The Event Center capability of TDMQ for CKafka (CKafka) supports centralized management, storage, analysis, and visualization of various Ops events, diagnostic events, and broker change events that occur when instances are running. This facilitates future querying, auditing, and retrospective analysis. It also supports event alarm capabilities. You can configure alarm rules for critical events (such as node offline and automatic disk capacity expansion) to send event details to target recipients, enabling Ops personnel to handle them promptly.The Event Center capability of TDMQ for CKafka (CKafka) supports centralized management, storage, analysis, and visualization of various Ops events, diagnostic events, and broker change events that occur when instances are running. This facilitates future querying, auditing, and retrospective analysis. It also supports event alarm capabilities. You can configure alarm rules for critical events (such as node offline and automatic disk capacity expansion) to send event details to target recipients, enabling Ops personnel to handle them promptly.
Event Types Supported by CKafka and Handling RecommendationsEvent Types Supported by CKafka and Handling Recommendations
CKafka Event Center provides three types of monitoring capabilities: instance events, diagnostic events, and broker change events. Its alarm support policy is implemented via Tencent Cloud Observability Platform (TCOP): Alarm policies can be configured for certain event types, while others only offer observability capabilities. The following table describes the supported event types.CKafka Event Center provides three types of monitoring capabilities: instance events, diagnostic events, and broker change events. Its alarm support policy is implemented via Tencent Cloud Observability Platform (TCOP): Alarm policies can be configured for certain event types, while others only offer observability capabilities. The following table describes the supported event types.
Instance EventsInstance Events
Alarmable Events
Observation-Only Events
Event Name
Event Description
Handling Method and Recommendation
Automatic Disk Capacity Expansion
High disk utilization triggers automatic expansion.
Evaluate whether to upgrade the disk specifications.
Dynamic disk message retention policy
High disk utilization triggers the dynamic retention policy.
Evaluate whether to upgrade the disk specifications.
Kafka version upgrade
The instance is upgraded to a newer version.
Check whether the version upgrade API is called via a cloud API or the console.
AZ change
The instance is changed to a different AZ.
Check whether the AZ change API is called via a cloud API or the console.
Event Name
Event Description
Handling Method and Recommendation
Automatic Partition Balancing
The instance automatically triggers partition migration to balance the load.
Pay attention to migration latency and avoid execution during peak business hours.
Manual partition balancing
Partition leader reassignment is manually initiated.
Verify traffic balance after the operation.
Kernel minor version upgrade
Emergency update to fix security vulnerabilities or feature defects.
Check version compatibility and check whether the upgrade is called via a cloud API.
Upgrade
Instance specifications, such as disk, bandwidth, and partition, are upgraded.
It is recommended to follow up and confirm the cost changes.
Downgrade
The instance specifications are downgraded to reduce costs.
It is recommended to follow up and observe the impact of the downgrade on the business for further evaluation.
Public Network Bandwidth Adjustment
The upper limit of public network bandwidth is modified.
Check whether the bandwidth change API is called via a cloud API. If an error occurs, it is recommended to troubleshoot accordingly.
Routing Policy Change
The routing rule for Virtual Private Cloud (VPC) network access is adjusted.
If an error occurs, verify the producer/consumer connectivity.
ACL Policy Change
An access control rule is added or deleted at the topic or IP level.
If an error occurs, it is recommended to check the client logs.
User adding/deletion
A Simple Authentication and Security Layer (SASL) authentication account is created or deleted.
It is recommended to synchronously update the authentication credentials for the producer and consumer.
Enable Elastic Bandwidth
Automatic scaling of bandwidth based on traffic peaks is enabled.
It is recommended to follow up on cost changes.
Turn off elastic bandwidth capacity
Automatic scaling of bandwidth is disabled.
It is recommended to refer to the peak traffic of the last 7 days and manually set the bandwidth upper limit.
Diagnostic EventsDiagnostic Events
Observation-Only EventsObservation-Only Events
Event Name
Event Description
Handling Method and Recommendation
Scheduled instance diagnosis
The system periodically performs instance health checks, covering network, disk, and broker status.
When an exception is detected, view the diagnostic report and perform targeted handling.
Instant instance diagnosis
The user manually triggers a real-time health check.
When an exception is detected, view the diagnostic report and perform targeted handling.
Broker Change EventsBroker Change Events
Alarmable Events
Observation-Only Events
Event Name
Event Description
Handling Method and Recommendation
Consumer group member heartbeat timeout
A heartbeat timeout occurs on a consumer.
It is recommended to check whether the consumer is functioning normally.
