3. vStream Configuration

3.1. vStream Distributed Coordination Server/Client Services

vStream provides an distributed configuration service, synchronization service, and naming registry for large distributed systems.

It is required to configure three zookeeper installations. Perform the following steps on each server.

  1. Edit the zookeeper configuration file.

user@host#: vi /etc/zookeeper/zoo.cfg

  1. Append the following at the bottom of the file. (Using IP address or Hostname)

server.1=<Server1.IP.Adress>:2888:3888
server.2=<Server2.IP.Adress>:2888:3888
server.3=<Server3.IP.Adress>:2888:3888

  1. Set the server id. Changing the # below to 1 for server 1, 2 for server 2, and 3 for server 3.

user@host#: echo # > /var/lib/zookeeper/data/myid

  1. Start the zookeeper service.

user@host#: service zookeeper start

Note

user@host#: tail -F /var/log/zookeeper/zookeeper.log to ensure zookeeper is running as expected.

3.2. vStream Distributed Commit Log Service

vStream is publish-subscribe messaging rethought as a distributed commit log.

It is required to configure kafka on each data tier server.

  1. Edit the kafka configuration file.

user@host#: vi /opt/kafka/config/server.properties

  1. Find and change the following items.

Find and change the broker id to a unique number starting from 0.

broker.id=0

Find and change the logs directory.

log.dirs=/data/vstream-logs-data01

Find and change the zookeeper connection string. Replacing the Server with correct IP address or Hostname from the zookeeper configuration steps above.

zookeeper.connect=<Server1.IP.Adress>:2181,<Server2.IP.Adress>:2181,<Server3.IP.Adress>:2181

  1. Make the logs directory and set the correct ownership.

user@host#: mkdir /data/vstream-logs-data01

user@host#: chown -R kafka:kafka /data/vstream-logs-data01

  1. Start the kafka service.

user@host#: service kafka start

  1. Repeat steps 1 through 4 for each data tier server.

Note

user@host#: tail -F /var/log/kafka/server.out to ensure kafka is running as expected.