
of server 1 would contain the text “1” and nothing else. The id must be unique within the ensemble and should have a value between 1 and 255.
Confluent Kafka (cp-kafka)¶
The Kafka image uses variables prefixed with KAFKA_
with an underscore (_) separating each word instead of periods. As an example, to set broker.id
, advertised.listeners
and zookeeper.connect
you’d run the following command:
docker run -d \ --net=host \ --name=kafka \ -e KAFKA_ZOOKEEPER_CONNECT=localhost:32181 \ -e KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://localhost:29092 \ -e KAFKA_BROKER_ID=2 \ confluentinc/cp-kafka:3.2.1Note
You’ll notice that we set the
KAFKA_ADVERTISED_LISTENERS
variable tolocalhost:29092
. This is an important setting, as it will make Kafka accessible from outside the container by advertising it’s location on the Docker host.
Required Settings¶
KAFKA_ZOOKEEPER_CONNECT
Tells Kafka how to get in touch with ZooKeeper.
KAFKA_ADVERTISED_LISTENERS
Advertised listeners is required for starting up the Docker image because it is important to think through how other clients are going to connect to kafka. In a Docker environment, you will need to make sure that your clients can connect to Kafka and other services. Advertised listeners is how it gives out a host name that can be reached by the client.
Confluent Enterprise Kafka (cp-enterprise-kafka)¶
The Enterprise Kafka image includes the packages for Confluent Auto Data Balancing and Proactive support in addition to Kafka. The Enterprise Kafka image uses variables prefixed with KAFKA_
for Apache Kafka and with CONFLUENT_
for Confluent components. These variables have an underscore (_) separating each word instead of periods. As an example, to set broker.id
, advertised.listeners
, zookeeper.connect
, confluent.support.customer.id
you’d run the following command:
docker run -d \ --net=host \ --name=kafka \ -e KAFKA_ZOOKEEPER_CONNECT=localhost:32181 \ -e KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://localhost:29092 \ -e KAFKA_BROKER_ID=2 \ -e CONFLUENT_SUPPORT_CUSTOMER_ID=c0 \ confluentinc/cp-enterprise-kafka:3.2.1Note
You’ll notice that we set the
KAFKA_ADVERTISED_LISTENERS
variable tolocalhost:29092
. This is an important setting, as it will make Kafka accessible from outside the container by advertising it’s location on the Docker host.If you want to enable Proactive support or use Confluent Auto Data Balancing features, please follow the Proactive support and ADB documentation at Confluent documentation.
Required Settings¶
KAFKA_ZOOKEEPER_CONNECT
Tells Kafka how to get in touch with ZooKeeper.
KAFKA_ADVERTISED_LISTENERS
Advertised listeners is required for starting up the Docker image because it is important to think through how other clients are going to connect to kafka. In a Docker environment, you will need to make sure that your clients can connect to Kafka and other services. Advertised listeners is how it gives out a host name that can be reached by the client.
Schema Registry¶
For the Schema Registry image, use variables prefixed with SCHEMA_REGISTRY_
with an underscore (_) separating each word instead of periods. As an example, to set kafkastore.connection.url
, host.name
, listeners
and debug
you’d run the following:
docker run -d \ --net=host \ --name=schema-registry \ -e SCHEMA_REGISTRY_KAFKASTORE_CONNECTION_URL=localhost:32181 \ -e SCHEMA_REGISTRY_HOST_NAME=localhost \ -e SCHEMA_REGISTRY_LISTENERS=http://localhost:8081 \ -e SCHEMA_REGISTRY_DEBUG=true \ confluentinc/cp-schema-registry:3.2.1
Required Settings¶
SCHEMA_REGISTRY_KAFKASTORE_CONNECTION_URL
ZooKeeper URL for the Kafka cluster.
