Why Is Basic Monitoring Data Inconsistent with Data Monitored by the OS?
Symptoms
CPU Usage under Basic Monitoring is close to 100%, which is very different from the CPU usage monitored by the OS (50%).
Possible Causes
- If you set idle to poll in the guest operating system (guest OS), and the guest OS is idle and enters the polling state, it consumes compute resources and does not proactively release CPU resources. As a result, the CPU usage is abnormal.
- In a HANA ECS, idle is set to mwait in the guest OS. When the guest OS is idle and enters the mwait state, the guest OS consumes less resources than that when idle is set to poll. However, the guest OS does not proactively release CPU resources, either. As a result, the CPU usage is abnormal.
Solution
Install and configure the Agent to view OS monitoring data.
Monitored Data Exceptions FAQs
- Why Is the Monitoring Data Not Displayed on the Cloud Eye Console?
- Why I Cannot See the Monitoring Data on the Cloud Eye Console After Purchasing Cloud Service Resources?
- Why Doesn't the Cloud Eye Console Display the OS Monitoring Data or Why Isn't the Data Displayed Immediately After the Agent Is Installed and Configured on an ECS?
- Why Is Basic Monitoring Data Inconsistent with Data Monitored by the OS?
- Why Are the Network Traffic Metric Values in Cloud Eye Different from Those Detected in ECS?
- Why Is the Metric Collection Point Lost During Certain Periods of Time?
- Why Are the Four Metrics Memory Usage, Disk Usage, Inband Incoming Rate, and Inband Outgoing Rate Not Displayed for an ECS?
- What Are the Impacts on ECS Metrics If UVP VMTools Is Not Installed on ECSs?
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.
more