Kubernetes is a complex distributed system, and a large number of events occur in daily operations, from node running status to workload deployment and updates. The IEC platform has the function of recording all events that occur in the Kubernetes cluster and querying these events through the EonKube. This feature plays a key role in system operation and troubleshooting. The event log can capture and display status changes, error and warning messages, etc. of various resources in the cluster. These records include the status of nodes, Pods, services and workloads, which can help users quickly respond and solve problems.

• Real-time monitoring: The event log supports real-time updates. When any important event occurs in the cluster, EonKube UI will automatically update the log to display the latest event information. This is very important for system administrators to understand system status changes in time and detect problems early.
  • Event sequencing: The event log provides multiple sorting methods to make it easier for users to find relevant events:
  • Sort by event type: Sort normal or warning events.
  • Sort by event source: You can choose to view related events of a specific resource (such as a node or Pod).
  • Sort by latest event time: Sort by the last time an event was recorded or detected, which is useful for backtracking and investigating past issues.
• Event details: Each event has detailed message records, including:
  • Timestamp: Tells the specific time the event occurred.
  • Trigger object: If the event is caused by a node, Pod or application, the details will show the specific resource from which the event originated.
  • Event type: Indicates whether the event is normal operation, or a warning or error message.
  • Detailed description: Give a specific description of the event, such as the status change of the resource or errors caused by certain operations (such as Pod deployment failure).
• Event history tracking: Event logs can be used not only to view real-time events, but also to track historical events. Through these historical records, users can look back into past operations to determine when problems began to occur.
• Recording process of events: When an event occurs on a resource (such as a Pod or a node), EonKube will automatically record the following stages of the event:
  • Event occurrence: When there is a state change in the Kubernetes cluster (such as Pod starting, stopping, failing, etc.), the event will be triggered and a record will be automatically generated.
  • Event capture: EonKube UI will continuously monitor the status of the Kubernetes cluster and present these changes to users in the form of event logs.
  • Event recording: Detailed information of each event, including time, source, type and description, will be recorded in the log and can be viewed through filtering.

Assume that a Pod in the cluster fails to start. The recording process of this event is as follows:

1. Event trigger: Pod cannot be started due to insufficient resources or configuration issues.
2. Event log capture: The event log will automatically record the startup failure event of the Pod and display the cause of the error (such as insufficient resources or the image cannot be downloaded).
3. Detailed description: Click the item in the event list to see the specific error message and recommended solutions (such as reconfiguring the Pod or adding resources).
4. Follow-up processing: Administrators can perform corresponding operations based on the information provided by the event log, such as re-deploying Pods or adjusting cluster configuration.

Procedure

When a pod failed to start, to check the related events, follow the procedure below.

1. Go to EonKube > Dashboard > Events. All events are displayed here in a chronological order
2. Click on Full Events list, you could open the Events page.
3. To filter events, use the top bar to filter specific events and sort by event type, last seen time, etc. to quickly find the events related to the problems.
4. To view event details, click the event Object:
   • Basic information: View the name of the pod, the namespace in which the pod is deployed, the time the pod has been running, the internal IP address of the pod, the node where the pod is running, the label and annotation of the pod, which is used to help manage and identify the pod.
   • Containers: View the current running status of the container (such as Running, Terminated or Waiting), whether the container is ready, the name of the container, the image name and version used by the container, the number of restarts of the container, and the startup of the container time, etc.
   • Metrics: View the resource usage of the pod, including the current usage of CPU and memory. This information can help administrators understand whether the pod is using too many resources and make appropriate resource adjustments.
   • Conditions: View the current status of the pod. These conditions describe the health of the pod.
   • Recent Events: View the recent events related to the Pod are listed, which includes all information related to the startup, stop, scheduling and other operations of the Pod. These events help administrators troubleshoot.
   • Related Resources: View other Kubernetes resource information related to the Pod, which may include services connected to the Pod. As well as storage resources such as persistent volumes used by the Pod.