SCHEMA_REGISTRY_HOST_NAME
The host name advertised in Zookeeper. Make sure to set this if running Schema Registry with multiple nodes. Hostname is required because it defaults to the Java canonical host name for the container, which may not always be resolvable in a Docker environment. Hostname must be resolveable because slave nodes serve registration requests indirectly by simply forwarding them to the current master, and returning the response supplied by the master. For more information, please refer to the Schema Registry documentation on Single Master Architecture.
Kafka REST Proxy¶
For the Kafka REST Proxy image use variables prefixed with KAFKA_REST_
with an underscore (_) separating each word instead of periods. As an example, to set the listeners
, schema.registry.url
and zookeeper.connect
you’d run the following command:
docker run -d \ --net=host \ --name=kafka-rest \ -e KAFKA_REST_ZOOKEEPER_CONNECT=localhost:32181 \ -e KAFKA_REST_LISTENERS=http://localhost:8082 \ -e KAFKA_REST_SCHEMA_REGISTRY_URL=http://localhost:8081 \ confluentinc/cp-kafka-rest:3.2.1
Required Settings¶
The following settings must be passed to run the REST Proxy Docker image.
KAFKA_REST_HOST_NAME
The host name used to generate absolute URLs in responses. Hostname is required because it defaults to the Java canonical host name for the container, which may not always be resolvable in a Docker environment. For more details, please refer to the Confluent Platform documentation on REST proxy deployment.
KAFKA_REST_ZOOKEEPER_CONNECT
Specifies the ZooKeeper connection string in the form hostname:port where host and port are the host and port of a ZooKeeper server. To allow connecting through other ZooKeeper nodes when that ZooKeeper machine is down you can also specify multiple hosts in the form hostname1:port1,hostname2:port2,hostname3:port3.
The server may also have a ZooKeeper
chroot
path as part of it’s ZooKeeper connection string which puts its data under some path in the global ZooKeeper namespace. If so the consumer should use the same chroot path in its connection string. For example to give a chroot path of /chroot/path you would give the connection string ashostname1:port1,hostname2:port2,hostname3:port3/chroot/path
.
Kafka Connect¶
The Kafka Connect image uses variables prefixed with CONNECT_
with an underscore (_) separating each word instead of periods. As an example, to set the required properties like bootstrap.servers
, the topic names for config
, offsets
and status
as well the key
or value
converter, run the following command:
docker run -d \ --name=kafka-connect \ --net=host \ -e CONNECT_BOOTSTRAP_SERVERS=localhost:29092 \ -e CONNECT_REST_PORT=28082 \ -e CONNECT_GROUP_ID="quickstart" \ -e CONNECT_CONFIG_STORAGE_TOPIC="quickstart-config" \ -e CONNECT_OFFSET_STORAGE_TOPIC="quickstart-offsets" \ -e CONNECT_STATUS_STORAGE_TOPIC="quickstart-status" \ -e CONNECT_KEY_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_VALUE_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_INTERNAL_KEY_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_INTERNAL_VALUE_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_REST_ADVERTISED_HOST_NAME="localhost" \ confluentinc/cp-kafka-connect:3.2.1
Required Settings¶
The following settings must be passed to run the Kafka Connect Docker image.
CONNECT_BOOTSTRAP_SERVERS
A unique string that identifies the Connect cluster group this worker belongs to.
CONNECT_GROUP_ID
A unique string that identifies the Connect cluster group this worker belongs to.
CONNECT_CONFIG_STORAGE_TOPIC
The name of the topic in which to store connector and task configuration data. This must be the same for all workers with the samegroup.id
CONNECT_OFFSET_STORAGE_TOPIC
The name of the topic in which to store offset data for connectors. This must be the same for all workers with the samegroup.id
CONNECT_STATUS_STORAGE_TOPIC
The name of the topic in which to store state for connectors. This must be the same for all workers with the samegroup.id
CONNECT_KEY_CONVERTER
Converter class for keys. This controls the format of the data that will be written to Kafka for source connectors or read from Kafka for sink connectors.
