Restarting Brokers of a RocketMQ Instance

Updated on 2024-12-25 GMT+08:00

View PDF

Restart one or more brokers of a RocketMQ instance at a time on the DMS for RocketMQ console.

During the restart, client requests for message retrieval and creation will be rejected for up to 30s. Restart brokers during off-peak hours. Restarting slave brokers does not affect services.

How do I know which broker is the master?

Each broker has one master node and two standby nodes. For example, to purchase one broker, two brokers will be created. Three brokers are deployed in raft mode with broker IDs 1, 2, and 3. When one broker becomes the master, its broker ID is 0. In this case, the broker whose ID is 0 is the master one.

Figure 1 Restarting brokers
Click to enlarge

Notes and Constraints

Available only for RocketMQ 4.8.0 and not for 5.x.

Prerequisite

The RocketMQ instance must be running.

Procedure

Log in to the console.
Click in the upper left corner to select a region.

NOTE:

Select the region where your RocketMQ instance is located.
Click and choose Application > Distributed Message Service for RocketMQ to open the console of DMS for RocketMQ.
In the row containing the desired RocketMQ instance, click Restart Broker.
Select the brokers to be restarted and click Yes.

It takes 3 to 15 minutes to restart the brokers.