Occasional heartbeat timeouts may be caused by fluctuations in consumer group members. If heartbeat timeouts occur frequently, it is recommended to check whether message processing is blocked based on the business logic. If blocking exists, adjust the downstream blocking points.
Alternatively, try adjusting the instance configuration.
Consumer group member update
A consumer group member is updated.
It is recommended to check the changes in consumer group members, such as whether there are newly released or added members.
Consumer group rebalancing
A consumer group rebalance occurs.
Occasional consumer group rebalances may be caused by normal fluctuations. If consumer group rebalances persist or occur frequently, it is recommended to perform further troubleshooting to prevent an impact on consumption.
Check whether there is a heartbeat timeout in the consumer group and handle it.
Check whether consumers are frequently created or deleted.
Troubleshoot other consumer group events.
Cluster node launch
A node comes online in the cluster.
It is recommended to first check in Event Center whether any configuration change events have occurred. When configuration changes are in progress, a cluster node going online/offline is normal and requires no special attention.
Events concerning the occasional node going online/offline may be caused by underlying server fluctuations and require no special attention.
If events concerning cluster node going online occur, with persistent impacts on production or consumption activities, and the situation has not recovered for an extended period, contact us.
Cluster node removal
A node goes offline in the cluster.
It is recommended to first check in Event Center whether any configuration change events have occurred. When configuration changes are in progress, a cluster node going online/offline is normal and requires no special attention.
Events concerning the occasional node going online/offline may be caused by underlying server fluctuations and require no special attention.
If events concerning cluster node going offline occur, with persistent impacts on production or consumption activities, and the situation has not recovered for an extended period, contact us.
Event Chinese NameEvent Chinese Name
Event NameEvent Name
Event DescriptionEvent Description
Handling Method and RecommendationHandling Method and Recommendation
Leader switchLeader switch
Leader switchLeader switch
The leader replica of a partition is transferred (may be triggered by configuration changes, replica balancing, disaster recovery failover, planned maintenance, or broker downtime).The leader replica of a partition is transferred (may be triggered by configuration changes, replica balancing, disaster recovery failover, planned maintenance, or broker downtime).
Occasional leader switches are normal and require no special attention.Occasional leader switches are normal and require no special attention.
If leader switch events persist, and production/consumption is continuously affected without recovery for an extended period, you can try restarting the client.If leader switch events persist, and production/consumption is continuously affected without recovery for an extended period, you can try restarting the client.
2. In the left sidebar, select In the left sidebar, select Instance ListInstance List. Click the ID of the target instance to go to the instance details page.. Click the ID of the target instance to go to the instance details page.
3. At the top of the page, select the At the top of the page, select the Event CenterEvent Center tab. Set the time range (last 7 days, last 30 days, or custom range) and select the event type you want to view to filter the corresponding event records. tab. Set the time range (last 7 days, last 30 days, or custom range) and select the event type you want to view to filter the corresponding event records.
4. On the event details page, click On the event details page, click View DetailsView Details in the Operation column of the target event. You can view the detailed event records in the right-side panel. in the Operation column of the target event. You can view the detailed event records in the right-side panel.
Click Click Configure Alarm PolicyConfigure Alarm Policy in the Operation column. You can configure a notification template for the event in EventBridge, sending event details to the target recipients. in the Operation column. You can configure a notification template for the event in EventBridge, sending event details to the target recipients.
Configuring an Event Alarm RuleConfiguring an Event Alarm Rule
2. In the left sidebar, choose In the left sidebar, choose Alarm ManagementAlarm Management > > Alarm ConfigurationAlarm Configuration. Click . Click Create PolicyCreate Policy in the upper-left corner. Enter the policy name, and configure an alarm rule. in the upper-left corner. Enter the policy name, and configure an alarm rule.
Alarm Target: Select the CKafka instance for which you want to configure the alarm policy.Alarm Target: Select the CKafka instance for which you want to configure the alarm policy.
Trigger Condition: Select Template and Configure Manually are supported. By default, Configure Manually is selected. Here, select Event Alarm and add the events for which you want to configure alarms.Trigger Condition: Select Template and Configure Manually are supported. By default, Configure Manually is selected. Here, select Event Alarm and add the events for which you want to configure alarms.
3. Click Click Next step: Configure Alarm NotificationNext step: Configure Alarm Notification. Select or create a notification template and set the alarm recipients and channels.. Select or create a notification template and set the alarm recipients and channels.
4. Click Click CompleteComplete to complete the alarm rule configuration. to complete the alarm rule configuration.
