Monitoring the Nutanix Acropolis Server
eG Enterprise provides a specialized model for monitoring Nutanix Acropolis servers with VMs that host server applications.
Figure 1 : Layer model of the Nutanix Acropolis server
Each layer of this model is mapped to tests that report on the resource usage of the hypervisor and the VMs inside-out, and point you to resource-hungry VMs and applications. Using the metrics reported by this model, administrators can find quick and accurate answers to the following performance queries:
- Is the Acropolis server available over the network? If so, how responsive is the server to requests?
- Is the content cache sized enough to handle requests, or are cache hits very low?
- How much memory did deduplication save in the content cache?
- Are any physical disks marked for removal? Has data from that disk been migrated?
- Which physical disk is currently offline?
- Is any physical disk being over-utilized?
- Are there any latent physical disks?
- How is the I/O performance of the physical disks?
- Is I/O load balanced across all physical disks?
- Is any physical disk consuming bandwidth excessively when processing I/O?
- How is the I/O load on the storage?
- Is the storage processing I/O requests quickly?
- Is too much bandwidth being consumed when processing I/O?
- Is the AHV sized with adequate storage resources? If not, what type of storage is running short of space - the SSDs? or the SATA HDDs?
- How many VMs are operating on the Acropolis server? What is their IP address and which OS are they running?
- Which VMs are currently powered off or suspended?
- Which VMs have been newly added and which ones were recently removed from the server?
- Which VM is the CVM of a cluster?
- Is any VM consuming the physical CPU, memory, network, and I/O resources of the Acropolis server, abnormally?
- Which VM is utilizing the CPU, memory, network, and I/O resources allocated to it, excessively? Which process on the VM is causing this abnormal resource consumption?
