This document aims to help users that are not familiar with all metrics exposed by different KubeVirt components.
All metrics documented here are auto-generated by the utility tool tools/doc-generator
and reflects exactly what is being exposed.
Version information.
Indication for an operating virt-controller.
Indication for a virt-controller that is ready to take the lead.
The vcpu affinity details.
Current balloon bytes.
Amount of usable
memory as seen by the domain.
The number of page faults when disk IO was required.
The number of other page faults, when disk IO was not required.
Resident set size of the process running the domain.
Swap in memory traffic in bytes.
Swap out memory traffic in bytes.
Amount of unused
memory as seen by the domain.
The amount of memory which can be reclaimed by balloon without causing host swapping in bytes.
The amount of memory in bytes used by the domain.
Network traffic receive in bytes.
Network receive error packets.
The number of rx packets dropped on vNIC interfaces.
Network traffic receive packets.
Deprecated.
Network traffic transmit in bytes.
Network transmit error packets.
The number of tx packets dropped on vNIC interfaces.
Network traffic transmit packets.
Indication for a VirtualMachine that its eviction strategy is set to Live Migration but is not migratable.
Indication for the number of VirtualMachineInstance workloads that are not running within the most up-to-date version of the virt-launcher environment.
Sum of VMIs per phase and node.
phase
can be one of the following: [Pending
, Scheduling
, Scheduled
, Running
, Succeeded
, Failed
, Unknown
]
Storage flush requests.
Total time (ms) spent on cache flushing.
I/O read operations.
I/O write operations.
Storage read operation time.
Storage read traffic in bytes.
Storage write operation time.
Storage write traffic in bytes.
Amount of time spent in each state by each vcpu. Where id
is the vcpu identifier and state
can be one of the following: [OFFLINE
, RUNNING
, BLOCKED
].
Amount of time spent by each vcpu while waiting on I/O.
After developing new metrics or changing old ones, please run make generate
to regenerate this document.
If you feel that the new metric doesn't follow these rules, please change doc-generator
with your needs.