Starting and Stopping the Ozone Cluster

This guide describes how to start and stop an Ozone cluster, assuming it has already been configured and initialized.

Startup the cluster

Run the following command on each SCM host:

ozone --daemon start scm

Run the following command on each OM host:

ozone --daemon start om

At this point Ozone's name services, the Ozone Manager, and the block service SCM are both running.

Now we need to start the Datanodes. Please run the following command on each Datanode.

ozone --daemon start datanode

Wait until SCM exits safe mode

ozone admin safemode wait -t 240

At this point SCM, Ozone Manager and Datanodes are up and running and are ready to serve requests.

If you need to start optional services like the Recon server, S3 Gateway, or HttpFS Gateway, you can start them separately:

ozone --daemon start recon
ozone --daemon start s3g
ozone --daemon start httpfs

If you want to make your life simpler, you can just run:

start-ozone.sh

This assumes that you have set up the workers file correctly and ssh configuration that allows ssh-ing to all Datanodes.

You are expected to list all hostnames or IP addresses in your ${OZONE_CONF_DIR}/workers file, one per line.
You have passwordless SSH access configured from the control node to all other nodes in the cluster.

To stop all core services, you can use the stop-ozone.sh script from your SCM or OM node:

stop-ozone.sh

To stop services individually, you can run the following commands on their respective nodes:

ozone --daemon stop scm
ozone --daemon stop om
ozone --daemon stop datanode

To stop optional services, run the following commands on their respective nodes:

ozone --daemon stop recon
ozone --daemon stop s3g
ozone --daemon stop httpfs