ZooKeeper on Docker

This Dockerfile creates a Docker image that can be used as the base for running ZooKeeper within a Docker container. The run script is responsible for creating the ZooKeeper configuration based on the container's environment and starting the ZooKeeper service.

The version of ZooKeeper is defined in the Dockerfile and generally points to the latest stable release of ZooKeeper.

Environment variables

The following environment variables are understood by the startup script to seed the service's configuration:

• SERVICE_NAME should contain the logical name of the service this container is an instance of;
• CONTAINER_NAME should contain the logical name of the container, which will be used for looking up links and ports informations from the other environment variables. For this, the name is uppercased and non-alphanumeric characters are replaced by underscores.
• <SERVICE_NAME>_<CONTAINER_NAME>_CLIENT_INTERNAL_PORT, which controls the clientPort configuration setting. Defaults to 2181;
• <SERVICE_NAME>_<CONTAINER_NAME>_PEER_PORT, which is used as the peer port specified in the server list for this node (and the others). Defaults to 2888;
• <SERVICE_NAME>_<CONTAINER_NAME>_LEADER_ELECTION_PORT, which is used as the leader election port specified in the server list for this node (and the others). Defaults to 3888;
• MAX_SNAPSHOT_RETAIN_COUNT, the maximum number of snapshots to retain on disk. Defaults to to 10;
• PURGE_INTERVAL, the interval, in hours, being transaction logs and snapshot purges. Defaults to 24 (daily).

If you plan to run a cluster of nodes, you also need to pass in the ZOOKEEPER_SERVER_IDS environment variable containing a comma-separated list of cluster nodes descriptors, as tuples of node:id. The node part being the container name of a node, from which its coordinates can be found.

For example, in three-node cluster, each node would expect the following set of environment variables:

SERVICE_NAME=zookeeper
CONTAINER_NAME=zk-node-{0,1,2}
ZOOKEEPER_SERVER_IDS=zk-node-0:1,zk-node-1:2,zk-node-2:3
ZOOKEEPER_INSTANCES=zk-node-0,zk-node-1,zk-node-2
ZOOKEEPER_ZK_NODE_0_HOST=host1
ZOOKEEPER_ZK_NODE_0_CLIENT_PORT=2181
ZOOKEEPER_ZK_NODE_0_PEER_PORT=2888
ZOOKEEPER_ZK_NODE_0_LEADER_ELECTION_PORT=3888
ZOOKEEPER_ZK_NODE_1_HOST=host2
ZOOKEEPER_ZK_NODE_1_CLIENT_PORT=2181
ZOOKEEPER_ZK_NODE_1_PEER_PORT=2888
ZOOKEEPER_ZK_NODE_1_LEADER_ELECTION_PORT=3888
ZOOKEEPER_ZK_NODE_2_HOST=host3
ZOOKEEPER_ZK_NODE_2_CLIENT_PORT=2181
ZOOKEEPER_ZK_NODE_2_PEER_PORT=2888
ZOOKEEPER_ZK_NODE_2_LEADER_ELECTION_PORT=3888

This will make zk-node-0 be ZooKeeper node ID 1 running on host1, etc.

If ZOOKEEPER_SERVER_IDS is not specified, the container will run in single-node mode.

Volumes

The ZooKeeper images uses the following volumes that you may want to bind from the container's host:

• /var/lib/zookeeper, for the ZooKeeper data snapshots.
• /var/log/zookeeper, for the ZooKeeper service logs (rotated 10 times by 100MB sections).

Usage

To build a new image, simply run from this directory:

$ docker build -t `whoami`/zookeeper:3.4.6 .

The Docker image will be built and now available for Docker to start a new container from:

$ docker images | grep zookeeper
mpetazzoni/zookeeper   3.4.6              294bef30310b        2 days ago          12.29 kB (virtual 856.7 MB)