Intel® Kubernetes Service Cloud Monitor¶
Use Cloud Monitor on the Intel® Kubernetes Service Control Plane, where you can view Kubernetes clusters.
Select the Kubernetes tab for the selected cluster in the Intel® Tiber™ AI Cloud console.
Select the appropriate cluster in the dropdown.
Kubernetes API Server¶
CPU Usage¶
This chart shows CPU utilization in % for the Intel® Kubernetes Service Control plane API Server by nodename. Use it to understand whether overall CPU usage is high or if there are overload scenarios during a specific time.
Memory Usage¶
This chart shows Memory utilization in megabytes (MB) for the Kubernetes Control plane API Server by nodename.
Requests by code¶
This chart shows the total number of requests handled by the Kubernetes API server by http code.
Requests by verb¶
This chart shows the total number of requests handled by the Kubernetes API server by HTTP verb. These verbs reflect standard API request names, like POST, GET, and PATCH.
Latency by node name¶
This chart shows the average latency by dividing the total latency by the number of requests, in milliseconds, grouped by the nodename.
Errors by nodename¶
This chart shows the percentage of error requests grouped by nodename.
An error request displays with error code 4xx, 5xx.
The number of error requests is divided by total number of requests to get the percentage error.
Errors by verb¶
This chart shows the percentage of error requests grouped by HTTP verb.
An error request displays with error code 4xx, 5xx.
The number of error requests is divided by total number of requests to get the percentage error.
HTTP Requests by nodename¶
This chart shows the total number of HTTP requests handled by the Kubernetes API server by nodename.
Kubernetes etcd Metrics¶
CPU Usage¶
This chart shows CPU utilization in % for the IKS Control plane etcd,
grouped by nodename.
Use it to understand whether overall CPU usage is high or if there are overload scenarios during a specific time.
Memory usage¶
This is a chart that which provides Memory utilization for the Kubernetes Control plane etcd grouped by nodename.
Client traffic In¶
This chart shows the rate of data received over gRPC by etcd from all clients grouped by nodename, measured in KiloBytes per second (KBps).
Client traffic Out¶
This chart shows the rate of data sent, in number of KiloBytes per second (KBps), over gRPC by etcd to its clients by etcd grouped by nodename.
Peer traffic In¶
This chart shows the rate of data received, in number of KiloBytes per second (KBps), over gRPC by etcd from all its peers grouped by nodename.
Peer traffic Out¶
This chart shows the rate of data number of bytes sent, in number of KiloBytes per second (KBps), over gRPC by etcd to its peers by etcd grouped by nodename.
DB Size in use¶
This chart shows the total size of the etcd database currently in use, in bytes. This metric reflects how much disk space the etcd key-value store is utilizing for storing data.
Total heartbeat send failures¶
This chart shows the total number of failed attempts by an etcd node to send heartbeats to its peers in the cluster grouped by nodename.
Next Steps¶
For more help and troubleshooting, see Cloud Monitor.