Cluster is a resource dimension in TDMQ for RocketMQ, and namespaces, topics, and groups of different clusters are completely isolated from each other. Each cluster has its own resource limits, such as the total number of topics and message retention period. It is a common practice to use respective dedicated clusters for development, test, and production environments.
Clusters are divided into virtual clusters and exclusive clusters:
Exclusive cluster: Physical resources are exclusive, and data is secure. There are almost no use limits.
Virtual cluster: Virtual computing and storage resources are used and automatically allocated based on usage. There are certain use limits.
TDMQ for RocketMQ resource hierarchy
Directions
Creating a cluster
1. Log in to the TDMQ console and enter the Cluster page.
2. On the Cluster Management page, select the region and click Create Cluster to enter the Create Cluster window.
Cluster Type: Select Exclusive cluster or Virtual cluster.
Exclusive cluster: Physical resources are exclusive, and data is secure. There are almost no use limits.
Virtual cluster: Virtual computing and storage resources are used and automatically allocated based on usage. There are certain use limits.
Billing Mode: Exclusive clusters are monthly subscribed, while virtual clusters are pay-as-you-go.
Region: Select a region closest to your business. Tencent Cloud products in different regions cannot communicate with each other over the private network. Select a region with caution, as it cannot be changed after purchase. For example, CVM instances in Guangzhou region cannot access clusters in Shanghai region over the private network. If you need cross-region communication over the private network, see Creating Intra-account Peering Connection.
Cluster Specification: Select an appropriate cluster specification as needed.
VPC: Authorize to bind the domain name of the new cluster's access point to the specified VPC.
Public Network Access: After the public network bandwidth is enabled, additional fees will be incurred, and you need to configure a security policy. If you don't configure any security policy, all IPs will be denied access by default.
Virtual cluster: Billing by the public network bandwidth hasn't been started, and we recommend that you use the public network access only in testing and debugging environments. The public network access is not enabled by default. To enable it, submit an application.
Cluster Name: Enter the cluster name, which can contain 3–64 digits, letters, hyphens, and underscores.
Tag: Tags are used to manage resources by category in different dimensions. For detailed directions, see Managing Resource with Tag.
3. Click Buy Now.
Next steps:
1. Create a namespace in the cluster and get the access address and connection information of the server. For more information, see Namespace Management.
2. Create a role in the cluster and grant it the production/consumption permissions of the namespace.
4. Write a demo as instructed in the SDK documentation and configure the connection information for message production/consumption.
Viewing cluster details
Exclusive Cluster
Virtual Cluster
On the Cluster Management list page, click View Details in the Operation column to enter the cluster details page. On the details page, you can query:
Cluster Statistics: This section displays the current cluster's message consumption speed, message production rate, number of heaped messages, cluster consumption traffic per second, and cluster production traffic per second in the selected time range.
Cluster's basic information (cluster name/ID, region, creation time, remarks, and resource tag)
Network: This section displays the VPC, public network bandwidth, as well as the private and public network access points for TCP and HTTP.
Cluster Overview: This section displays the number of resources, resource quota usage, and message type distribution in the current cluster.
Top Cluster Resource Consumption: This section displays the ranking of groups and topics using major resources in the current cluster, including the ranking of heaped message count and dead letter count of the groups, the ranking of production and consumption speed of the topics, and the ranking of occupied storage space of the topics.
After the exclusive cluster is created, you can modify the message retention time of the cluster based on the metrics of storage usage and business peak conditions. On the cluster list page, click Edit to modify the message retention time on the cluster information page.
On the Cluster Management list page, click View Details in the Operation column to enter the cluster details page. On the details page, you can query:
Cluster overview (numbers of topics, currently heaped messages, and month-to-date API calls)
Cluster's basic information (cluster name/ID, region, creation time, remarks, and resource tag)
Cluster configuration:
Instance Configuration
Configuration Description
Max TPS
The upper limit for production TPS does not apply to consumption TPS.
Max Namespaces
Maximum number of namespaces that can be created.
Max Topics
Maximum number of topics that can be created.
Max Groups
Maximum number of groups that can be created.
Max Retention Period
Maximum message retention period that can be configured.
Network: This section displays the private and public network access points for TCP and HTTP.
Top Cluster Resource Consumption: This section displays the ranking of groups and topics using major resources in the current cluster, including the ranking of heaped message count and dead letter count of the groups as well as the ranking of production and consumption speed of the topics.
Upgrading the cluster specification
If the current cluster specifications cannot meet your business needs, you can increase the node specification, node quantity, and storage specification in the console.
Note
Currently, cluster specification upgrade is supported only for exclusive clusters.
Currently, you can only upgrade but not downgrade node count and node storage specification.
During the upgrade of exclusive clusters, whether the operations involve adding new nodes, upgrading node specifications or upgrading storage specifications, Tencent Cloud ensures that the upgrade process is smooth and imperceptible, without causing downtime on the client-side applications.
1. On the Cluster list page, click Edit in the Operation column of the target cluster.
2. Select the target node specification and click Confirm the Adjustment.
Target Node Count: You can only increase but not decrease it.
Single-Node Storage Specification: You can only increase but not decrease it. The new single-node storage specification takes effect for all nodes in the cluster.
Downgrading the cluster specification
After purchasing an exclusive cluster, you can adjust its node specifications based on the business volume and cluster usage.
1. On the Cluster list page, click Downgrade in the Operation column of the target cluster.
2. Select the target node specification and click Confirm the Adjustment.
Note
Currently, the single-node storage specification cannot be downgraded because of the constraints of the underlying components. In order to avoid data loss during the downgrade, you can only modify the node specifications of the cluster, but not the number of nodes.
Adjusting the public network bandwidth
Note
Currently, this is supported only for exclusive clusters.
After creating the exclusive cluster, you can enable/disable public network access by adjusting the public network bandwidth, modify the public network bandwidth value, and configure the public network security policy to restrict user access.
There are two ways to adjust the public network bandwidth:
Option 1: On the cluster list page, click More > Adjust Public Network Bandwidth in the Operation column.
Option 2: On the cluster's Basic Info page, click Edit for Public Network Bandwidth in the Network section.
Public Network Bandwidth: Select the target public network bandwidth.
Public Network Access Policy: Enter the IPs that are allowed to access. If you don't configure any security policy, all IPs will be denied access by default. If the newly added rule and an existing rule overlap, the new rule will be matched first.
Editing a cluster
1. On the Cluster list page, click Edit in the Operation column of the target cluster.
2. Enter the cluster name and description in the pop-up window and click Submit.
Deleting a cluster
Note
After a cluster is deleted, all its configurations will be cleared and cannot be recovered. Therefore, proceed with caution.
You can delete a cluster in the following steps:
1. On the Cluster list page, click Delete in the Operation column of the target cluster.
2. In the deletion confirmation pop-up window, click Delete.
Note
In order to prevent misoperations from causing internal cluster data (such as topic and group) to be deleted, when you delete a cluster, the resources in your cluster will be verified in the console. If there are undeleted resources such as namespaces, topics or groups, the cluster cannot be deleted.
When deleting a namespace, you need to delete the topic and group in the namespace first. Currently, you can batch delete topics and groups in the console. Before deleting a namespace, you also need to delete the role bound to the namespace. To unbind the role from the namespace, go to the namespace list page, click Configure Permissions.
After entering the permission configuration page, click Delete to unbind the two from each other, and then you can go to the namespace page again to delete the namespace.
Yes
No
Was this page helpful?