Restarting the VMWare Management Agent
Virtual servers on your vCenter collect data from ESXi servers by using the VMware Management Agent installed on your Virtual Machines (VMs). In cases where the Agent shuts down or otherwise fails, your vCenter can no longer provide performance data for VMs that are running on the host.
The solution is to simply restart the Management Agent, and then to validate that it's performing.
Note: For supporting documentation, please see the following:
- For more information about ESXi, see the article VMware ESXi.
- For help with diagnosing and troubleshooting issues with the reporting of performance data on your vCenter servers, see Troubleshooting gaps in performance data or missing performance data in vCenter Server, particularly Step #1.
- For guidance on restarting the Management Agent, see Restarting the Management agents in ESXi.
To validate your solution, you must observe whether performance data is available for VMs running on impacted vCenter hosts. To do so, do the following:
- Open your cCenter client and navigate to Home > Inventory > Hosts and Clusters, and then select a host name that was known to be missing performance data.
- Select the Performance tab, then go to Advanced.
- Select a CPU performance metric, then select a past Month.
- If you can view performance data, the Management Agent has successfully restarted. Note, however, that if you see only a partial set of performance data, the Agent is probably still not working properly.
- If you receive an error message reading "Performance data is currently not available for this entity," then the connection between the ESXi server and VMs on your vCenter is not working properly.
Note that a crashed management Agent might not impact the performance of the VMs. However, VMs running on these hosts will be excluded from HealthCheck for Azure assessments. Once the Agent is restarted, be sure to check whether the impacted VMs have been excluded from the assessment by going to the Settings page in the HealthCheck for Azure application, and confirming that the impacted VMs are enabled.