This topic describes the precautions and methods for deploying OceanBase Migration Service (OMS) of the current version.
Deploy OMS on a single node
OMS deployed on a single node can provide all OMS features. If you do not need to use a high-availability environment, you can deploy OMS on a single node. For more information, see Deploy OMS on a single node.
Deploy OMS on multiple nodes in a single region
You can deploy OMS on multiple nodes to achieve high availability (HA). In the HA deployment architecture, each OMS node provides all OMS features, and the backend task framework implements distributed scheduling based on the database. When an OMS node becomes unavailable, the framework automatically schedules tasks to other available OMS nodes. OMS also supports the deployment on multiple nodes in a single region. For more information, see Deploy OMS on multiple nodes in a single region.
Note:
HA does not enable the automatic restoration of the checker.
We recommend that you apply multi-node deployment in the production environment.
Deploy OMS on multiple nodes in multiple regions
OMS supports multi-region deployment. You can deploy OMS across multiple regions. For more information, see Deploy OMS on multiple nodes in multiple regions.
OAT-based deployment
If OceanBase Admin Toolkit (OAT) is installed in your environment, you can deploy OMS on the server by using OAT. For more information, see the OAT documentation.
Precautions
If you use OMS V2.1.0 and later to collect and display historical monitoring data, you need to deploy an InfluxDB database.
By default, HA is disabled in OMS. To enable HA, use one of the following methods:
If you use OMS earlier than V3.1.0, run the following command:
supervisorctl start oms_drc_haIf you use OMS V3.1.0 and later, go to the OMS console, find the
ha.configparameter, and set theenableparameter totrue. For more information, see Modify HA configurations.