CONNECT_VALUE_CONVERTER
Converter class for values. This controls the format of the data that will be written to Kafka for source connectors or read from Kafka for sink connectors.
CONNECT_INTERNAL_KEY_CONVERTER
Converter class for internal keys that implements theConverter
interface.
CONNECT_INTERNAL_VALUE_CONVERTER
Converter class for internal values that implements theConverter
interface.
CONNECT_REST_ADVERTISED_HOST_NAME
Advertised host name is required for starting up the Docker image because it is important to think through how other clients are going to connect to Connect REST API. In a Docker environment, you will need to make sure that your clients can connect to Connect and other services. Advertised host name is how Connect gives out a host name that can be reached by the client.
Optional Settings¶
All other settings for Connect like security, monitoring interceptors, producer and consumer overrides can be passed to the Docker images as environment variables. The names of these environment variables are derived by replacing .
with _
, converting the resulting string to uppercase and prefixing it with CONNECT_
. For example, if you need to set ssl.key.password
, the environment variable name would be CONNECT_SSL_KEY_PASSWORD
.
The image will then convert these environment variables to corresponding Connect config variables.
Confluent Control Center¶
The Confluent Control Center image uses variables prefixed with CONTROL_CENTER_
with an underscore (_) separating each word instead of periods. As an example, the following command runs Control Center, passing in its ZooKeeper, Kafka, and Connect configuration parameters.
docker run -d \
--net=host \
--name=control-center \
--ulimit nofile=16384:16384 \
-e CONTROL_CENTER_ZOOKEEPER_CONNECT=localhost:32181 \
-e CONTROL_CENTER_BOOTSTRAP_SERVERS=localhost:29092 \
-e CONTROL_CENTER_REPLICATION_FACTOR=1 \
-e CONTROL_CENTER_CONNECT_CLUSTER=http://localhost:28082 \
-v /mnt/control-center/data:/var/lib/confluent-control-center \
confluentinc/cp-control-center:3.2.1
Docker Options¶
- File descriptor limit: Control Center may require many open files so we recommend setting the file descriptor limit to at least 16384
- Data persistence: the Control Center image stores its data in the /var/lib/confluent-control-center directory. We recommend that you bind this to a volume on the host machine so that data is persisted across runs.
Required Settings¶
The following settings must be passed to run the Confluent Control Center image.
CONTROL_CENTER_ZOOKEEPER_CONNECT
Specifies the ZooKeeper connection string in the form hostname:port where host and port are the host and port of a ZooKeeper server. To allow connecting through other ZooKeeper nodes when that ZooKeeper machine is down you can also specify multiple hosts in the form
hostname1:port1,hostname2:port2,hostname3:port3
.The server may also have a ZooKeeper
chroot
path as part of it’s ZooKeeper connection string which puts its data under some path in the global ZooKeeper namespace. If so the consumer should use the same chroot path in its connection string. For example to give a chroot path of /chroot/path you would give the connection string ashostname1:port1,hostname2:port2,hostname3:port3/chroot/path
.
CONTROL_CENTER_BOOTSTRAP_SERVERS
A list of host/port pairs to use for establishing the initial connection to the Kafka cluster. The client will make use of all servers irrespective of which servers are specified here for bootstrapping; this list only impacts the initial hosts used to discover the full set of servers. This list should be in the form host1:port1,host2:port2,.... Since these servers are just used for the initial connection to discover the full cluster membership (which may change dynamically), this list need not contain the full set of servers (you may want more than one, though, in case a server is down).
CONTROL_CENTER_REPLICATION_FACTOR
Replication factor for Control Center topics. We recommend setting this to 3 in a production environment.
Optional Settings¶
CONTROL_CENTER_CONNECT_CLUSTER
To enable Control Center to interact with a Kafka Connect cluster, set this parameter to the REST endpoint URL for the Kafka Connect cluster.
