Monitoring vSphere

Instana enables you to monitor the metrics and configuration data on vSphere. The vSphere sensor is automatically deployed and installed after you install the Instana host agent.

This is an optional feature, disabled by default in the Instana backend. To enable this optional feature, see the page for your Instana deployment: SaaS, Self-Hosted Custom Edition (Kubernetes or Red Hat OpenShift Container Platform), or Self-Hosted Classic Edition (Docker)

Supported Versions

Confirmed support for metrics and configuration data for versions v6.5, v6.7, v7.0 and v8.0. To monitor vSphere v7.0 and higher, Instana agent needs to run with JDK 8u242 or higher.

Configuration

To enable vSphere metric monitoring, add the vCenter credentials to the agent config file <agent_install_dir>/etc/instana/configuration.yaml:

com.instana.plugin.vsphere:
  host: https://<INSERT_VCENTER_URL_HERE>/sdk
  username: <INSERT_USERNAME_HERE>
  password: <INSERT_PASSWORD_HERE>
  enabled: true
  poll_rate: 20 # metrics poll rate in seconds

The configured user needs to have read-only role at least.

Remote monitoring is supported, so the agent can be installed on a vSphere machine, which ensures minimal latency and high performance, or somewhere else. In case of a Kubernetes environment, we recommend installing Instana agent separately on a dedicated machine to ensure only one agent monitors vSphere entities. The machine where the agent is running needs to be able to establish a network connection to the vCenter SDK. Additionally, the Instana agent (without vSphere monitoring enabled) can be deployed to each ESXi host. This will correlate the VM the infrastructure Host.

vSphere metrics are polled every 20 seconds by default or as configured in the poll_rate parameter. In case the metrics update cycle does not finish within the poll_rate interval, it will not overlap with the following one, as it waits until the previous one completes.

If the vSphere dashboards are not visible on your environment, you need to make a request to Instana to enable this feature.

Metrics collection

To view the metrics, select Infrastructure in the sidebar of the Instana User interface, click a specific monitored host, and then you can see a host dashboard with all the collected metrics and monitored processes.

Datacenter

Configuration data

  • Name
  • Status
  • Number of ESXi Hosts
  • Number of Virtual Machines

Performance metrics

Data point Description Granularity
Uptime Total time elapsed, in seconds, since last operating system start 20 seconds
CPU usage This metric indicates the percentage of CPU that was used out of all the CPU that was allocated to the datacenter 20 seconds
CPU wait Amount of time spent on system processes on each virtual CPU in the virtual machine 20 seconds
CPU system Total CPU time spent in wait state wait on each virtual CPU in the virtual machine 20 seconds
Memory usage Average percentage of each host physical memory that has been consumed usage 20 seconds
VM ballon memory usage Average amount of guest physical memory reclaimed from the virtual machine by the balloon driver in the guest 20 seconds
Net usage Average network utilization (combined transmit-rates and receive-rates) during the interval 20 seconds

Host

Configuration data

  • Name
  • Number of CPU cores
  • Total memory
  • List of Virtual Machines
  • List of mounted datastores

Performance metrics

Data point Description Granularity
CPU usage CPU usage as a percentage during the interval usage 20 seconds
Memory usage Percentage of host physical memory that has been consumed 20 seconds
Memory shared Amount of guest physical memory that is shared within a single virtual machine or across virtual machines 20 seconds
Memory active Amount of guest physical memory that is being actively read or written by guest 20 seconds
Memory vmmemctl Amount of guest physical memory reclaimed from the virtual machine by the balloon driver in the guest 20 seconds
Network transmitted Average amount of data transmitted per second 20 seconds
Network received Average amount of data received per second 20 seconds
Network total Average amount of data transmitted + received per second 20 seconds
Network packets received Number of broadcast packets received during the sampling interval 20 seconds
Network packets transmitted Number of broadcast packets transmitted during the sampling interval 20 seconds
Network packets total Number of broadcast packets received + transmitted during the sampling interval 20 seconds

Datastore

Configuration data

  • Name
  • Mounted URL
  • Type
  • Max file size
  • Capacity

Performance metrics

Data point Description Granularity
Free space Total free space for given datastore 20 seconds
IOPS read Storage DRS datastore read I/O rate 20 seconds
IOPS write Storage DRS datastore write I/O rate 20 seconds
IOPS total Storage DRS datastore read + write I/O rate 20 seconds
Number of reads Average number of read commands issued per second to the datastore during the collection interval 20 seconds
Number of writes Average number of write commands issued per second to the datastore during the collection interval 20 seconds
Read latency Storage DRS datastore normalized read latency 20 seconds
Write latency Storage DRS datastore normalized write latency 20 seconds
Total latency Storage DRS datastore normalized read + write latency 20 seconds

These performance metrics are displayed in the Filesystems table on the ESXi Host dashboard.

Virtual Machine

Configuration data

  • Name
  • Number of CPU cores
  • Total memory
  • Link to Infrastructure Host where Instana agent is running
  • Guest OS
  • State

Performance metrics

Data point Description Granularity
CPU usage CPU usage as a percentage during the interval usage 20 seconds
CPU system Total CPU time spent in wait state wait on each virtual CPU in the virtual machine 20 seconds
CPU ready Time that the virtual machine was ready, but could not get scheduled to run on the physical CPU during last measurement interval 20 seconds
CPU wait Total CPU time spent in wait state wait 20 seconds
CPU readiness Percentage of time that the virtual machine was ready, but could not get scheduled to run on the physical CPU 20 seconds
CPU latency Percent of time the virtual machine is unable to run because it is contending for access to the physical CPU(s) 20 seconds
Memory usage Percentage of host physical memory that has been consumed 20 seconds
Memory active Amount of guest physical memory that is being actively read or written by guest. Activeness is estimated by ESXi 20 seconds
Memory swapped Amount of guest physical memory that is swapped out to the swap space swapped 20 seconds
Memory granted Amount of host physical memory or physical memory that is mapped for a virtual machine or a host 20 seconds
Memory vmmemctl Amount of guest physical memory reclaimed from the virtual machine by the balloon driver in the guest 20 seconds
Network transmitted Average amount of data transmitted per second 20 seconds
Network received Average amount of data received per second 20 seconds
Network total Average amount of data transmitted + received per second 20 seconds
Network packets received Number of broadcast packets received during the sampling interval 20 seconds
Network packets transmitted Number of broadcast packets transmitted during the sampling interval 20 seconds
Network packets total Number of broadcast packets received + transmitted during the sampling interval 20 seconds

vMotion events collection

Instana agent collects the following vMotion event types and reports them as Instana events:

vMotion Instana
VmFailedMigrateEvent Issue event
VmRelocateFailedEvent Issue event
VmBeingHotMigratedEvent Change event
VmBeingMigratedEvent Change event
VmBeingRelocatedEvent Change event
VmRelocatedEvent Change event
VmMigratedEvent Change event
VmRemovedEvent Change event
VmStoppingEvent Change event

The events' collection starts right after the agent collects data about all virtual machines and repeats every 60 seconds or as configured in the poll_rate.