Confluent Enterprise Replicator¶
Confluent Kafka Replicator is a Kafka connector and runs on a Kafka Connect cluster. The image uses variables prefixed with CONNECT_
with an underscore (_) separating each word instead of periods. As an example, to set the required properties like bootstrap.servers
, the topic names for config
, offsets
and status
as well the key
or value
converter, run the following command:
docker run -d \ --name=cp-enterprise-replicator \ --net=host \ -e CONNECT_BOOTSTRAP_SERVERS=localhost:29092 \ -e CONNECT_REST_PORT=28082 \ -e CONNECT_GROUP_ID="quickstart" \ -e CONNECT_CONFIG_STORAGE_TOPIC="quickstart-config" \ -e CONNECT_OFFSET_STORAGE_TOPIC="quickstart-offsets" \ -e CONNECT_STATUS_STORAGE_TOPIC="quickstart-status" \ -e CONNECT_KEY_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_VALUE_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_INTERNAL_KEY_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_INTERNAL_VALUE_CONVERTER="org.apache.kafka.connect.json.JsonConverter" \ -e CONNECT_REST_ADVERTISED_HOST_NAME="localhost" \ confluentinc/cp-kafka-connect:3.2.1
The following example shows how to create a Confluent Kafka Replicator connector which replicates topic “confluent” from source Kafka cluster (src) to a destination Kafka cluster (dest).
curl -X POST \ -H "Content-Type: application/json" \ --data '{ "name": "confluent-src-to-dest", "config": { "connector.class":"io.confluent.connect.replicator.ReplicatorSourceConnector", "key.converter": "io.confluent.connect.replicator.util.ByteArrayConverter", "value.converter": "io.confluent.connect.replicator.util.ByteArrayConverter", "src.zookeeper.connect": "zookeeper-src:2181", "src.kafka.bootstrap.servers": "kafka-src:9082", "dest.zookeeper.connect": "zookeeper-dest:2181", "topic.whitelist": "confluent", "topic.rename.format": "${topic}.replica"}}' \ http://localhost:28082/connectors
Required Settings¶
The following settings must be passed to run the Kafka Connect Docker image:
CONNECT_BOOTSTRAP_SERVERS
A unique string that identifies the Connect cluster group this worker belongs to.
CONNECT_GROUP_ID
A unique string that identifies the Connect cluster group this worker belongs to.
CONNECT_CONFIG_STORAGE_TOPIC
The name of the topic in which to store connector and task configuration data. This must be the same for all workers with the samegroup.id
CONNECT_OFFSET_STORAGE_TOPIC
The name of the topic in which to store offset data for connectors. This must be the same for all workers with the samegroup.id
CONNECT_STATUS_STORAGE_TOPIC
The name of the topic in which to store state for connectors. This must be the same for all workers with the samegroup.id
CONNECT_KEY_CONVERTER
Converter class for keys. This controls the format of the data that will be written to Kafka for source connectors or read from Kafka for sink connectors.
CONNECT_VALUE_CONVERTER
Converter class for values. This controls the format of the data that will be written to Kafka for source connectors or read from Kafka for sink connectors.
CONNECT_INTERNAL_KEY_CONVERTER
Converter class for internal keys that implements theConverter
interface.
CONNECT_INTERNAL_VALUE_CONVERTER
Converter class for internal values that implements theConverter
interface.
CONNECT_REST_ADVERTISED_HOST_NAME
Advertised host name is required for starting up the Docker image because it is important to think through how other clients are going to connect to Connect REST API. In a Docker environment, you will need to make sure that your clients can connect to Connect and other services. Advertised host name is how Connect gives out a host name that can be reached by the client.
Optional Settings¶
All other settings for Connect like security, monitoring interceptors, producer and consumer overrides can be passed to the Docker images as environment variables. The names of these environment variables are derived by replacing .
with _
, converting the resulting string to uppercase and prefixing it with CONNECT_
. For example, if you need to set ssl.key.password
, the environment variable name would be CONNECT_SSL_KEY_PASSWORD
.
The image will then convert these environment variables to corresponding Connect config variables.